Hci 634 G

Undergraduate Lecture Notes in Physics
For further volumes:

http://www.springer.com/series/8917
Undergraduate Lecture Notes in Physics (ULNP) publishes authoritative texts
covering topics throughout pure and applied physics. Each title in the series is
suitable as a basis for undergraduate instruction, typically containing practice
problems, worked examples, chapter summaries, and suggestions for further
reading.
ULNP titles must provide at least one of the following:

• An exceptionally clear and concise treatment of a standard undergraduate
subject.
• A solid undergraduate-level introduction to a graduate, advanced, or non-
standard subject.
• A novel perspective or an unusual approach to teaching a subject.
ULNP especially encourages new, original, and idiosyncratic approaches to physics

teaching at the undergraduate level.
The purpose of ULNP is to provide intriguing, absorbing books that will continue to
be the reader’s preferred reference throughout their academic career.
Series Editors:
Neil Ashby, Professor, Professor Emeritus, University of Colorado Boulder, CO,

USA
William Brantley, Professor, Furman University, Greenville, SC, USA
Michael Fowler, Professor, University of Virginia, Charlottesville, VA, USA
Michael Inglis, Associate Professor, SUNY Suffolk County Community College,
Selden, NY, USA
Elena Sassi, Professor, University of Naples Federico II, Naples, Italy
Helmy Sherif, Professor Emeritus, University of Alberta, Edmonton, AB, Canada
C. R. Kitchin
Telescopes and Techniques

Third Edition
C. R. Kitchin
ISSN 2192-4791 ISSN 2192-4805 (electronic)

ISBN 978-1-4614-4890-7 ISBN 978-1-4614-4891-4 (eBook)
DOI 10.1007/978-1-4614-4891-4
Springer New York Heidelberg Dordrecht London
Library of Congress Control Number: 2012947024
# Springer Science+Business Media New York 1995, 2003, 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts
in connection with reviews or scholarly analysis or material supplied specifically for the purpose of being
entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication
of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s location, in its current version, and permission for use must always be obtained from
Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center.
Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

For Christine
Preface to the First Edition
This book arose from the need to introduce first-year astronomy students at the
University of Hertfordshire to the basic techniques involved in using a telescope
and finding their way around the sky. It soon transpired that many aspiring
astronomers, not just those in a university environment, needed similar guidance.
The aim of the book is therefore to be introductory in the sense that prior knowledge
is not assumed, but not in the usual sense that mathematics is avoided. Anyone
wanting to learn about telescopes, how they work, how to use them, and how to
choose a telescope for their own use should find this book helpful. It provides the
information on how to set up a telescope from scratch, to find objects in the sky,
both those bright enough to be seen with the naked eye and those for which the
telescope must be set on to the object before it can be seen. It explains such things as
sidereal and solar time, right ascension and declination, light grasp, aberrations,
etc., in sufficient detail for useful work to be undertaken. The techniques involved
in visual work with a telescope and imaging with photography and CCDs are also
explained to a similar depth. Ancillary work and instrumentation such as data
processing, photometry, and spectroscopy are outlined in somewhat less detail. I
hope readers find the book useful and as interesting as I found it to write.
vii
Preface to the Second Edition
In the 7 years since the first edition was published, Telescopes and Techniques has
shown itself to fill a requirement for an explanation of all aspects of the use of
telescopes, detectors, etc. However, it has also become apparent that some topics
that were omitted first time around, such as radio telescopes, should have been
included. Many techniques, especially relating to detectors, have also undergone
rapid changes over that interval.
This second edition has therefore been expanded to include a good deal of new
material and updated. I hope that it will continue to be useful to astronomers of all
kinds, and to provide the background to operating telescopes successfully.
May you all have clear skies!
ix
Preface to the Third Edition
The passage of another 9 years has seen many more changes and advances in
practical astronomy. The text of this book has therefore been thoroughly updated –
both by adding new material and deleting or reducing the coverage of certain
things, such as the use photographic emulsions, that have fallen or are falling out
of use. Two of the appendices have been dropped since the information that they
contained (telescope manufacturers and suppliers and astronomical societies) can
now readily be found using the Internet. A few items have been retained that are no
longer used with large instruments but which still do find applications with the
owners of smaller telescopes.
In response to comments from some readers, worked examples have been added
to illustrate the applications of some of the mathematical techniques and formulae
that are discussed. Also in response to comments, photographs of telescopes,
instruments, and astronomical phenomena have been included, where these are
helpful in illustrating the points being made in the text.
I hope that you continue to find the book useful, a handy source of reference, and
an aid to getting the best out of your telescope whether it is 0.1 or 10 m in diameter.
xi
About the Author
Chris Kitchin is Professor Emeritus of Astronomy at the University of Hertfordshire

and has had a lifelong interest in astronomical instrumentation, using telescopes of
all sizes, studying the rapidly growing number of planets known to orbit other stars,
and trying to communicate his own excitement at the recent explosion of
discoveries about the universe. When not a 100 light years from Earth (at least in
his imagination), Chris enjoys being taken for long country walks by his two border
collies.
He is the author of numerous scientific articles and papers and 16 books by
various publishers, including Springer’s Illustrated Dictionary of Practical Astron-
omy, Solar Observing Techniques, Seeing Stars, Photo-Guide to the Constellations,
Exoplanets, Galaxies in Turmoil, and of course the first two editions of Telescopes
and Techniques, which sold in excess of 12,000 copies.
xiii
Contents
Preface to the First Edition .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. vii

Preface to the Second Edition . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . ix
Preface to the Third Edition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Part I Telescopes
1 Types of Telescopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Historical Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Modern Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Mountings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
The Schmidt Camera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Multi-Mirror and Space Telescopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Atmospheric Compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Radio Telescopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Interferometers . .. . . . . . . . .. . . . . . . .. . . . . . . .. . . . . . . . .. . . . . . . .. . . . . . . .. . . . . . . . .. . . 27
2 Telescope Optics . .. . . . . . . . .. . . . . . . . . .. . . . . . . . .. . . . . . . . . .. . . . . . . . .. . . . . . . . . .. . 31
Point Sources . . .. . .. . .. . .. . .. . . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . . .. . .. . .. 31
Extended Images . .. . .. . .. . .. . .. . .. .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. 36
Telescope Objectives . . . . . . . .. . . . . . . . .. . . . . . . . . .. . . . . . . . .. . . . . . . . . .. . . . . . . . .. . 38
Eyepieces . . . . . . . . .. . . . . . . . . .. . . . . . . . .. . . . . . . . . .. . . . . . . . .. . . . . . . . . .. . . . . . . . .. . . . 40
Accessories . . . . . .. . . . . . . . .. . . . . . . .. . . . . . . . .. . . . . . . . .. . . . . . . .. . . . . . . . .. . . . . . . . .. 42
Star Diagonal . . . .. . . . .. . . . .. . . .. . . . .. . . . .. . . . .. . . .. . . . .. . . . .. . . . .. . . . .. . . .. . 42
Solar Diagonal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Barlow Lens .. . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . . .. . . . 45
Telecompressor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Major Telescopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Aberrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
xv
xvi Contents
Interferometers . .. . . . . . . . .. . . . . . . .. . . . . . . .. . . . . . . . .. . . . . . . .. . . . . . . .. . . . . . . . .. . . 51
Mountings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Observatories and Observing Sites . .. . .. . .. . .. . . .. . .. . .. . .. . . .. . .. . .. . . .. . .. 56
3 Modern Small Telescope Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Obtaining a Telescope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Making Your Own Telescope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Brief Survey of Commercially Produced Telescopes . . . . . . . . . . . . . . . . . . . . . 63
Binoculars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Part II Positions and Motions
4 Positions in the Sky . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Spherical Polar Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Celestial Sphere . . .. . . .. . . .. . . .. . . . .. . . .. . . .. . . .. . . .. . . .. . . . .. . . .. . . .. . . .. . . .. . 71
Altitude and Azimuth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Rotation . . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . 76
Solar and Sidereal Days . .. . . .. . . .. . . .. . . .. . . .. . . .. . .. . . .. . . .. . . .. . . .. . . .. . .. . 76
Declination and Hour Angle . . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . .. . . . . .. . 77
Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Mean Solar Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Solar Time . . .. . . . .. . . . .. . . .. . . . .. . . . .. . . . .. . . .. . . . .. . . . .. . . . .. . . .. . . . .. . . . .. 81
Civil Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Sidereal Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Right Ascension and Declination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Other Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Heliocentric Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Julian Date . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Spherical Trigonometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5 Movements of Objects in the Sky . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Diurnal Motion .. . .. .. . .. .. .. . .. .. . .. .. . .. .. . .. .. . .. .. . .. .. . .. .. .. . .. .. . .. .. . .. 101
Circumpolar Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Seasons and Annual Motions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 104
Planispheres and Astrolabes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Movements of the Moon and Planets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
The Moon .. . .. .. . .. . .. .. . .. . .. . .. .. . .. . .. .. . .. . .. .. . .. . .. . .. .. . .. . .. .. . .. . .. 109
Other Solar System Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Proper Motion . . . . .. . . . . . . .. . . . . . . .. . . . . . .. . . . . . . .. . . . . . . .. . . . . . .. . . . . . . .. . . 111
Precession . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Parallax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Aberration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Relative Planetary Positions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Contents xvii
Eclipses and Related Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Position in an Orbit . . . . . .. . . . .. . . . . .. . . . . .. . . . .. . . . . .. . . . . .. . . . . .. . . . .. . . . . .. . 122
Synodic Period .. . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . . .. . . . 125
6 Telescope Mountings .. . .. . .. . .. . . .. . .. . .. . .. . .. . .. . . .. . .. . .. . .. . .. . .. . . .. . .. 129
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Equatorial Mountings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Alt-Az Mountings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Other Types of Commercial Mountings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Making Your Own Mounting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Setting Circles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
Guiding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Modern Commercial Mountings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Part III Observing
7 Electromagnetic Radiation . . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. 145

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Intensity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Photons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Polarization . .. . .. . .. . . .. . .. . .. . .. . . .. . .. . .. . .. . .. . . .. . .. . .. . .. . . .. . .. . .. . .. . . .. 147
Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Measurements . . . . .. . . . . . . .. . . . . . .. . . . . . .. . . . . . .. . . . . . .. . . . . . .. . . . . . .. . . . . . . .. . 148
Photometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Polarimetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
8 Visual Observing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
General and Practical Considerations and Safety . .. . . .. . . .. . . .. . .. . . .. . . .. 152
Finding Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
The Moon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
The Planets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
The Sun . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . 162
Finding the Sun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Observing the Sun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Stars . . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . . .. . . . .. . 166
Stellar Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
Magnitudes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
Observing Stars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
Nebulae and Galaxies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
Daytime Observing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
False Observations .. . .. . .. .. . .. . .. . .. .. . .. . .. . .. . .. .. . .. . .. . .. .. . .. . .. . .. . .. .. 178
xviii Contents
9 Detectors and Imaging .. . .. . .. . .. .. . .. . .. .. . .. . .. .. . .. . .. . .. .. . .. . .. .. . .. . .. 181

The Eye . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . 181
Charge Coupled Devices (CCDs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Image Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
Flat Fielding . . . . . . . .. . . . . . . . .. . . . . . . . .. . . . . . . .. . . . . . . . .. . . . . . . . .. . . . . . . . .. . . 191
Cosmic Rays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Contrast Stretching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Background and Dark Signal Subtraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Hot Spots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Image Compression . . .. . .. . .. . .. . .. .. . .. . .. . .. . .. . .. . .. .. . .. . .. . .. . .. . .. . .. 193
Photography . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . 193
Other Detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
10 Data Processing . . . . .. . . . . . . .. . . . . . . .. . . . . . .. . . . . . . .. . . . . . . .. . . . . . .. . . . . . . .. . . 197
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Data Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
Errors and Uncertainties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Linear Regression . .. . .. . . .. . .. . . .. . .. . . .. . .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. 202
Correlation Coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Student’s t Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Archives, Data Mining, Virtual Observatories and Citizen Science
Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
11 Photometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
CCD Photometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Absolute Magnitude . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
Wavelength Dependence . . .. . . .. . . .. . . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. 216
Bolometric Magnitude . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Spatial Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Photometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Observing Techniques . . .. . .. . .. . .. . .. .. . .. . .. . .. . .. . .. . .. .. . .. . .. . .. . .. . .. . .. 220
12 Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Spectroscopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Spectral Type . .. . .. . .. .. . .. .. . .. . .. .. . .. .. . .. . .. .. . .. .. . .. . .. .. . .. .. . .. . .. .. 226
Luminosity Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
Radial Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
Spectrophotometry . .. . . .. . . .. . . .. . . .. . . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . 229
Contents xix
Appendix A For Further Reading . .. . . . . . .. . . . . . .. . . . . .. . . . . . .. . . . . . .. . . . . .. . 231
Appendix B Constellations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
Appendix C Answers to Exercises . . . . .. . . . . . . . .. . . . . . . . . .. . . . . . . . .. . . . . . . . . .. 237
Appendix D SI and Other Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
Appendix E The Greek Alphabet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Index . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . 245
Part I
Telescopes
Chapter 1
Types of Telescopes
Historical Introduction
Until recently, the invention of the telescope was generally attributed to a Dutch
spectacle maker called Hans Lippershey (or Johann or Lipperhey or Lippersheim,
1570–1619). He worked at Middelburg on the island of Walcheren, some 130 km
southwest of Amsterdam. The probably apocryphal story has it that in 1608 his
children discovered, while playing with some of his spare lenses, that one combi-
nation made a distant church spire appear much closer. The exact combination of
lenses they and he used is no longer known. It could have been a pair of converging
lenses, though these would produce an upside-down image, so a converging lens
and a diverging lens such as Galileo’s invention is also possible.
An example of the resulting telescope was duly presented to the States-General,
Prince Maurice. The news of the discovery spread rapidly, reaching Venice and
Galileo only a year later. The details received by Galileo concerned just the effect
of observing with the instrument, and not the details of its design. However, he had
at that time been working extensively on optics, and in a few hours was able to
design an optical system that reproduced the reported distance-shortening effect of
the instrument. Galileo’s design used a long focal length converging lens and a
shorter focal length diverging lens placed before the focus (Fig. 1.1), and had the
advantage for terrestrial use of producing an upright image. This optical system we
now know as a Galilean refractor, after its (first known) inventor. The Galilean
telescope still finds a use today in the form of opera glasses.
The invention of the telescope, however, may well predate Lippershey. Lately,
historical research has suggested that a form of telescope may have been discovered
by the Englishman Leonard Digges (?–1571?), who also invented the theodolite,
sometime around 1550. His design appears to have used a long focal length lens and
a short focal length concave mirror for what we would now call the eyepiece, and
therefore was probably used to look at objects behind the observer (Fig. 1.2). It is
possible that a garbled account of such a system could have led to the idea of the
crystal ball used by fortune-tellers. Even earlier telescopes may be ascribable to
C. R. Kitchin, Telescopes and Techniques, Undergraduate Lecture Notes in Physics, 3

DOI 10.1007/978-1-4614-4891-4_1, # Springer Science+Business Media New York 2013
4 1 Types of Telescopes
Fig. 1.1 Optics of the Galilean refractor
Fig. 1.2 Suggested optical system of Digges’ possible telescope

Historical Introduction 5
Fig. 1.3 Optics of the

astronomical refractor
Bishop Robert Grossetest (ca. 1175–1253), Friar Roger Bacon (ca. 1214–94, who
certainly knew about spectacles for the correction of long-sight), and to
Giambattista della Porta (1535–1615, who investigated the camera obscura), but
these attributions are more dubious.
If Lippershey had not invented the telescope when he did, someone else would
have done so within a matter of months. Indeed, Lippershey was refused a patent on
his invention partly because Jacob Metius (ca. 1571–1628) also applied for a patent
on a telescope just a few weeks later. Metius’ application could, of course, have
been the result of an early variety of industrial espionage, but in the early 1600s the
telescope was an idea ripe for invention anyway. Optical science had become
sufficiently advanced that the instrument could be designed theoretically – as
Galileo demonstrated – and lenses (for spectacles and magnifying glasses) were
in plentiful supply so that experimentation could also lead to its discovery, as
Lippershey’s children showed.
Lippershey’s and Galileo’s first telescopes magnified by about three times, but
later Galileo produced instruments with magnifications up to 30. Pointing his
telescopes towards objects in the sky revealed the craters on the Moon, that planets,
unlike stars, were not point sources, and that the four largest satellites of Jupiter,
which he called the Medicean stars, but which we now know as the Galilean
satellites, orbited that planet. Most importantly, he observed the phases of Venus,
and its angular size changes proving that it at least orbited the Sun and not the Earth,
so giving strong support to the heliocentric theory of Copernicus. However, the
magnification of the Galilean refractor was limited in the early seventeenth century
by the difficulty of producing short focal length diverging lenses with their deep
concave surfaces, and so was soon replaced for astronomical purposes by the
astronomical refractor. This design is still in use today and has a long focal length
converging first lens (or objective), followed by a much shorter focal length con-
verging lens as the eyepiece placed after the focus of the objective (Fig. 1.3). It gives
an upside-down image, of course, thus limiting its use for terrestrial purposes, but
causing little disadvantage when looking at objects in the sky.
Although the astronomical refractor could more easily be made to give higher
magnifications than the Galilean telescope, both designs suffered from other
problems, which considerably limited their use. These problems are now known
Fig. 1.4 Spherical aberration

in a mirror: rays from
different radii of the mirror
come to foci at different
points along the optical axis
as “aberrations,” and are discussed in more detail in Chap. 2. Two of the six primary
aberrations affected these early telescopes particularly badly, and these were
spherical aberration and chromatic aberration. Spherical aberration derives its
name from the way in which a spherical mirror brings parallel light rays to a
“focus” (Fig. 1.4). The light rays at different distances from the optical axis are
brought to foci at different distances along the optical axis. Simple lenses produce a
similar effect (Fig. 1.5). Thus there is no point at which the eyepiece can be placed
to give a sharp image.
Chromatic aberration is a similar effect in which light rays of differing wavelengths
are brought to differing foci along the optical axis (Fig. 1.6). Since the law of reflection
is independent of wavelength, however, this aberration applies only to lenses. Like
spherical aberration, there is no point at which the image will be sharply in focus.
However, since the eye has a sensitivity that peaks in the yellow-green part of
the spectrum, on focusing a telescope in which chromatic aberration is present, the
tendency is to put the yellow-green part of the image into the sharpest focus.
In addition to these two aberrations, the quality of glass in the seventeenth
century was very poor. Lenses could thus contain debris from the furnace in
which the glass had been melted, bubbles of air, un-annealed stresses, etc. The
surfaces of the lenses could also be poorly polished and deviate markedly from the
correct spherical shape. The images in these early telescopes were thus of very poor
quality indeed by today’s standards. Saturn, for example, was not observed
correctly as a planet surrounded by rings until Christiaan Huygens’ (1629–95)
work in 1656, almost 40 years after the application of the telescope to astronomy.
Prior to Huygens, Saturn had been variously observed as a triple planet, an elliptical
planet and a planet with handles. If we add to the poor quality of the images
the difficulty of finding and tracking objects owing to the tiny fields of view of
the telescopes and their inadequate or non-existent mountings, then the discoveries
Fig. 1.5 Spherical aberration in a simple lens
Fig. 1.6 Chromatic aberration in a simple lens. For simplicity the effects of the spherical
aberration that will also be present have been omitted
made by the early telescopists become truly remarkable. It also becomes much less
surprising that Galileo’s discoveries and their support for the Copernican model of
the Solar System were not accepted immediately and universally. Skeptics looking
through one of his instruments could easily, and with some justification, come to
Fig. 1.7 Simple terrestrial

telescope (upright image)
quite different interpretations of the faint fleeting blurry images from those reached
by heliocentrists.
The problems caused by chromatic and spherical aberrations were soon found to
be reduced if the focal length of the lens was long compared with its diameter, or, as
we would now express it, if lenses with large focal ratios were used. This led to
increasing lengths for telescopes and to the use of nested drawtubes so that the
instrument could be collapsed to a convenient length when not in use. Even for
terrestrial telescopes (which were provided with an upright image by the addition of
a relay lens, Fig. 1.7), lengths of 3–5 m became common.
For astronomical work, this trend, in the late seventeenth century, eventually
resulted in the aerial telescope (aerial in the sense of “belonging to the air,” not
today’s common meaning of a radio antenna). Johannes Hevelius of Gdansk
(1611–87), Huygens, and others constructed telescopes up to 60 m long. Telescopes
of such a size could not take the same form, nor be handled in the same way, as
smaller instruments. In one version, the objective would be mounted in a small tube
near the top of a mast. The eyepiece would be in a separate tube, and the two tubes
connected and aligned by a taut line. The eyepiece was held in the hand, and the
observer moved around the mast to keep the line in tension and to look at different
parts of the sky. When required, the objective could be raised or lowered on the
mast by assistants to make a major change in the altitude to be observed.
It was the open nature of these telescopes that led to their name. Such
instruments would clearly be incredibly difficult to use, swinging and trembling
in the lightest of breezes, and with very tiny fields of view. Nonetheless, they could
be used to great effect; using one Giovanni Cassini (1625–1712), for example,
found four new satellites of Saturn, and the break in the rings of Saturn that is still
known as Cassini’s division.
Since reflection, unlike refraction, is independent of wavelength, an alternative
approach to overcoming the problem of chromatic aberration using mirrors in place
of the main lenses was suggested in the mid-seventeenth century. It was also
Fig. 1.8 Ray paths for

reflection from a
paraboloidal mirror
(compare with Fig. 1.4)
Fig. 1.9 Gregorian telescope
realized that by deepening the curve of the primary mirror from a spherical to a
paraboloidal shape, spherical aberration could additionally be eliminated (Fig. 1.8).
James Gregory (1638–75) of Aberdeen proposed the first such design in 1663. His
system, now called the Gregorian telescope, used a paraboloidal mirror as the
objective (usually now called the primary mirror), and a small ellipsoidal mirror
placed after the focus of the primary mirror (the secondary mirror). Light from a
distant object hitting the primary mirror would be reflected to the secondary mirror
that would reflect it down the telescope tube again to a second focus. The light
would emerge from the telescope through a small central hole in the primary mirror
and be observed with an eyepiece made with lenses (Fig. 1.9). Gregory had a
telescope made to his design by two London-based opticians, but was dissatisfied
Fig. 1.10 Newtonian

telescope
with the results, and there is no record of any successful observations ever being
made with the instrument.
The first actual working reflector was therefore designed and built by Isaac
Newton (1642–1727) 5 years after Gregory’s work. Newton’s telescope was of a
different design from the Gregorian, with a parabolic primary but a flat mirror for
the secondary set at 45 to the axis of the telescope. The light was thus brought to a
focus at the side of the telescope, and that is where the eyepiece had to be positioned
(Fig. 1.10).
Newton’s system is still in widespread use today and is called the Newtonian
telescope. It is mostly found in the smaller telescopes by today’s standards, these
being used and often made by amateur astronomers. The first Newtonian telescope,
however, had a diameter of only about one and quarter inches (30 mm), so even the
smallest amateur’s telescope today dwarfs the one that Newton used.
All the merits of Newton’s design were not immediately obvious in practice,
because the mirrors he used were spherical, not parabolic, and made from speculum
metal (also known as bell metal, an alloy of variable composition but normally
about 75% copper, 25% tin, plus sometimes some zinc and a little arsenic).
Speculum metal has a reflectivity of only about 60% when freshly polished, and
much less when tarnished, which happens quite quickly. Thus with an aperture of
just over an inch, and two such mirrors, Newton’s telescope would only have
delivered to the eye a little more light than could be obtained looking directly at
the object. Since the instrument had a magnification of about 25, then except for
point sources such as stars, this light would be spread over a 600 times greater area,
making objects seen through the telescope appear very dull and faint, and so only
the brightest objects could have been observed.
In 1672, only 4 years after Newton’s invention, yet another design for a
reflecting telescope was produced, this time by the Frenchman Laurent
Fig. 1.11 Cassegrain telescope
Cassegrain.1 This optical design, or variations on it, is still almost universally

used for large telescopes. Both the Hubble Space Telescope and the Keck 10 m
telescope are of this basic design. It is very similar to the Gregorian telescope, but
with the elliptical concave secondary replaced by a convex hyperboloidal mirror,
and with that mirror placed before instead of after the focal point of the primary
mirror (Fig. 1.11).
A major reason for the continuing popularity of the Cassegrain design is its
telephoto property. That is, it has a focal length much greater than its physical
length. This arises, as can be seen from Fig. 1.11, because the secondary mirror
produces a much narrower converging cone of light than that originally coming
from the primary. The short length that this gives to the telescope reduces the cost
of the instrument by far more than might be expected. There is economic gain at
every point in the overall design; the shorter tube length means it can be made from
thinner materials, the mounting and drive can be smaller and lighter, the dome can
be smaller, and so on. The Gregorian design has a similar telephoto effect, but with
less of a reduction in overall telescope length because the secondary mirror is
placed after the focus of the primary, not before it.
The Gregorian also suffers through having a very small field of view, making it
difficult to use in comparison with other designs. Nonetheless, in the eighteenth
century, it was the Gregorian telescope that was the more popular of the two
designs, and instrument makers such as James Short (1710–68) and others made
numerous examples with apertures of a few inches. The popularity of the design at
that time arose from its relative ease of manufacture and testing. The concave
secondary of the Gregorian was much more straightforward to produce than the
1
Little is known for certain about Cassegrain. Research by Andre Baranne and Francois Launay in
1997 suggests that he might have been the priest and teacher Laurent Cassegrain (c. 1629–1693,
probably born near Chartres). Other suggestions include Guillaume Cassegrain (a sculptor),
Jacques Cassegrain (a doctor) and Nicolas Cassegrain (nothing further known). Quite possibly
some of these names may belong to the same individual.
convex secondary of the Cassegrain. The first reflectors of any design to be made
with correctly shaped paraboloidal primaries were produced by John Hadley
(1682–1744) in the early 1720s.
Much later, William Herschel (1738–1822), with his large telescopes, dispensed
altogether with the secondary mirror in order to eliminate its light losses. He simply
tipped the primary mirror to one side and placed the eyepiece immediately after
the primary focus. This type of design has become known as a Herschelian
telescope. In the form that Herschel used it, the images it produces are seriously
degraded by aberrations. Not only does spherical aberration reappear, but also other
aberrations such as coma and astigmatism (Chap. 2) become significant. Herschel’s
problems with the design arose because at that time it was only possible to produce
symmetrically shaped mirrors. Today the design is in widespread and successful
use, not least at microwave wavelengths for the ubiquitous satellite television
receiving aerials, but with asymmetrical, or off-axis mirrors.
In the early eighteenth century, however, small reflectors and small refractors
were of roughly equal (and fairly poor) quality. In 1729 Chester Moor Hall
(1703–71) invented the achromatic lens, but it was not until 1754 that the situation
for refractors was transformed when John Dollond (1706–61) began manu-
facturing high quality achromatic lenses commercially. This invention reduced
the effects of chromatic aberration on images by a large amount. It relied for its
effect on the fact that the light deviating property of a glass (refraction) and the
light splitting property of a glass (dispersion) vary from one type of glass to
another, but not always by equal amounts. Thus it was possible to find two
varieties of glass that were such that if one were made into a converging lens
and the second into a diverging lens, then when they were combined together the
dispersion of one was canceled out by the other, but the overall combination still
acted as a converging lens.
The glasses chosen by Dollond remain in common use today, and were crown
glass for the converging lens and flint glass for the diverging lens. The combina-
tion of the two lenses is called an achromatic doublet. One of the most widely
used designs for such a doublet has the two inner (contact) surfaces with the same
radius so that the two lenses may be cemented together to produce a robust
combination not easily misaligned (Fig. 1.12). However, many variations on
this basic theme are possible, including separating the two components by a
small distance, and with care some of the other aberrations can also be reduced.
A two-element achromat typically reduces the chromatic aberration by a factor of
10 but does not eliminate it completely. The colored images of the simple lens are
folded around so that at any point on the optical axis two wavelengths are
simultaneously in focus (Fig. 1.13).
Choosing the coincident wavelengths appropriately can produce a lens optimized
for a particular purpose. Thus for visual work, wavelengths around 400 and 600 nm
should coincide, while for photographic work, having 350 and 500 nm together may
be more useful. A third lens may be added to the combination, and this folds the color
spread a second time, reducing the remnant of chromatic aberration even further and
producing a lens called an apochromat. Yet more lenses may be added to continue to
Fig. 1.12 Achromatic doublet
Fig. 1.13 Chromatic aberration for a simple lens, an achromatic doublet and an apochromat
Fig. 1.14 Modern

astronomical refractor
improve the color and other aberrations in the image, but these are not economically
viable for the large lenses required for telescope objectives; modern camera and
other lenses, however, may go on to have a dozen or more components.
With the invention of the achromat for the objective, and the use of multi-lens
eyepieces (see later), we have the refractor in essentially its modern form
(Fig. 1.14). From the mid-eighteenth century onwards, the refractor was devel-
oped rapidly. The other major contribution to its advance came from the German
astronomer, Joseph von Fraunhofer (1787–1826), who developed greatly
improved glass making techniques, thus providing large, high-quality blanks
for the production of the lenses.
The development of the refractor culminated at the end of the nineteenth century
in the 1 m refractor at Yerkes. The refractor at this size, however, is limited by
several factors: the lenses are several inches thick, thus becoming very heavy and
absorbing light to a significant extent; they can only be supported at their edges and
can therefore sag under their own weight; the residual chromatic aberration
becomes intrusive; and, by no means finally, the long focal ratios required to reduce
other aberrations produce very lengthy telescopes requiring massive mountings and
huge and expensive domes – the Yerkes telescope, for example, is 18 m in length
and is housed in a dome 27 m in diameter.
While refractors were being pushed towards their limits, reflectors also
continued to be used, because mirrors could be made far larger than lenses. This
was due to several factors: firstly, the casting of speculum metal, though by no
means trivial, was easier than that of glass; secondly, even if the casting did contain
bubbles, etc., this was of relatively little importance compared with the glass for a
lens, because light does not pass into the material forming a mirror, so only its
surface needs to be of high quality. Finally, mirrors can be supported on their backs,
as well as at the sides, so distortion due to gravitational loading is less significant.
Thus even with the poor reflectivity of speculum metal, the total light gathered
could be superior to that of a refractor. William Herschel, for example, produced his
1.2-m reflector in 1787, over 100 years before the rather smaller Yerkes refractor.
The largest ever speculum-mirror based telescope, a 1.8-m Newtonian reflector,
was built by William Parsons (1800–67, third Earl of Rosse) in 1845.
Modern Instruments 15
Speculum metal mirrors suffer from an additional serious drawback apart from
their low reflectivity. Repolishing to remove the tarnish will also spoil the precise
shape of the surface of the mirror. When mirrors were of poor optical quality anyway,
this was not of great importance; but once their surfaces could be shaped (figured)
closely to the required paraboloid, then repolishing became a very lengthy process –
perhaps half as long as producing the original mirror from its blank in the first place.
Thus William Parson’s telescope (which made some significant discoveries, including
the spiral nature of some galaxies) was the culmination of its type.
Modern Instruments
In the latter part of the nineteenth century, metal-on-glass mirrors replaced the
speculum metal used for the previous two centuries. These had the advantages of
high reflectivities (95% or more) and the fact that renewing a tarnished surface
could be done by removing the old reflecting layer with a chemical solution, leaving
the glass and its precisely shaped optical surface untouched and then adding a new
coating of the metal. Initially, silver was used as the reflector, and this could be
deposited chemically. However, silver tarnishes very quickly, necessitating new
coatings at intervals of a few weeks, so that most telescopes now use a layer of
aluminum evaporated on to the glass surface while it is inside a vacuum chamber.
Aluminum oxidizes, but the thin layer of aluminum oxide produced when the
mirror is first exposed to the atmosphere is transparent and furthermore seals the
surface, protecting the aluminum from additional oxidation. Nowadays, it is com-
mon practice to overcoat the aluminum with a further protective layer of silicon
dioxide. Even so, most major telescopes have to have their mirrors recoated at
intervals of a year or so.
By the early part of the twentieth century, most new major telescopes being built
were recognizably the ancestors of today’s instruments – that is, they were metal-
on-glass reflectors of the Cassegrain design or one of its derivatives, such as the
Ritchey-Chrétien. In the latter, the primary mirror is deepened from a paraboloidal
to an hyperboloidal shape, the secondary remains hyperboloidal, but with a steeper
curve. This change slightly degrades the on-axis image but not outside the diffrac-
tion limits of the telescope (Chap. 2), but very considerably improves the image
away from the optical axis. Thus telescopes of the Ritchey-Chrétien design can
produce good quality images over a field of view several tens of minutes of arc
across, compared with the basic Cassegrain design in which the field of sharp focus
may be limited to only a few minutes of arc.
Most large, and many smaller, telescopes can be used in several modes. One of
the commonest is as a Cassegrain system, already mentioned. Alternatively the
Cassegrain secondary mirror can be replaced by a flat mirror set at 45 to produce a
Newtonian. In the larger telescopes, the instrumentation to be used, such as a
camera, photometer or spectroscope, may be no larger than the Newtonian secon-
dary, and so that mirror may be dispensed with altogether and the instrumentation
Fig. 1.15 Telescope used

at prime focus
placed directly at the focus of the primary mirror. Such a system is called the
primary focus of the telescope and is similar to the Herschelian system except that it
is on-axis. In telescopes over 3 or 4 m in diameter, even the observer can occupy the
prime focus position in order to operate the instrumentation without obstructing too
great a proportion of the incoming light. Usually weak lenses, sometimes of
complex shapes and/or made from exotic materials, have to be placed before the
prime focus to produce a reasonable quality image. These are known as correcting
lenses (Fig. 1.15).
Mountings
All telescopes must have a mounting to enable them to be pointed at the required
part of the sky, and then to follow (track) the object being observed as it changes
position because of Earth’s rotation. Little has been said about the design of
mountings in the previous discussion, though they are at least as important for the
successful use of a telescope as the optics themselves. Some mounting designs are
discussed later. Here we just note that with most mountings, the position in space of
the focus of the telescope changes as the telescope moves. Any instrumentation
placed at that focus therefore has to be relatively lightweight, and capable of being
operated at varying angles of inclination. Large ancillary instruments, or
instruments under development, are therefore difficult to attach to telescopes.
A third mode therefore exists for many major telescopes, known as the Coudé
focus in the case of an equatorial mounting (Fig. 1.16), or the Nasmyth focus in the
case of an alt-azimuth mounting (Figs. 1.17 and 1.18). This provides a focus that is
fixed (for the Coudé system), or remains horizontal (in the case of the Nasmyth
system). In both cases, therefore, large and/or delicate equipment can be used with
the telescope.
Mountings 17
Fig. 1.16 The Coudé system
Fig. 1.17 The Nasmyth

system
Fig. 1.18 The 8.2-m Antu telescope is one of four identical instruments making up the European
Southern Observatory’s Very Large Telescope. All four telescopes have alt-azimuth mountings. One
of Antu’s two Nasmyth platforms is seen in the foreground. The large metallic cylinder with a red-
colored end is ISAAC (infrared spectrometer and array camera), which receives its light through the
Nasmyth focus. (Reproduced by kind permission of ESO/G Hudepohl, atacamaphoto.com)
The Schmidt Camera
There is one exception to the almost universal use over the last century of reflectors
of Cassegrain or Ritchey-Chrétien design, and that is the Schmidt camera, which
was invented in 1930 by the Estonian optician, Bernhard Schmidt (1879–1935).
This instrument has one great advantage over the conventional telescope – its very
wide field of sharp focus. The field of sharp focus for a Cassegrain design can be as
little as a few minutes of arc, that for a Ritchey-Chrétien, 10 or 20 min of arc; the
Schmidt camera, by contrast, can cover a field of view 6–10 across. This wide field
of view makes the instrument ideal for survey work, since large areas of sky can be
covered quickly.
However, the instrument does have several drawbacks. Firstly, the focus is inac-
cessible, making it only usable normally as a camera. Secondly, it has the opposite of
the Cassegrain’s telephoto property in that its physical length is at least twice its focal
length. Other problems are that it uses a lens, with the attendant difficulties of support,
and that the mirror is considerably larger than the useful aperture. An instrument that
uses both lenses and mirrors in the primary light-gathering capacity, like a Schmidt
The Schmidt Camera 19
Fig. 1.19 Optical layout

of the Schmidt camera
camera, is called a catadioptric instrument (reflectors can also be called catoptric

instruments, and refractors, dioptric instruments).
The superior optical performance of the Schmidt camera lies in the apparently
retrograde step of using a spherical mirror in place of the parabolic mirrors of other
telescopes. This reintroduces the problem of spherical aberration. However, the
other aberrations, such as coma, astigmatism, etc. (see later), are avoided. Spherical
aberration is then eliminated by using a correcting lens placed at the radius of
curvature of the primary mirror (Fig. 1.19). This lens is thin, and the chromatic
aberration that it produces can be kept small enough to be negligible. The only
remaining aberration therefore in the system is that the field of sharp focus is not flat
(field curvature), and this is overcome by bending the photographic plate or other
detector into the same shape as the sharp focus surface.
The largest Schmidt camera is the 1.34-m instrument at Tautenburg in Germany.
As with ordinary refractors, the size is limited by the maximum size of lens that can be
given an adequate support. Although there are a number of other smaller cameras at
other observatories, there is relatively little need for them since, with their wide fields of
view, two major instruments (one in each hemisphere) can quickly cover the whole sky.
The original Schmidt design can only be used as a camera. However the design
has been adapted to allow an accessible focus in recent years. The resulting
instrument is known as a Schmidt-Cassegrain, and forms the basis for several
types of excellent small telescopes produced for the amateur and education
markets. The Schmidt-Cassegrain uses a correcting lens close to the focus of the
primary mirror, and a secondary mirror (which is often attached to the lens, so
reducing diffraction effects) to send the focus out through the back of the telescope
through a hole in the primary mirror, as with the Cassegrain design (Fig. 1.20).
A rather similar design, called the Maksutov, uses spherical surfaces for the optical
components (Fig. 1.21). These instruments have the telephoto advantage of the
Cassegrain, allied with a wider field of sharp focus. Their production in large
quantities also means that they are relatively cheap. Small Schmidt-Newtonian
telescopes are also to be found. Essentially Newtonian telescopes that use a
spherical mirror and corrector lens, they seem to have no advantages over the
Schmidt-Cassegrain design.
Fig. 1.20 Optical layout

of a Schmidt-Cassegrain
telescope
Fig. 1.21 A 180-mm

Maksutov telescope.
The correcting lens with the
secondary mirror coated onto
it can clearly be seen.
(Copyright # C. R. Kitchin,
2011)
Multi-Mirror and Space Telescopes
The history of telescopes shows a recurring theme:

Existing designs are found to be limited in some way (image quality, size, cost, etc.).
#
A new development in telescope design occurs, which supersedes the existing
designs.
#
That new design is then used as the basis of telescopes of ever-increasing size,
until it too reaches its limitations in some way.
#
Further development of telescope performance has to await a new breakthrough
in design.
We have seen this with the early refractors, then the early reflectors, refractors
again and most recently yet again with reflectors.
The current limitation is that of the difficulty of producing large mirrors.
Although some recent telescopes have relatively conventional primary mirrors a
Multi-Mirror and Space Telescopes 21
Fig. 1.22 Optical layout of

the Keck 10-m multi-mirror
telescope
little over 8 m in diameter (Fig. 1.18), that size is nearing or at the upper limit of
what is possible with current technology. The fifth and latest cycle in this process of
telescope development is to replace the single main mirror of the telescope, and
sometimes the secondary mirror as well, with an assemblage of smaller mirrors.
This stage is now well established, and many current and most proposed major new
optical telescopes are based upon multi-mirror designs of one sort or another.
The first multi-mirror telescope of any significance was the Mount Hopkins
instrument that used six 1.8-m primary mirrors in a relatively conventional
Cassegrain design, but on a single mounting and feeding a common focus.
Although this gave the equivalent of a 4.4-m telescope in area, at perhaps a third
of the normal cost of such an instrument, it did not in practice prove successful, and
the instrument now uses a single 6.5-m monolithic mirror.
More successful approaches use a number of hexagonal mirrors mounted in an
array to synthesize the effect of a single large mirror. Each hexagonal mirror has to
be correctly positioned with respect to the others, and this is accomplished by active
supports that are continually adjusted by a computer to keep the alignment within
the required limits of about 50 nm. Each segment also has to have the correct shape
for its reflecting surface. In most cases this means it has to be an off-axis part of a
paraboloidal or hyperboloidal form. Special techniques, such as distorting the glass
blank while it is being polished so that when the stresses are released it springs into
the required shape and/or stressing the mirror segment actively while it is in use, are
needed for this purpose.
The 10-m Keck 1 and Keck 2 telescopes, built in the early 1990s, were the first
very large telescopes to have primary mirrors formed from hexagonal segments in
this way. These telescopes use thirty-six 2-m mirrors as the segments of their
primary mirrors (Fig. 1.22).
The Gran Telescopio Canarias (GTC), completed in 2009, also uses thirty-six
hexagonal component mirrors for its 10.4-m mirror. The Hobby Eberly Telescope
(HET) and its twin, the South African Large Telescope (SALT), use ninety-one
hexagonal segments for their 11 9.2-m primary mirrors. The shapes of the
segments, though, and so also of the mirrors as a whole, are spherical. This reduces
the problems of manufacturing the segments but means that the spherical aberration
(Fig. 1.5) in the images produced by the telescopes has to be corrected using four
smaller mirrors close to the focus.
The latter two telescopes also have mountings that rotate only in azimuth and
point to fixed altitudes of 55 (HET) and 53 (SALT). In order to follow an image as
it moves across the sky, the correcting mirrors and prime focus instrumentation (or
fiber-optic feeds for other instruments) therefore have to track along the focal plane.
The financial savings brought about by the identical mirror shapes and the
simplified mountings are, however, enormous – HET and SALT were constructed
for about a fifth of the cost of a similar sized conventionally built telescope.
The design of the Large Binocular Telescope (LBT) harks back to that of the Mount
Hopkins instrument. It uses two 8.4-m monolithic mirrors on a single mounting. Its
light-gathering power is thus equal to that of a single 11.9-m mirror – making it the
largest optical telescope in the world at the time of the writing this third edition. The
Giant Magellan Telescope (GMT) that is currently under construction will also use
8.4-m monolithic mirrors. It will utilize seven mirrors on a single mounting arrayed as
a single central mirror surrounded by the other six. Unlike the LBT, though, the outer
mirrors will be shaped to be off-axis paraboloids, so that the mirrors form a single
optical surface. Unusually the GMT will be of the Gregorian design (Fig. 1.9). It will
have an area equivalent to that of a single 21.4-m diameter mirror.
Two even larger telescopes are in their early stages of planning at the time of
writing, with possible completion dates in the next 10–15 years. The Thirty-Meter
Telescope (TMT) is to have 492 hexagonal segments for its primary mirror. Each
segment will be 1.45 m across and the mirror, as the telescope’s name implies, will
be 30 m in diameter. The European Extremely Large Telescope (E-ELT) will also
be based upon 1.45-m hexagonal segments, but will use 798 of them and be almost
40 m across (Fig. 1.23).
Yet another approach is to link separate telescopes by fiber optics, and this is the
principle of the European Southern Observatory’s (ESO) Very Large Telescope (VLT)
project. This has four 8.2-m conventional telescopes that can operate independently of
each other or be linked to give the light-gathering power of a 16-m mirror, and they can
also operate as an interferometer (see the later section on radio telescopes).
In a separate development, we have various space telescopes, such as the Hubble
Space Telescope (HST) and its planned replacement, the James Webb Space
Telescope (JWST). The instruments are designed to overcome the limitations
imposed on ground-based telescopes by Earth’s atmosphere. This both degrades
the image through scintillation, and absorbs radiation strongly in the ultraviolet and
infrared parts of the spectrum.
The HST’s primary mirror at 2.4 m in diameter is tiny compared with the
telescopes that we have just been discussing. However its diffraction-limited
Multi-Mirror and Space Telescopes 23
Fig. 1.23 A mock-up (using cardboard hexagons) of the 39.3-m primary mirror for the E-ELT.
The human figures give some idea of what will be the scale of the instrument. (Reproduced by kind
permission of ESO)
resolution (2.3) of 0.0500 2 has enabled it to make some outstanding discoveries.

There is currently some doubt over whether or not the JWST will be completed. Its
original cost estimate of $1 billion in 2001 has already spiraled to $9 billion, and it
seems unlikely to stop there. Furthermore the JWST will observe from the orange
part of the visible spectrum to the mid-infrared. Ground-based instruments are now
able to operate with near diffraction-limited resolution over this part of the
spectrum by correcting the distorting effects of the atmosphere (see next section).
2
Angular measure is normally in degrees ( ), minutes of arc (0 ) and seconds of arc (00 ), with
360 ¼ a full circle

600 ¼ 1
6000 ¼ 10
For some purposes, though, radians need to be used, and

2p radians ¼ 360 ¼ a full circle
so that
1 radian ¼ 57.2958 ¼ 57 170 4500 ¼ 206,26400 and 1 ¼ 0.01745 radians.
The use of hours, minutes and seconds as an angular measure is discussed in Chap. 4.
The scientific justification for the JWST is thus far less compelling than it seemed to
be a decade ago.
If completed (launch is currently planned for 2018), the JWST will have a 6.5-m
mirror made up of from eighteen 1.32-m segments. In this case, though, the mirror
will be segmented in order to reduce its weight and to allow the mirror to be folded
up so that it will fit into the payload bay of the launch rocket.
Atmospheric Compensation
The images produced by ground-based telescopes are affected by the passage of the
incoming light through Earth’s inhomogeneous and turbulent atmosphere. Part of
the atmosphere’s effect is visible even to the unaided eye as the well-known
‘twinkling’ of stars. Twinkling (or scintillation, as it is usually called by
astronomers) is the small rapid change in the brightness of a star. The overall effect
of the atmosphere on images is called ‘seeing’ and in addition to scintillation
includes blurring and movement of the image. Unless an image of a star is obtained
with a very short exposure (around a thousandth of a second) it is thus many times
the size of the diffraction limit of the telescope, and the resolution of the telescope is
far poorer than optical theory would suggest that it should be.
From an average observing site, the images of stars are likely to be several
seconds of arc across – equal to the theoretical resolutions of 10–100-mm diameter
telescopes. At the best observing sites, such as Chile’s Atacama desert and the
summits of Hawaii and the Canary Islands, the star images may get down to a
quarter of a second of arc at times, but more typically are 1–2 s of arc. Even quarter-
of-a-second-of-arc seeing, though, means that the large telescopes found on such
sites are performing many times worse than should be the case. For about the last
two decades therefore increasingly successful attempts have been made to counter-
act the atmospheric distortion of the images. The technique is called active atmo-
spheric compensation.
The main effect of the atmosphere is to delay some parts of the wave front
coming from the object more than others. What should be a flat wave front therefore
becomes one with bumps and hollows. To counteract these imperfections a small
mirror is added to the telescope, whose shape is the inverse of that of the wave front.
After reflection from that mirror, therefore, the wave front is again flat.
There are two problems with this procedure. One is finding the shape of the wave
front so that the correcting mirror can be made to the right profile. The second is that
the distortions produced by the atmosphere change on a time scale of a few
milliseconds.
The shape of the distorted wave front is determined by monitoring its shape from
a guide star close to the object actually being observed. There are several
techniques for determining the shape of the wave front, and one of those used
most commonly is the Hartmann sensor. This consists of a grid of small lenses, each
of which projects an image of the guide star onto a CCD (Chap. 9). The movements
of those images from their correct positions then provide the details of the wave
Radio Telescopes 25
front distortions. The guide star needs to be within a few seconds of arc of the object
whose image is to be corrected. If such a star is not available, some compensation
systems produce an artificial star by shining a powerful laser upwards to cause
sodium atoms to glow at a height of about 90 km.
Once the form of the wave front has been found, the correcting mirror is adjusted
to be its inverse shape. The correcting mirror may be constructed from a number of
small independent segments, or it may be very thin so that its shape can be bent. In
either case the mirror or mirror segments are mounted on computer-controlled
actuators whose position can be changed in a fraction of a millisecond in order to
produce the correct profile for the mirror.
With such an active atmospheric compensation system operating, large ground-
based telescopes can reach resolutions of a tenth of a second of arc or so. In the visible
region, this is nearly as good as the Hubble Space Telescope, while in the near infrared,
ground-based telescopes can equal the Hubble Space Telescope’s performance.
Whether the guide star is real or artificial, the region of the sky over which the
atmospheric distortion can be corrected is small, 1500 or less in diameter. This is too
small to obtain sharp images of the whole of many galaxies and nebulae and of
Venus, Mars, Jupiter or Saturn. Recently therefore the use of several artificial guide
stars, two or more wave front sensors and two or more correcting mirrors has been
pioneered, particularly for the 8.1-m Gemini South telescope. At the time of writing
a constellation of five artificial guide stars has successfully been demonstrated.
Plans for the TMT envisage up to nine such guide stars being used. This technique
is dubbed ‘Multi-Conjugate Adaptive Optics’ and results in atmospheric distortions
being corrected over an area of the sky 1 min of arc or more across.
Clearly a full atmospheric compensation system is complex and expensive. It is
unlikely to become available to amateur astronomers. A significant improvement to
the sharpness of an image, however, can be obtained at quite a modest cost by
correcting just for the overall slope introduced to the wave front by the atmosphere.
That slope may be corrected by using a tip-tilt mirror. This is a plane mirror that can
be tilted rapidly in any direction to compensate for the overall inclination of the
wave front. Commercially produced devices are now available for use on small
telescopes.
Two or more telescopes can also achieve high resolution by being used together
as an interferometer or as an aperture synthesis system (see the section on radio
telescopes, following).
Radio Telescopes
The atmosphere is all but fully transparent only over two regions of the spectrum:
the optical and radio regions (Fig. 7.2 in Chap. 7). Although most amateur
astronomers use visual telescopes, it is quite easy to build a small radio telescope
to detect the brighter sources, such as the Sun and the center of our galaxy. Large
radio dishes, interferometers, and aperture synthesis systems seem likely, though,
largely to remain the territory of professional astronomers and major observatories.
Fig. 1.24 Three (out of 66) of the large parabolic dishes that form the Atacama Large Millimeter
Array (ALMA) on the Chajnantor Plateau. (Reproduced by kind permission of ESO/José
Francisco Salgado, josefrancisco.org)
The huge dishes (Fig. 1.24) that form most peoples’ idea of a radio telescope are
only a part of the whole instrument. The purpose of the dishes is to concentrate the
radiation onto the detector, and also to shield the detector from unwanted radio
emissions. The dishes are large partly because of the intrinsic faintness of radio
sources. The unit used for intensity at radio wavelengths is the jansky3 (defined as
1 Jy ¼ 1026 W m2 Hz1), and some astronomical sources have millijansky
intensities. The dishes are also large because the resolution of a telescope (Chap.
2) varies directly with the wavelength (2.3). A 10-m optical telescope thus has a
resolution (ignoring atmospheric degradations) of 0.0100 , while a 10-m radio tele-
scope operating at 21 cm wavelength would have a resolution of 5,25000 , or nearly
1.5 . Most radio dishes have similar optical principles to visual telescopes and
operate either at prime focus or as Cassegrain systems.
Once the radio waves have been brought to a focus, they must be converted to an
electrical signal in order to be detected. This is done by the feed. The feed is often
the Yagi antenna that is well known as the ubiquitous television aerial, or a half-
wave dipole. The latter is just the active element of the Yagi antenna and comprises
two in-line conducting strips each a quarter of the operating wavelength long.
3
Named for Karl Jansky (1905–50) who, in 1932, started off radio astronomy by detecting radio
emission from the galactic center.
Interferometers 27
At high frequencies, waveguides may need to be used and the radiation detected by
Schottky diodes, superconducting tunnel junctions or bolometers. Details of these
latter detectors are beyond the scope of this book, and the interested reader is
referred to sources listed in Appendix A.
Once the radio wave has been converted to an electrical signal it is then
amplified and processed by the radio receiver. Most receivers for radio telescopes
operate on the same super-heterodyne principle used in domestic radios. Super-
heterodyne receivers mix the main signal with a signal from a local oscillator at a
similar but different frequency. The beat frequency between the two, usually known
as the intermediate frequency, is then amplified. The receiver may also filter the
signal and convert it into a voltage proportional to the input power. The final output
nowadays is usually to a computer that can then produce the image in whatever
format may be desired.
Interferometers
The resolution of even the largest radio dish in the world, the 305-m Arecibo
telescope,4 is poor compared with that of the smallest optical telescope. Since it
is impractical to build even larger single instruments, high resolution at radio
wavelengths, and increasingly at optical wavelengths, is achieved through
interferometry.
The basic operating principle of the interferometer is to combine the outputs
from two or more telescopes when they are observing the same object. The outputs
interfere with each other to an extent that depends upon the instantaneous path
differences between the signals. As the object moves across the sky, those
path differences change, and so the output from the interferometer oscillates
(Fig. 1.25). The form of the oscillation may then be used to provide high resolution
data about the object (Chap. 2). The angular separation of two point sources may be
easily determined using an interferometer, but more comprehensive imaging
requires the use of advanced mathematics that is beyond the scope of this book.
The interested reader is referred to sources listed in Appendix A.
The resolution of an interferometer can be very high since it depends upon the
separation of the two telescopes (Chap. 2), not the diameters of the telescopes. With
very long base-line interferometry, that separation can be thousands of kilometers,
leading to resolutions of milli-arc seconds or less. However, there is no increase in
sensitivity, since no more light or radio waves are gathered.
The sensitivity as well as the resolution of a telescope with a diameter equal to
the separation of the two telescopes making up the interferometer may be achieved
4
The RATAN600 radio telescope in the Caucasus has a diameter twice that of the Arecibo dish,
but consists of just a thin annulus made up from 895 7.4 m 2 m mirrors. Its maximum practical
collecting area is thus around 10,000 m2 compared with Arecibo’s 73,000 m2.
Fig. 1.25 An interferometer and its output
through aperture synthesis. This technique was also first applied at radio
wavelengths, but again is starting to be used in the infrared and visible regions. It
relies upon Earth’s rotation to change the orientation in space of a pair of
components of an interferometer through 360 over a 24-h period. The two
components thus trace out an annulus with a diameter equal to their separation
and a thickness equal to their apertures over that period (Fig. 1.26).
If one component is then moved through its own diameter, the next annulus can
be synthesized in the next 24-h interval. Over a period of time, by moving the two
components from being next to each other to their maximum separation, or vice
versa, adjacent annuli can be synthesized to give the effect of observing with a
normal telescope with an aperture equal to the maximum separation.
Interferometers 29
Fig. 1.26 Aperture synthesis showing the first two annuli swept out during the first 2 days of
observing. Further movement of the second telescope through its own diameter during subsequent
observing sessions will complete the synthesized telescope
The technique can only be used on objects that will remain unchanged over the
period of observation. The process may, however, be speeded up by using many
telescopes. For example, an interferometer with five telescopes could provide ten
different separations simultaneously (Fig. 1.27), one with ten telescopes would
provide 45 separations, etc. Also, in practice, only 12 h of observation are needed
since the other 12 h can be reconstructed within the computer recording the
observations. Even shorter periods of observation can be achieved by using several
arms to the interferometer, oriented at different angles to each other. Thus the Very
Large Array (VLA) in New Mexico has a Y-shaped configuration and twenty-seven
25-m diameter radio dishes with a maximum separation of 36 km. ALMA
(Fig. 1.24), which observes the sky over the 0.3–9.6 mm part of the radio spectrum,
Fig. 1.27 An aperture synthesis system with five telescopes can synthesize ten annuli
simultaneously
uses over 60 antennas laid out in a complex and changeable array and can simulate a
14-km diameter instrument.
If all the annuli required for the equivalent normal dish are synthesized, then it is
a filled aperture system. If some annuli are omitted, then it is an unfilled aperture
system. This latter is common for the larger base-line systems such as the UK’s
e-MERLIN,5 which has a maximum separation of 217 km. Unfilled aperture
systems synthesize the resolution but not the complete sensitivity of the equivalent
normal telescope.
5
Originally MERLIN stood for ‘Multi-Element Radio Linked Interferometer Network.’ It now
uses dedicated fiber optic cables to link the telescopes, but the original acronym has been retained.
Chapter 2
Telescope Optics
Point Sources
Telescopes normally use two lenses and/or mirrors to produce a magnified, and in
the case of point sources, brighter image. There are many different designs, the
basic properties of several of those more commonly encountered having been
discussed in Chap. 1. In this chapter we take a closer look at the details of some
of those designs.
Astronomical telescopes operate with the object at an effectively infinite dis-
tance. When used visually by someone with normal eyesight, they produce a beam
of parallel rays of light from the eyepiece, which is then focused by the eye to
produce the observed image. The focal points of the objective and the eyepiece
therefore coincide in normal use. It can easily be seen from Fig. 2.1 that the angle
for the exit rays (b) is related to the angle for the incoming rays (a) by the ratio of
the focal lengths of the objective and the eyepiece. This ratio, for visual work, is the
magnification of the telescope:
b fo
Magnification ¼ M ¼ ¼ (2.1)
a fe
For point sources, that is, images that are smaller than the detecting elements of
the detector (the rod and cone cells in the case of the eye), the telescope also gives a
brighter image. The situation for extended sources (images larger than the detector
elements) is more complex and is considered later in this chapter. The increase in
brightness for a point source is given by the light grasp, G, of the telescope. This is
the ratio of the collecting area of the telescope to that of the pupil of the eye. The
pupil of the dark-adapted eye is typically about 7 mm in diameter. The light grasp,
ignoring any losses of light in the telescope, is therefore given by
pD2 =4
Light grasp ¼ G 20000D2 (2.2)
p0:0072 =4

32 2 Telescope Optics
Fig. 2.1 Optical paths in

an astronomical telescope
where D is the diameter of the objective in meters. A typical amateur astronomer’s

0.2-m telescope thus has a light grasp of 800, the Keck telescopes (D ¼ 10 m) have
light grasps of 2,000,000, while the E-ELT (D ¼ 39.3 m) will have one of
30,000,000. In the highly unlikely event that the E-ELT should ever be used for
visual observing, Pluto would appear as luminous as does Venus to the unaided eye
when it is at its brightest – and that is sufficient to cast shadows on Earth.
Example Calculation 2.1

Q. What is the magnification of a 0.5-m f4 Newtonian telescope when used with
an eyepiece of focal length 15 mm and what is its light grasp?
A. Since the focal ratio (denoted by the “f4”) is given by the ratio of the
objective’s focal length to its diameter, we have
fo
Focal ratio ¼
D
and so
f o ¼ Focal ratio D ¼ 4 0:5 ¼ 2 metres
Thus from (2.1)
fo 2
Magnification ¼ ¼ ¼ 133
f e 0:015
From (2.2) we have
Light grasp 20000D2 ¼ 20000 ð0:5Þ2 ¼ 5000
The ultimate limitation on telescope performance is due to the wave nature of

light. At any edge (such as the rim of an objective lens or mirror) the light waves
will be diffracted (Fig. 2.2). Even for a point source, therefore, in an otherwise
Point Sources 33
Fig. 2.2 Diffraction

of light at an aperture
Fig. 2.3 Diffraction-limited

image of a point source (Airy
disc and fringes)
perfect telescope, some of the light will be spread away from the center of the
image, and the actual image will not be a point. In such a case there will also be
interference effects, and the image will consist of a circular bright central region,
known as the Airy disc, surrounded by light and dark fringes (Fig. 2.3). If the optics
of a telescope are diffraction-limited, then its angular resolution, A, is given by the
radius of the Airy disc:
1:22l
A¼ radians (2.3)
D
where l is the operating wavelength and D is the diameter of the objective (in the
same units).
Fig. 2.4 Exit pupil and

eye relief of an eyepiece
Thus, for a telescope used visually, with l ¼ 500 nm, we have
0:122
A seconds of arc (2.4)
D
where D is now in meters. This measure of the resolution of a telescope is often
called its Rayleigh limit.
Since the resolution of the eye is typically 3–5 min of arc, if a telescope is to be
used at its diffraction limit, then it must magnify the image sufficiently (A M) to
exceed the eye’s resolution. Thus a minimum magnification of about 1,300D would
appear to be needed for a telescope used visually to realize its potential resolution in
practice. However, in reality, Earth-based telescopes have their resolutions limited
by the atmosphere. From an average site, that resolution might be as poor as 2–1000 .
From a good site, the resolution might reach 100 occasionally. The very best sites,
such as Mauna Kea on Hawaii, with all possible precautions taken to minimize the
effects of the dome and telescope, etc., can reach 0.2500 . The Hubble Space
Telescope, with its diameter of 2.5 m, thus betters our resolution of astronomical
objects by a factor of about 5. Atmospheric compensation (Chap. 1) allows ground-
based telescopes to reach resolutions of 0.100 on occasion.
The eyepiece, since it consists of one or more lenses, produces an image of
the objective, known as the exit pupil (Fig. 2.4). All the light passing through the
objective must thus pass through the exit pupil, and as can be seen from the diagram,
the exit pupil is also the point at which the emerging pencil of rays has its smallest
diameter. The best place for the eye when observing with a telescope is therefore to
be coincident with the exit pupil. For all the light to pass into the eye, the exit pupil
must clearly be smaller than the eye pupil. The diameter of the exit pupil, Dep, is
given by
Df e Df D
Dep ¼ e¼ (2.5)
fo þ fe fo M
(since fo >> fe). The minimum magnification for a telescope, if all the light it
gathers is to be used by the eye, is therefore given by
D
0:007 > Dep ¼ (2.6)
M
Point Sources 35
or
Mmin 140D (2.7)
The distance from the eyepiece to the exit pupil is known as the eye relief.
For the simple lens shown in Fig. 2.4, the eye relief is clearly approximately given
by fe. For comfortable viewing, its value should be about 6–10 mm. Practicable
eyepiece designs (see below), using several lenses, may have values of eye relief
from 2 to 20 mm.

Q. For the telescope featured in Example calculation 2.1, assuming that it was at
a site where it could operate at its diffraction limit, what eyepiece focal lengths
would be needed (1) to ensure that all a star’s light entered the observer’s eye
and (2) to ensure that the full resolution of the telescope could be utilized by the
observer?
A. (1) From (2.7) and assuming that the observer’s eyes are dark adapted, we
have
Mmin 140D ¼ 140 0:4 ¼ 70
Rearranging (2.1) (see also (2.18)) we have
fo 2
fe ¼ ¼ ¼ 0:029 m ¼ 29 mm
M 70
Thus ideally a 29 mm focal length eyepiece would be needed to ensure that

all a star’s light entered the observer’s eye. In practice this would be likely to
mean that an eyepiece with a focal length of 25 or 30 mm would be used.
(2) From the above discussion, a minimum magnification of about 1,300D
would be needed for a telescope used visually to realize its potential resolution
(see also 2.19). For a 0.5-m telescope this would be a minimum magnification
of 650 and so
fo 2
fe ¼ ¼ ¼ 0:003 m ¼ 3 mm
M 650
In practice few observers would be likely to have an eyepiece with such a

short focal length, so a longer focal length eyepiece would need to be combined
with a Barlow lens (see later) to give the required magnification.
The field of view of an eyepiece varies with its design and can range from 30 to

70 , or more. The field of view of the telescope is then (if not restricted by other
parts of the instrument) given by
Field of view of eyepiece

Field of view of telescope ¼ (2.8)
M
Extended Images
The light grasp of a telescope (2.2) shows that point sources are brighter when viewed
through a telescope, and this is in line with most people’s expectations of what a
telescope does. However, the same is not true for extended sources, which in this context
means any source whose image is larger than the size of the elements of the detector.
Thus even stars at high magnifications and/or in poor atmospheric conditions can
become extended sources. For extended sources, the greater amount of light gathered
by the telescope must be spread over the larger area of the magnified image. Thus
Surface brightness through the telescope G

¼ 2 (2.9)
Surface brightness to the eye M
Since from (2.2) to (2.5), we have
D2o
G¼ (2.10)
D2eye
and
Do
M¼ (2.11)
Dep
we get
Surface brightness through the telescope D2ep

¼ 2 (2.12)
Surface brightness to the eye Deye
Now at the minimum magnification (2.7), we have
Dep ¼ Deye (2.13)
and so
Surface brightness through the telescope ¼ Surface brightness to the eye:

(2.14)
Also, at higher magnifications we have
Dep < Deye (2.15)

Extended Images 37
and so
Surface brightness through the telescope < Surface brightness to the eye;
(2.16)
while at magnifications lower than the minimum (2.6)
Dep > Deye (2.17)
so that not all the light from the telescope enters the eye. Effectively, in this case,
the objective is reduced in size by the ratio Deye/Dep. The ratio of surface
brightnesses therefore remains unity as the magnification is reduced below the
minimum needed to fill the pupil of the eye. The surface brightness of an extended
object is thus, at best, no brighter than when seen with the naked eye, and will
usually be fainter than this. Other losses will ensure that in all circumstances the
object is in fact fainter through a telescope than when seen directly!
This result is, to most people, very surprising, and perhaps contradicts their
experience of using a telescope. Thus faint extended objects, like comets or the
Andromeda Galaxy, M31, do appear brighter through a telescope, and sometimes
binoculars are advertised for night vision, whereas in fact their images cannot be
brighter than those seen with the unaided eye. The answer to this paradox lies in the
structure of the eye. The retina contains two types of detectors, cells known as rods
and cones (from their shapes). The cones are of three varieties, sensitive to the red,
yellow and blue parts of the spectrum, respectively, and enable us to see in color. The
rods are sensitive to only the yellow part of the spectrum. In bright light the rods
are almost depleted of their light-sensitive chemical (visual purple, or rhodopsin), and
we therefore see via the cones. At low light levels, the rhodopsin in the rods
regenerates slowly. When the rhodopsin is fully replaced, the rods are about 100
times more sensitive than the cones. We then see principally via the rods in the retina.
This behavior results in two commonly experienced phenomena. Firstly, that of
dark adaptation: our vision on going from, say, a brightly lit room into the dark
slowly improves. This is partly because the pupil of the eye increases in size, but
mostly because the regeneration of the rhodopsin is restoring the sensitivity of the
rods. After about half an hour in dark conditions, one can often see quite clearly,
whereas nothing at all had been visible to begin with. The second phenomenon is
that of loss of color in objects at night, and this occurs because the rods respond
only to a single waveband of light.
Now, the rods and cones are not evenly distributed throughout the retina.
The cones are concentrated in a region called the fovea centralis, which is the
point where the image falls on to the retina when we look at it directly. The rods
become more plentiful away from this region. If we look directly at a faint extended
object, its image will fall on to the fovea centralis, with its concentration of low
sensitivity cones. However, on looking at the same image through a telescope, it
will be magnified and some portions of it will have to fall on to parts of the retina
away from the fovea centralis, and thus be detected by the higher sensitivity rods.
Thus faint extended sources do appear brighter through a telescope, not because
they are actually brighter but because more sensitive parts of the eye are being used
to detect them.
This property of the eye can also be used deliberately to improve the detection of
faint objects through the use of averted vision. In this, one deliberately looks a little
way to the side of the object of interest. Its image then falls on to a region of the
retina richer in rods, and becomes quite noticeably easier to see.
Averted vision is a difficult trick to acquire, because as soon as the object flicks
into view, one’s normal reaction is to look directly at it again, and so it disappears.
However, with practice the technique will become easier, and for any aspiring
astronomer it is well worth the effort required to perfect it.
Telescope Objectives
The objective of a telescope is the primary light-gathering element of the instrument,

and it can be either a lens or a mirror, or sometimes both. Its function is to gather the
light and to focus it. The main properties required of both lenses and mirrors when
used as objectives have been covered in Chap. 1, and there is no need to add to that
discussion. Here, the practical requirements for telescope objectives in current use or
under construction are considered.
Except for Schmidt cameras and small telescopes for the amateur market (which
have been discussed in Chap. 1), a telescope objective today is invariably a metal-
on-glass mirror. Its surface is a paraboloid, or for the Ritchey-Chrétien design, a
hyperboloid. The surface must be true to its correct shape to better than about one-
eighth of its operating wavelength if it is to perform to its diffraction limit (2.3).
For visual work this requirement is therefore about 60 nm. For some purposes, such
as observations of the planets with their low levels of contrast, even more stringent
limits may be necessary. The primary mirror of the Hubble Space Telescope
deviated from its correct shape by 2 mm, which by everyday standards is a very
tiny amount. However, it is over 30 times the deviation acceptable on even the
cheapest of telescopes, hence the early problems the instrument experienced.
The surface of the glass or other ceramic material used for the substrate of the
mirror is still shaped by techniques that Newton and Herschel would recognize
(Chap. 3). The mirror starts off as a disc of material called a blank. A second blank
(called the tool) is placed beneath the mirror blank with some grinding powder,
such as carborundum, between. The mirror blank is then moved backwards and
forwards over the tool in as random a motion as possible. With a coarse grade of
grinding powder, the mirror blank quickly becomes concave and the tool convex.
The process is continued until the mirror blank is hollowed out to the required
depth. Sometimes this stage may be replaced by milling the surfaces to the
required shapes with a diamond cutter in a numerically controlled machine. With
the mirror blank of roughly the correct shape, the pits left by the grinding have to be
removed. This is done by using a finer grade of grinding powder. The pits left by that
powder then have to be removed with a yet finer grade of powder, and so on. Six or
seven such stages may be needed before the mirror blank is smoothed sufficiently
Telescope Objectives 39
Fig. 2.5 Optical layout of the Foucault test and an example of the shadow patterns to be seen
to be polished. Polishing is a similar process to grinding but using a softer powder.

Usually ferrous oxide or cerium oxide powder is used for this process, and the tool
is covered with a layer of pitch to provide a slightly malleable surface.
When the mirror is finally polished, it then has to be figured. This is the process
whereby the required shape for the surface is produced. It consists of a continued
polishing of the surface, but with the polishing effect concentrated on those areas
that need to be reduced. Only a micron1 or so of material can be removed by
figuring, and so the mirror must be close to its finally required shape before the
process starts. The shape of the mirror must be tested during figuring to ensure that
it is being altered towards that required. A variety of tests are available. One of the
most straightforward is the Foucault test which, despite being simple enough for the
equipment it needs to be constructed in a home workshop, can measure the shape of
the mirror surface to a few tens of nanometers (more than accurate enough to have
shown the problems for the Hubble Space Telescope mirror, at a cost of 0.00001%
that of the COSTAR corrective optics!).
The optical layout of the Foucault test is shown in Fig. 2.5, and the shape of the
surface is judged by the shadows appearing on the mirror as the knife-edge cuts
the reflected beam. With large mirrors the tool may be considerably smaller than the
mirror, and moved over the surface by a numerically controlled machine. With
unusual shapes, such as the off-axis hyperboloids required for the segments of the
Keck telescope, the blank may be stressed during the figuring, so that it may be
1
A thousandth of a millimeter. It is usually symbolized as ‘m’ or ‘mm’ so that 1,000 mm ¼ 1 mm.
polished to a simple curve, and then it will relax to the more complex curve that is
required, when those stresses are released.
The mirror, once correctly figured, must be sufficiently rigid to keep its shape
under changing gravitational loadings as the telescope moves around the sky. With
older telescopes, and with small telescopes, this was and is accomplished by
making the mirror very thick. At larger sizes, however, the mirror then becomes
very heavy, requiring a massive and expensive mounting. It will also have a large
thermal inertia, and thus take a long time to match the local ambient temperature.
While its temperature is changing it will have thermal stresses within it, which
except for very low expansion materials, such as Zerodur and ULE (Ultra Low
Expansion), will lead to distortions of the surface. The Russian 6-m telescope, for
example, which has a Pyrex mirror, often gives poor quality images because the
temperature of its mirror has not stabilized even by the end of the night. These
problems can be partly overcome by making the mirror with a honeycomb back, as
is the case with the 5-m Mount Palomar telescope, thus retaining most of the
rigidity of a thick mirror, but reducing the weight and thermal inertia. More recently
this has been extended to fabricating the mirror blank inside a rotating furnace. Two
thin sheets of material are used for the front and back of the blank separated by
numerous “struts” made from the same material. The whole is then fused together in
the furnace, which because it is rotating, produces a dished front surface for the
blank that is close to the finally required shape. An alternative approach is to make
the mirror thin, but to mount it on active supports that are rapidly and frequently
adjusted under computer control to distort the mirror so that its front surface
remains in the required shape.
Eyepieces
“Eyepieces matter very little for real astronomy.”

“Eyepieces are the most important accessories you will ever buy for your
telescope.”
These two extreme views of the importance of eyepieces summarize the confusion
that reigns over this topic, for both are perfectly valid, and what is more, perfectly true
in certain circumstances. It is the last caveat that is the important consideration when
choosing among the enormous range of types of eyepiece available. The type of
eyepiece that you will need depends upon the purpose for which you intend to use
your telescope. Eyepieces are not cheap. At the time of writing one can pay over 15%
of the cost of a complete 0.2 m Schmidt-Cassegrain telescope system for a single
eyepiece. Since several eyepieces with a range of focal lengths will normally be
needed, it is clear that they need to be chosen with care if money is not to be wasted.
The first quotation is applicable when the telescope will normally be used with a
camera, CCD detector, photometer, etc. Then, the eyepiece is only needed for
visually aligning the telescope on to the object to be studied before the real work
Eyepieces 41
begins. The performance of the eyepiece is almost immaterial; it can suffer badly
from aberrations (see below) and still be quite adequate for its purpose. The same
comment applies to an eyepiece to be used for guiding, though a somewhat higher
quality will not go amiss in order to reduce eye strain, since it may be in use for
considerable lengths of time. A guiding eyepiece will also normally need to have
illuminated cross-wires.
The second quotation applies when a telescope is to be used primarily for visual
work. Even for this type of work, however, it is easy to waste money on expensive
eyepieces. One of the major factors in the cost of an eyepiece is its field of view, and
this can range from 30 at the bottom end of the range to 85 or more for ultra wide-
angle eyepieces. Yet if we consider how a telescope is used in practice, a large field
of view is rarely necessary. The object being studied will always be moved to the
center of the field of view for critical study whatever type of eyepiece is used, and
on the optical axis eyepieces differ little in their performances. Thus the only
applications that require eyepieces with a wide field of view will be searching for
new comets and novae, and for “gawping” at large extended objects such as the
Orion Nebula. A wide field of view is useful when trying to find objects at the start
of an observing session, but this can be achieved much more cheaply by using a low
power eyepiece rather than a wide angle eyepiece (2.8).
Another application that might appear to require a wide angle eyepiece is
micrometry. This is a technique for measuring the angular sizes and separations of
objects in the sky. The micrometer eyepiece has a fixed, centered set of cross-wires,
and a second movable set whose position may be determined accurately. To measure
the angular separation of a double star, for example, the fixed set of cross-wires
would be set on to one star by moving the telescope, and the other set moved to the
second star. However, it is more important to have an undistorted image for this
purpose, and so the simpler designs of eyepiece are again to be preferred.
The second consideration in choosing eyepieces is their focal lengths, and hence
the resulting magnifications of the telescope (2.1). It is usual to have eyepieces with
a range of focal lengths to suit different observing conditions and different objects.
Four eyepieces, well chosen, will normally be sufficient for most purposes. As we
have seen, there is a minimum magnification if all the light gathered by a telescope
is to be utilized by the eye (2.7). This translates into a maximum focal length for an
eyepiece for a particular telescope of
fo
fe (2.18)
140D
or a focal length of 70 mm and a magnification of 30 for the widely found f10,
0.2 m Schmidt-Cassegrain telescope. Of course, a lower power eyepiece may still
be used, and be of use when searching for objects, but some of the light gathered by
the telescope will be wasted. The atmosphere will be sufficiently steady to allow
telescopes less than about 0.25 m in diameter to perform at their diffraction limit a
few times a year from most sites. There is then a minimum magnification of about
1,300D if that limit is to be realized visually. This again translates into a maximum
focal length, given by
fo
fe (2.19)
1; 300D
For the same telescope as in the previous example, this would be a focal length of
8 mm and a magnification of 250. High power, however, is more usually a
disadvantage, since it reduces the contrast in the image and often results in less detail
being able to be seen. Under normal observing conditions, magnifications greater than
200 would rarely be used. Thus for the f10, 0.2 m Schmidt-Cassegrain previously
considered, a suitable range of eyepieces would have focal lengths of about 40 mm
(50), 25 mm (80), 12.5 or 15 mm (160 or 130) and 6 mm (330).
Other items to consider in choosing an eyepiece include the eye relief (Fig. 2.4),
which for comfortable viewing should lie between about 6 and 10 mm, and whether
or not the eyepieces are parfocal. This last term means that, when pushed fully
home into the eyepiece mount on the telescope, the eyepieces have their foci at the
same point. Exchanging eyepieces therefore does not require the telescope to be
refocused. Parfocal eyepieces will normally have to be bought as a set from a single
manufacturer and will be more expensive than equivalent non-parfocal eyepieces
bought separately. However, the extra expense is usually worthwhile because of the
increased ease of use of the telescope. Some eyepiece designs may have up to eight
lenses. Plain glass reflects about 5% of the light incident on to it, so with 16
surfaces, such an eyepiece could lose about 55% of the light entering it. It is thus
essential to have the lenses in eyepieces covered with an anti-reflection coating to
reduce the losses at each surface to 1% or less.
Finally, we come to a consideration of the choice of eyepiece optical design.
There are a huge number of eyepiece designs, many of which are only slight
variations on each other. Only the most commonly encountered designs are consid-
ered here, and their optical layouts are shown in Fig. 2.6. We may divide them into
three groups on the basis of their cost and performance. At the bottom end we have
class C designs such as the Ramsden and Huyghenian. These are not achromatic and
have small fields of view. Though cheap, they are generally not worth considering.
In the middle range, class B, we have eyepieces such as the Kellner, Orthoscopic,
Monocentric and Plössl. These designs are well corrected for aberrations, have a
wider field of view and are only a little more expensive than the class C eyepieces.
They are the eyepieces of choice for most people for most purposes. The most
expensive group, class A, are the wide-angle eyepieces, such as the Erfle and Nagler.
As already discussed, these have few applications that justify their cost.
Accessories
Star Diagonal
The star diagonal is just a device to reflect the light from the telescope through
90 (Fig. 2.7). It can make the observing position more comfortable when using a
small telescope close to the zenith, but it will generally degrade the image to
Fig. 2.6 Optical layout of
some commonly encountered
eyepieces. In each case the
eye end of the eyepiece is
on the right
Fig. 2.7 Star diagonal

Fig. 2.8 Solar diagonal

(Herschel wedge)
a noticeable extent. It is possible to purchase a diagonal in which several eyepieces

can be mounted simultaneously. The eyepieces are in a rotating carousel, which
enables them to be interchanged very rapidly. Parfocal eyepieces are almost
essential for use in such a device.
Solar Diagonal
In the past, the reduction in intensity caused by reflection from plain glass has been
used as the basis of a device for observing the Sun known as a solar diagonal
(Fig. 2.8) or Herschel wedge. (Note that great care should always be exercised in
solar observing. Make sure that you have read about and applied all the precautions
discussed in Chap. 8 before undertaking any solar observation.) The reflecting
surface for the diagonal is just a plain piece of glass that diverts about 5% of the
incoming light into the eyepiece. The glass is wedge-shaped to prevent reflections
from the second surface from entering the eyepiece.
Solar diagonals are no longer recommended for observing the Sun. The reason
for the change in the recommendation on the use of solar diagonals is that, in order
to be safe, they must be used on a 50-mm (2-in.) or smaller telescope, and at a
minimum magnification of 300. This arises because plain glass reflects about 5%
of the incident radiation, and the solar intensity must therefore be reduced by a
further factor of 1,700 in order for the final intensity of the image to reach the
recommended safe limit of 0.003% of the unfiltered solar image. Since nowadays
few people use telescopes as small as 50 mm except as finders, and rarely will
seeing conditions in the daytime be good enough to allow such high magnifications,
there is a great temptation to use the solar diagonal on a telescope that is too large
and with too low a magnification. A solar diagonal on a 75-mm (3-in.) telescope
used at 100 for example will produce an image that is nearly 20 times brighter
than the safe limit.
Accessories 45
Fig. 2.9 Barlow lens
Additional problems with the solar diagonal arise from its use at the eyepiece end
of the telescope. The objective will therefore form a real image of the Sun inside the
telescope, and this may cause damage to the telescope structure. The wedge itself is
“fail-safe” in that if it shatters due to the heat from the Sun, the eye is not exposed to
the full solar brightness. However, even in normal use some 90% of the solar energy
passes through the diagonal and emerges from the back of the telescope. It is easy to
forget this when observing, and only to remember when the smell of scorching can
no longer be ignored!
Full aperture filters consisting of aluminized Mylar film are better than solar
diagonals, since the light from the Sun is cut down before the telescope concentrates
it (Chap. 8).
Barlow Lens
The Barlow lens is a diverging lens placed a short distance before the eyepiece.
It serves to increase the effective focal length of the objective (Fig. 2.9), and therefore
(2.1) increase the magnification obtained from a given eyepiece. When using a
camera on the telescope a larger image may be obtained by imaging through the
eyepiece. The method is called eyepiece projection, and it is discussed in Chap. 8.
Telecompressor
A telecompressor is the inverse of a Barlow lens. It is a positive lens placed before

the eyepiece in order to decrease the effective focal length of the telescope. Its main
use is to increase the field of view when imaging large objects.
Filters
Filters are used in photometry (Chap. 11) to define the wavelengths over which a
star’s energy is being measured. They can also be used to enhance visual observing,
and to improve CCD and photographic images.
The light pollution rejection filter is a filter designed to absorb strongly at those
wavelengths where the light pollution is most intense. If the local street lighting is
predominantly low pressure sodium lamps, then a light pollution rejection filter
absorbing the region around 590 nm (the sodium D lines) can be very effective.
Unfortunately high pressure sodium and mercury lamps emit light over most of the
visible spectrum and so cannot so easily be eliminated.
The nebula filter is a narrowband filter designed to transmit the strong emission
lines from gaseous nebulae. These lines are principally the forbidden lines of doubly
ionized oxygen at 495.9 and 500.7 nm or the H-b line at 485.6 nm resulting in
oxygen III and H-b filters, respectively. Use of the filter enables longer exposures to
be used and hence images obtained with improved signal to noise ratios, since much
of the light pollution is eliminated but most of the light from the nebula retained.
The H-a cut-off filter is a filter that cuts out wavelengths shorter than about
650 nm, thus allowing through the H-a line at 656.28 nm, but eliminating most light
pollution. It may be used for imaging gaseous nebulae. Note that it is not the same
as the H-a narrow band filter and must not be used for solar observing.
The Comet filter is a filter that transmits light from about 460 to 550 nm so that
the emission lines of molecular carbon (known as the Swan bands) are transmitted.
Such a filter will help to enhance the views of the ion tail of a comet.
Major Telescopes
Many small and all large telescopes are used without eyepieces, and the detector,
camera, photometer, spectroscope or other instrument, is placed directly at the focus
of the telescope. In some cases a fiber optic feed may be placed at the telescope’s
focus and the light then taken to the instrument, which is housed elsewhere.
Magnification2 and the other visual observing criteria are now no longer relevant.
Instead the performance of the telescope depends upon the size of the objective and
its focal length.
The focal length determines the physical size of the telescope’s image – known
as the image scale. When a telescope is used at prime focus the focal length is that
of the objective. At secondary foci (Cassegrain, Coude, Gregorian, Nasmyth, etc.)
the effective focal length is required. This is usually much longer than the focal
length of the objective because of the telephoto effect of the secondary mirror
(Chap. 1). The effective focal length will be determined by the optical properties of
the secondary mirrors, and often a large telescope will have several different
secondary mirrors so that its effective focal length may be altered.
2
Indeed it is clearly inappropriate to talk of the “magnification” of an image of (say) the Moon
when that image has a physical size at the telescope’s focal plane of perhaps 1 cm while the Moon
itself is 3,476 km in diameter.
Major Telescopes 47
With the focal length or effective focal length (f – measured in meters) known,
the image scale (IS) is then given simply by
1
IS ¼ radians per meter (2.20)
f
or, more usefully
206
IS ¼ seconds of arc per millimeter (2.21)
f
Thus a 10-m telescope with an effective focal length of 80 m will have an image
scale of 2.600 /mm (NB – image scale is sometimes given as the inverse of the
definition used here so that this figure would be expressed as 0.38 mm/00 ). So, if
through the use of atmospheric compensators it were able to operate at its visual
diffraction limit of 0.0100 (2.4), then the physical size of the image of a point
source (star) would be 26 mm. Now the physical size of individual detectors (pixels–
Chap. 9) in CCD arrays is typically 20–30 mm. Thus the diffraction limited image of
a star would be comparable in size with the detector elements. This would be the
optimum situation for a given telescope-detector combination since if the image
scale were to be smaller than the optimum value, then resolution would be lost,
while, if it were to be larger, then the field of view would be less than it could be.
The light grasp of a telescope used visually (2.2) becomes the light-gathering
power of a telescope when used non-visually, and this just depends upon the
collecting area of the objective. The difference between point and extended
sources is less clear-cut however than was the case for the eye. Many detectors
(Chap. 9) used in astronomy, at all wavelengths, count individual photons (Chap. 7).
It therefore matters little whether or not the image is concentrated onto one pixel or
spread over many3 – the number of photons counted will be the same. Thus an
extended source will have more photons gathered by a larger telescope than by a
smaller one, and if the two telescopes have the same focal lengths it will be detected
as being brighter in the larger one.
The field of view of a telescope used non-visually depends mainly upon the image
scale and the size of the detector to be used. Most major telescopes use mosaics of
CCD arrays (Chap. 9). The 8.2-m Subaru telescope’s Suprime-Cam for example
uses ten 2,048 4,096 CCD arrays arranged in a 2 5 mosaic and covering an area
about 125 155 mm. The telescope’s focal length at prime focus and using
corrector lenses is 16.4 m. The image scale is therefore 12.600 /mm and so the field
of view when using Suprime-Cam is about 270 340 .
3
The signal-to-noise ratio (Chap. 10), though, will be reduced when the image is spread over
several pixels compared with being concentrated onto one.

Q. Determine the field of view of the Schmidt camera used by the Kepler
spacecraft. It has an effective focal length of 1.4 m and uses a mosaic of forty-
two 1,0242,200 CCD arrays, each of which is 29.559 mm. The arrays are
grouped in pairs to form 5959 mm squares, and those small squares are then
arranged in the form of a larger 55 square that has had the four corner small
squares omitted. The gaps between the small squares are 11 mm across.
A. Ignoring the missing corners, the mosaic would be a square about with sides
339 mm long (¼ 559 mm + 411 mm). For a focal length of 1.4 m we have
(2.21)
206 206
IS ¼ ¼ ¼ 147 seconds of arc per millimetre
f 1:4
Thus 339 mm at the cameras’ focal plane corresponds to 49,80000 (¼ 13.8 )

on the sky. The field of view however is actually a circle 16.1 across (This is the
circle into which the square mosaic with its corners missing can just be fitted).
Aberrations
Those deviations of an image from perfection that are not due to diffraction
(or grossly poor optics) are known as aberrations. We have already encountered
the problems that early astronomers found with chromatic and spherical aberration,
but those are not the only aberrations. There are six primary (Seidel) aberrations:
• Spherical aberration
• Coma
• Astigmatism
• Distortion
• Field curvature
• Chromatic aberration.
All, except the last, affect both lens and mirror systems. Chromatic aberration
affects only lenses. There is not space for a detailed treatment of aberrations in this
book, so just a brief summary of their effects will be given.
In spherical aberration, as we have seen (Figs. 1.4 and 1.5), annuli of the lens or
mirror of differing radii have differing focal lengths, while in chromatic aberration
(Fig. 1.6), light rays of differing wavelengths have differing focal lengths. In both
cases, it is impossible to focus all the light rays simultaneously, and the image of a
point source will consist of a bright center surrounded by a halo of out-of-focus rays
(Fig. 2.10), these being colored in the case of chromatic aberration. As we have seen,
however (Fig. 1.12), combining two lenses of different glasses can produce
an achromatic lens in which the chromatic aberration is significantly reduced.
Aberrations 49
Fig. 2.10 Image of a point

source affected by spherical
or chromatic aberration
The condition for achromatism in a cemented achromat, with two wavelengths la

and lb coinciding, is given by
R1 þ R2
Dnc ¼ Dnf (2.22)
R1
where Rl is the radius of the first surface; R2 is the radius of the second and third
surfaces, which are in contact (Rl and R2 are simply positive numbers – the usual
sign convention of optics does not apply); Dnc is the difference in refractive indices
for the glass making up the first (usually crown glass) lens at wavelengths la and lb;
Dnf is the difference in refractive indices for the glass making up the second
(usually flint glass) lens at wavelengths la and lb.
Coma is an effect whereby images of off-axis objects from different annuli of the
lens or mirror are displaced by increasing amounts away from (or towards) the
optical axis, and consists of rings of increasing sizes. The resulting image is a fuzzy
blob of triangular shape pointing towards or away from the optical axis (Fig. 2.11),
and reminiscent of the image of a comet (hence the name).
Astigmatism is differing focal lengths for rays in the vertical plane compared
with rays in the horizontal plane (Fig. 2.12). A cylindrical lens thus produces
images with 100% astigmatism. The image of a point source from an ordinary
lens or mirror with astigmatism will vary from a vertical line, through a uniform
circle (known as the circle of least confusion, and which is the point of best focus),
to a horizontal line (Fig. 2.13). Distortion is differential transverse magnification for
different distances of the image away from the optical axis. Decreasing magnifica-
tion results in barrel distortion, increasing magnification in pin-cushion distortion
(Fig. 2.14). Finally, field curvature has been encountered earlier in the discussion of
Fig. 2.11 Image of a point

source affected by coma
Fig. 2.12 Astigmatism
the Schmidt camera (Chap. 1) and is caused because the focal “plane” is no longer a
plane, but a curved surface. A flat detector such as a single CCD array will therefore
not be in focus over its entire area. The large mosaics of CCD arrays (Chap. 9) used
on major instruments can, however, be curved to match the image surface.
Interferometers 51
Fig. 2.13 Image of a point source affected by astigmatism at various points along the optical axis
Fig. 2.14 Image of an extended source affected by distortion
Interferometers
Interferometers (Chap. 1) at both radio and optical wavelengths can provide much
higher resolutions than single telescopes. A simple interferometer, however, does
not produce a direct image, and aperture synthesis systems require a great deal of
data processing in order to do so. Nonetheless, the basis of the high resolution of
interferometers can be appreciated from considering the case of two close point
sources.
Fig. 2.15 An interferometer

observing two point sources
The output from a two-element radio interferometer as it tracks a single point

source across the sky is oscillatory (Fig. 1.25). The equivalent for an optical
interferometer is a series of fringes. If either type of interferometer should be
observing two point sources close together in the sky, then each source will give
rise to an oscillation or fringe pattern. When the path difference to the two elements
of the interferometer from one source differs from that for the other source by a
whole number of wavelengths, plus half a wavelength, a maximum from one fringe
pattern will coincide with a minimum from the other. Assuming the sources are of
equal brightness, the fringe pattern, or oscillations, will then disappear. When the
path difference to the two elements of the interferometer from one source differs
from that of the other source by exactly a whole number of wavelengths, the
maximum from one fringe pattern will coincide with the maximum for the other
and the minimum from one with the minimum for the other. The fringe pattern or
oscillations will then be doubled in intensity.
Thus when the two sources are perpendicular to the line of the interferometer, if
the difference between their path differences is exactly half a wavelength, then the
fringe pattern or oscillations will disappear, while for other positions in the sky it
will be present to a greater or lesser extent. In that situation, we may see from the
geometry in Fig. 2.15 that
ðl=2Þ
sin b ¼ (2.23)
S
or, since b is small and in radians
l
b (2.24)
2S
Mountings 53
This is defined as the resolution of the interferometer. Comparing it with (2.3)

we may see that the resolution of an interferometer betters that of a conventional
telescope with a diameter equal to the separation of the elements of the interferom-
eter by a factor of 2.44. The main improvement in resolution of an interferometer
over a conventional telescope, though, arises from the ease with which large
separations for the elements may be used.
The operation of an interferometer to measure the separation of two point
sources requires the separation of the elements to be changed. Since there may be
path differences for the radio or light beams that arise from, for example, delays in
connecting cables or atmospheric effects, the fringes or oscillations may not die
away when the two telescopes are placed side by side. In practice therefore it is the
difference between the separations of the two telescopes when the fringes or
oscillations are at their maximum and minimum values that determines ‘S.’ Putting
the value of that telescope separation into (2.24) then gives the separation of the
point sources along the line parallel to the axis of the interferometer.
A second interferometer perpendicular to the first can provide the separation
parallel to that axis, and so the actual separation of the objects in the sky can be
found. Alternatively an aperture synthesis system can produce the image directly
after suitable data processing. Filled aperture synthesis systems, though, have the
same resolution (2.3) as a conventional telescope but with a diameter equal to
the separation of the elements of the interferometer.
Mountings
The function of a telescope mounting is to support the optical components, point

them in the required direction and then to track the object as it moves. The mounting
may be divided into two components: the telescope tube (which is often nothing like
a tube), and the support for the tube (which we shall now call the mounting – see
Chap. 6).
The tube in the case of a refractor is usually exactly that, with the lenses,
supported at their edges, at the top and bottom of the tube. The main problems
are the strains set up in large lenses by their weight, about which little can be done,
and flexure in the lengthy tubes.
For reflectors, the mirror can be supported at the back, as well as around the
edges and, as previously discussed, this is one of the main reasons for the larger
sizes of reflectors over refractors. The mounts for primary mirrors are often active,
changing in response to the changing loads as the telescope orientation alters.
In telescopes constructed half a century ago, complex mechanical linkages would
adjust the mounting; nowadays, the adjustment is normally through computer-
controlled actuators (Fig. 2.16).
The optics and their immediate supports in a reflector are then also mounted in a
“tube” – the name is retained, even though the structure is often an open frame.
Many such tubes are based upon the Serrurier truss (Figs. 2.17 and 1.19) because
Fig. 2.16 Active supports

for a primary mirror
Fig. 2.17 Reflector “tube”

based upon Serrurier trusses
this flexes in a known manner. Both the primary and secondary mirror supports flex
as the telescope changes its orientation, but, if designed correctly, the flexures are
identical and so the optics remain in mutual alignment, even though the optical axis
moves within the tube. With CAD (computer aided design) programs it is now
possible to design and manufacture far more complex tubes than those based upon
Serrurier trusses and to ensure that flexure, etc., is controlled to within the
constraints required for the optimum performance of the optics.
There are two main types of mountings for telescopes, called equatorial and
alt-azimuth (alt-az) (Chap. 6). Examples are shown in Figs 1.16 and 1.17. Although
there are a number of different versions, the equatorial mounting always has one
axis, the polar axis, parallel to Earth’s rotational axis. This is so that a single,
constant-speed motor driving the telescope around this axis at one revolution per
sidereal day in the opposite direction to Earth’s rotation will suffice to provide
tracking (Fig. 2.18). The other axis, at right angles to the polar axis, is called the
declination axis. This system thus also has the advantage of giving simple direct
Mountings 55
Fig. 2.18 Equatorial

telescope mounting
Fig. 2.19 Alt-az

telescope mounting
readouts of hour angle (HA) or right ascension (RA) and declination (Dec) (Chap. 4).
The main disadvantages of the equatorial mounting are that it is relatively expensive
to construct, and the gravitational loads change in complex ways, making compen-
sation for flexure difficult.
The alt-az (which name derives from the allowed motions in altitude and azimuth)
mounting has axes in the horizontal and vertical planes (Fig. 2.19). This design is
much cheaper to build, and simplifies the change in gravitational loading to just the
vertical plane. In order to track an object in the sky, however, the telescope must be
driven in both axes at varying speeds. Nowadays, with cheap computers, this
requirement is no problem, and so many of the recently constructed major
instruments and also small commercially produced telescopes use alt-az mountings.
One remaining problem is that the image rotates as the telescope tracks; the detector
mounting must therefore also be rotated during any exposure, to produce sharp
images. For some objects the field rotation produced by an alt-az mounting will
briefly cease, and sharp images can then be obtained without the use of a rotating
detector. The conditions that are required for zero field rotation are discussed in
Chap. 5.
A camera with a moderate focal length lens, such as might be used to image
constellations, may be very easily mounted so that it can track objects in the sky.
The device is known as a barn door, Haig or Scotch mounting. It may simply and
cheaply be made by any DIY enthusiast. It comprises two flat boards that are joined
by a hinge along one of their edges. The hinge axis is aligned parallel to Earth’s
rotation axis (i.e., on the north celestial pole in the sky). The camera is mounted on
the upper board, and the tracking movement is produced by a bolt through the lower
board that bears on the upper board. The bolt is turned by hand at intervals of a few
seconds. The amount and rate of motion required for the bolt will depend upon its
thread angle and the distance from the hinge, and can be calculated or found by trial
and error.
Observatories and Observing Sites
Even small telescopes are better when inside a purpose-built observatory, which
provides protection from the wind when the instrument is in use and some shielding
from unwanted lights. Having an observatory can also enable the telescope to be
brought into action quickly so that advantage can be taken of brief clear spells. The
classic design of a hemispherical roof with an open slot, rotating on a circular wall,
has much to recommend it, but it needs to be built with precision if it is to function
well. Small observatories are often made as DIY jobs, and therefore can come in
many shapes and sizes. The design is only limited by the skill, imagination and
facilities available. Medium-sized observatories can also be home built, but may
also be bought from specialist suppliers. In the latter case, the cost is likely at least
to equal that of the telescope itself. The housing for a major telescope requires very
careful planning and design and is likely to form a significant fraction of the cost of
the whole installation.
There is often little choice over the observing site, the observer’s back garden
usually being the only possibility. Sometimes, however, it is possible to make a
small telescope completely transportable so that it may be taken by car or van to
a better site. Most major instruments constructed nowadays cluster together in a few
Observatories and Observing Sites 57
sites, where the observing conditions are best. The principal requirements for a
good site may be summarized as:
• Away from light pollution;
• Low dust content in the atmosphere, to reduce scattering;
• Low water content in the atmosphere, to facilitate infrared observations;
• Low diurnal temperature changes;
• Steady atmosphere;
• Height, to reduce the depth of the atmosphere above telescopes to a minimum;
• Political stability of the host country;
• Accessibility.
This long list limits suitable sites to a very few, such as Hawaii, La Silla, the
Canary Islands, and Kitt Peak. As some measure of the determination of astronomers
to get the best out of their instruments, recently telescopes have begun to be operated
from the Antarctic plateau, where all conditions except perhaps the last are fulfilled.
Exercises
2.1. William Herschel produced his ‘40-ft telescope’ in 1789. It was of the
Herschelian design and had a 1.26-m f9.7 primary mirror made from speculum
metal. One of the eyepieces that he used still survives and has a focal length of
1.5 mm.
(a) What magnification would the eyepiece have given on this telescope?
(b) Assuming a 60% reflectance for speculum metal, what would have been
the telescope’s light grasp?
2.2. A 150-mm f5 refractor is to be used both visually and with a CCD camera at its
prime focus. The detecting elements of the CCD chip are 20 mm in diameter.
The telescope’s optics are of diffraction-limited quality, and the telescope is to
be used from a site where it can reach this limit some of the time.
(a) What would be the minimum magnification of the telescope if no light
from it is to be wasted, and what eyepiece focal length would give this
magnification?
(b) What would be the minimum magnification for the telescope if it is to be
used visually at its diffraction-limited resolution, and what eyepiece focal
length would give this magnification?
(c) What would be the angular size of the diffraction-limited image of a star,
what would be its physical size when focused onto the CCD and what
would the telescope’s effective focal length need to be for the CCD to
work at the diffraction limit?
2.3. An exoplanet and its natural satellite belonging to a star 20 pc (5.12) away
from us are twins of Earth and the Moon in all respects. What separation would
be needed for the elements of a two-element interferometer operating at a 500-
nm wavelength in order to resolve the two objects? (The Moon’s mean
separation from Earth is 384,000 km).
Chapter 3
Modern Small Telescope Design
Introduction
The modern aspiring astronomer is faced with a bewildering choice of commercially

produced telescopes, including all the designs considered in the preceding chapter.
A quick search of a popular Internet supplier reveals over 500 individual telescopes
from over 50 different manufacturers. This popularity is undoubtedly a conse-
quence of the enormous interest that has grown in astronomy since the start of the
Space Age and with the mind-boggling discoveries of the past 50 years.
Obtaining a Telescope
Except for some of the very small instruments that are, unfortunately, often heavily
promoted in general mail order catalogs, camera shops and the like, the optical
quality of these commercially produced telescopes is almost uniformly excellent.
Although one product may be slightly better for some types of observations, or
more suited to the personal circumstances of the observer than another, most of
them will provide excellent observing opportunities. The same general praise
cannot be applied, however, to the mountings with which many of these telescopes
are provided, and those problems are covered in Chap. 6.
When the first and second editions of this book were published, the text said:
“The second problem associated with commercially produced instruments is that of
cost. A new 0.2 m Schmidt-Cassegrain telescope can cost around a quarter to half
the price of a small car. Other designs may be rather cheaper for the same aperture,
but will nonetheless still represent a substantial outlay.”
Although if you cannot afford it, the price of anything is a problem, the relative
price of telescopes has come down significantly in the last decade. At the time
of writing a new 0.2-m Schmidt-Cassegrain telescope including a tripod alt-az

60 3 Modern Small Telescope Design
mounting, GPS receiver to give the telescope’s location, semi-automatic alignment

and a database of the positions of 40,000 celestial objects can be had for about
15–20% of the price of a small car.
Very recently a completely robotic system has been put on the market by one
manufacturer. The telescope has an integrated camera that is used when the
telescope is turned on to image three parts of the sky automatically. These images
are compared with a database and are used to calculate the time and the telescope’s
position and alignment. The setting-up takes less than 5 min, thereafter any object
in the database may be found and tracked simply by selecting it or a sky tour chosen
that looks at the interesting objects currently visible in the sky from the observer’s
position and for the time of night. So far the largest telescope in this range is just
0.15 m, but the cost of this instrument is well under 10% of the price of a small car.
If purchasing a new instrument is out of the question, then there are alternative,
perhaps more affordable, ways to become the owner of a respectable astronomical
telescope. One solution is to look at the second-hand market. The popular astron-
omy journals (Appendix A) all have small classified sections, and telescopes can
usually be found in there. Typically, the second-hand price of an instrument will be
half its new price, and since telescopes have long lifetimes, this represents very
good value for the money. Indeed the ads often include phrases such as “hardly
used” or “only used twice,” perhaps because the owners had not read Chap. 8 of this
book first!
There are few problems in buying a second-hand telescope, unless it has clearly
been very severely treated. Mirrors can be re-aluminized, optics realigned, eyepieces
replaced or the whole instrument refurbished by the original manufacturers or their
agents at a relatively small cost. The main points to watch are therefore concerned
with the mounting and its drives: if these are worn or damaged then they can be very
difficult to correct, especially with older instruments which may no longer be in
production.
An alternative to second-hand instruments and a means of further reducing costs,
which is open to the reasonably skilled DIY enthusiast, is to purchase the optical
components of the telescope and its tube, and then to construct a mounting. This
may well be the preferred option anyway (Chap. 6) for telescopes with some of the
more flimsy commercially produced mountings. The popular astronomy journals
abound with articles giving detailed designs and constructional requirements for
mountings from people who have chosen this route. No two such mountings are the
same, of course, but the observer interested in such a project will find numerous
ideas adaptable to his or her circumstances in only a few editions of these
magazines.
The ultimate in astronomical DIY is to produce the optics for the telescope as well
as the mounting. Indeed, for some people telescope making becomes more important
than observing, and they produce highly sophisticated designs to far higher than
commercially available standards. There is no need to go to such lengths, however,
and a simple telescope can be made from scratch that will give really excellent
images. Assuming a certain degree of ingenuity on the part of the builder, and
depending upon the proportion of the materials that can be rescued from the
Making Your Own Telescope 61
scrap heap, a 0.2-m Newtonian telescope can be made for 5–20% of the cost of a
similarly sized commercially produced Schmidt-Cassegrain instrument. The author
made just such a telescope at the age of 14, working largely from his own bedroom!
The requirements of such a project are briefly outlined below, but the interested
reader is referred to more specialized literature for details (Appendix A).
Making Your Own Telescope
Although it is possible to make the optics for any design of telescope, in practice it
is only the Newtonian reflector that is likely to be attempted by the vast majority of
telescope makers, and this design should certainly be chosen for the first such
project. The Newtonian requires only two mirrors, and the secondary is flat and
available at low cost for purchasing. Only the primary mirror therefore needs to be
produced, involving just a single concave paraboloidal surface.
Generally speaking, when it comes to acquiring a telescope, the bigger the
better! Size in this case refers to aperture, not to length. However, the difficulties
in making your own mirror scale as about the diameter cubed (i.e. D D D - so
that a 0.5-metre mirror would be about sixteen times more difficult to produce than
a 0.2-metre mirror). The telescope maker should therefore be content to start with a
0.15 or 0.2 m mirror, even though larger sizes might be financially possible.
The basic processes required to produce a mirror have been discussed in Chap. 2.
For a 0.2-m mirror, the rough grinding would take about 10–20 h, depending upon
the depth required. It is therefore well worth getting the blank and tool diamond-
milled to the shape to start with, if this is possible. The cost of such pre-shaped
blanks is usually not much more than that of unshaped blanks.
For a small mirror such as this, Pyrex is adequate as a material for the mirror, and
there is no need to use the much more expensive very low expansion materials such
as Zerodur or ULE. Thereafter the builder can expect to have to go through eight or
so smoothing stages, at 2–3 h per stage, polishing requiring 5–20 h, and figuring
requiring from 5 h upwards. The total time required to produce a mirror is thus in
the region of 40–60 h, assuming that no mistakes are made. In addition, the
telescope maker will also have to make a Foucault or other type of tester
(Fig. 2.5), and probably a stand on which the mirror may be worked. The latter
stages of figuring are particularly time consuming, because the polishing process
heats the mirror blank, and so it has to be left to cool down each time before it may
be tested. Thus 5 or 10 min actual work on the blank may take over an hour to
accomplish.
The surface of the mirror should normally be within an eighth of a wavelength of
light (0.000 000 06 m or 60 nm) of the correct paraboloidal shape, if it is not to
degrade images beyond the normal diffraction limit (2.4), and most specialist books
on telescope making (Appendix A) go to great lengths to describe how this may be
achieved. However, for a first attempt, far poorer quality than this can be accepted.
Deviations of a wavelength or more will still yield reasonable images in the
telescope; the Hubble Space Telescope mirror, after all, is incorrect by two microns
(four wavelengths) and was still able to be used even before the COSTAR (Correc-
tive Optics Space Telescope Axial Replacement) and later correctors were added.
The best strategy for a first-time telescope maker is therefore to produce a mirror
of moderate quality, and to observe with it until its defects become a limitation, then
to refigure it to a higher accuracy. Alternatively, a second higher quality mirror can
be produced while the first is being used – something to do on cloudy evenings!
Once the mirror has been shaped, it must have its reflective coating applied.
A silver coating can be applied at home using some simple chemicals. Details of the
process may be found in a reasonable chemistry textbook, where it will be described
as the ‘silver mirror test.’ A silvered surface will, however, oxidize rapidly and will
need reapplying at intervals of a few weeks; but this may be acceptable if the mirror
is to be refigured after a short while, as suggested above. Normally, the reflective
coating is aluminum, and this is often over-coated with a protective layer of
transparent silicon oxide. The lifetime of such a coating, if the mirror is stored in
dry conditions, is several years. However, both the aluminum and the over-coating
have to be applied by evaporation onto the glass support inside a vacuum chamber,
and the process is therefore unsuitable for undertaking at home. The cost of having
a mirror professionally aluminised is not high, and suitable firms advertise in the
popular journals. Local or national astronomy societies may also have lists.
The secondary mirror for a Newtonian telescope is a flat mirror set at 45 to the
optical axis (Fig. 1.10). It is quite small and can be purchased fairly cheaply. Flat
mirrors can be produced, if wished, by a variation of the techniques used for the
primary mirror. However, unless a flat is already available, three mirrors have to be
produced at the same time, so that they may be tested against each other. For a first
try, with a less than perfect primary mirror, the flat can be cut from a piece of plate
glass (not ordinary window glass), and then silvered or aluminized. An elliptically
shaped secondary obstructs the least amount of light, but again, for a first try, a
rectangular one (much easier to cut) will only cause the loss of a small additional
amount of light.
The mirrors, once produced, have to be supported in their correct relative
positions and orientations within the telescope tube. The supports should be firm
enough to prevent movement of the mirrors, and yet not apply sufficient pressure to
cause stresses, which might distort the surfaces. For a 0.2-m primary, a mirror cell
such as that shown in Fig. 3.1 will be adequate. The weight of the mirror is taken on
the bottom and side supports, and these may be adjusted to align the mirrors. The
top restraints just hold the mirror in place, for those occasions when the telescope is
used at large zenith distances. A small secondary mirror may have a similar cell or
be more simply attached to the supporting arms (often called the spider) using glue.
The lenses for eyepieces can be made if the telescope maker has access to a
lathe; but several identical lenses have to be produced at the same time, and
specialized equipment is needed, so that eyepieces are generally best purchased.
Initial costs can be minimized by searching the Internet and antique stores, where
eyepieces from old microscopes and binoculars are often to be found. The eyepiece
mount needs to hold the eyepiece, to allow interchange of eyepieces, and to enable
Brief Survey of Commercially Produced Telescopes 63
Fig. 3.1 Primary mirror

cell for a small mirror
the eyepiece position to be moved smoothly in and out for focusing. A good
machinist can construct such a mount, but otherwise it will need to be purchased.
Costs, however, can again be reduced by searching the second-hand market.
The mirrors in their cells or mounts and the eyepiece holder need supporting, and
this is conventionally done by means of a tube. For a small telescope, plastic tubes
can often be found of sufficient size and rigidity for this purpose. Alternatively, the
tube can be constructed from wood or metal. A tube with a square cross-section is
quite adequate and is much easier to construct than one with a circular cross-section.
Open tubes (Fig. 2.17) can be used but have few advantages for small telescopes.
Once in the tube, the mirrors need aligning correctly, a process known as
collimating the telescope. The eyepiece should be removed from its mount and
the mirror positions and orientations adjusted until they have the appearance shown
in Fig. 3.2 when seen through the empty eyepiece mount, with all the components
and their reflections concentric.
Finally, the tube and its optics need supporting on a mounting to enable them to
be pointed at the sky and to follow objects as they move across the sky. Details of
such mountings are to be found in Chaps. 1, 2, and 6.
Brief Survey of Commercially Produced Telescopes
Even the briefest of perusals of one of the popular astronomy magazines will reveal
a bewildering array of advertisements for commercially produced telescopes.
The optical qualities of these instruments are generally excellent, and the choice
Fig. 3.2 View through the

eyepiece holder of a correctly
collimated Newtonian
telescope: (a) eye reflected in
the primary and the secondary
(twice); (b) eyepiece holder
reflected in the primary and
the secondary (twice);
(c) secondary reflected
in the primary and the
secondary; (d) primary
reflected in the secondary;
(e) secondary seen directly;
(f) eyepiece holder seen
directly; (g) spider reflected
in the primary and the
secondary
usually depends upon the available funds and the purpose for which the telescope is
to be used. It goes almost without saying that anyone intending to purchase a
telescope should not only read all the available literature carefully but should also
try out the instrument, preferably by finding someone else (perhaps in a local
astronomy club) with the same telescope.
We may divide the vast majority of commercially available telescopes into four
main groupings: Newtonian reflectors, refractors, Schmidt-Cassegrains and
Maksutovs, and Dobsonian telescopes. Smallish Newtonian reflectors on equatorial
mountings are probably the cheapest way to acquire a brand-new telescope, and
they will perform well, provided that they are kept collimated properly. Their
main disadvantages are the ease with which the optics can become misaligned,
their size and the general awkwardness of viewing with the eyepiece high on the
side of the tube.
Modern refractors, Schmidt-Cassegrains and Maksutovs, are very comparable in
their performances. They all provide highly corrected images and have closed tubes
with rigidly mounted optics. The refractors, for a given aperture, are generally
lengthier, and therefore more awkward to transport, but have the advantages of not
needing a secondary mirror to block some of the light and to (slightly – see Fig. 8.4
in Chap. 8) reduce the image quality. If light grasp is your main consideration, then
large aperture Newtonians on Dobsonian mountings are the best bet. These have the
usual disadvantages of the Newtonian, coupled with the disadvantages of an alt-az
mounting, but for the same money will provide two to three times the aperture
(four to nine times the light grasp) of a refractor or Schmidt-Cassegrain, etc. It is
possible to attach a Dobsonian-mounted telescope to a platform that can then be
driven to track the stars for a short interval (Fig. 3.3). Cassegrain and Ritchey-
Chrétien telescopes are also available commercially, but generally only to special
order and in quite large sizes, and are usually more suited to educational or small
research applications.
Binoculars 65
Fig. 3.3 An equatorial

platform for a Dobsonian
telescope
Binoculars
Binoculars are just a pair of matched telescopes held together within a single
framework that allows them to be pointed at and to focus on an object simulta-
neously. Both eyes may be thus be used for observing. Many people have difficulty
in closing one eye, and so find binoculars easier to use than a single telescope.
Binoculars provide an upright image through the use of internal prisms that also
serve to shorten the length of the instrument.
The performance of binoculars is usually specified using two numbers, such as
7 30 or 10 50, etc. Here the first number is the magnification (i.e., 7 or 10),
and the second the diameter of the objective lenses in millimeters (i.e., 30 or
50 mm). Recently gyro-stabilized binoculars have appeared on the market. These
are much easier to use, since minor tremors are eliminated.
Aspiring astronomers are often advised to purchase binoculars instead of a
telescope. While it is true that good quality binoculars can be purchased for the
same cost as a cheap telescope, they are likely to disappoint if used for astronomical
observing. This is because hand-held binoculars have too low a magnification to
show very much in the sky. Even a magnification of 10, which is about the highest
that can easily be hand-held, will not show lunar craters or the rings of Saturn.
Higher magnifications are available, but can only be used if the binoculars are
stabilized, or on some type of mounting. Then, however, they become as expensive
as a good telescope.
Binoculars are also often advertised as “night vision glasses,” implying that they
provide a brighter image than that seen by the naked eye. However, as we have seen
in Chap. 2 (2.16, etc.), this can never be the case for extended objects. Stars, though,
will appear brighter through binoculars just as they do in telescopes.
Part II
Positions and Motions
Chapter 4
Positions in the Sky
Spherical Polar Coordinates
Most people are familiar with the idea of plotting a graph. This is one example of a
coordinate system, the x and y coordinates (abscissa and ordinate) providing a
means of specifying the position of a point within the two-dimensional surface
occupied by the graph. It is a simple extension of the idea to give completely the
position of an object in space using three coordinates, x, y and z (Fig. 4.1).
In astronomy there is frequently the need to specify the position of an object;
however, neither the two or three-dimensional Cartesian coordinate systems above
are convenient. Instead, the three-dimensional position in space of an object is
specified in a spherical polar coordinate system. This gives the position of an object,
P, with respect to a point in space (the center of the sphere) and a reference direction
and plane, and results in coordinates, R, y, f (instead of x, y, z), where y and f are
not linear distances, but angles (Fig. 4.2).
A familiar example of the use of spherical polar coordinates is to give positions on
Earth. Here, however, we normally only specify two of the three coordinates because
the radial distance is almost constant at about 6,370 km. Thus only the latitude and
longitude need be given in order to fix a point on the surface of Earth. Latitude is
measured in degrees north or south of the equator, from 0 at the equator to 90 at
the poles. Other systems of spherical polar coordinates may go from 0 to 180
(i.e., from South Pole to North Pole) for this coordinate. Longitude is measured in
degrees from 0 to 180 east or west of the Greenwich meridian (Fig. 4.3). In other
systems, the equivalent coordinate may go from 0 to 360 . For the latitude and
longitude position coordinate system, the reference plane is Earth’s equator, and the
reference direction, the direction along the equator through the Greenwich meridian.
Thus the old observatory at Greenwich is at a latitude of 51 280 N, and a longitude of
0 , while Mount Palomar Observatory is at a latitude of 33 210 N and a longitude
of 116 520 W.

70 4 Positions in the Sky
Fig. 4.1 Three-dimensional

coordinate system based on
orthogonal axes
Fig. 4.2 Spherical

polar coordinates
Fig. 4.3 Latitude

and longitude
Celestial Sphere 71
Celestial Sphere
Unlike objects on the surface of Earth, those in the sky are distributed through-
out the whole of three-dimensional space. Nonetheless we may still use a system
of spherical polar coordinates analogous to latitude and longitude. We do this by
imagining a huge sphere, centered on Earth and large enough to contain every
object in the universe. We then project an object in three-dimensional space on
to the surface of that sphere. Then just the two angular coordinates give its
position (Fig. 4.4). In other words, we ignore the radial coordinate of an object
when it comes to assigning it a position in the sky. This accords with our
common sense view of looking at the sky, where we are concerned with
where to point a telescope and the object’s distance is immaterial for this
purpose.
Any material object, such as a star, planet, galaxy, Sun, Moon, gas cloud, etc.,
can thus have its position in the sky represented by its projected position on the
celestial sphere. The same, however, may also be done for less material objects. The
north and south celestial poles (NCP and SCP, the points in the sky about which
everything appears to rotate because of Earth’s counter-rotation), for example, can
easily be added. These are the points where Earth’s rotational axis would intersect
the celestial sphere if it were extended far enough. Thus we get the positions of the
north and south celestial poles in the sky (Fig. 4.5). Most of the time, however, we
can drop the qualifications “position of” and “celestial,” and just talk about the
north and south poles, etc., without any ambiguity. There is rarely likely to be any
confusion with the real objects, and so we shall generally adopt this custom from
now on.
We may also add other items in a similar fashion. Thus we have the celestial
equator. This we may obtain by imagining a line extending from the center of Earth
Fig. 4.4 Celestial sphere

Fig. 4.5 Celestial equator
in the plane of Earth’s equator. If that line is swept around, it will mark a great
circle1 on the celestial sphere, which is where the plane of Earth’s equator meets
the celestial sphere (Fig. 4.5). Naturally this is perpendicular to the line joining the
poles. The zenith and nadir (the points in the sky directly overhead and directly
underfoot) for a particular observer can also be added.
It might appear that we may plot in the horizon for that observer, in a similar
manner to the way in which the celestial equator was obtained. Here, though, we
normally deviate from strict accuracy because the true horizon will be affected by
buildings, trees, hills, etc., and will also not be a plane because the observer’s eyes
are above ground level (Fig. 4.6). Thus the celestial horizon is actually defined by
the plane through the center of Earth perpendicular to the line joining the zenith and
nadir (Fig. 4.7). Conventionally, the horizon is shown horizontally, and the poles,
zenith and nadir are shown on the outlining circle of the celestial sphere. Even
though this is incorrect perspective, it simplifies geometrical calculations (Fig. 4.8).
The compass (cardinal) points on the horizon are called the north point
(to distinguish it from the North Pole), the east point, the south point (to distinguish
it from the South Pole) and the west point (Fig. 4.8). Magnetic North (which is
actually a magnetic South Pole!) is displaced from true north by an amount that
varies with place and time. The angle between the direction of true north
and magnetic north is called the magnetic variation or magnetic declination.2
1
The term “great circle” is used to denote any circle drawn on a sphere that divides that sphere into
two exactly equal hemispheres. The name originates because great circles are the largest circles
that can be drawn onto a sphere. The distance along a great circle that joins two points on the
surface of the sphere is the shortest distance between those two points when moving over
the surface of the sphere (hence the importance of great circle navigation when moving large
distances over the surface of a nearly spherical Earth).
2
The magnetic deviation is the angle between the north needle of a compass and magnetic north.
It arises from local magnetic fields such as those from the magnetized steel making up a ship.
Fig. 4.6 True horizon
Fig. 4.7 Celestial horizon
Fig. 4.8 Conventional

representation of Fig. 4.7
Fig. 4.9 The ecliptic
The magnetic north pole is currently about 400 km off the northern coast of Ellesmere
Island and moving at about 50 km/year NW, while the magnetic south pole is about
125 km off the coast of Adélie Island and moving at about 15 km/year NW. The
first line of zero magnetic variation runs just past the western edge of Hudson’s
Bay, then close to New Orleans, Guatemala city and then south of Buenos Aires.
The second line has a far more complex route, skirting the western coast of
Norway, passing near Barcelona, Algiers and Nairobi before looping north again
across India and China into Siberia then returning south across the western Pacific,
past Singapore and along the western edge of Australia. To the East of the first line
and as far as the second line (i.e., roughly the North and South Atlantic and their
borders) the variation is negative, so that the direction of magnetic North is to the
west of true north, while over much of Europe and Asia the variation is positive
(the direction of magnetic north is to the east of true north). Representative values
of the variation in 2010 were about
Beijing 6
Buenos Aires 8
Canberra 12
Chicago 4
Johannesburg 10
London 2
Los Angeles 14
New Delhi 0
New York 16
Paris 0
We can also plot out the yearly path of the Sun across the sky, which is known as
the ecliptic, and along which lie the zodiacal constellations (Fig. 4.9). The zodiac
covers a band of the sky about 20 wide, centered on the ecliptic, within which are
Altitude and Azimuth 75
Table 4.1 The zodiacal Constellation Dates of solar passage

constellations
Sagittarius 19 Dec to 21 Jan
Capricornus 22 Jan to 16 Feb
Aquarius 17 Feb to 12 Mar
Pisces 13 Mar to 8 Apr
Aries 19 Apr to 14 May
Taurus 15 May to 21 June
Gemini 22 June to 21 July
Cancer 22 July to 11 Aug
Leo 12 Aug to 17 Sep
Virgo 18 Sep to 31 Oct
Libra 1 Nov to 22 Nov
Scorpius 23 Nov to 30 Nov
Ophiuchus 1 Dec to 18 Dec
found the Sun, Moon and planets. Traditionally there are 12 constellations to be
found within the zodiac, but the modern constellation boundaries place a thirteenth,
Ophiuchus, into the region as well. The full list of zodiacal constellations is given in
Table 4.1.
The conventional 12 zodiacal constellations are used in astrology to label twelve
30 -wide bands (or signs) of the zodiac. However, precession (Chap. 5) means that
these now have little correlation with the actual constellations. Thus, for example,
the Sun is in the zodiacal sign of Sagittarius from November 22 to December 21, an
overlap of only 2 days with the true times that the Sun is within the constellation.
The motion of the Sun across the sky is of course actually due to Earth’s motion
around its orbit. The ecliptic is therefore also the plane of Earth’s orbit marked on to
the celestial sphere. We can imagine this being drawn by taking the line from the Sun
to Earth and extending it to the celestial sphere, and then making a mark as Earth
moves. More formally we may say that the ecliptic marks the intersection of the
plane of Earth’s orbit extended outwards in all directions with the celestial sphere.
Altitude and Azimuth
We may now return to the problem of the positions of objects in the sky. The first of
several spherical polar coordinate systems (with R ignored) that we shall examine is
known as altitude and azimuth3 (abbreviated to alt-az). This system uses the horizon
as the reference plane, and the north point as the reference direction. Altitude is then
measured from 0 on the horizon to 90 at the zenith. Zenith distance is also used at
3
Occasionally azimuth may be referred to as the bearing, especially for objects on the surface of
Earth.
Fig. 4.10 Altitude and

azimuth coordinate system
times and is just (90 – altitude). The azimuth is the angle around the horizon from
the north point to the great circle from the zenith passing through the object. It is
measured from 0 to 180 east or west (Fig. 4.10) or from 0 to 360 going from the
north point through east, south, west and back to north. Since the horizon and zenith
are individual to each place, clearly the altitude and azimuth of the same object will,
in general, be different as seen from different places. More importantly, the altitude
and azimuth will change with time as Earth rotates, and so cannot be used to identify
a specific object in the sky except at a specific time.
Rotation
The change in altitude and azimuth of an object in the sky as seen by a particular
observer is due to the movement of that object in the sky. Of course, this motion is
actually due to the rotation of Earth. For the discussions in this chapter, however, it is
convenient to take the geocentric viewpoint and simply talk about the rotation of the sky
or celestial sphere, while always remembering that it is actually Earth that is rotating.
Thus the celestial sphere rotates on an axis running through the North and South
poles once every 23 h 56 min 4 s. This period is shorter than the day because of Earth’s
orbital motion. In the time that it takes Earth to rotate through 360 , it has moved nearly
a degree around its orbit. To return the Sun to its starting point as seen in the sky (the
definition of the day), Earth has to rotate that further fraction of a degree (Fig. 4.11).
Solar and Sidereal Days
The period of 24 h is called the mean solar day, since it is based on the average time
required to return the Sun to its starting place in the sky. The period of 23 h 56 min 4 s
is called the sidereal day and is based on the time required to return a fixed object
Declination and Hour Angle 77
Fig. 4.11 Rotation of Earth

(much exaggerated orbital
motion)
(star, galaxy, etc.) to its starting place in the sky. A clock that goes through 24 h on its
dial in 23 h 56 min 4 s is called a sidereal clock, and it registers sidereal time. Two
(24 h) clocks, one keeping mean solar time and the other sidereal time, would agree
with each other at the autumnal equinox (usually about September 21, see later), and
thereafter the sidereal clock would gain about 4 min a day. By the winter solstice
(December 21), the sidereal clock would be 6 h ahead, by the spring (or vernal)
equinox (March 21) 12 h ahead, by the summer solstice (June 21) 18 h ahead, and by
the next vernal equinox, the sidereal clock would have gained 24 h and would again
be in agreement with the mean solar clock. There are thus 366.25 sidereal days in a
year, compared with 365.25 solar days.
Declination and Hour Angle
With altitude and azimuth as measures of the positions of objects in the sky, we
found that they changed both with time and with the position of the observer, and
that the manner of the change was quite complex. Using the above definition of the
sidereal day, we may now look at a new system of coordinates in which the changes
occur in a simpler manner. This second system of coordinates is based upon the use
of the celestial equator as the reference plane. The reference direction is obtained
from the line in the plane of the equator that also goes through the prime meridian
(Fig. 4.12). The prime meridian is individual to each observer and is the great circle
through the poles and the zenith (Fig. 4.12).
The two angular coordinates are thus rather like latitude and longitude: up or
down from the equator (declination, dec), and around from the prime meridian
(hour angle, HA). This set of coordinates has the advantage over altitude and
azimuth in that one coordinate for an object in the sky (declination) is now fixed,
and does not change with position on Earth, or with time. The other coordinate
(hour angle), however, does still change, but in a simple manner compared with the
Fig. 4.12 Declination

and hour angle
Fig. 4.13 Hour angle of 0 h
changes in altitude and azimuth. Like latitude, declination is the angle from 0 to
90 north or south of the equator (usually indicated through the use of “þ” for north,
and “” for south).
The hour angle, though the equivalent of longitude, is different from it in two
important ways. First it is measured only westwards from the prime meridian rather
than east or west from the Greenwich meridian. Second, the angular measure used
is hours, minutes and seconds of time, not degrees, minutes and seconds of arc (for
reasons that will shortly become apparent), with
1 h ¼ 15
1 min ¼ 150
1 s ¼ 1500 .
Declination and Hour Angle 79
Fig. 4.14 Hour angle of 2 h
Fig. 4.15 Effect of

longitude on hour angle
Hour angle thus goes from 0 to 24 h. For example, an hour angle of 6 h 45 min
20 s corresponds to the more normal angular measure of 101 200 , etc.
The value of the hour angle changes by increasing uniformly with time as the sky
rotates. The HA is 0 h when the object is on the prime meridian (Fig. 4.13). Then,
provided that we measure time in sidereal units, 1 h later, the HA will be 1 h. Two
sidereal hours after meridian passage, the HA will be 2 h (Fig. 4.14), and so on.
Thus the reason for the units used for HA is through its direct relationship to the
sidereal time elapsed since the meridian passage of the object.
The use of h, m, and s to measure HA facilitates the calculation of the effect of
time on its value. However, it complicates slightly the calculation of the effect
of changing longitude, because the latter is measures in degrees, minutes and
seconds of arc, and East and West of the Greenwich meridian. Clearly if an object
has an HA of 1 h from one place, however, it will be on the prime meridian at the
same instant for a second place that is 15 to the West (¼ 1 h) of the first place
(Fig. 4.15). The relationship between HA and longitude is thus given by
HA2 ¼ HA1 DLong(W) (4.1)
¼ HA1 þ DLong(E) (4.2)
where HAl is the HA from point one; HA2 is the HA from point two at the same
moment of time; DLong (W) is the difference in the longitudes between the two
points, when the second point is west of the first, measured in h, m, s, etc.; DLong
(E) is the difference in the longitudes between the two points, when the second
point is east of the first, measured in h, m, s, etc.

Q. A distant galaxy is observed to be on the prime meridian from the Mount
Palomar Observatory. What is its hour angle at the same instant from the Mauna
Kea Observatory on Hawaii?
Data:
Mount Palomar Observatory longitude 117 W
Mauna Kea Observatory longitude 155 W
A. From the question the galaxy’s hour angle from Mount Palomar (HA1) is
00 h 00 min.
Mauna Kea is to the West of Mount Palomar so that we have
DLong(W) ¼ 155 117 ¼ 38 ¼ 2 h 32 min
Thus from (4.1)
Hour angle from Mauna Kea ¼ HA2 ¼ 00 h 00 min 2 h 32 m ¼ 2 h
32 min ¼ 21 h 28 min
Time
Having defined hour angle, we may now be more precise about our definitions
of time.
Mean Solar Time
Mean solar time is defined as the hour angle of the mean Sun plus 12 h. The extra
12 h is so that we get the conventional usage of 0 h being midnight, when the
HAMS (hour angle of the mean Sun) is actually 12 h. The mean Sun is an imaginary
body that moves around the equator (not the ecliptic) at a constant velocity that
takes it through 360 in exactly 1 year. Mean solar time is thus a uniform measure of
time, also known as civil time (see below for a more precise definition) and is what
we customarily use in normal life (except when it is adjusted for summer time, etc.).
Time 81
For convenience civil time is taken to have the same value over a range (usually
15 ) of longitudes that are known as time zones. Mean solar time will agree exactly
with civil time only at the center of a time zone. Civil time will be 30 min ahead
(faster) than mean solar time for inhabitants on the western edge of a time zone and
30 min behind (slower) on the eastern edge of the time zone. On the equator
15 time zones are 1,670 km wide so that moving 28 km west (east) across the
zone delays (advances) the mean solar time by an additional minute compared with
the civil time. At latitudes of 45 a linear movement of just 20 km west or east
suffices for the gain or loss of a minute by civil time compared with mean solar time
and at latitudes of 60 a movement of only 14 km is needed.4 Similarly moving 1
of longitude west (east) across the zone delays (advances) the mean solar time by an
additional 4 min compared with the civil time.
Solar Time
Solar time is time by the actual Sun (as given by a sundial, etc.) and thus generally
differs from mean solar time because (a) the actual Sun moves around the ecliptic
(not the equator), and (b) the actual Sun moves at a non-uniform velocity because
Earth’s orbit is elliptical, and thus Earth’s orbital motion, which produces the
annual motion of the Sun across the sky, is non-uniform.
The difference between mean solar time and solar time is known as the equation
of time (E):
E ¼ Solar time Mean solar time: (4.3)
The convention of adding 12 h also applies to solar time, and so this does not
contribute to the difference. The value of E can be up to 16 min (Fig. 4.16). The
position of the Sun in the sky at civil midday (or any other fixed civil time, ignoring
summer time) thus varies either side of the meridian. This variation, when com-
bined with the Sun’s motion in declination, results in a figure eight shape known as
the analemma, often to be found on antique maps and globes (Fig. 4.17).

Q. Calculate the time shown by a (correctly aligned) sundial at the Anglo-
Australian Observatory (AAO) at midday local civil time on March 21
(the vernal equinox) to within 1 min.
(continued)
4
Thus at latitudes of 60 , an aircraft moving west at 14 km every minute (840 km/h, 525 mph)
would cause the mean Sun, and so also to a good approximation for a short while, the true Sun, to
appear stationary in the sky. At a latitude of 89.9 you would be able walk westward fast enough
to get the same effect.
(continued)
AAO data: Longitude 149 E

Time zone – 10 h ahead of GMT.
A. The center of the local time zone is at 10 15 ¼ longitude 150 E. The
AAO is at longitude 149 E and so is 1 to the west of the center of the time zone.
From the earlier discussion therefore mean solar time is delayed with respect to
civil time by 4 min. At midday within the time zone, the actual (local) mean
solar time at the AAO is therefore
Local mean solar time ¼ 12 h 00 min – 04 m ¼ 11 h 56 min
From Fig. 4.16 on March 21, Equation of Time ¼ 7 min
So from (4.3):
0 h 7 min ¼ Local solar time 11 h 56 min
Giving
Local solar time ¼ Sundial time ¼ 11 h 49 min
So comparing the sundial with your watch, you would find the sundial to be
11 min ‘slow.’
Fig. 4.16 Equation of time.

The difference between time
given by a sundial and by a
domestic clock
Civil Time
Sadly, astronomy’s traditional role as guardian of the world’s time keeping has now
been usurped by atomic physics. The current unit of time is the second, which has
been defined since 1967 as “the duration of 9 192 631 770 periods of the radiation
corresponding to the transition between the two hyperfine levels of the fundamental
state of the atom caesium 133”. The average results from a large number of atomic
clocks around the world are used to provide “International Atomic Time” (TAI),
which is used as the starting point for all other types of time scales.
Time 83
30 Solar
Declination
(degrees)
July June
20
August
May
10
September
Equation
April Of
Time
(minutes)
0
-20 -10 0 10 20
October
March
-10
November
February
-20
December
January
-30
Fig. 4.17 The analemma. The variation of the equation of time over a year plotted against the
changing declination of the Sun
On this basis, mean solar time as defined above is now known as Universal Time
(UT), or Greenwich Mean Time (GMT). However, UT is still based upon Earth’s
rotation and is therefore affected by changes in the rotation of Earth such as the
Chandler wobble.5 The UT corrected for the Chandler wobble is called UT1 and is
5
A small irregular motion of Earth’s geographical poles, probably arising from movement of
material deep inside Earth.
the basis of civil time keeping. It is kept to within 0.9 s of TAI by the occasional
insertion or removal of a leap second.
At the time of the writing of the third edition there is some debate about
removing the link between atomic time and the Earth’s rotation completely. The
outcome of that debate remains to be seen, but in immediate practical terms it will
make little difference to most people. Of course over centuries the disparity
between Earth’s rotation and civil time could become large, but doubtless that
problem will be addressed in some ad hoc fashion when needed.
Sidereal Time
Just as mean solar time was precisely defined via the hour angle of a particular point in
the sky (determined by the imaginary mean Sun), so sidereal time can be defined in
terms of the hour angle of a point in the sky. Any fixed point on the celestial sphere
could be chosen to define sidereal time. The point actually chosen is known as the first
point of Aries (FPA), or the vernal equinox. It is the point at which the equator and
ecliptic intersect, and at which the Sun in its yearly motion passes from the southern
hemisphere to the northern (Figs. 4.17, 4.18 and 4.19). Unlike the mean Sun, the FPA is
fixed with respect to the sky (but see precession, Chap. 5), and so, as we have seen, the
sidereal day is shorter by about 4 min than the mean solar day. The addition of 12 h to
solar time to give 0 h at midnight means that sidereal and mean solar times agree at the
autumnal equinox (about September 21), instead of at the vernal equinox (about March
21), when the HAMS and HAFPA (hour angle of the first point of Aries) are the same.
Just as was the case for mean solar time, sidereal time varies with longitude.
From (4.1) and (4.2), we may see that the sidereal time at a particular place (local
sidereal time, LST) is given by
LST2 ¼ LST1 DLong(W) ðwhen point 0 20 is to the West of point 0 10 Þ (4.4)
¼ LST1 þ DLong(E) ðwhen point 0 20 is to the East of point 0 10 Þ (4.5)
where LST1 is the local sidereal time from point 1 and LST2 is the local sidereal
time from point 2 and DLong is the difference between the longitudes of points ‘1’
and ‘2’ expressed in units of hours, minutes and seconds of time.
For sidereal time the main reference time (just as with mean solar time) is the
LST at Greenwich, known as Greenwich Sidereal Time (GST). Equations (4.4) and
(4.5) thus become
LST ¼ GST DLong(W) ðfor points west of the Greenwich meridianÞ (4.6)
¼ GST þ DLong(E) ðfor points east of the Greenwich meridianÞ (4.7)
and GST is tabulated for midnight (GMT) in the Astronomical Almanac (Appendix A)
for every night of the year.
Time 85
Fig. 4.18 Analemma above the Temple of Apollo (Corinth). This image is the result of multiple
exposures of the Sun onto a single negative obtained at the same local civil time (9 a.m.) on 47 days
over a 348-day period in 2003. The north–south and east–west displacements of the Sun that lead to
the slanting orientation of the figure eight both result from the Sun’s changing declination (from
23 260 N at the summer solstice to 23 260 S at the winter solstice and back again). The changing
declination itself results from the inclination of Earth’s rotational axis to its orbital plane (66 340 ).
The figure eight shape is the result of the differing values of the equation of time over the year.
(Reproduced by kind permission of A. Ayiomamitis – see http://www.perseus.gr for many more of
his gorgeous analemma and numerous other images)
The calculation to find the LST for a particular place and at a particular civil time
often seems to cause problems for the student, but is quite straightforward if it is
approached calmly and logically:
1. Calculate the local mean time, LMT(0), corresponding to the previous midnight
at Greenwich:
LMTð0Þ ¼ 24 h TðWÞ (4.8)
¼ 0 h þ TðEÞ (4.9)
Fig. 4.19 The definition

of sidereal time
where T(W) and T(E) are the time zone corrections from Greenwich to the zone
containing the place of interest, ignoring any summer time corrections, for places
to the west and east of Greenwich, respectively (e.g., Mount Palomar is 8 h west
of Greenwich so that T(W) ¼ 8, and LMT(0) ¼ 24 h 8 h ¼ 16 h ¼ 4 p.m.).
2. Calculate the sidereal time interval, DLST, corresponding to the solar time
interval, DLMT, from LMT(0) to the local mean time of interest, LMT(t).
To avoid a common source of error, you should note that the numerical value
of a sidereal time interval is always longer than that of a solar time interval. The
one is obtained from the other by multiplying by the ratio of the number of
sidereal days in a year to the number of solar days in a year:
½LMTðtÞ LMTð0Þ 366:25

DLST ¼ ¼ DLMT 1:0027379 (4.10)
365:25
3. Look up in a current copy of the Astronomical Almanac the GST corresponding

to the previous midnight, GST(0).
4. Calculate the GST at the local time of interest, GST(t), subtracting 24 from the
result if it exceeds 24 h:
GSTðtÞ ¼ GSTð0Þ þ DLST (4.11)
5. Correct for the difference in longitude from Greenwich (4.6 and 4.7), to obtain
the local sidereal time, LST(t), corresponding to the local mean time of interest.
(Subtract or add 24 if the time goes over 24 h or under 0 h.)
Time 87
Table 4.2 Local sidereal time at midnight civil time (corrected for any summer time changes)
LST (h) Date LST (h) Date LST (h) Date LST (h) Date
at midnight (1 at midnight (1 at midnight (1 at midnight (1
civil time day) civil time day) civil time day) civil time day)
0 (or 24) Sep 21st 6 Dec 22nd 12 Mar 22nd 18 June 22nd
0.5 Sep 29th 6.5 Dec 29th 12.5 Mar 30th 18.5 June 30th
1 Oct 7th 7 Jan 6th 13 Apr 7th 19 July 7th
1.5 Oct 14th 7.5 Jan 13th 13.5 Apr 15th 19.5 July 15th
2 Oct 22nd 8 Jan 21st 14 Apr 22nd 20 July 22nd
2.5 Oct 29th 8.5 Jan 28th 14.5 Apr 30th 20.5 July 30th
3 Nov 6th 9 Feb 5th 15 May 7th 21 Aug 7th
3.5 Nov 14th 9.5 Feb 13th 15.5 May 15th 21.5 Aug 14th
4 Nov 21st 10 Feb 20th 16 May 23rd 22 Aug 22nd
4.5 Nov 29th 10.5 Feb 28th 16.5 May 30th 22.5 Aug 29th
5 Dec 6th 11 Mar 7th 17 June 7th 23 Sep 6th
5.5 Dec 14th 11.5 Mar 15th 17.5 June 14th 23.5 Sep 14th

Q. Find the LST at 10.30 p.m. (LMT) on December 31, 2011, for the Mount
Palomar Observatory.
Data
Position 116 210 3000 W (longitude)
+33 210 22.400 (latitude)
Time zone 8 h behind Greenwich
A
(1) LMT(0) ¼ 24 h 8 h ¼ 16 h ¼ 4 p.m.
(2) DLMT ¼ 22 h 30 m 16 h ¼ 6 h 30 min ¼ 6.5 h
DLST ¼ 6.5 1.002 7379 ¼ 6.517 7054 h ¼ 6 h 31 min 4 s
(3) GST(0) ¼ 6 h 40 min 16 s (from the Astronomical Almanac for 2011)
(4) GST(t) ¼ 6 h 40 min 16 s + 6 h 31 min 4 s ¼ 13 h 11 min 20 s
(5) DLong(W) ¼ 116 210 3000 ¼ 7 h 45 min 26 s W
LST(t) ¼ 13 h 11 min 20 s 7 h 45 min 26 s ¼ 5 h 25 min 54 s
If only an approximate value of the local sidereal time is desired, then it can be
estimated quickly from Table 4.2. This lists the values of the local sidereal time
throughout the year at midnight local civil time. The dates listed are approximations
only, because of the width of time zones and the slight shuffle of days throughout the year
arising from the insertion of a leap day once every 4 years. Approximate local sidereal
times for other times and dates may be obtained by adding the elapsed time since the
previous midnight (the difference between sidereal hours and mean solar hours can be
ignored for rough calculations) and by interpolating between the dates listed in the table.
A correction for the observer’s position within a time zone can also be made if wished.

Q. What is the approximate local sidereal time at 8 p.m. on December 3rd (there
are no summertime changes involved)?
A. From Table 4.2. the LST is 4.5 h at midnight on November 29th and 5.0 h at
midnight on December 6th. Interpolating gives the LST at midnight on Decem-
ber 3rd as about 4 h 47 min.
8 p.m. is 20 h after midnight, so the LST at 8 p.m. is about 24 h 47 min
(¼ 00 h 47 min).
Right Ascension and Declination
With the definition of a fixed point in the sky, the first point of Aries, we may return
to the problem of the positions of objects in the sky, and now arrive at a much more
satisfactory solution. We continue to use declination, for as we have seen, this is
constant with time and position on Earth. We replace hour angle, however, with a
new coordinate, right ascension (RA), which is defined as the angle, measured in
hours, minutes and seconds of time, around the equator from the first point of Aries
in an easterly direction (Fig. 4.20). This is in the opposite sense to the direction of
Fig. 4.20 Right ascension

and declination
Right Ascension and Declination 89
Fig. 4.21 Relationship

between LST, RA and HA
measurement of hour angle. Since the first point of Aries is fixed in the sky and
rotates with it, the right ascension of another point fixed with respect to the sky is
constant with time. Thus RA and Dec form a system of spherical polar coordinates
for mapping the positions of objects in the sky. The equator is the reference plane
and the direction of the first point of Aries is the reference direction.
RA and Dec produce the commonly used coordinate system in astronomy, and
most objects have their positions tabulated in terms of these coordinates. For
example, we have the positions of two well-known stars:
Sirius (a CMa) RA 6 h 45 min 9 s Dec 16 420 5800

Betelgeuse (a Ori) RA 5 h 55 min 10 s Dec +7 240 2500
We may easily see (Fig. 4.21) that there is a relationship between RA and HA for
a specific object, the LST being given by
LST ¼ RA þ HA (4.12)
since LST is just the HA of the FPA.

This relationship (4.12) is fundamental to finding an object through a telescope.
Many telescopes are mounted on equatorial mounts (Chap. 2), which have one axis
parallel to the Earth’s rotation, the other then being at right angles to the first. This
system is chosen because it allows the telescope to be driven at a constant speed in
one axis (one revolution per sidereal day about the polar axis in the opposite sense
from the direction of Earth’s rotation) in order to counteract the rotation of the sky
and to track an object being viewed in the telescope (Fig. 2.18). It is therefore
relatively cheap and simple to construct. Other mountings, such as the alt-az
mounting (Fig. 2.19) may be used, but these, though simpler to construct, require
the telescope to be driven in both axes and at variable rates in order to track an
object. Only recently, with the availability of cheap computers, have such
mountings been used to any extent.
With an equatorial mounting, rotating the telescope around the polar axis changes
the HA at which the telescope is pointing without changing the declination. Simi-
larly, rotating the telescope around the declination axis changes the Dec at which the
telescope is pointing without changing the HA. Thus, if the axes are equipped with
angle-measuring devices (usually called setting circles – essentially very large 360
protractors, see Chap. 6), the telescope can be set directly to the declination of
the object. By going through the calculation of HA of the object (as above) for the
position of the observatory, for an instant say 10 min ahead of the current time,
the telescope can then be set to that HA, and the telescope drive turned on when the
correct time arrives. The object (given sufficiently accurate and well aligned axes
and setting circles, and no mistakes) will then be found centered in the field of view.
Thus can faint objects be found that cannot be seen, except in the main telescope.
Most observatories will have a clock giving LST to enable this procedure to be
carried out without having to go through the calculation every time. Many telescopes,
including all larger instruments, automate the process to a greater or lesser extent,
primarily by incorporating the LST into the telescope setting circles and read-outs to
enable RA to be set directly. This of course makes life very much simpler; indeed, it is
now quite common for even smallish telescope systems to contain catalogues of
positions so that you just have to type in the name (e.g., Canopus, M31, Saturn, etc.)
of the object you are interested in and the computer sets the telescope for you.
Other Coordinate Systems
There are two other coordinate systems used to give the positions of objects in the
sky, which may be encountered, and of which it is therefore worth being aware:
celestial latitude and longitude, and galactic latitude and longitude.
Celestial latitude and longitude use the first point of Aries as a reference
direction, and the ecliptic as a reference plane (compared with RA and Dec that
use the first point of Aries and the equator). The symbols l (celestial longitude) and
b (celestial latitude) are commonly used, and the units for both are degrees, minutes
of arc and seconds of arc.
Galactic latitude and longitude use the direction towards the center of our galaxy
(at RA 17 h 45 min 37 s, Dec 28 560 1000 , in Sagittarius6) as a reference direction,
and the plane of the galaxy (roughly the median line through the Milky Way) as the
reference plane. The symbols l (galactic longitude) and b (galactic latitude) are
commonly used and the units for both are degrees, minutes of arc and seconds of arc.
6
The actual center of the galaxy is now known to be about 40 away from this point, but this position
is still used for the galactic coordinates. Prior to 1958 a rather different definition of galactic
latitude and longitude was employed. For a while the old system was distinguished by the use of
“I” as a superscript and the new one by the use of “II” as a superscript (i.e. lI and bI or lII and bII).
Recently the superscripts have been dropped, since almost all positions now in use are based upon
the newer system.
Spherical Trigonometry 91
Heliocentric Time
For some purposes, such as observing variable stars, time as “seen” from the Sun is
used in order to eliminate the varying travel time for the starlight in crossing different
parts of Earth’s orbit. Any of the times discussed earlier may be converted to the
heliocentric equivalent. The correction from the terrestrial time, in seconds, is given by
500ðsin d sin dSun cos d cos dSun cosðaSun aÞÞ (4.13)
where a and d are the RA and Dec of the object being observed and aSun and dSun
are the RA and Dec of the Sun (obtainable from the Astronomical Almanac –
Appendix A) at the same moment.
Julian Date
Mention needs to be made of another convention used widely within astronomy for
representing the time. This is based upon the solar day and is called the Julian date.
It is used particularly when long time intervals are involved (for example when
measurements of a visual binary star are being used to calculate its orbit), since it
avoids the problems caused by the differing numbers of days in the months and of
leap years.
The Julian date is a running day number that provides a simple way of calculating
the time interval between two calendar dates. The Julian date starts at midday so that
there is no change of date throughout the night, and began at midday on January 1,
4713 B.C. on the Julian calendar (November 24, 4714 B.C. on the Gregorian calendar).
Midday on January 1, 2000, saw the start of JD 2451 545 (see Table 4.3). Times other
than midday are shown as decimal days. The Julian day number is the whole number
part of the Julian date. The heliocentric Julian date is the Julian date corrected for the
light travel time difference between Earth and the Sun. The running number of days
throughout the year is shown in Table 4.3.
There is also a variation of the Julian date that started at midnight on November
17, 1858, and called the modified Julian date (mJD). The modified Julian date is
thus the Julian date minus 2400 000.5 days. It is sometimes used for data on
spacecraft orbits.
Spherical Trigonometry
Many people will have encountered trigonometry in some form or other as a means
of dealing with geometry on flat or Euclidean surfaces. A triangle, for example, that
has its internal angles designated A, B and C and the lengths of the sides opposite to
those angles designated by a, b and c (Fig. 4.22) then has the following relationships
between its internal angles and the lengths of its sides:
Table 4.3 The Julian date Date (midday, 1 Jan,

Gregorian calendar) Julian date
2050 2469 808.0
2025 2460 677.0
2000 2451 545.0
1975 2442 414.0
1950 2433 283.0
1925 2424 152.0
1900 2415 021.0
1850 2396 759.0
1800 2378 497.0
1750 2360 234.0
1700 2341 972.0
1650 2323 710.0
1600 2305 448.0
1200 2159 351.0
800 2013 254.0
400 1867 157.0
0 1721 060.0
400 B.C. 1574 963.0
800 B.C. 1428 866.0
1200 B.C. 1282 769.0
1600 B.C. 1136 672.0
2000 B.C. 990 575.0
4000 B.C. 260 090.0
4714 B.C. (Nov 24) 0.0
Table 4.4 Running number Date (midday) Normal year Leap year
of days throughout the year
Feb 1 31 31
Mar 1 59 60
Apr 1 90 91
May 1 120 121
Jun 1 151 152
Jul 1 181 182
Aug 1 212 213
Sep 1 243 244
Oct 1 273 274
Nov 1 304 305
Dec 1 334 335
Jan 1 365 366
1. Sine rule
a b c
¼ ¼ (4.14)
sin A sin B sin C
Fig. 4.22 Euclidean triangle
2. Cosine rule
a2 ¼ b2 þ c2 2bc cos A (4.15)
Since it is arbitrary which angles are labeled A, B or C, these relationships can be

used to inter-relate any angles and side lengths.
However, when we are dealing with positions in the sky, for example trying to find
the angle between two objects or relating RA and Dec to altitude and azimuth, then we
are concerned with the geometry that takes place on a spherical surface. This is the
subject of spherical trigonometry, which is very similar to Euclidean trigonometry.
The two significant differences between Euclidean and spherical trigonometry are:
(a) The straight lines involved in the geometry occurring on a flat surface are
replaced by great circles, or segments of great circles. (Note that spherical
trigonometry does not deal with small circles.)
(b) The lengths of lines (i.e., segments of great circles) are given by the angle that
they subtend at the center of the sphere (Fig. 4.23), not by their actual linear
extents. The size of the sphere is thus not a concern of spherical trigonometry,
and its results apply just as much to figures drawn on the surface of a child’s ball
as to figures on the surface of Earth, or on the celestial sphere.
With these definitions, a triangle formed by the intersections of three great
circles on a sphere (Fig. 4.23), that has its internal angles designated A, B and C
and the angular lengths of the sides opposite to those angles designated by a, b
and c, then has the following four relationships between its internal angles and the
lengths of its sides. (Note that only three of the relationships are independent;
Fig. 4.23 Spherical triangle
the fourth can always be derived from the other three. Note also the similarity to the
Euclidean relationships for the first two equations.)
1. Sine rule
sin a sin b sin c
¼ ¼ (4.16)
sin A sin B sin C
2. Cosine rule
cos a ¼ cos b cos c þ sin b sin c cos A (4.17)
3. Four parts rule (since it involves four of the six angles and sides)
sin a sin C
cos a cos C ¼ (4.18)
tan b tan B
Fig. 4.24 Occultation
4. Five parts rule (since it involves all the quantities except one of the spherical
triangle’s internal angles):
sin a cos B ¼ cos b sin c sin b cos c cos A (4.19)
Thus, for example, in order to determine if a star will be occulted by the Moon at
their closest mutual approach, we must show that its angular distance from the
center of the Moon is less than the Moon’s angular radius. Given the RA and Dec of
the Moon and star as aM, dM, aS and dS, we may use the spherical triangle between
the celestial pole, the Moon and the star (Fig. 4.24) since lines of constant RA are
great circles. (Note that RA, HA, etc. must always be converted to degrees before
inserting into trigonometrical formulae). Hence
A ¼ aM aS
a ¼ separation of the star from the center of the Moon
b ¼ 90 dM
c ¼ 90 dS.
The cosine rule then gives us
cosðseparationÞ ¼ cosð90 dM Þ cosð90 dS Þ þ sinð90 dM Þ sinð90

dS Þ cosðaM aS Þ (4.20)
and for the actual values (where the star is Regulus, a Leonis)
aM ¼ 10 h 10 m 10 s ¼ 152 320 3000
dM ¼ þ12 000 0000
aS ¼ 10 h 08 m 30 s ¼ 152 070 3000
dS ¼ þ11 580 2000
we get
cosðseparationÞ ¼ cosð90 12 000 0000 Þ cosð90 11 580 2000 Þ

þ sinð90 12 000 0000 Þ sinð90 11 580 2000 Þ
cosð152 320 3000 152 070 3000 Þ
¼ 0:9999094 (4.21)
and so
separation ¼ 460 1700 : (4.22)
Since the lunar radius is about 150 , this particular appulse will thus not give rise
to an occultation. (An appulse is the instant when the separation of two moving
objects passing by each other is at its smallest. The term ‘conjunction’ is often used
for this moment, but strictly speaking conjunction is the instant when the two
objects have the same celestial longitude).

Q. Find formulae to convert the RA and Dec of an object into its celestial
latitude and longitude.
A. From the angles shown in the expanded view of the triangle formed by
the object in the sky, the North Pole (NP) and the pole of the ecliptic (Fig. 4.25)
and using the cosine rule we have
cosð90 bÞ ¼ cosðeÞ cosð90 dÞ þ sinðeÞ sinð90 dÞ cosð90 þ aÞ
giving
sin (b) ¼ cos (e) sin (d) sin (e) cos (d) sin (a)
or
Celestial latitude ¼ b ¼ sin1 {cos (e) sin (d) sin (e) cos (d) sin (a)}
Similarly using the four parts rule
(continued)
Ecliptic NP
Pole
Object
δ
β
λ
α
Equator
FPA
Ecliptic
NP
ε
90 + α
90 - λ
Ecliptic
Pole
90 - δ
90 - β
Object
Fig. 4.25 The relationship between RA and Dec (in blue) and celestial latitude and celestial
longitude (in red). e is the obliquity of the ecliptic (the angle between the planes of the equator and
ecliptic; 23 260 )
(continued)
cos (e) cos (90 + a) ¼ sin (e)/tan (90 d) sin (90 + a)//tan (90 l)
giving
cos (e) sin (a) ¼ sin (e)/cot (d) cos (a)/cot (l)
cos (a)/cot (l) ¼ sin (e)/cot (d) + cos (e) sin (a)
tan (l) ¼ [sin (e) tan (d) + cos (e) sin (a)]/cos (a)
or
Celestial longitude ¼ l ¼ tan1 {[sin (e) tan (d) + cos (e) sin (a)]/cos (a)}
Exercises
4.1. Convert the following angles into hours, minutes and seconds:
00 0
43 17 23
16 0 03
92 57 29
291 33 57
4.2. Convert the following angles into degrees ( ), minutes (0 ) and seconds (00 ) of
arc:
h m s
03 06 05
14 46 0
18 09 28
21 13 13
4.3. What is the hour angle (in h, min, s) of an object 6 h 30 min of solar time after
crossing the prime meridian?
4.4. If an object is on the prime meridian at the European Southern Observatory
at La Silla, what is its HA from Greenwich? (La Silla (ESO): latitude 29 150
2600 S; longitude 70 430 4800 W).
4.5. Calculate the local sidereal time at La Silla for 8 p.m. local time on November
5, given that the previous GST(0) is 2 h 56 min 20 s. (La Silla time zone: 5 h
West).
4.6. What is the HA of (a) Sirius, (b) Betelgeuse, at 8 p.m. GMT on November
5 from La Silla (see Exercise 4.5, ignore precession)? Comment on whether
either star would be visible. (Sirius: RA(2000) 6 h 45 min 09 s; Dec(2000) 16
420 5800 . Betelgeuse: RA(2000) 5 h 55 min 10 s; Dec(2000) þ7 240 2600 ).
4.7. (a) Use the cosine rule of spherical trigonometry to show that the altitude of an
object is related to its RA and Dec (a and d), the local sidereal time (LST)
and the observer’s latitude (f) by:
sinðAltitude) ¼ sin f sin d þ cos f cos d cosðLST aÞ

(b) Use the five parts rule similarly to derive the formula for azimuth. (Hint:
Remember that sin2x + cos2x ¼ 1.):
cos f sin d sin f cos d cosðLST aÞ

cosðAzimuthÞ ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 ðsin f sin d þ cos f cos d cosðLST aÞÞ2
(c) Hence determine the altitude and azimuth at 05 h local sidereal time of an
object at RA 3 h and Dec +50 for an observer at a latitude of þ60 .
Chapter 5
Movements of Objects in the Sky
Objects in the sky, including the so-called “fixed” stars, actually move in various
ways, some quite complex, for a variety of reasons.
Diurnal Motion
First, and obviously, the whole celestial sphere rotates in one sidereal day due to the
counter-rotation of Earth. Objects in the sky therefore move as seen from a fixed
point on Earth (Fig. 5.1), and telescopes have to be driven around their polar axes to
counteract this motion. An immediate consequence of this motion is that most
objects in the sky are only visible for the fraction of the sidereal day between rising
and setting. The hour angles1 of an object at rising and setting are related to its
declination, d, and to the latitude, f, of the observer:
H R ¼ 24 cos1 ð tan f tan dÞ (5.1)
H S ¼ cos1 ð tan f tan dÞ (5.2)
where HR is the HA at rising and HS is the HA at setting.
1
This is the time that the object rises or sets on the celestial horizon (Fig. 4.7). The actual time that
you see an object rise or set will generally differ from this because (1) the “true” horizon is below
the celestial horizon in the sky (Fig. 4.6) due to the observer’s height above the surface of Earth,
(2) the actual horizon is defined by hills, mountains, valleys, trees, buildings, etc., (3) refraction by
Earth’s atmosphere (5.7) lifts objects close to the horizon by about half a degree towards the zenith
and (4) in the case of the Sun and the Moon the positions tabulated for them in almanacs are for
their centers. Both objects however have angular radii of about 150 and we naturally in real life
count sunrise (moonrise) as the first appearance of the upper limb of the Sun (Moon) and sunset
(moonset) as the last glimpse of the edge of the Sun (Moon) going below the horizon.

102 5 Movements of Objects in the Sky
Fig. 5.1 Diurnal motion
The right ascension of the object, a, then gives the local sidereal times for its rising
or setting using (4.12). Reversing the calculation to obtain local sidereal time (Chap.
4) then gives the local civil time of the object’s rising or setting. The position on the
horizon of rising or setting can also be found. Its azimuth, AR or AS, is given by:
AR ¼ cos1 ðsin d sec fÞ East (5.3)
AS ¼ cos1 ðsin d sec fÞ West (5.4)

Q. Find the azimuth of the rising Sun on the summer solstice as seen from the
array of standing stones at Carnac (Brittany). Comment on the fact that the
alignment of the lines of standing stones is towards an azimuth of ~70 E.
Data:
Carnac latitude: 47 360 N
Solar declination at the summer solstice: 23 260
A. From (5.3)
Sunrise azimuth ¼ AR ¼ cos1 (sin d sec f) E
AR ¼ cos1 (sin 23 260 sec 47 360 ) E
¼ cos1 (0.3977 1.4830) E
¼ 53 510 E
Comment: Clearly the lines of standing stones at Carnac are NOT aligned
towards the sunrise at summer solstice – unlike the alignments of many other
megalithic constructions. For example at Stonehenge the Heel stone is at an
azimuth of ~51 E as seen from the center of the circle of stones while the azimuth
of summer solstice sunrise from Stonehenge (latitude 51 110 ) is 50 400 E.
Circumpolar Objects 103
Circumpolar Objects
Equations (5.1), (5.2), (5.3), and (5.4) reveal that some stars never rise or set
because the inverse cosine only has meaning for values in the range 1, and both
(tan f tan d) and (sin d sec f) can lie outside this range. For example, with d ¼ 60
and f ¼ 52 (the latitude of the old Greenwich Observatory), we find sin d sec
f ¼ 1.4067 and so we cannot obtain the inverse cosine in order to find the rising or
setting points. This is an example of a circumpolar object, or an object that is always
above the horizon, and it is easily pictured (Fig. 5.2). For northern hemisphere
observers, the best known circumpolar object is the pole star, Polaris. Similarly,
there are objects close to the South Pole (for a northern observer) that never rise.
The minimum declination for an object never to set is (Fig. 5.3):
d 90o f (5.5)
Fig. 5.2 Circumpolar object
Fig. 5.3 Cross-section

through the celestial sphere
for an object that is just
circumpolar
and similarly
d f 90o (5.6)
for an object never to rise.
Seasons and Annual Motions
Earth’s orbital motion around the Sun, like its rotation, causes movements in the sky.
For the more distant objects, the main effects are small and are dealt with later as
aberration, precession and parallax. For objects within the Solar System, however, the
changes due to Earth’s orbital motion can be large. Those of the planets and the Moon
are also dealt with later; here we just consider the effect on the Sun’s position in the sky.
Since Earth moves completely around the Sun in its orbit, the geocentric viewpoint
is of the Sun moving completely around the sky, its path being called the ecliptic. This
is not normally obvious, because when the Sun is in the sky, the other celestial
reference points are invisible. Thus it is not easy to see that in early September, say,
the Sun is to be found in the constellation of Leo, while by late September it is in Virgo.
Instead of being aware directly of the Sun’s movement, we are more aware of its
consequences in the changing constellations visible at night and the seasons.
The constellations visible at night are clearly those opposite to the Sun in the
sky, and so these change throughout the year. This change is reflected in names such
as the “Summer Triangle” for the brightest stars in Cygnus, Lyra and Aquila, visible
on summer nights. The solar movement is also marked at particular points along the
ecliptic. One that we have already encountered is the first point of Aries, or vernal
equinox (which is actually in Pisces; its position has moved owing to precession
since it was first labeled by the early Greek astronomers – see later). The Sun passes
through this point on or about March 21 each year.
When the Sun is at the vernal equinox, it is also on the equator, and so we have
days and nights of equal length. The other equinox, the autumnal, is opposite the
first point of Aries in the sky, in Virgo, and the Sun passes through it on or about
September 21 each year (Fig. 5.4). The two other points commonly noted are the
summer and winter solstices on or about June 21 and December 21. These occur
when the Sun is at its northernmost or southernmost points on the ecliptic; on the
Gemini-Taurus border, and in Sagittarius, respectively. These points also mark
the longest and shortest days, though because of the equation of time (Fig. 4.16), the
latest evening and the earliest morning are about June 26 and June 16, and the
earliest evening and latest morning about December 12 and December 31.
Our perception of day length, of course, is generally different from just sunrise to
sunset, because of twilight. Twilight is due to light from the Sun reaching the
surface after the Sun has set, as seen from that point, because the light has been
scattered in Earth’s atmosphere. Thus the sky remains bright after the Sun has set
(or before it has risen). For astronomical purposes, twilight ceases when the Sun
Seasons and Annual Motions 105
Fig. 5.4 Annual movement

of the Sun
is 18 below the horizon. Civil twilight ends or begins when the Sun is 6 below the
horizon, and nautical twilight when it is 12 below. At the height of summer,
therefore, for any place north of latitude of 48.5 N (or south of a latitude of
48.5 S), twilight lasts throughout the night on the astronomical definition, and it
never really gets dark. This period of “undark” nights increases as the latitude
increases, until above 66.5 N, or below 66.5 S (the Arctic and Antarctic circles)
there are occasions when the Sun becomes a circumpolar object.
This consideration of rising and setting times and points, and of the declinations
of circumpolar objects, which we have just encountered, is correct in the absence of
an atmosphere. The presence of the atmosphere, however, alters the calculations
slightly, because refraction in the atmosphere causes objects to appear slightly
closer to the zenith than their true positions. The effect is greatest for an object
on the horizon, when it will appear to be about half a degree above the horizon.
Sunrise and sunset will thus occur about 2 minutes earlier or later than predicted.
The date on which the day length and night length are in reality most nearly equal is
known as the equilux and it occurs two to three days before the vernal equinox or
after the autumnal equinox. For objects higher in the sky than those just rising
or setting, the increase in altitude, R, is given approximately by
R 58:200 tan z (5.7)
where z is the zenith distance of the object.

The other effect of the changing position of the Sun is that of the seasons, and
this arises from the changing declination of the Sun in its path around the ecliptic,
not, as is commonly mistaken, from the changing distance of Earth from the Sun2 as
it moves around its elliptical orbit.
2
In fact Earth is closest to the Sun (perihelion) on or about January 4th each year – which is indeed
the height of summer in the southern hemisphere but is the depths of winter for the northern
hemisphere.
Fig. 5.5 Seasonal variation

in the energy per unit area
from the Sun
There are three contributory factors to the larger amount of energy received from
the Sun at a particular place in summer when compared with winter. The first is that
the solar radiation is spread over a smaller area of the ground in summer than in
winter (Fig. 5.5), the second that the day is longer in summer, and the third that the
lower altitude of the Sun in winter results in more energy being lost due to
absorption by the atmosphere.
Planispheres and Astrolabes
Approximate rising and setting times of stars and their rising and setting points on
the horizon as well as which constellations are visible at any time of the night
throughout the year may be found using planispheres (sometimes called star
wheels) and astrolabes. The planisphere is a modern version of the very ancient
instrument called the astrolabe. Both have to be constructed for a particular latitude,
but both can still be used by some 5 either side of that latitude, especially if
precise results are not needed.
The planisphere (Fig. 5.6) has a circular disk with the brighter stars, the
constellations, the ecliptic, the equator, etc., printed onto it. A second circular disk
overlays the first and has an oval transparent window that permits those stars and
constellations visible in the observer’s sky at a particular time of the night and on a
particular night of the year to be seen. The outer edges of the two disks have the days
of the year and the time of day or night printed around their edges. To see which
constellations will be in the sky and which stars will just be rising or setting at (say)
10 p.m. on February 14, the disks are rotated until the marker for 10 p.m. (or 22.00 h)
on one disk aligns with the marker for February 14 on the other disk.
The shapes of the constellations and the distances between stars are quite
distorted on the planisphere map, since the entire hemisphere of the sky has had
Planispheres and Astrolabes 107
Fig. 5.6 A selection of commercial and home-made (From a pattern downloaded via the internet)
planispheres
to be squashed to fit onto the flat disk, but they can usually be recognized after a bit
of practice. Planispheres are produced commercially at fairly low cost, and free
versions to make up yourself can be downloaded via the Internet. One such
example of the latter is shown in Fig. 5.6. It was sourced from the astronomy
magazine Sky & Telescope’s’ Internet site at http://www.skyandtelescope.com/
letsgo/learningthesky/Make_a_Star_Wheel.html if this site is no longer avail-
able when you read this paragraph, then an Internet search for ‘Planisphere Free
Download’ will soon pick up numerous alternatives. Whether purchased or
downloaded most observers will find a planisphere very useful as a quick source
of reference.
The astrolabe (Fig. 5.7) was used in mediaeval times, not just to show which
stars were in the sky but also (and more importantly) to determine the time at night
and the latitude of a ship at sea. The front of the astrolabe (left hand image of
Fig. 5.7) in the modern version shown here is clearly a variation on the planisphere.
Since plastic sheet was not available eight centuries or so ago, in the medieval
instruments a small number (12 or so) of star positions were indicated using a
pierced metal disk (called the Rete) by the ends of needle-shaped pointers. The back
of the astrolabe has a rotating bar (the alidade) that is used to sight a star at night and
determine its altitude. The other engraved circles on both sides of the instrument
then enable the time, latitude and various other quantities to be calculated.
Fig. 5.7 Left and center: a modernized astrolabe based upon Chaucer’s 1391 design. The pattern
was downloaded via the internet and constructed in a couple of hours. The main modernization is
that the star map, known as the Rete, is printed onto a transparent plastic sheet and shows a 100 or
so stars. The Rete for a medieval instrument would have indicated the positions of only a dozen or
so stars by means of individual pointers on a pierced brass disk. Right: The front of an astrolabe
and its ruler and alidade (From ‘A treatise on the Astrolabe’ by G. Chaucer (1391), edited and re-
printed in 1872 by W. Skeat)
An astrolabe can be purchased today, but a modern version will cost hundreds of
dollars and an antique medieval instrument will cost 10 to a 1,000 times more than
that. However an article3 by Dominic Ford giving a history of the astrolabe,
diagrams of its various components and instructions on how to make and use the
astrolabe may be found at http://dcford.org.uk/astrolabe/index.html. The astrolabe
design at that site is based upon the instrument described by Geoffrey Chaucer in
1391. The site gives details for astrolabes for use at latitudes ranging from 85 S to
85 N at 5 intervals. The astrolabe shown in Fig. 5.7 was downloaded from the site
and made into a working model in a couple of hours.
Movements of the Moon and Planets
Unlike the Sun, some objects in the sky can easily be seen to move against the
background of the (relatively) fixed stars. Many of these objects have been known
since antiquity: studies of the planets, from Mercury to Saturn; the Moon; and the
occasional comet probably represented the first stage of astronomy as a recogniz-
able science. The motivation for their study lay in the mistaken idea that they could
foretell events on Earth, and coincidences between notable events such as battles,
famines, deaths of kings, etc., and events in the sky would have seemed to support
the possibility. Today, instead, the movements of the planets are studied in their
own right for what they may reveal about gravity, the nature of the planets
themselves and their satellites, plus, of course, the rather practical reason of
knowing where they are in the sky so that they can be observed.
3
‘Building a Model Astrolabe,’ Journal of the British Astronomical Association, Vol. 122, Nos 1,
February, 2012.
Movements of the Moon and Planets 109
Fig. 5.8 Lunar rotation
The Moon
The Moon changes its position with respect to the background stars in a very short
time; indeed its motion can be seen with the naked eye in an hour or so. This is due
to the Moon’s motion in its orbit around Earth, though from an external viewpoint it
would be truer to say that the Moon and Earth both have orbits around the Sun but
with mutual perturbations, since the Moon’s motion around the Sun is always
concave inwards.
The Moon is often said to keep the same face towards Earth the whole time
(Fig. 5.8). This arises because the Moon’s orbital and rotational periods are
identical. That this is so is not a chance coincidence, but because in the past Earth’s
tides produced in the Moon by Earth have dissipated the Moon’s rotational energy
and slowed it down. The process only halts when the Moon has stopped rotating
with respect to the Earth. Its rotation is then said to be tidally locked on to Earth.
The tides in the Moon have not disappeared when this has happened but are now
stationary with respect to the Moon, and so do not dissipate energy.
The phases of the Moon occur from the changing proportion of its illuminated
surface that we can see (Fig. 5.18, later), and not, as is often mistakenly thought,
from Earth’s shadow falling onto the Moon. The latter situation is a lunar eclipse
(see later). The Moon’s rotational motion is at a constant rate, but its orbital motion
changes because the orbit is elliptical. Thus sometimes the Moon’s rotation is ahead
of its orbital motion and sometimes behind, and we can see a little way “around the
corner.” This effect is called libration, and it allows some 59% of the Moon’s
surface to be seen from Earth.
Other Solar System Objects
Under this heading we include the major and minor planets and comets. All these
objects move against the background stars at greater or lesser rates, owing to the
combined motions of Earth and of the object around their orbits (Fig. 5.9). Their RA
and Dec have therefore to be calculated from their orbital parameters for a given
moment in time, if they are to be found in a telescope. Fortunately this calculation
has already been done for the principal objects and the results tabulated in, for
example, the Astronomical Almanac. Likewise, the production of position
predictions (an ephemeris) is the first urgent task of astronomers after the discovery
of a new comet, so that it may be followed across the sky.
The paths of planets around the sky are complex (Fig. 5.10), and their explana-
tion caused many problems while Earth was still regarded as the center of the solar
system. The main riddle was the way in which the normal motion of some of the
planets across the sky from west to east (direct motion) would sometimes reverse
Fig. 5.9 Movements of Earth and other planets

Other Solar System Objects 111
Fig. 5.10 Motion of a planet

in the sky (Schematic)
(retrograde motion), and then after a while the planets would again resume their
usual direction of travel.
The difficulties with understanding the motions of the planets, however, were not
just due to the geocentric ideas, but also to the notion that the circle was the most
perfect geometrical shape, and that therefore objects in the “perfect heavens” must
move in circles. These two preconceptions eventually resulted in a model of the
Solar System described in A.D. 140 in the Megale syntaxis tes astronomias by
Ptolemy (about A.D. 100–170). This book is better known as the Almagest (from
the Arabic, Al magiste: the greatest) and had a planet moving around one circle
(the epicycle), the center of which in turn moved around a second circle (the
deferent). To get good agreement with the observations, the center of the deferent
had to be displaced from the center of Earth, and in turn sometimes moved around a
small circle. Although such constructions now seem outlandish, it should be
remembered that Ptolemy’s geocentric model of the Solar System is actually
the most successful scientific theory ever proposed, giving reasonably accurate
predictions of the positions of the planets for some fourteen centuries. Eventually,
of course, in 1543, the heliocentric model of the Solar System was suggested by
Nicolas Copernicus (1473–1543); but it was not until the publication of Astronomia
Nova in 1609 by Johannes Kepler (1571–1630) that circular motions were replaced
by elliptical orbits and thus finally solved the problem of the movements of the
planets in the sky in terms of the relative motions of Earth and the planets (Fig. 5.11).
Proper Motion
We have so far regarded the stars for all practical purposes as fixed in position.
However, in many cases this is not quite true. Stars, including the Sun, are moving in
orbits around the Milky Way Galaxy with speeds typically of 200–300 km/s.
Some stars are sufficiently close and/or have high enough velocities for this motion
to lead to an observable change in their position. This change in position is called the
proper motion of the star and is usually symbolized by the Greek letter m. Typical
values for the proper motion are in the range of 0.000100 –100 /year. The bottom limit is
due to the degree of precision with which stellar positions can be measured and is not
a true cut-off. As higher precision positional measurements are obtained we may
Fig. 5.11 Copernican explanation of the motion of the planets in the sky
expect smaller proper motions to be determined. The Gaia spacecraft (due for launch
in 2013), for example, is expected to measure proper motions to accuracies of a few
millionths of an arc second per year.
These values for typical proper motions do not lead to changes in the star
patterns, etc., on human time scales, but on longer time scales, the constellations
will change (Fig. 5.12). The more distant stars and galaxies are effectively fixed in
position except on time scales of hundreds of millions of years, although, in special
cases, motions can be detected even for objects tens of millions of parsecs away.
Thus very long base-line radio interferometry has revealed motions across the line
of sight for relativistic jets in active galactic nuclei, and changes may be observed in
some quasars owing to the motion of an intervening galaxy acting as a gravitational
lens. Radio measurements of water masers in the galaxy M33 have recently
achieved levels of accuracy better than ten millionths of an arc second per year
and suggest a space velocity of about 190 km/s relative to the Milky Way –
although this may not be the true velocity of the galaxy as a whole, since the masers
may be moving about inside it.
Fig. 5.12 Constellation

of Ursa Major
100,000 years ago, now
and in 100,000 years’ time
The proper motions of stars are usually tabulated in catalogues as the changes in
the stars’ RA and Dec – ma and md respectively. There is then the possibility of some
confusion over the meaning of the change in RA – it may be the actual change in
RA or it may be the real movement of the star across the sky perpendicular to the RA
meridian.4 These two differ because the RA meridians approach each other as the
declination increases (like the shape of the segment of an orange). If ma is the change
in RA, then the movement across the sky is given by ma cos d. In making calculations
involving proper motions the student must thus take care to ascertain which value
has been tabulated. Some indication may be given by the units – normally s/year for
ma and 00 /year for ma cos d, but this cannot be relied on completely. There is no such
confusion about md – it means both the actual change in declination and the real
movement in the sky. The total proper motion is given by
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi00
m¼ m2d þ m2a cos2 ðdÞ =year (5.8)
4
The great circle running from the celestial north pole to the celestial south pole and along which
the RA value is constant.
Fig. 5.13 Precession in a

spinning top and Earth (Note
that the imbalance between
the gravitational forces on the
two halves of Earth has been
much exaggerated)
Precession
One of the first “motions” of the stars was noticed by Hipparchus of Nicaea (about
190–115 B. C.) in 130 B. C. He noticed that Spica (among other stars) was some
2 further from the equinoctial points than had been the case 150 years earlier when
observed by Timocharis. This “motion” turns out not to be a movement of the stars
themselves but of the reference point from which their positions are measured.
As we have seen, that reference point is the first point of Aries, which we earlier
took to be a fixed point on the celestial sphere. In fact, it is not fixed but moves
backwards around the ecliptic once every 25,700 years. The movement of the first
point of Aries is due to the changing position in space of the equator, and this in turn
arises because Earth’s rotation axis is swinging through space. Like a spinning top,
Earth’s rotational axis swings (precesses) through space under the effect of the solar
gravity on Earth’s equatorial bulge (Fig. 5.13). The precessional motion causes the
north and south rotational poles to trace out circles around the pole of the ecliptic
(Fig. 5.14). Thus it is just chance that we have a pole star (Polaris) at the moment.
Fig. 5.14 The movement of the north rotational pole due to precession
As may be seen from Fig. 5.14, no other bright star is approached by the pole in its
motion as closely as Polaris, and so for most of the 25,700 year cycle there is no
pole star.
The effect of precession upon the coordinates of an object is given by
Da ¼ 3:4DTðcos e þ sin e sin a tan dÞ seconds of time (5.9)
Dd ¼ 50:4DT sin e cos a seconds of arc (5.10)
where Da is the change in RA, Dd is the change in declination, DT is the time interval
between the epoch (the time at which the position is tabulated) and the time for
which it is required in years, e is the obliquity of the ecliptic (23 260 ) and a and d are
the RA and Dec of the object at the epoch. Currently most catalogs of objects’
positions in the sky use the epoch 2000.0, catalogs based upon 2025.00 are likely to
start coming into use in the next few years.

Q. Find the position of a Orionis (Betelgeuse) on January 1, 2100, by allowing
for the effects of precession and proper motion.
Data:
RA (epoch 2000.0): 05 h 55 min 10 s
Dec (epoch 2000.0): +07 240 2500
Proper motion RA: ma ¼ 0.00168 s/year (change in RA values with time)
Dec: md ¼ 0.0096 00 /year
(continued)
(continued)
Obliquity of the ecliptic, e ¼ 23 260
A Converting RA to degrees, we have
RA ¼ 5 15 + 55 150 + 10 150 ¼ 88 470 3000
From (5.9)
Da ¼ 3.4 100 (cos 23 260 ) + sin (23 260 ) sin (88 470 3000 )
tan (þ07 240 2500 )
¼ 340 (0.9175 + 0.3977 0.9998 0.1300)
¼ 329.5 s ¼ 5 m 29.5 s
and from (5.10)
Dd ¼ 50.4 100 sin (23 260 ) cos (88 470 3000 )
¼ 5040 0.9998 0.9917
¼ 499700 ¼ 1 230 1700
In 100 years the proper motion will have changed the positions by
Da ¼ 100 0.00168 ¼ 0.168 s
Dd ¼ 100 0.0096 ¼ 0.9600
Thus in 2100 a Orionis will have the position
RA (epoch 2100.0) ¼ 05 h 55 min 10 s + 5 min 29.5 s + 0.168 s
¼ 06 h 00 m 39.668 s
¼ 06 h 00 m 40 s
Dec (Epoch 2100.0) ¼ +07 240 2500 + 1 230 1700 + 0.9600
¼ +08 470 42.9600
¼ +08 470 5000
Parallax
Earth’s orbital motion causes its position in space to change by 300,000,000 km

every 6 months. This is sufficient to cause the nearer stars to shift their positions
slightly owing to parallax (Fig. 5.15). The movement is small, only just over 1 s of
arc even for the nearest star (Proxima Cen), but it is sufficient to form the basic
method of determining the distances of the stars:
1
D¼ (5.11)
P
where D is the distance in parsecs (1 pc ¼ 3.3 light-years ¼ 3 1016 m) and P is

the parallax angle in seconds of arc (note that this is the half angle, not the whole
parallax).
Fig. 5.15 Parallax motion

due to Earth’s orbital motion
Over a year the object being observed will trace out a small ellipse centered on
its position as seen from the Sun.
Aberration
This is the last of the motions of objects in the sky that we need to consider. Like
parallax, aberration is an effect of Earth’s orbital motion. However, its magnitude is
considerably larger than that of parallax – up to 2000 , and it does not depend upon the
distance of the object. It arises from Earth’s motion through space. Light travels at a
finite speed (300,000 km/s ¼ 1 m in 3.3 ns); thus, Earth’s motion carries the eyepiece
(or detector, etc.) of a telescope along a short distance in the few nanoseconds it takes
light to travel down the telescope (Fig. 5.16). The observed position of the object is
hence shifted towards the apex of Earth’s velocity at that instant of observation by a
small amount. (The apex is the point in the sky towards which Earth’s orbital motion is
directed at any given instant. It lies on the ecliptic and follows the Sun over a period of a
year. At aphelion and perihelion it is exactly 90 behind the Sun; at other times it gains
or loses some distance on the Sun due to the ellipticity of Earth’s orbit.) Since Earth’s
velocity changes direction around the orbit, an object observed throughout a whole year
would move in a small ellipse centered on the position the object would have if Earth
were stationary. The aberration ellipse may be distinguished from the parallax ellipse
because the instantaneous shift of the object is towards the position of the Sun in the
case of parallax, and towards a point on the ecliptic 90 behind the position of the Sun in
the case of aberration.
Fig. 5.16 Aberration
Relative Planetary Positions
The movements of Earth and planets mean that their relative positions in space are
continually changing. Several relative positions of Earth, the Sun and another planet
are sufficiently important to be given specific labels (Fig. 5.17). The most useful of
these terms are opposition, when an outer planet may most easily be observed and
greatest elongation, which is the optimum time for observation of an inner planet. The
best time for observing an inner planet is not inferior conjunction as might be expected,
when the planet is at its closest to us, because we can then only see the un-illuminated
half of the planet. The relative positions result in a greater or lesser extent of the planet’s
surface being seen as illuminated. For the Moon, this results in the familiar phases
(Fig. 5.18). Mercury and Venus go through similar phases, and also noticeably change
their angular sizes (Fig. 5.19). The outer planets show phases, but they are always
gibbous, and normally the effect is unnoticeable except, at times, for Mars.
Eclipses and Related Phenomena
Some relative positions of Earth and other Solar System objects are of particular
scientific significance, and these are eclipses, occultations and transits.
Sometimes the relative orbital motions of Earth and two other objects cause one
of those objects to pass in front of or behind the other. An eclipse occurs when the
two objects are of comparable angular size, for example, the Sun and Moon, two of
Jupiter’s Galilean satellites, or two stars in a binary system. An occultation occurs
when an object of much larger angular size passes in front of one that is angularly
much smaller, such as the Moon in front of a star. A transit occurs when an object of
Eclipses and Related Phenomena 119
Fig. 5.17 Relative planetary

positions
much smaller angular size passes in front of one that is angularly much larger, for
example, Mercury or Venus in front of the Sun, or a Galilean satellite in front of
Jupiter. All these phenomena are of great interest because they may reveal details
of the objects not otherwise easily observable, such as the solar corona or the rings
of Uranus (discovered during observations of an occultation of a star by Uranus).
Near approaches to mutual alignments result in conjunctions and appulses when
two planets, say, may come within a degree or two of each other in the sky, but
these events, like quadrature, are of only aesthetic interest.
A solar eclipse occurs when the Moon passes in front of the Sun as seen from
Earth (Figs. 5.20, 5.21, 5.22, and 5.23) or from a spacecraft in near-Earth orbit.
By accident the Sun’s and Moon’s angular sizes are almost equal. Most of the time,
the angular size of the Moon is slightly larger than that of the Sun, but at apogee
(see below) its angular size is slightly smaller, and this results in annular eclipses.5
When the Moon passes close to the center of the Sun, we get a total or annular
eclipse; otherwise, we get a partial eclipse. Because the plane of the Moon’s orbit is
inclined to the ecliptic by about 5 (Fig. 5.24) and because that plane rotates in
space due to terrestrial and solar perturbations, the number of solar eclipses can
vary from 2 to 5 per year.
5
The tides that the Moon raises on Earth also have the effect of slowly increasing the size of the
Moon’s orbit. Currently the increase is by about 38 mm/year. In the absence of any other changes
in the Moon’s orbit, in around 450 million years therefore the Moon will be too far away from
Earth for any total solar eclipses to occur. All solar eclipses then will be annular or partial.
Fig. 5.18 Phases of the Moon
Lunar eclipses are quite different from all the other phenomena discussed in this
section. For a lunar eclipse to be comparable with the other eclipses, it would have
to be viewed by an astronaut on the surface of the Moon – and should be called a
solar eclipse by Earth! A lunar eclipse occurs when the Moon passes into the
shadow cast by Earth (Figs. 5.25 and 5.26). For the same reason, therefore, as
solar eclipses, lunar eclipses can only occur when the line of nodes of the Moon’s
orbit passes close to the Sun (Fig. 5.24). There can thus be up to three lunar eclipses
per year. Lunar eclipses are much rarer than solar eclipses, but they appear to be
more common because they can be seen from half of Earth at a time.
Transits occur for Venus and Mercury against the Sun, and for planetary
satellites against the parent planet. They are no longer very significant, but in the
past they were important as a means of measuring the astronomical unit, and for Ole
Rómer’s (1644–1710) determination of the velocity of light by timing the transits of
the Galilean satellites.
Eclipses and Related Phenomena 121
Fig. 5.19 Phases of Mercury and Venus
Sun
Moon Earth
Total
eclipse
Partial
region
eclipse
region
Fig. 5.20 The alignment of the Sun, Moon and Earth required for a solar eclipse to occur (Not to
scale)
Fig. 5.21 A total solar eclipse enabling the inner corona to be seen (Reproduced by courtesy of
ESO and Phillipe Duhoux)
In the last couple of decades many planets belonging to stars other than the Sun
(exoplanets) have been discovered as they transit their host stars. We cannot
actually see the planet silhouetted against the disk of the star, of course, but the
total observed intensity of the star dips slightly as a small portion of its disk is
obscured by the planet. The change in brightness of the star is small – 1% or less
even for exoplanets larger than Jupiter and down to 0.01% for Earth-sized
exoplanets. However the Kepler spacecraft (Fig. 5.27), which is designed, inter
alia, for detecting exoplanet transits together with other related ground- and space-
based instruments can currently detect brightness changes down to 0.002%. Thus,
of the over 700 exoplanets that have been discovered to date, nearly 30% were
found through their transits.
In an occultation, the angularly smaller body disappears behind the angularly
larger body. Lunar occultations are of use in identifying close double stars and
measuring stellar diameters. Occultations by (Solar System) planets can give infor-
mation on the outer parts of the planets’ atmospheres, since the light from the star
can still be detected even when passing through an appreciable amount of material.
Position in an Orbit
In an elliptical orbit of one object around another, the separation of the two objects
varies; this also leads to certain relative positions being identified by name. The
maximum and minimum separation points are called the apsides (singular, apsis),
Fig. 5.22 A partial solar eclipse (Reproduced by courtesy of C. E. Danes)
Fig. 5.23 An annular solar eclipse (Schematic)

Fig. 5.24 Limits on the occurrence of eclipses
Moon
Partial (partially
eclipse eclipsed)
Sun
region
Earth
Partial
eclipse
region Total
Moon eclipse
(totally region
eclipsed)
Fig. 5.25 The alignment of the Sun, Moon and Earth required for a lunar eclipse to occur (Not to
scale)
and the line joining them, the line of apsides, or more commonly, the major axis of
the orbit. The maximum separation is signified by the prefix “Ap,” the minimum by
the prefix “Peri,” followed by a suitable signifier for the object involved. Thus for
an object in orbit around Earth, we have apogee and perigee. Similarly, for other
objects in orbits around the Sun, we have aphelion, perihelion; around the Moon we
have apocynthion, pericynthion; around Jupiter, we have apojove, perijove; and
around a star, we have apastron, periastron, etc.
Synodic Period 125
Fig. 5.26 Left: A total lunar eclipse. The central part of totality can be seen encroaching over the
Moon on its left while the lunar surface is still partly illuminated on the right. The Moon appears
reddish because light refracted in Earth’s atmosphere is reaching it. The shorter wavelengths (blue)
have been preferentially scattered out of the light beam, leaving the red wavelengths to pass through
the atmosphere (the Sun appears reddish when low on the horizon for the same reason) (Copyright
# C.R. Kitchin, 1989). Right: the normal waxing Moon showing roughly the same appearance as
the eclipsed Moon. It can be distinguished from the eclipsed Moon by (1) the details, such as craters,
etc., to be seen on its surface (this is because the eclipsed Moon, being full, lacks visible shadows to
highlight the same features), (2) the sharper curve of the boundary (terminator) between its light and
dark regions compared with that on the eclipsed Moon and (3) the alignment of the terminator with
the Moon’s north and south polar caps compared with the slant for the boundary on the eclipsed
Moon (Reproduced by kind permission of ESO and Andy Strappazzon)
Synodic Period
The time taken for a planet to go around its orbit once, the orbital period, is
normally called the sidereal period of the planet. It is the fundamental quantity
required in determining the orbit of the planet. However, our observations of the
planets are made from the moving Earth, and so we do not observe the sidereal
period directly but the period required for the planet to return to the same relative
position with respect to Earth and the Sun. The interval between successive returns
to the same relative position (e.g., between two oppositions) is called the synodic
period of a planet. Since we are observing from Earth, it is the synodic period that
Fig. 5.27 An artist’s impression of the Kepler spacecraft. It is shown silhouetted against a distant
exoplanetary system. However this is completely fictional. Kepler remains in orbit around the Sun
– it does NOT travel to distant stars in order to make its observations (Reproduced by kind
permission of NASA, the Kepler mission and Wendy Stenzel)
we are able to measure, and the sidereal period must then be derived from this.
From Fig. 5.28, we may find the sidereal period:
360 360
Relative angular velocity ¼ (5.12)
E P
360
¼ (5.13)
S
where E is the sidereal period of Earth (1 year), P is the sidereal period of the planet,
S is the synodic period of the planet, which gives, for an outer planet, working in
units of years (the sidereal period of Earth, so that E ¼ 1):
S
P¼ (5.14)
S1
and we may similarly find for an inner planet
S
P¼ (5.15)
Sþ1
Synodic Period 127
Fig. 5.28 Relationship between synodic and sidereal periods

Q. Find the orbital (sidereal) period of Mercury
Data:
Length of Earth’s year: 365.25 days
Dates of Mercury’s superior conjunctions:
Date Time (h)

2013 Jan 18 09
May 11 21
Aug 24 21
Dec 29 06
2014 Apr 26 03
Aug 8 16
Dec 8 10
A. From the tabulated data we find the following values for Mercury’s synodic
period:
Interval between 2013 Jan 18th at 09h and 2013 May 11th at 21h ¼ 113d 12h ¼ 0.3107 yr
Interval between 2013 May 11th at 21h and 2013 Aug 24th at 21h ¼ 105d 00h ¼ 0.2875 yr
Interval between 2013 Aug 24th at 21h and 2013 Dec 29th at 06h ¼ 126d 09h ¼ 0.3460 yr
Interval between 2013 Dec 29th at 06h and 2014 Apr 26th at 03h ¼ 117d 21h ¼ 0.3227 yr
nterval between 2014 Apr 26th at 03h and 2014 Aug 8th at 16h ¼ 104d 13h ¼ 0.2862 yr
Interval between 2014 Aug 8th at 16h and 2014 Dec 8th at 10h ¼ 121d 18h ¼ 0.3333 yr
(continued)
(continued)
Mercury’s average synodic period is thus 0.3144 years.
Hence from (5.15)
Mercury’s orbital period ¼ Mercury’s sidereal period

¼P
¼ S/(S þ 1)
¼ 0.3144=ð0.3144 þ 1Þ
¼ 0.2392 years
¼ 87.4 days
The correct orbital period for Mercury is 87.97 days. The difference between
this figure and the one calculated, and also the reason for the varying values
found for the synodic period, lies in the highly elliptical nature of Mercury’s
orbit. Its angular velocity varies from 6.3 /day at perihelion to 2.8 /day at
aphelion. Thus the time interval between superior conjunctions (etc.) varies
depending upon whereabouts within Mercury’s orbit the successive
conjunctions occur.
Exercises
5.1. What is the azimuth of the Sun on rising at the summer solstice as seen from
the old Greenwich Observatory? (latitude 51 290 ; longitude 0 ).
5.2. From above what latitude is Betelgeuse a circumpolar star (ignore precession)?
(Betelgeuse: RA(2000) 5 h 55 min 10 s; Dec(2000) +7 240 2600 .)
5.3. Calculate the position of Sirius on January 1, 2025, by allowing for the effects
of precession. (Sirius: RA(2000) 6 h 45 min 09 s; Dec(2000) 16 420 5800 ;
obliquity of the ecliptic, e ¼ 23 26).
5.4. Calculate the synodic periods of
(a) Venus
(b) Jupiter
(c) Pluto
(Sidereal orbital periods of Venus 0.615 years; Jupiter 11.862 years; Pluto
247.7 years.)
5.5. Can
(a) A lunar eclipse occur at new Moon?
(b) A solar eclipse occur at the summer solstice?
Chapter 6
Telescope Mountings
Introduction
The requirements for a telescope mounting are divided into two distinct sections.
The first is the set of mechanical components that are required to hold the optics
of the telescope in their correct relative positions, and allow them to be collimated
and focused. This section is normally called the telescope tube, and reference is
made to it in Chaps. 1, 2, 3, and 8.
Here, therefore, we are concerned with the second section, which is the set of
mechanical components that enables the telescope tube and the optics to be pointed
at the object to be studied, and in most cases then to compensate for Earth’s rotation
(i.e., to track the object) automatically. Reference to this section of the mounting
has also been made elsewhere (Chaps. 1, 2, 3, 4, 5, and 8), but we now look at its
specifications in more detail. In this chapter, we shall use the term mounting to refer
only to this second section from now onwards.
The basic requirements for the mounting are:
1. To enable the object to be observed to be found in the sky as quickly and easily
as possible;
2. To enable the object to be held steadily in the field of view as eyepieces are
focused or changed, and cameras, photometers, spectroscopes, etc., are attached
to the telescope tube;
3. To enable the telescope to follow the object as it moves across the sky, because
of Earth’s rotation, for periods of time up to several hours;
4. To provide the observer with convenient access to the eyepiece or ancillary
instrumentation.
All these objectives are achievable to almost any desired level of performance,
but, as ever, the cost of the mounting will rise steeply as the specifications on its
performance are increased. For a small telescope, it is generally necessary to
compromise at some point, or the cost of the mounting may become many times
the cost of the telescope. With major telescopes, initial acquisition may need to be

130 6 Telescope Mountings
to an accuracy of less than a second of arc, and tracking to better than a small
fraction of a second of arc over an hour. There are innumerable designs for
telescope mountings, but they mostly fall into the two main categories of equatorial
(Fig. 2.18) or alt-az (Fig. 2.19), and we shall consider the equatorial first.
Equatorial Mountings
The objectives for a telescope mounting are most straightforward to achieve with
designs of this type. The telescope tube can be moved about two perpendicular axes,
one of which is parallel to Earth’s rotational axis. Movement of the telescope around
this axis (the polar axis) therefore changes only the hour angle or right ascension
(Chap. 4) that is observed, while movement around the other axis (the declination axis)
changes only the declination that is observed. The position of the telescope in RA (or
HA) and Dec can easily be determined from suitably graduated 360 protractors
mounted on the two axes known as setting circles. Earth’s rotation can be compensated
for by a motor that drives the telescope around the polar axis, from east to west at a
constant rate of one revolution in 23 h 56 min (the sidereal day – see Chap. 4).
The main variations on the basic design are shown in Fig. 6.1. Any of these
designs, or their many variants, can make a suitable mounting for a small telescope.
However, a prime requirement of any mounting is for stability against vibrations,
and the designs with single supports and unbalanced weights such as the German
and fork mountings, will, all other things being equal, be poorer in this respect than
the English and modified English designs. In selecting a mounting, if there are
alternatives, then in general the thicker the axles, the larger the bearings and the
more massive the construction, the better.
For meter-class and larger telescopes, equatorial mountings are now rarely used.
The largest telescope with an equatorial mounting is the 5-m Hale instrument on
Mount Palomar, which was constructed in the late 1940s. It uses a horseshoe
mounting, which is a variant of the English mounting in which the upper bearing
takes the shape of a large horseshoe. The opening of the horseshoe is large enough for
the telescope to swing into it in order to observe near the north celestial pole. More
recently the 4-m Mayall and Blanco telescopes (1973 and 1974, respectively) and the
ESO 3.6-m telescope (1976 – Fig. 6.2) have also used horseshoe-type mountings.
Alt-Az Mountings
Alt-az mountings also enable the telescope to be moved around two perpendicular
axes, but this time one of the axes is vertical and the other horizontal (Fig. 2.19).
There is thus no simple way of relating the position of the telescope with respect to
its mounting to the point in the sky that is being observed. There is no simple means
of enabling the telescope to track an object either, since this will require motions in
both axes and at non-uniform rates.
Alt-Az Mountings 131
Fig. 6.1 Variations on the equatorial mounting design: (a) German mounting, (b) modified
English mounting, (c) English or yoke mounting, (d) fork mounting
Fig. 6.1 (continued)

Fig. 6.2 The ESO’s 3.6-m telescope showing its horseshoe-type equatorial mounting.
(Reproduced by courtesy of ESO)
The alt-az mounting, however, does have one big advantage – it is much easier to
make it stable and robust than any of the equatorial designs, because the weight, and
therefore the stresses, always acts in the same direction. In other words, a given
level of performance in terms of stability can be achieved at lower cost with an alt-
az than with an equatorial mounting. This has resulted in most of the major
telescopes commissioned recently and some smaller commercially produced
telescopes having alt-az mountings. The extra cost of the motors for both axes
and the computers to control them is more than made up for by the reduction in the
overall cost of the mounting for such instruments.
Most alt-az designs are like the fork mounting (Figs. 1.18 and 6.1d) with the polar
axis pointing to the zenith, and so there is little to say about them. One particular
variant, however, has become popular recently because of its low cost, suitability for
DIY construction and portability, and that is the Dobsonian design (Fig. 6.3), which
uses Teflon strips as the bearing material, giving very smooth adjustment of the
telescope’s position. The whole mounting can be placed on a low platform which, by
rotation over an inclined plane, will give tracking of objects in the sky over short
intervals of time, thus overcoming the main drawback of the alt-az.
Fig. 6.3 Dobsonian

mounting
There remains one problem with the alt-az mounting, though: if an object is
followed as it moves across the sky, the image plane rotates with respect to the
telescope. For visual observing this field rotation is of little consequence, but when
images are obtained with exposures of more than a minute or two, their outer edges
will be blurred because each point there has become slightly trailed. The rate of field
rotation varies in a complex manner with the position of the object in the sky and the
latitude of the observer (three examples are shown in Fig. 6.4). With large telescopes
on alt-az mountings the field rotation is countered by rotating the camera or other
instrument in the opposite direction or by putting counter-rotating optical
components, such as Dove prisms,1 into the beam of light before it reaches the focus.
Similar systems are manufactured commercially for small telescopes, but their
cost is high, their use difficult, and the results often leave a lot to be desired. From
Fig. 6.4, however, it can be seen that for an observer at a latitude of 45 N the field
1
Dove prisms have a trapezoidal cross section. Light entering one of the sloping faces is refracted
down to the base of the prism, where it is reflected. After refraction at the second sloping face, the
light beam emerges with an unchanged direction. Rotating the prism causes the light beam passing
through it to rotate at twice the prism’s speed.
Fig. 6.4 The field rotation rates for alt-az-mounted telescopes for objects at declinations of 30 ,
45 , and 60 , for an observer at a latitude of 45 N
rotation reduces to zero for objects that have declinations near 30 when those
objects have hour angles of 3.5 h (about 3.5 h before or after meridian passage).
Similarly for that same observer, the field rotation of objects at a declination of 45
reduces to zero when their hour angle is 0 h (i.e., on the prime meridian). Thus by
selecting the time of exposure so that it coincides with a period when the field
rotation for his/her chosen object is zero or small, an observer can obtain sharper
images without going to the expense of buying an image de-rotator. Observers
wishing to utilize this phenomenon to improve their images can obtain the times of
zero field rotation for their latitudes from Fig. 6.5. Of course, if you are observing
from, or close to, Earth’s north or south poles, then an alt-az mounting becomes an
equatorial mounting, and so field rotation is then zero for objects at all
declinations.
Note that for objects with declinations higher (lower) than the latitudes of northern
(southern) observers, there are NO times when the field rotation reduces to zero.

Q. An observer at a latitude of 30 N desires to image an object whose position is
a ¼ 4 h, d ¼ +20 at a time when the field rotation is zero. If the date is October
22 and the observer is 5 E of the center of the observer’s time zone, at what time
(s) should the observer make the observation?
A. From the 30 N curve in Fig. 6.5, the hour angles for zero field rotation at 20
declination are about 4 h.
From (4.12) the LST when the object is 4 h before or after the meridian are
LST ¼ RA þ HA ¼ 4h 4h ¼ 00h or 08h:
(continued)
(continued)
From Table 4.2 the LST at midnight on October 22nd is 02 h.
Roughly therefore (taking a sidereal hour to equal a solar time hour)
LST ¼ 00 h corresponds to about 10 pm civil time
and
LST ¼ 08h corresponds to about 6 am civil time:
However because the observer is 5 E of the center of the time zone, the hour
angles of 4 h will occur 20 min earlier than at the center of the time zone. The
required times for the observations are thus about 9 h 40 min p.m. and 5 h 40 min a.
m.
Since in October it will still be dark at 5 h 40 min a.m., the observer can
obtain images at either time.
Fig. 6.5 Times of zero field rotation for alt-azimuth mounted telescopes for various observers’
latitudes and objects’ declinations. To use this graph, (i) select the graph nearest to your latitude
(or interpolate if you wish for higher accuracy), (ii) at the declination (vertical axis) of the object
that you wish to observe draw a horizontal line across to meet the curve, (iii) at the intersections of
this line with the curve draw vertical lines down to the hour angle (horizontal) axis, (iv) read off the
hour angles from the intercepts of those lines with the hour angle axis when the field rotation for
your selected object will be zero, (v) determine the civil times corresponding to those hour angles.
A planisphere or celestial globe will probably suffice for this purpose, or you could use (4.12) and
Table 4.2 to find the local sidereal times and then convert these to civil times.
Alignment 137
Other Types of Commercial Mountings
Rarely a telescope may be put onto a mount that is of neither the equatorial nor the
alt-az design. Some telescopes, such as the 300-m radio dish at Arecibo and
telescopes with liquid mirrors,2 are fixed and point towards the zenith. Objects may
then be observed and tracked for a short time by moving the detector around the field
of view. The Hobby-Eberly and SALT telescopes can rotate in azimuth but have fixed
altitudes of 55 (HET) and 53 (SALT). Again objects can be observed with these
telescopes for a short time by moving the detectors. A very few radio dishes can be
moved in altitude in two mutually perpendicular planes (an alt-alt mounting). Finally
some telescopes, especially specialist solar telescopes, are fixed in position, and light
is fed to them by a pair of driven flat mirrors called a coelostat (see Chap. 8).
Making Your Own Mounting
Just as it is possible to make your own optics for a telescope to a standard as good as or
better than many commercially produced items (Chap. 3), so it is possible for a
competent DIY enthusiast to produce a more than adequate mounting. There is no
standard approach, however, to producing a mounting that is comparable to the
grinding and polishing of the optics (which varies little in its essentials from one
telescope to another). The intending builder of a telescope mount should take to heart
the requirements for a mounting (described above), see as many different designs as
possible, perhaps by joining a local astronomical society, consult the frequently
published articles in the popular astronomy journals on the subject, and then finally
produce a design suited to his or her needs, abilities, available equipment and materials.
Alignment
An equatorial mounting needs to be aligned so that its polar axis is precisely parallel
to Earth’s rotational axis if it is to track objects well, and if its setting circles are to be
accurate. For a permanently installed telescope, this alignment only has to be done
once, and so the observer can afford to spend sufficient time to get it done well. For a
portable telescope, the alignment will have to be done each time the telescope is set
up, and will therefore normally be accomplished to much lower accuracy.
If the telescope has a cross-wire eyepiece and setting circles, the initial prepara-
tion can be done by setting the telescope to the coordinates of the pole star, then
finding the pole star and centering it on the cross-wires by moving the polar axis in
2
A liquid mirror telescope uses a rotating bath of mercury as its primary mirror. NB mercury and
its compounds are highly poisonous – do not attempt to make such a telescope yourself.
azimuth and altitude. Even without setting circles, aligning the telescope by eye to
be parallel to the polar axis and then finding Polaris will set the mounting up to
within a degree or two. Some commercially produced telescopes are provided with
a rifle sighting scope, which may be attached to the mounting in order to sight on the
pole star in a similar way. An arrangement of this sort is almost essential for a
portable telescope if considerable amounts of time are not to be wasted every time it
is used.
A permanently installed telescope mounting can be aligned much more precisely
after this initial setting up, using observational tests. Corrections to the altitude of
the polar axis are found by observing a star between 30 and 60 declination, about
6 h east or west of the observer’s meridian. The star should be centered on the cross-
wires of a high-power eyepiece and tracked using the telescope drive.
If, after a while, a star to the east of the meridian is observed to drift northwards
in the eyepiece, then the altitude of the polar axis is too high. A star to the west of
the meridian would drift to the south in the same circumstances. If the easterly star
drifts south, or the westerly star north, then the polar axis is set at too low an
altitude. After a suitable adjustment to the axis, the procedure is repeated until any
remaining drift is acceptable.
A similar procedure may be used to align the polar axis in azimuth. Two stars are
selected that are 10 or 20 north and south of the equator, and differing in right
ascension by a few minutes of arc. (Reference to a reasonable star catalog will
probably be necessary for this – see Appendix A of this book.) When the stars are
within an hour or so of transiting the observer’s meridian, with the drive off, the
telescope is set just ahead of the leading star (the one with the smaller right
ascension). A stopwatch is started as that star transits the vertical cross-wire. The
telescope is then moved to the declination of the second star, and the time interval
until that star transits is measured. The correct time interval between the two
transits can be found from the difference in their right ascensions. If the measured
time interval is too small, then the polar axis is aligned to the east; if the time
interval is too large, then the polar axis is to the west of its true direction.
Adjustments to the position of the polar axis are made, and the procedure again
repeated until satisfactory. Alignment will clearly have to be done much more
precisely if the observer intends, say, to undertake photography with exposures of
several hours rather do than visual work.
Setting Circles
With a well-aligned equatorial mounting, 360 protractors called setting circles

suitably graduated can easily be mounted to allow the position of the telescope in
the sky in terms of right ascension or hour angle and declination to be measured;
these are then a great aid to finding objects in the sky. The declination can be read
simply from a protractor graduated from 0 to 90 attached to the telescope and
with a pointer fixed to the mounting (or vice versa – see Fig. 6.6).
Guiding 139
Fig. 6.6 Declination circle
Even without RA or HA setting circles, the declination setting circle is very

useful. The telescope can be set accurately to the declination of the required object,
and pointed roughly in the right direction in RA by looking at the constellations.
The telescope can then be swept in RA, without changing its declination, and the
required object will often be found rapidly.
Hour angle is equally easily determined using a protractor graduated from 0 to
24 h and attached to the polar axis of the mounting. Unfortunately, the hour angles
of objects vary with time and longitude, and the correct setting has to be calculated
(4.1, 4.2, and 4.12) every time it is wanted from the local sidereal time. A setting
circle reading RA directly is therefore much more useful. This also is a protractor
graduated from 0 to 24 h, but driven at sidereal rate to allow for the motion of the
sky (Fig. 6.7). It will normally need to be set at the start of an observing session.
Its correct setting can be calculated from first principles, but usually it is simpler to
find a bright object with a known right ascension in the telescope, and then to set the
RA circle to read that right ascension correctly. Thereafter, the circle will read RA
directly, unless its drive is turned off at any time. Some modern small mountings
use digital readouts for the telescope’s position, but the principle does not differ
from the use of the circles.
Guiding
During long exposures, it will be found with most mountings and drives that the
object being observed wanders about in the field of view. This may be due to
inadequate alignment of the mounting (see above), to changing flexure in the
Fig. 6.7 RA circle
mounting, or to non-uniform or incorrect drive speeds. Even a perfect mounting and

drive would not track an object precisely over several hours, though, because
refraction in Earth’s atmosphere changes the apparent positions of objects in the
sky (5.7). For visual work, such inaccuracies in the tracking are unimportant unless
their magnitude is huge; but if images are being obtained by photography or a CCD
detector, or the light is being fed into a photometer or spectroscope, then the
inaccuracies have to be corrected or the images will be trailed, or the light lost
from the instrument. Such correction of the drive is known as guiding.
Conventionally, guiding is undertaken by the observer, who looks at the object
through a second telescope attached to the main instrument called the guide tele-
scope. With a cross-wire eyepiece in the guide telescope, the observer manipulates
the slow-motion adjustments of the mounting to keep the object centered on the
cross-wires throughout the exposure.
This procedure requires considerable practice for success and can become
tedious during long exposures. It also requires the guide telescope to be not much
smaller than the main telescope, or the fainter objects will not be visible. For these
reasons, many observers use autoguiders of various designs. An autoguider is a
device that can detect the object being observed, and then generate an error signal if
the object wanders away from its correct position. The error signal is used to
operate the slow motions in RA and Dec in such a manner that the object is brought
back to its correct position. The autoguider may be attached to the guide telescope,
or the field of view of the main telescope may be sampled, and a nearby object to the
one of interest used for guiding. This latter process of off-axis guiding is also often
useful for fainter objects generally, when a nearby brighter object can be selected
for easier guiding. Autoguiders are available commercially and are advertised in the
popular astronomy magazines (see Appendix A of this book).
With CCDs, another approach is possible and is based upon the ease with which
many CCD images may be added together. An exposure is chosen that is suffi-
ciently short, so that unguided, the images show no trace of a trail. Multiple images
Modern Commercial Mountings 141
are then obtained with this basic exposure, aligned with each other and added
together to give a single image with the required total exposure. Another approach
adopted for some small commercial CCD astronomical cameras is to use two CCD
chips, one of which is used to obtain the required image, the other to autoguide on
an off-axis object.
Modern Commercial Mountings
Commercially produced telescopes (Chap. 3) are invariably provided with

mountings. Though the optics of such instruments are generally excellent, they
often leave something to be desired in terms of the quality of their mountings.
At the worst there are the “single pivot,” “ball and claw” and related types of
mounting (Fig. 6.8), which should be avoided completely. Fortunately this type
of mounting is now less commonly found than in earlier times. Variations on the
mounting designs discussed above are to be found for most commercial telescopes,
with the German and fork designs probably being most popular. Unfortunately, they
are often too lightweight for the size of telescope they are required to carry. This
does not mean that they are likely to break, but that vibrations will be only slowly
damped out, thus limiting their usefulness for photography or CCD imaging.
Mountings have generally become heavier and more stable in the last few years,
and with some manufacturers, heavy-duty German mounts are now available as
options that will provide a reasonable level of rigidity. Nonetheless, the prospective
purchaser of a telescope should still pay close attention to the mounting and make
use of any opportunity to try it out in practice.
Many modern mountings are linked to small computers, which then provide the
readouts of the RA and Dec of the telescope. Usually the positions of thousands of
Fig. 6.8 Single pivot

mounting (to be avoided!)
interesting objects in the sky are also stored on the computer, and these can therefore
be found rapidly in the telescope. Slight changes to the drive speed from the sidereal
rate to match the motions of the Sun and Moon are also often available.
In at least one case, the computer can be used to compensate for inaccurate
alignment. Two stars are found in the telescope, and their details fed into the
computer that thereafter will correct the displayed positions to correspond to
the actual point in the sky that the telescope is observing. With an alt-azimuth
mounting, it is usually necessary to set onto three stars at the start of the observing
session so that the computer can calculate the telescope’s position thereafter.
Recently GPS receivers have been incorporated into some of these systems so
that the observer’s position on Earth is automatically registered as well. Robotic
systems that set themselves up completely without any intervention from the
observer apart from pressing the ‘on’ switch are just coming onto the market at
the time of writing.
Exercise
6.1. An observer at a latitude of 50 N desires to image an object whose position is
a ¼ 17 h, d ¼ +45 at a time when the field rotation is zero. If the date is
February 5 and the observer is 3 W of the center of his or her time zone, at
what time(s) should the observer make the observation?
Part III
Observing
Chapter 7
Electromagnetic Radiation
Introduction
Almost all our information in astronomy is obtained by the electromagnetic radiation

traveling from the object to the observer. Apart from those few objects within the
Solar System that we have been able to investigate directly, cosmic rays, neutrinos, in
due course, gravity waves and just possibly (if they exist) dark matter and dark energy
are the only other information carriers likely to tell us about the universe as a whole.
The electromagnetic radiation wave consists of a magnetic wave and an electric
wave whose directions are orthogonal and which vary sinusoidally. The frequency
of the sinusoidal variation is called the frequency of the wave and denoted usually
by n. The separation of successive crests or troughs gives the wavelength, l. The
product of n and l gives the wave’s velocity. In a vacuum this has a constant value,
denoted by c, of 299 792 500 m s1.
ln ¼ c (7.1)
In other media the velocity is reduced from c by an amount given by the

refractive index, m, of the material:
c
v ¼ (7.2)
m
Since even materials normally regarded as opaque have less than an infinite
absorption, light penetrates them to a certain extent, and so they have a refractive
index. We can thus talk, for example, about the velocity of light in a brick!
In a medium where the electromagnetic radiation has a velocity less than c, the
frequency remains constant, so from (7.1), with v replacing c, the wavelength
becomes shorter.1
1
An effect used by oil immersion optical microscopes to increase their resolution.

146 7 Electromagnetic Radiation
As is well known, the special theory of relativity postulates that the velocity of
electromagnetic radiation in a vacuum is the maximum possible for any normal
particle, and experiments have confirmed this many times. However, because of the
reduction in velocity given by (7.2), in any medium other than a vacuum it is
possible for, say, a particle to exceed the local velocity of light. When the particle is
charged, as for example is the case with most cosmic ray particles, a type of
radiation is produced known as Čerenkov radiation. This is the electromagnetic
equivalent of the sonic boom of a supersonic aircraft. It is of significance in that it
results in noise spikes in CCD and other detectors (see Chap. 10).
Intensity
The equation for the electric component of an electromagnetic wave is

2px
Eðx; tÞ ¼ E0 sin þ 2ptn þ f (7.3)
l
where E(x,t) is the magnitude of the electric vector at position x and time t; Eo is the
amplitude of the wave and f is the phase at t ¼ 0, x ¼ 0.
The intensity of the radiation is given by Eo2. It has units of energy per unit area
and per unit wavelength or frequency interval, and for convenience various differ-
ent units are used over the spectrum. Thus, in the radio region, the units are janskys
(1 Jy ¼ 1026 W m2 Hz1) while at shorter wavelengths, the units are just
W m2 Hz1. At the shortest wavelengths (X-rays and g rays), electron volts
(1 eV ¼ 1.6022 1019 J, the energy that an electron gains in falling through
1 V) are often used for the energy of one photon.
Determination of the intensity of the electromagnetic radiation from a source is
the most fundamental measurement made in astronomy. Even for visual work,
although we do not normally express it in this fashion, an image is registered because
the eye determines different intensities falling onto different parts of the retina.
Photons
Although for many astronomical purposes we can regard electromagnetic radiation

as a wave, it actually has a dual nature and sometimes behaves as a particle, known
as a photon or quantum of light. The converse is also true: electrons, which we
usually regard as particles, can sometimes behave as waves, allowing among other
things the production of electron microscopes.
For astronomy at visual wavelengths, the particle nature of light is mostly of
significance in the way most types of detectors work. Thus the eye, photographic
emulsion, CCD and photomultiplier all detect photons rather than waves. Quantum
Range 147
mechanics is required to deal with the complete situation; here we just note that
when behaving as a particle, the photon of radiation has an energy, E, given by
hc
E ¼ hn ¼ (7.4)
l
where h is Planck’s constant (6.6262 1034 J s).

The units for photon energy are joules, but, as noted above, electron volts are
often used because they lead to more convenient numbers. Thus the visual spectrum
ranges from 1.8 eV (red) to 3 eV (blue).
Polarization
A second fundamental property of radiation, but one less commonly studied in

optical astronomy, is the state of polarization of the radiation. If the direction of the
electric vector (or magnetic vector) in space in a beam of radiation changes
randomly on the time scale of 1/n, then the radiation is not polarized. If the direction
of the electric vector is constant, then the radiation is plane polarized. If the
direction rotates at the frequency of the radiation, it is circularly polarized (clock-
wise or counter-clockwise, according to the direction of rotation of the vector).
If the amplitude also varies at the frequency of the radiation, then we have ellipti-
cally polarized radiation (the tip of the electric vector would trace out an ellipse if
you were to look down the beam). Most radiation from astronomical sources is not
polarized, or nearly so. Some sources, however, may be highly polarized (for
example, the Crab Nebula and active galaxies), and much information can then
be gleaned from the details of the polarization.
Range
Electromagnetic radiation extends, in theory, from an infinite wavelength (n ¼ 0,

E ¼ 0) to a zero wavelength (n ¼ 1, E ¼ 1). In practice, we can detect from
wavelengths of a few tens of kilometers to 1014 m or so. By tradition and for
convenience, different parts of the spectrum are labeled by different names
(Fig. 7.1), although the nature of the radiation is the same at all wavelengths.
The nature of the interaction of radiation with matter, however, does change in
different parts of the spectrum, and this results in the differing detectors and
telescopes used by radio, optical and X-ray astronomers. It also means that we
can obtain different types of information about the universe from different parts of
the spectrum. Thus, as a generalization, at long wavelengths, we have direct
induction of electric currents, free-free and synchrotron radiation; in the microwave
and infrared regions, the interaction is with rotating and vibrating molecules; in the
148 7 Electromagnetic Radiation
Fig. 7.1 Range

of electromagnetic
radiation
optical region, it is with the outer electrons of atoms and molecules; in the ultravio-
let and X-ray region, with inner electrons and ionizations; and at the shortest
wavelengths, directly with nuclei. Thermal emission can occur at all wavelengths
but is generally more important at longer wavelengths.
Astronomy started off with visual observations because that is the spectral
region to which our eyes are sensitive, and because the atmosphere is reasonably
transparent in that region. Observations of the Sun in the infrared and ultraviolet
regions were already being made in the early nineteenth century and of the Moon
and other objects 50 years later, again because the atmosphere is reasonably
transparent in those regions. The radio region was developed from the 1940s, yet
again because the atmosphere is reasonably transparent in that region. All other
spectral regions had to await balloons, rockets and spacecraft from the 1960s
onwards for their exploitation because there the atmosphere is opaque (Fig. 7.2).
Measurements
From (7.3) we may see that the only parameters of electromagnetic radiation that
we may measure are its amplitude (Eo), wavelength or frequency, phase, plus its
direction and state of polarization. Only in the radio region are all these quantities
routinely determined. At all shorter wavelengths, the measurements are generally of
intensity (Eo2), and wavelength, plus occasionally linear polarization. Observations
are divided on a practical basis, however, into several distinctive areas.
Polarimetry 149
Fig. 7.2 Atmospheric transparency
Photometry
This is theoretically the measurement of intensity at a single wavelength. In practice,

it is the intensity integrated over the pass-band of the filter being used. Separate
photometric measurements may be made at several different wavelengths, which in
the limit as the number of filters increases and the pass-band widths decrease
approaches spectroscopy. The details of photometry are covered in Chap. 11.
Numerous photometric measurements over a two-dimensional field (i.e., measuring
direction as well) give us an image of that field.
Spectroscopy
This is the measurement of intensity as a function of wavelength; as previously

mentioned, it is the limiting case for photometry over numerous pass-bands.
Occasionally the spectrum may be obtained simultaneously at a number of points
over the field, usually leading to a transect of the object in the field of view (long slit
spectroscopy). Also many spectroscopes on large telescopes now obtain the spectra
of many objects simultaneously (multi-object spectroscopy) or of everything over
the entire field of view (integral field spectroscopy). Spectroscopy is considered in
more detail in Chap. 12.
Polarimetry
This is the measurement of the degree and direction of linear polarization, or, much
more rarely, of the state of elliptical polarization. The measurements are often
extended over two dimensions to give a polarization map of an object. It is a highly
specialized technique and is not considered further in this book.
Chapter 8
Visual Observing
Introduction
In this chapter it is assumed that the reader has access to a small- to medium-sized
telescope (10–30 cm, or 4–12 in.) on a motor-driven mounting. For visual work the
field rotation inherent to alt-az mounted telescopes (Figs. 6.4 and 6.5) matters little.
If the observer intends to go on to do significant amounts of imaging, though, an
equatorial mounting will be the significantly better choice (Chap. 9).1
Descriptions of how to find and track objects in the sky are based upon what is
needed for a basic mounting with just simple setting circles. Owners of instruments
with computer-controlled mountings will probably not need much of the advice that
is given here – although when confronted with finding objects not in their
computer’s databases, such as novae, supernovae or comets, such observers will
still have to resort to the old-fashioned methods. If the telescope is smaller than
10 cm, then many of the fainter objects will be difficult or impossible to see. If the
telescope is larger than 30 cm – congratulations! Other types of observing such as
imaging, photometry and spectroscopy are considered in later chapters.
Visual work is an aspect of imaging (Chap. 9); nowadays it is sometimes
regarded as inferior to methods providing a permanent record. It certainly can be
a subjective process (see later), and is limited in the wavelengths and intensities
detectable. Nonetheless, with care and practice it can still give very valuable
results; after all, for half the time since their invention, telescopes could be used
with no other type of detector than the eye. Furthermore, one of the principal joys of
astronomy is seeing for yourself the mysteries and wonders of the universe, and it
1
Some, but not all, commercially produced alt-az mountings can be converted to equatorials
simply by attaching them to a wedge whose apex angle equals 90 minus the observer’s latitude.
The vertical (azimuth) axis then becomes the polar axis and the horizontal (altitude axis) the
declination axis. Computer-controlled mountings are likely to have an equatorial option available
or even automatically to recognize their orientation during setting-up.

152 8 Visual Observing
should never be forgotten that one of the great strengths of the science is that the
vast majority of astronomers study the subject for pleasure and interest (how many
amateur solid-state physicists do you know?).
General and Practical Considerations and Safety
Whether in an observatory or not, telescopes are often used in damp conditions, and
obviously at low levels of illumination. It is therefore advisable for all equipment to
be of low voltage. Any high voltages should always be taken through rapid circuit
breakers (RCBs – also known as Residual Current Devices [RCDs], Residual
Current Circuit Breakers [RCCBs], Ground Fault Circuit Interrupters [GFCIs],
Ground Fault Interrupters [GFIs], Appliance Leakage Current Interrupters [ALCI]
and trip switches) to avoid the possibility of electrocution.
Other safety considerations are to wear appropriate clothing, never to rush
around the site, and if not working with someone else (see below) to let a responsi-
ble person know when and where you will be observing and when you expect to
finish. That person should have instructions to take appropriate action if you do not
return on time. Heavy items, such as counter-weights, should be handled carefully
and attached securely to the telescope.
Certain factors apply to all types of observations. One of the most important of
these is the selection of observing site. This has already been discussed in Chap. 2.
Here, there is just the reminder that if there is any choice in the matter, the site
should be as well away from artificial light sources as possible. In choosing a site
for, say, a small portable telescope, the question of personal safety should not be
neglected, since being away from artificial light will usually mean that the site is
isolated. It is a sensible precaution therefore to take a companion in such
circumstances, to carry mobile phones and always to obtain prior permission if
using private ground.
If the telescope is not in some sort of observatory, then movable screens to
protect it from the wind and perhaps to obscure lights will prove to be very useful.
A commonly encountered problem is that of misting of the optics, or dewing up, as
it is known. If observations seem to be becoming more difficult, always check to see
if dew is forming, since even a very thin layer will ruin the image. The dew should
never be wiped off the optics, since coatings can be damaged and surfaces
scratched. In the cases of refractors and Schmidt-Cassegrain telescopes, a long
extension of the telescope tube beyond the objective, known as a dew cap, can
reduce the formation of dew. It is also possible to add a low voltage heating cable
around the objective, but if this is too powerful it may induce convection currents
and cause the images to deteriorate. If these precautions fail, then a low voltage hair
drier, of the type sold for campers, can be used to evaporate the dew. Although this
also runs the risk of inducing convection currents, they are likely to die away
shortly after the drier has ceased to be used. In heavy dewing conditions, it may be
necessary to clear the optics at intervals as short as a few minutes.
Finding Objects 153
Even on clear nights, images will often be poorer than the diffraction limit of the
telescope. This is due to turbulence in the atmosphere causing the image to
scintillate, an effect also known as twinkling. The magnitude of the effect is
known as the seeing. From the very best observing sites under the very best
conditions, it can mean that a stellar image is about 0.2500 across. The best that
can be expected from an average site is about 100 , and it will more typically be
200 and on poor nights as much as 500 .
As discussed in Chap. 2, the choice of eyepiece is important. As a general rule,
the lowest power that will reveal the objects of interest should be used. The
maximum usable magnification will reduce sharply as the seeing deteriorates,
since except under the very best conditions, high powers will give fuzzy images
that lack detail. Eyepieces can also dew up, and the observer should be careful to
avoid breathing on them. The problem can be reduced if eyepieces not in use are
kept in individual lint-free cloth bags in an inside pocket, so that they are warmer
than their surroundings when brought out for use.
Finding Objects
Before anything can be observed, the telescope must be pointed in the right
direction – that is, the object must be found. If the mounting has setting circles
(Chap. 6), and the right ascension (or hour angle) and declination (Chap. 4) of the
object are known, then the telescope can be set directly on to the object. Some
telescopes are linked to computers that store the positions of numerous objects and
can therefore be set very quickly. If such a device is not available, then the positions
of the stars, galaxies, etc., can be looked up in catalogues or read off star charts
(Appendix A of this book). A planisphere is useful for ascertaining quickly which
constellations are in the sky at a particular time of the night throughout the year.
The positions of the Sun, Moon and planets can be found for the date of
observation from the Astronomical Almanac, or from the popular astronomical
journals (Appendix A of this book). The positions of ephemeral objects such as
comets and novae are sometimes listed in the popular journals or even in
newspapers. The national astronomy societies and the IAU (International Astro-
nomical Union) may also send out details, though sometimes this requires an
additional subscription. The observing details for a new object such as a comet
will also often be found quickly by a web search.
If the mounting has been set up correctly and the setting circles are accurate, then
the object should be in the field of view. However, most setting circles on small
telescopes can only be set to a quarter of a degree or so. The eyepiece that gives the
widest field of view with the telescope, normally the lowest power eyepiece
available (2.8), should therefore always be used for finding. If the main instrument
has a finder telescope, then this will normally have an even larger field of view and
should be used before trying the main telescope. If the object is not in the field of
view initially, then it will usually be possible to find it by nudging the telescope by
small amounts to search around the area. If it still does not come into view, then
check for something being amiss. Common problems include:
• Forgetting to take off a mirror or lens cover;
• One or more of the optical components being covered in dew;
• Misreading the setting circles;
• Looking up the details of the wrong object in the catalog;
• Miscalculating the sidereal time (do not forget to subtract any summertime
corrections to the local civil time);
• Not correcting or correcting the wrong way for precession if the data are from an
old source;
• Misalignment of the mounting.
Once the object has been found, then it may be centered in the eyepiece by
adjusting the position of the telescope, and higher-power eyepieces substituted if
required.
If setting circles or the accurate position of the object are not available, then it
must be found by a search. This is not difficult for any object bright enough to be
seen with the naked eye. The telescope is roughly aligned towards the object
by sighting along its tube. The object should then be visible in the finder, or if
not, findable by a search around the area while looking through the finder.
Centering the object in the finder should then enable it to be found in the main
telescope.
For objects too faint to be seen with the naked eye, but bright enough to be seen
through the finder telescope, the procedure is similar, but the initial setting will
have to be done by estimating where to point the telescope by comparison with a
nearby object bright enough to see. For objects visible only in the main telescope, a
finder chart needs to be drawn from a star chart or catalog. Sometimes ready-
prepared charts will be published in the popular astronomy journals for comets, etc.
There are also several books that contain finder charts for the Messier objects, and
some other nebulae (Appendix A of this book). The finder chart should show the
pattern of stars around a nearby bright object, especially those along the line to the
object of interest (Fig. 8.1). The bright object is then found, and the star pattern
identified.
Remember most telescopes give an inverted image, so for this purpose it is
useful to draw the finder chart on tracing paper so that it can be viewed from either
side. Also remember that the field of view will rotate as the telescope points to
different parts of the sky when it is on an alt-az mounting. The telescope is then
moved along the star pattern until the desired object comes into view, a process
sometimes called star hopping. For this sort of procedure it is useful to know the
size of the actual field of view of the eyepiece in the sky so that the finder chart is of
the appropriate scale. This can be calculated (2.8) or measured. To measure the field
of view, set the telescope on to any object whose declination is known, position that
object to one side of the eyepiece and turn off the telescope drive. The time taken
for the object to drift centrally across the eyepiece should be measured. (If the
object is set to the wrong side of the eyepiece, so that it drifts out of the field of
The Moon 155
Fig. 8.1 Schematic finder

chart
view, just bring it back and set it to the other side.) The angular size of the field of
view is then given by
Field of view ¼ 4:2 103 t cos d degrees (8.1)
where t is the time, in seconds, for the object to drift across the field of view.
It is helpful to note prominent asterisms (easily recognizable groups of stars) on
the star chart and move from one such to the next to ensure positive identification at
each stage of the route, or it is easy to get confused and lose the way.
The Moon
Although familiar to everyone and sufficiently close and bright for some details to be
seen with the naked eye, the Moon remains one of the most rewarding objects for
telescopic observation. It is ideal as a first object to look at, being easy to find, and with
a wealth of features discernible by even the most inexperienced observer. It is also an
object that may be observed from the worst imaginable observing sites, even through
an open window from an apartment in the center of a brightly lit city, provided
only that the window faces in the right direction. Yet at the same time, it is an object
worthy of continuing and detailed study. For despite the various spacecraft missions
to and around the Moon, there remain puzzles about it that deserve attention.
All the features visible on the Moon from Earth have long ago been mapped and
named, and as an initial task the observer should become familiar with a few of the
more prominent ones. These would be the lunar maria and some of the larger
craters, and they will then provide reference points for finding more obscure
features. For this purpose a map of the Moon will be needed.
Care needs to be taken in the choice of such a map for two reasons. First, in most
astronomical telescopes, the image is inverted (Chap. 2). The Moon is therefore
seen with its south pole at the top of the field of view. Most maps of the Moon
published in astronomical sources show the Moon in this orientation, and so are
easy to correlate with what is seen in the telescope. Other sources, however,
especially those with space-related interests, plot maps of the Moon as it would
be seen with the naked eye, with the north pole at the top. Such maps are very
confusing as references for observing and should be avoided.
The second problem is that the appearance of the lunar surface changes very
considerably with the phase. Thus near full Moon, many of even the largest craters
are almost invisible (see Fig. 5.26). Small craters may only be detectable for 1 or
2 days at exactly the right phase when their shadows make them prominent. For this
reason, some lunar maps show features over the whole surface with a constant angle
of illumination (i.e., with an identical altitude for the Sun as seen from each point on
the Moon). They are therefore quite misleading as a guide to the appearance of the
whole Moon, though useful for more detailed studies.
Craters are by no means the only type of object visible on the Moon; there are
the various other features, known as rilles, domes, scarps, valleys, mountains,
lava flows, crater rays, etc. To become reasonably familiar with the Moon is
therefore a task likely to take a year (or a lifetime) of dedicated observing. During
such a program it is of interest to determine the smallest objects that can be
discerned. Under the best observing conditions, these are likely to be about 1 km
or so in size. But, as already mentioned, small features will often only be
detectable for a short time while the shadows are exactly right, and since such
a time must also coincide with very good observing conditions, many small
objects may only actually be visible on one or two occasions throughout the
whole year.
It can also be of interest to try and detect the new Moon as early as possible, or
the old Moon as late as possible. It might seem that the Moon should be detectable
right up to the instant of new Moon, but because it is then close to the Sun in the sky,
it is swamped by the scattered solar light. Detections within 30 h either side of new
Moon can generally be regarded as an above average achievement.
Lunar eclipses (Chap. 5) do not have the same significance as solar eclipses but
can nonetheless be interesting to observe. The Moon is, of course, full at the time of
an eclipse, and so few craters will be visible. The main interest is in the details of
Earth’s shadow. Depending upon atmospheric conditions and events such as major
volcanic eruptions, this can vary from almost black, so that the Moon disappears, to
quite a bright reddish hue. The color in the latter case is due to light being scattered
in Earth’s atmosphere towards the Moon.
An apparently somewhat similar phenomenon is “the old Moon in the new
Moon’s arms.” When the Moon is a narrow crescent, the dark portion can some-
times still be seen. This, however, is due to reflected light from the illuminated part
of Earth reaching the Moon (Fig. 8.2). If you were actually on the “dark” part of the
Moon at such a time, then since Earth is over five times as reflective as the Moon
and three and a half times larger, the amount of light received would be about 70
times as much as we receive from the full Moon – more than bright enough to work,
read and move around the surface.
There are several aspects to lunar observing suitable for more advanced work.
As mentioned in Chap. 5, though the Moon nominally keeps the same portion of its
The Moon 157
Fig. 8.2 The old Moon in the new Moon’s arms
surface towards Earth, its speed around its orbit varies owing to the ellipticity of the
orbit, and so sometimes its rotation gains or loses some of its orbital motion. We can
thus see about 59% of the total surface area of the Moon. This slight wobble is
known as libration, and observing the limbs of the Moon at times of maximum
libration enables glimpses to be had of what is normally the far side. The heights
and depths of features on the Moon can be measured relative to their immediate
surroundings quite simply (Fig. 8.3). Ideally a micrometer eyepiece (Chap. 2) is
used to measure the angular size of the shadow of the object, and its distance from
the lunar limb. However, in the absence of such an eyepiece, an ordinary cross-wire
eyepiece, normally used for guiding, can be used instead. The angles are then
measured by drifting the image with the telescope drive turned off, timing how
long it takes to pass over the cross-wire and converting the time into an angle using
(8.1), or more conveniently in seconds of arc:
Angular distance ¼ 15t cos d seconds of arc (8.2)

Shadow of
Central
Mountain
range
Crater
D
P
R
Fig. 8.3 Finding the height of a feature on the Moon
The height of the feature in meters is then given in terms of the angles shown in
Fig. 8.3 by
" pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi#
R D ðR PÞ ð2DR D2 Þ
Height ¼ 1900L pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi meters (8.3)
R R ð2PR P2Þ
where L, D, P and R are in units of seconds of arc.

The Moon 159

Q. The central peak of a lunar crater is observed to have a shadow with a length
(L) of 0.500 . The peak is 80 from the limb of the Moon (D), and the distance of the
terminator from the limb is 120 (P). What is the height of the mountain? (Take
the lunar angular radius (R) to be 15.50 ).
A. In units of seconds of arc, we have:
R ¼ 93000
P ¼ 72000
D ¼ 48000 L ¼ 0.500
Thus
(2DR D2)0.5 ¼ (892,800 230,400)0.5 ¼ (662,400)0.5
¼ 813.9 (2PR P2)0.5
¼ (1,339,200 518,400)0.5 ¼ (820,800)0.5 ¼ 906.0
From (8.3) therefore
Height of central mountain
¼ 1,900 0.5 {[(930 480)/930] [(930 720)/930]
[813.9/906.0]}
¼ 950 {0.4839 0.2258 0.8983}
¼ 950 0.2811
¼ 267 m
¼ 250–300 m given likely errors of measurement.
One controversial area in which visual observations can be valuable is that of

transient lunar phenomena, or TLPs. There have been frequent claims of
observations of short-term changes to features on the Moon, but many astronomers
remain skeptical. The typical TLP consists of an apparent veiling, as though by a
thin mist or dust cloud, of a small area on the Moon. The observation is generally
therefore that the feature is less clearly defined than usual rather than a specific
alteration, which accounts for some of the skepticism.
However, on a few occasions, the obscuration has been much more pronounced,
and light emissions have sometimes been claimed. The events cluster in a few areas
such as around the central mountains of the craters Aristarchus and Alphonsus.
They are perhaps more common when the Moon is near perigee, which could be
due to the increased tidal stresses at that time. If they occur, then they may be due to
trapped gases making their way to the surface and producing a short-lived dust
cloud at the surface from dust entrained along the way. If the light emissions are
real, then these could be glow discharges produced by the build-up of static
electrical potentials (i.e., the lunar equivalent of lightning). The whole subject is
a fascinating one, and an area where observations made with small telescopes can
potentially make a real contribution to science.
Occultations (Chap. 5) of stars by the Moon are frequent because of its large
angular size and rapid motion across the sky. They are studied using high-speed
photometers (Chap. 11) as a means of determining a star’s angular size, and to
obtain details of close double stars. However, the rapid flickering upon which such
measurements are based will generally not be detectable by the eye. On some
occasions, though, especially if the relative motion is close to parallel to the limb
of the Moon, it may be possible to see the double dimming as a double star goes
behind the Moon, and very occasionally the star may reappear briefly when seen
through a valley at the edge of the Moon. Apart from these events, the main interest
in observing occultations is the way in which the rapid motion of the Moon across
the sky is dramatically revealed.
The Planets
Like the Moon, the brighter planets are generally easy to find, especially if the
observer has some familiarity with the constellations. Jupiter and Venus, near
opposition and at greatest elongation, respectively, are far brighter even than Sirius,
the brightest of the stars in our sky. Mars and Saturn are also bright enough to be
readily found most of the time.
Mercury, however, is quite a difficult object to locate. Although occasionally as
bright as Sirius, it is always close to the Sun in the sky; in addition, its orbit is quite
elliptical, so that its greatest elongation can vary considerably but has a maximum
value of only 28 . For this reason, Mercury is often also known as the evening or
morning star, because it can be best seen soon after sunset when it is to the east of the
Sun, and just before dawn when it is to the west of the Sun. In either case, it has to be
observed against a bright twilight sky, and with the added haze and light pollution
that is present at most urban or semi-urban observing sites, it then becomes almost
impossible to see with the naked eye. Thus Mercury generally requires a telescope
with accurate setting circles in order to be found. IF OFF-SETTING FROM THE
SUN, THEN ALL THE PRECAUTIONS MENTIONED IN THIS BOOK IN
RESPECT OF SOLAR OBSERVING MUST BE FOLLOWED.
Uranus is just bright enough to be seen with the naked eye at times. However,
this would only be from a good observing site, under good conditions and for an
observer with good eyesight. Neptune, Pluto and the asteroids are all too faint to be
seen without a telescope, and so along with Uranus must usually be found from their
tabulated positions using setting circles or by star hopping. The positions of the
planets and the brighter asteroids may be found from several sources, and these are
listed in Appendix A of this book. Most computer-controlled mountings can
calculate the positions of the planets and sometimes of the brighter asteroids and
so set directly onto them.
The observer must be prepared for some disappointment on first observing a
planet through a small telescope. This is for two reasons. First, articles and books on
astronomy are often filled with magnificent pictures of the planets obtained from
The Planets 161
Fig. 8.4 Diffraction-limited

image of a star from a
reflecting telescope that has a
secondary mirror supported
by a cross
spacecraft. Not only are these obtained without the blurring effect of Earth’s
atmosphere and from close to the planet, but also they are often computer-enhanced
to make the colors and contrasts stronger. Sometimes, indeed, the image may be
completely false color, either a monochromatic image with the different intensity
levels being given different colors or an image obtained partially or wholly in the
ultraviolet or infrared and then visible colors substituted for those wavelengths. If
such images are what the observer expects to see, then reality will be rather
different. The second reason is that visually observing the planets requires a skill
that can only be developed with practice. The skill required is that of taking
advantage of the brief instants of clarity in the atmosphere to note the features of
the planet.
The turbulent effects of the atmosphere that degrade stellar images (see seeing,
above) also affect planetary images. Since the planet is an extended object, the
observed effect of seeing is slightly different from that of stars. When seen with the
naked eye, planets do not supposedly twinkle in the manner of stars (though the
author has never found this to be a reliable way of identifying planets). Through a
telescope, each point making up the planetary image twinkles individually, and so
the image as a whole is blurred and reduced in contrast. However, there will be
occasional instants, lasting from a fraction of a second to a few seconds, when the
atmosphere is uniform along the line of sight. The planet will then be seen as clearly
as the optics of the telescope allow.
Thus, with practice, the observer can disregard the fuzzy image normally visible
and concentrate on what can be seen only under the best conditions. It is for this
reason also that photographs or CCD images (Chap. 9) only rarely show the detail
that can be discerned by a skilled visual observer of the planets. Nonetheless, visual
observations of the planets can be very rewarding. Even the inexperienced observer
will find the sight of Saturn floating inside its incredible rings quite breathtaking.
With experience, the observer will soon be able to distinguish the broad surface
features and polar caps of Mars, details of Jupiter’s spots and belts, transits and
occultations of its Galilean satellites, the major divisions of Saturn’s rings, and
perhaps even check the disputed presence of light on the dark side of Venus (known
as the ashen light).
Any type of telescope can be used to observe the planets, although some designs
are likely to give better results than others. Features on the planets are of low
contrast, and therefore the best telescope to use is one that degrades the available
contrast least. Reflectors have secondary mirrors, and these are often supported by
arms from the sides of the telescope tube. Both the secondary mirror and its support
arms affect the diffraction-limited image, broadening it and producing the spikes
sometimes to be seen on images of stars (Fig. 8.4). This has the effect of reducing
the contrast of an extended object because each point of that object has an image in
the shape of that shown in Fig. 8.4, and these then overlap to blur the image.
Thus other things being equal, a Schmidt-Cassegrain or Maksutov telescope,
which does not have support arms for its secondary, will be better for observing
planets than a Cassegrain or Newtonian telescope, and a refractor, without any
secondary mirror at all, will be better still. Other factors, of course, will influence
the clarity of the image as well as that of the design of the telescope. Any factor that
reduces contrast or increases the background illumination is deleterious. Thus the
lowest magnification consistent with showing the required detail should be used,
and observations should be made when the Moon is not around. Open designs for
telescope tubes (Fig. 2.17) will usually be worse than closed tubes and should be
covered if possible. Baffles should be carefully designed and positioned, all trans-
mission surfaces should be anti-reflection coated, and all surfaces should be clean
and dust-free. Serious planetary observers may even go to the extent of refiguring
the optical surfaces of their telescope’s objective to a far higher standard than usual.
Improvements in the image are claimed to be found if surface accuracies of a
twentieth of the operating wavelength or better are used, rather than the more usual
eighth of a wavelength.
The Sun
CAUTION
THE SUN SHOULD NEVER BE OBSERVED, EITHER WITH THE NAKED

EYE, OR THROUGH A TELESCOPE, WITHOUT THE USE OF APPROPRI-
ATE FILTERS AND STOPS OR OTHER DEVICES DESIGNED TO REDUCE
ITS INTENSITY IN THE VISUAL AND INFRARED PARTS OF THE SPEC-
TRUM. ONLY COMMERCIALLY PRODUCED FILTERS DESIGNED FOR
THE TYPE AND SIZE OF TELESCOPE SHOULD BE USED. SMOKED
GLASS, EXPOSED BLACK AND WHITE FILM, CDS, SPACE BLANKETS,
ALUMINIZED HELIUM BALLOONS, ALUMINIZED POTATO CRISP
PACKETS, FLOPPY DISKS, CROSSED POLAROID FILTERS, SMOKED
The Sun 163
PLASTIC, SUNGLASSES, MIRRORS, X-RAY PHOTOGRAPHS AND SO ON

ARE NOT SUITABLE.
PERMANENT DAMAGE TO THE EYE AND/OR TO THE TELESCOPE
CAN BE CAUSED IF THESE PRECAUTIONS ARE IGNORED. IF USING A
COMMERCIALLY PRODUCED INSTRUMENT THE CONDITIONS OF THE
MANUFACTURER’S WARRANTY SHOULD BE CHECKED SOMETIMES
THIS CAN BE INVALIDATED IF THE TELESCOPE IS USED FOR SOLAR
OBSERVATIONS.
The observer should take the above warning to heart, but with the appropriate
precautions should not let it stop him or her from observing the Sun. It is a most
rewarding object, and has the not inconsiderable advantages of being observable in
daytime and often under warm and pleasant conditions!
Finding the Sun
As the brightest object available to an astronomer, finding the Sun might not seem
to be a problem. However, the warning given above applies just as much to looking
through the finder telescope as through the main telescope, and even to sighting
along the telescope tube with the naked eye to point it near the Sun. The finder can
be used if it has an appropriate filter (see below).
Do not forget to cover the objective of the finder telescope if it does not have a
filter. Alternatively, the setting circles can be used if the position of the Sun is known,
though since the Sun moves quickly, the Astronomical Almanac (Appendix A)
will be needed to give its position. In the absence of either of these facilities, the
Sun is best found by circularizing the telescope’s shadow; that is, the telescope is
pointed very roughly towards the Sun, and then its shadow observed. As the
telescope is moved, the size of the shadow of the tube will change and be at its
smallest when the telescope is pointing towards the Sun. For telescopes with a
cylindrical tube, the shadow will then be circular rather than elliptical (Fig. 8.5).
If done carefully, this procedure is accurate enough to bring the Sun into the field of
view of the main telescope.
Observing the Sun
Full aperture filters are the best way of observing the Sun and are widely available
from the telescope suppliers who advertise in popular astronomy journals
(Appendix A). These filters cover the objective and reduce the solar intensity before
it becomes concentrated by the telescope optics. There is therefore no danger of
damage to the telescope if, when finding the Sun, its image falls onto the mechani-
cal parts of the instrument. This latter point is of considerable importance for
Fig. 8.5 Finding the Sun by

circularizing the telescope’s
shadow
Cassegrain, Schmidt-Cassegrain and Maksutov designs, because the focal ratio of

the primary mirror in these is often quite small, leading to a primary image with
great burning potential.
It is important to ensure that any filter used not only cuts down the visible light
but also eliminates the infrared part of the spectrum. If the image is too bright in the
visible there is not much danger to the observer, because it will not be possible to
look through the telescope for any length of time. However, a filter that cuts out the
visible light, but not the infrared, will lead to the retina of the eye being burned and
to possible blindness because there is no automatic physiological mechanism to
protect the eye from excess infrared radiation. For safety, the solar surface bright-
ness must be reduced by a factor of at least 30,000 (i.e., to 0.003%) across the
whole spectrum.
A full aperture filter consists of a very thin film of plastic, covered by a thin
deposit of aluminum. The whole film is thin enough to cause little distortion of the
image, so that it does not need to be optically flat, and the aluminum reflects most of
both the visible and infrared radiation. The reflectivity of aluminum falls off
towards the blue, so images seen through such a filter have a slight bluish cast.
Other metals are also used such as nickel, chrome and stainless steel, resulting in
other colors for the filtered Sun. Full aperture filters using optically flat glass are
also produced, though they are generally significantly more expensive than the
metal-on-plastic type.
The Sun 165
Fig. 8.6 Observing the Sun by eyepiece projection
An alternative way of adapting a telescope to enable the Sun to be observed is

via eyepiece projection. Note that eyepiece projection should not be used with
telescope designs such as the Schmidt-Cassegrain or Maksutov, where a short
focal-ratio primary mirror can form a real image of the Sun inside the telescope.
Indeed many manufacturers’ guarantees will be voided if you use such telescopes
for solar eyepiece projection. Always check with the manufacturer of your tele-
scope before undertaking any solar work.
The telescope should be stopped down to about 50 mm if it is larger than this, by
covering the objective using a securely attached piece of cardboard with a 50-mm
hole cut into it. The telescope is then pointed at the Sun (see above), and a sheet of
white cardboard placed 30–40 cm behind the eyepiece. The bright disc of the Sun
will be visible, projected on to this card (Fig. 8.6), and the eyepiece position should
be adjusted to bring the image into sharp focus. It will normally be helpful to attach
additional cardboard around the telescope tube to act as shielding.
At one time filters were produced that could be placed directly into the eyepiece.
However, these should be avoided for several reasons. First, the filter will get very
hot and the thermal stress may cause it to shatter, allowing the undiluted radiation
from the Sun to enter the eye. Second, the heat may damage the eyepiece; and third,
the heat from the filter will lead to convection currents and possibly distort the
optics of the eyepiece, ruining the image quality.
Specialized solar telescopes at major observatories are often fixed, and the light
is then fed into them by a coelostat. This is a device that has an arrangement of two
plane mirrors, both on driven mountings, so that the light from any part of the sky
can be reflected towards a fixed direction and the object (usually the Sun) tracked as
it moves across the sky.
With any of the above methods, it should be possible to observe limb darkening,
sunspots, plages, etc., and to measure the solar rotation period, which varies with
solar latitude. Sunspot counts can easily be made in order to monitor the activity of
the Sun and to observe the 11-year solar cycle, though some practice is required to
obtain consistency. The Zurich sunspot number, R, is obtained from the formula
R ¼ kð10g þ sÞ (8.4)
where g is the number of visible sunspot regions, either groups or single spots, s is
the number of individual sunspots (both the isolated spots and those within the
groups) and k is a personal correction factor to take account of the instrument used,
the quality of the observing site, and the idiosyncrasies of the observer in allocating
spots into groups. It has to be determined by comparing the sunspot counts obtained
by the observer over a period of a few months with the published values.
At a very considerably higher cost than the full aperture filters just considered,
Ha filters are obtainable. These are very narrow-band filters centered on the red line
of hydrogen at 656 nm (which is called the Balmer-a or Ha line); these are
primarily intended to enable prominences to be observed at the edge of the Sun,
and filaments, flares, etc., on its disc.
A reasonably skilled DIY enthusiast can make an alternative device for observing
prominences for little cost. This is the prominence spectroscope. It is a spectroscope
with an additional slit centered on one of the strong solar absorption lines, usually
Ha. The entrance slit to the spectroscope is aligned on the limb of the Sun, and
the observer looks through the second slit. Prominences may then be seen because
the prominence has an emission line spectrum and so emits strongly at that wave-
length, while the solar photospheric emission is spread over the whole spectrum.
Stars
Stars have been observed by astronomers since before recorded history. Only in
the last few centuries, however, have they been observed in their own right;
before that they were just a background against which the Sun, Moon and planets
moved. The two basic visual observations possible for stars are of their positions and
brightnesses. Positions have been considered in Chaps. 4 and 5; here therefore we
should look at stellar brightnesses. First, however, let us try and reduce the confusion
that many beginners experience over the apparently haphazard naming of the stars.
Stellar Nomenclature
The naming of stars appears to be haphazard because of the many different systems
that are in use at the same time. Some stars, especially the fainter ones, may only
have one label, but most of the brighter stars can have several names. Thus Sirius is
also known as a CMa, 9 CMa, GSC 0594902767, HIP 32349, HR 2491, BS 2491,
ADS 5423, HD 48915, and SAO 151881, among many other possibilities.
Stars 167
Half a millennium ago, stars were identified by being given proper names such
as Betelgeuse, Albireo or Deneb, or their position in the sky described, as in “The
left-hand star of the ‘W’ of Cassiopeia.” The latter method was cumbersome, and
neither method was systematic.
The first systematic method of naming stars originated in the star catalog
Uranometria, published in 1603 by Johan Bayer (1572–1625). The Bayer system
used a Greek letter in the order of the star’s brightness within the constellation
followed by the constellation name. For example the previously named stars
become a CMA (Sirius), a Ori (Betelgeuse), b Cyg (Albireo) and a Cyg (Deneb).
There are often anomalies to the system arising from changes to the constellations,
or mistakes, with some constellations missing some letters, or brighter stars having
later letters than fainter ones, etc. Thus a Orionis (0.9m see below for a description
of the stellar magnitude system) is fainter than b Ori (0.1m). After the Greek letters
have been used, then the lower case letters, a, b, c, . . . may be used in a similar way,
and after the lower case letters, upper case letters from A to P. The letters from R to
Z are reserved for variable stars (see below).
A second systematic method of identifying stars, the Flamsteed number,
originated in John Flamsteed’s (1646–1719) catalog published in 1725. The stars
are numbered within each constellation in order of increasing right ascension – for
example; 57 UMa, 88 Leo or 15 Pic.
Over the last 100 years, thousands of star catalogs have been produced, many
containing just a few stars of a particular type, but some with hundreds of thousands or
millions or more stars. The second version of the Hubble Space Telescope Guide Star
Catalog (GSC 2), for example, contains data on 945,592,683 stars. Many stars are
therefore identified by their numbers in one or more of these catalogs. The label then
usually takes the form of an abbreviation of the catalog’s name followed by the star’s
number. Commonly encountered usages of this are based upon the Henry Draper
Catalogue (HD), the Bright Star Catalog (BS), the Smithsonian Astrophysical Obser-
vatory Catalog (SAO), the Hipparcos Catalog (HIP) and the Hubble Space Telescope
Guide Star Catalog version 1.2 (GSC). A slight exception to this practice occurs with
the Bonner Durchmusterung and Córdoba Durchmusterung catalogs (BD and CD).
These catalogs are subdivided by the stars’ declinations. The label therefore takes the
form of the declination followed by a running number, e.g. BD þ12 1234.
Variable stars may well be named within one or more of the above systems. The
well-known eclipsing binary, Algol, for example, is additionally b Per, 26 Per, BS
936, HD19356, SAO38592, etc. However, there are also several systems specifi-
cally designed for naming variable stars. An extension to the Bayer system is the
most widespread approach. This labels the variables in each constellation with
capital letters, starting at R for the brightest. Letters from R to Z suffice for only
nine variables in each constellation, and so thereafter double capitals are used. RR
to RZ is used for the next nine variables discovered, then SS to SZ for the next eight,
TT to TZ for the next seven, and so on to ZZ, giving another 54 labels in all. After
that the sequences AA to AZ, BB to BZ, CC to CZ, . . . , QQ to QZ are used. A total
of 334 variables in each constellation may thus be identified, with the letters
followed by the constellation abbreviation; S Cyg, RR Lyr, FG Sag, etc.
When more than 334 variables are found the simpler practice of using the letter
V followed by a number starting from 335 is used. Thus we have V861 Sco, V444
Cyg, and V432 Her, etc.
Pulsars and other special groups of variables may use other systems of labeling;
a common practice is to use a label to indicate the variable type and combine it with
an approximate right ascension and declination. Thus we have the pulsar, PSR
1257 þ 12, which is to be found near RA 12 h 57 min and Dec þ12 . Gamma ray
bursters are named for the date of their occurrence as GRB yymmdd. Thus GRB
970228 occurred on February 28, 1997.
It is thus not surprising that the naming of stars appears haphazard at first (or
even second) sight. However, it does all become familiar after a while!2 It is worth
adding, though, that with only a few exceptions, such as Barnard’s star, stars are not
named after people.3 The numerous advertisements that one sees in the popular
press purporting to name stars after you for the payment of a fee are therefore
spurious. The operators of such a scheme will doubtless write your name against
your chosen star on their star map, but it will not be officially recognized and no one
else will use it. So do not waste your money.
Magnitudes
The first known eye estimates of the brightness of stars occurred in Hipparchus’ star
catalog of about 150 B.C., though there had been at least one earlier catalog produced
by Timocharis and Aristyllus around 300 B.C. Hipparchus divided the 1,000 or so
stars in his catalog into six groups based upon their brightness, with group 1 being
the brightest, and group 6 those only just visible. This classification remained
adequate until the invention of the telescope revealed stars fainter than those visible
to the naked eye. Such stars were then placed into groups 7, 8, 9 and so on. By the
middle of the nineteenth century, however, more precise measurements of stellar
brightness (photometry, see Chap. 11) were being pioneered by William Herschel’s
son John (1792–1871) and others.
2
The Centre de Données astronomiques de Strasbourg houses many astronomical data bases and is
accessible to anyone via the Internet (http://cdsweb.u-strasbg.fr/). Its Simbad search engine (http://
simbad.u-strasbg.fr/simbad/sim-fbasic) enables a great deal of data about stars to be accessed on
online – including lists of all the names that have been given to each star.
3
Comets and asteroids can be named after people, but you have to discover a new one yourself to
have the right to name it after yourself. On discovery, comets are designated by the year of
discovery and a running letter plus, usually, the discoverer’s name (e.g., Comet Kohoutek 1973f).
Once an orbit has been determined, then the name changes to the year of perihelion passage plus a
running Roman numeral giving its order of perihelion passage (e.g., Comet West 1976 VI). The
discovery of a new asteroid, after which you are entitled to name it subject to certain restrictions,
such as not using an existing name, is somewhat more complex. First you have to obtain at least
three separate positions for the object so that its orbital parameters can be determined. It then has to
be retrieved at two successive oppositions before it can be added to the list of known asteroids.
Stars 169
Now, Hipparchus’s grouping of stars according to their brightness tried to divide

the total range in a uniform manner. But the response of the eye to intensity is
geometric rather than arithmetic – that is, if the eye perceives stars A and B to differ
by a similar amount to the difference between stars C and D, then it will be the ratio
of the energy from A to that from B that will be similar to the ratio of the energy
from C to that from D, and not their energy differences. This is known as the Weber-
Fechner law. Thus the difference from one of Hipparchus’s groups to the next was a
constant factor in energy terms, not a constant difference. This was codified in 1856
by Norman Pogson (1829–1891) into the system of stellar magnitudes still in use
today. The system is defined by Pogson’s equation:

E1
m1 m2 ¼ 2:5log10 (8.5)
E2
where ml is the magnitude of the first star and El its energy; m2 is the magnitude of
the second star and E2 its energy.
Equation (8.5) gives a constant factor in energy terms of 2.512 . . . (¼ 100.4)
between one magnitude and the next as required, and was chosen to ensure that
measurements under the new system corresponded as closely as possible to those
under the previous groupings. The equation is a relative one, however, and gives
the magnitude of one star with respect to that of another. The magnitude scale
must therefore be calibrated by assigning an arbitrary value for the magnitude of at
least one star (known as a standard star). The magnitudes of all other stars will
then follow from (8.5) by comparison with that standard. In practice, there are
many standard stars distributed over the whole sky, so that there is always one
reasonably close to any star that is being observed. The magnitudes assigned to the
standards are again chosen so that the magnitudes correspond as closely as
possible to the previous estimates, and are such that a star of magnitude 6 should
just be visible to the unaided eye. This calibration results in a star of magnitude 1
having an energy (irradiance) of 9.87 109 W m2, at the top of Earth’s
atmosphere or about the perceived brightness of a 20 W energy-saving light
bulb at a distance of 9 km.
Because of these historical roots to the stellar magnitude scale, it is rather
awkward to use. Thus against all normal practice, the brighter the star, the smaller
the resulting number (magnitude). Setting magnitude 6 stars to be those just visible
results in the brightest objects being outside the range, and so it has to be extended
to zero and negative magnitudes; Sirius, for example, has a magnitude of 1.45.
Since it is a logarithmic scale and thus very compressed, the range from the
brightest object in the sky (the Sun) to the faintest detectable in the largest
telescopes is only about 55 magnitudes, but this corresponds to an actual difference
in their energies of a factor of 1022.
The relationship between magnitude difference and energy difference is
tabulated in Table 8.1, and some example magnitudes are listed in Table 8.2
below. The magnitudes listed in that table are all apparent magnitudes, that is, the
Table 8.1 Magnitude/energy relationship

Dm E1/E2
0.1 1.1
0.2 1.2
0.3 1.3
0.4 1.4
0.5 1.6
0.6 1.7
0.7 1.9
0.8 2.1
0.9 2.3
1.0 2.5
2.0 6.3
3.0 15.9
4.0 39.8
5.0 100 (exact)
10.0 10,000 (exact)
15.0 1,000,000 (exact)
20.0 1,000,000,000 (exact)
50.0 100,000,000,000,000,000,000 (exact)
Table 8.2 Examples of apparent magnitudes

Object m
Sun 26.7
Full Moon 12.7
Venus (maximum) 4.3
Jupiter (maximum) 2.6
Mars (maximum) 2.02
Sirius A 1.45
Betelgeuse 0.73
Polaris þ2.3
Uranus (maximum) þ5.5
Faintest object visible to the unaided eye þ6.0
Faintest object visible in a 0.3 m telescope þ13
Faintest object visible in a 1-m telescope þ16
Faintest object visible in a 10-m telescope þ21
Faintest object detectable using the very best of current techniques þ28
Faintest object detectable using the E-ELT þ36
magnitude of the star or other object as it appears in the sky, and therefore the
important quantity to know when observing the object. Apparent magnitudes,
however, take no account of the distance of the object, and so are unrelated to the
object’s actual luminosity. Thus Sirius appears to be twice as bright as Betelgeuse,
but Betelgeuse is much further away and is actually over 900 times brighter than
Stars 171
Sirius. A second magnitude scale is therefore also used that takes account of
distance. This is the absolute magnitude and is discussed in Chap. 11.
Observing Stars
For stellar observations there is no substitute for aperture – the bigger the better!
This is because for point sources (2.2), unlike extended sources, the telescope does
increase the brightness. From the definition of stellar magnitudes, a star of magni-
tude 6 is one that should just be visible to the unaided eye. However, that would be
from a good observing site, under good conditions and for an observer with good
eyesight. Magnitude 5 is therefore often a more realistic limiting magnitude for
unaided observations, and from a poor site it could be considerably worse. We thus
get as a realistic limiting magnitude for a particular telescope used visually,
roughly that
mlim ¼ 16 þ 5log10 D (8.6)
where D is the objective diameter in meters.

Q. If the proposed 39.3-m E-ELT should ever be used visually, what would be
its limiting visual magnitude?
A. From (8.6)
Mlim ¼ 16 þ 5 log10 D ¼ 16 þ 5 log10 39:3 ¼ 16 þ 5 1:59 ¼ 24:0m
Visual observations of a single star can give information on its brightness, and in
a few cases, its color. The brightness can be estimated if another star of known
magnitude is within the field of view, and with practice an accuracy of 0.l or 0.2
magnitude can be achieved. This is very useful for observing and monitoring
variable stars, especially irregular variables and the explosive variables such as
novae. An eye estimate will reveal whether or not the star has changed appreciably
from a previous observation, and may be of sufficient accuracy in itself, or indicate
whether or not it is worth making a more precise measurement (Chap. 11).
A few stars are bright enough to trigger the color vision of the eye. Thus Antares
(a Sco) is clearly red, and Sirius (a UMa) and Vega (a Lyr) bluish-white, even to
the naked eye. Even more spectacular are double stars with components of widely
differing temperatures so that the colors show up strongly by contrast. A few of the
brighter examples of these are listed in Table 8.3.
Table 8.3 Double stars with Star Colors (brightest star first)a
strong color contrasts
g And Red, green and blue
e Boo Yellow and blue
x Boo Orange and red
a CVn Yellow and blue
i Cas Yellow, red and green
b Cyg Yellow and blue
g Del Yellow and green
a Her Orange and green
b Ori Blue-white and blue
Z Per Red and blue
a Sco Red and green
b Sco Green and blue
a
Stars appear green or genuine blue only by contrast and perhaps
through partial dark adaptation by the eye’s color receptors. The
true colors of stars range from deep red for the coolest through
orange-yellow, yellow, white-yellow, white to white-blue for the
hottest
Much of the interest in visually observing stars, however, is not for the single
stars but when two or more are closely associated. This can range from just a pair of
stars to very large groupings, called globular clusters, containing up to a million
members. The latter are more analogous to small galaxies and are considered under
that heading below. Pairs of stars are divided into double stars and binary stars.
There is no difference between the two in their visual appearance. In a double star,
though, the two stars just happen to lie along the same line of sight, but are actually
separated in space by a very large distance. In a binary, the two stars are physically
close together as well as apparently close in the sky, and one will usually be in orbit
around the other. The probability of the chance alignment of two unassociated stars
is small, so binary stars greatly outnumber genuine double stars in practice.
Most visual binaries have very long orbital periods, but in some cases the orbital
motion can be detected over a period of a few years. The relative positions of the
two stars will need to be measured accurately if the orbital motion is to be found,
and a micrometer eyepiece will be needed for this purpose (Chap. 2). The separa-
tion of the binary is measured by aligning the horizontal cross-wire between the two
stars, moving the telescope to set the intersection of the fixed cross-wires on to one
star, and adjusting the moving cross-wire to set it on to the other star (Fig. 8.7). The
scale of the micrometer can be calibrated from the rate of drift when the telescope
drive is turned off (8.1) and (8.2). The position angle is the angle from the north
direction measured in the sense N-E-S-W, and can be read directly from the angular
scale of the micrometer (Fig. 8.7), with the east–west direction as seen in the
eyepiece being established from the drift direction. A few of the brighter visual
binaries are listed in Table 8.4.
Stars 173
Fig. 8.7 Micrometer

eyepiece
Table 8.4 Visual binary stars Star Magnitudes Period (years)

e Boo 4.7 6.7 150
a CMa (Sirius) 1.5 8.6 50
Z Cas 3.4 7.2 480
a Cen 0.1 1.4 80
z Her 2.9 5.5 34
70 Oph 5.1 8.5 88
85 Peg 5.8 8.9 26
Both double and binary stars are useful as tests of the performance of a telescope,
the skill of the observer, and the quality of the observing conditions. The
diffraction-limited resolution of a telescope given by (2.3) and (2.4) is for two
stars of equal brightness. It is also an arbitrary, though generally realistic, definition.
An experienced observer though can sometimes detect that the image of a double
star with a separation of only a half or a third of the diffraction limit appears to be
different from the image of a single star, and it has therefore in some sense been
resolved. Conversely, for stars of widely differing magnitudes, the separation may
need to be ten or more times the theoretical resolution before the stars can actually
be split. Remembering that high magnifications are needed to reach the diffraction
limit (2.19), observations of close double stars under the best observing conditions
will thus test the quality of the telescope; under poorer observing conditions,
the closest double star separable will provide a measure of how good or bad
those conditions may be. Lists of double stars may be found in many sources
(see Appendix A of this book).
Multi-star asterisms ranging from triplets to galactic clusters, such as the

Pleiades, containing hundreds of stars are also to be found, though less widespread
than doubles and binaries. Many are listed in the same sources as the double stars
(see Appendix A of this book). They may again just be chance alignments or be
physically linked, but as the number of stars in the asterism increases, the probabil-
ity of there being a physical association rises very rapidly. Similar observations
may be made for these objects as those described for the single and double stars.
The larger and denser clusters, however, are also well worth finding and observing
for their beauty alone.
Nebulae and Galaxies
Observations of diffuse interstellar nebulae (H II regions, supernovae remnants,

planetary nebulae, absorption nebulae, etc.), star clouds, globular clusters and
galaxies have much in common with each other. Despite the very different physical
natures of the objects, in the telescope their visual appearance is as faint diffuse
extended objects. It is worth repeating here the warning given in respect of
observing the planets – that images of these objects to be found in books, etc., far
exceed their visual appearances in small telescopes. Even the brightest of them
hardly trigger color vision in the eye (remember a telescope does not increase the
surface brightness of an extended object, 2.16), and so they will appear milky white,
not resplendent in all their spectacular colors. Furthermore, the images in books are
the result of long exposures, sometimes with CCD and other detectors that have a
hundred or more times the sensitivity of the eye, so that they reveal far more of the
objects’ extents than can be seen visually. Some idea of the difference between
visual observing and imaging may be given by Figs. 8.8, 8.9, and 8.10.
Fig. 8.8 The Sombrero Galaxy (M104) as it appears to a visual observer in a smallish telescope
(based upon a sketch by the author). (Copyright # C.R. Kitchin, 2012)
Nebulae and Galaxies 175
Fig. 8.9 The Sombrero Galaxy (M104) – a CCD image obtained on a 0.4-m SCT (Reproduced by
kind permission of R. W. Forrest)
Fig. 8.10 The Sombrero Galaxy (M104) – an image obtained using the 8.4-m ANTU telescope
(Reproduced by kind permission of ESO and P. Barthel)
Nonetheless, observations of these objects can still be fascinating and highly

rewarding. With galaxies, one is looking far into the depths of the universe. With a
0.3 m telescope, for example, the limiting magnitude for a distant galaxy can range
from magnitude 9 to 12, depending upon the galaxy’s angular size and the sensitiv-
ity of the observer’s eyes (not magnitude 13, as given by (8.6), because of the
extended nature of the galaxy). Thus galaxies can be seen directly that are over 200
million light years away, and the photons actually entering the eye have then been
traveling through space since before the time during which dinosaurs roamed Earth.
H II regions, such as the Orion Nebula, are the birthplaces of stars and planets, and
supernovae remnants, such as the Crab Nebula, their most spectacular death throes.
Given the faintness and low contrast of many of these objects, observing them is
one of the most severe tests possible of an observer’s skill. The lowest-power
eyepiece consistent with showing the object as extended should always be used.
The optics of the telescope and the eyepiece should be as clean and dust-free as
possible. Any trace of dew will be fatal to finding even the brightest nebulae or
galaxies. Except for the very brightest, such as the Orion Nebula or the Andromeda
Galaxy, these objects have to be found from their tabulated positions and the use of
accurate setting circles, or by star hopping from a nearby bright star or using the
telescope’s computer database. Even when the telescope is pointing at the object, it
may be necessary to use averted vision (Chap. 2) to see it at all. For really difficult
objects, it can be helpful to nudge the telescope back and forth by small amounts,
and the object may then be found to show up while moving, even though it
disappears again when stationary.
Lists of extended objects and their positions may be found in many places
(Appendix A). Two catalogs in particular, however, are to be noted because they
are used to give names to many extended objects: these are the Messier list and the
New General Catalogue. Charles Messier (1730–1817) was a comet hunter who got
exasperated with continually mistaking nebulae and other objects for comets, and
therefore produced a list of such objects in order to avoid them! There were 103
objects in Messier’s original list, but one was a double star and another either a
duplication or even a comet. Later additions to the list have increased the number to
110, of which 39 are galaxies, 29 globular clusters, 27 galactic clusters, 11 gaseous
nebulae, and one each a double star, an asterism and a bright patch of the Milky
Way.
The Messier objects are identified by an “M” and a number from the catalog;
thus the Crab Nebula is M1, the Orion Nebula M42, the Pleiades M45, and the
Andromeda Galaxy M31. For the observer using a small telescope, the Messier list
is useful since it contains many of the extended objects likely to be observable. It
can also be used as a challenge; initially just to try and observe as many of the
objects as possible. M74 is probably the most difficult of the Messier objects, so if
you can find it, you should be able to see all the others. With more experience, the
challenge can be to find as many Messier objects in a single observing session as
possible, and this can be particularly interesting as a competition among members
of an astronomy society or at a “star party,” etc.
The second of the catalogs originated in 1786 as William Herschel’s Catalogue
of Nebulae, and was extended by his son John and then by Johann Dreyer in 1888
into the New General Catalogue of Nebulas and Clusters of Stars, containing nearly
8,000 objects. This catalog is abbreviated as NGC, and objects within it are
referenced by this identifier and their catalog number. Thus under this system of
nomenclature, the Crab Nebula is NGC1952, the Orion Nebula NGC1976, and the
Daytime Observing 177
Andromeda Galaxy NGC224. The Pleiades are not listed in the NGC catalog since
the cluster is close enough to us for the member stars to be observed individually,
though other galactic clusters are included.
A newer catalog, the Caldwell catalog, produced by Sir Patrick Moore (whose
full surname is Caldwell-Moore), has found some favor. The catalog mirrors the
Messier catalog in that it also has 109 objects within it. The Caldwell catalog,
however, covers the whole sky, and the running numbers increase consistently from
north to south. C plus the running number designate objects; for example, the star
cluster h and w Per is C14.
Daytime Observing
Many people will have noticed that the Moon is often visible to the naked eye
during the daytime, and can obviously therefore also be observed through a
telescope. The brighter planets and stars can also often be found. For any daytime
observing, the prime requirement is for a haze-free sky to reduce the background
light. This requirement does not necessarily mean cloud-free. The most transparent
skies often occur after a rainstorm, when the dust has been washed out of the
atmosphere. The gaps between the clouds can then be used on an opportunistic basis
for observing.
Allied to this, the further the object is away from the Sun in the sky, generally the
easier it will be to observe. The scattered solar light (i.e., the blue sky) is polarized,
approaching 100% linear polarization at 90 from the Sun under good conditions. A
sheet of Polaroid, aligned orthogonally to the sky polarization, can therefore help to
reduce the background light considerably. Objects have to be found either from
their positions and the use of setting circles, or by off-setting from the Sun (NOTE
THAT ALL THE PRECAUTIONS FOR OBSERVING THE SUN DISCUSSED
EARLIER MUST BE EMPLOYED, EVEN WHEN IT IS ONLY TO BE USED AS
THE STARTING POINT) or utilizing the database in computer-controlled
mountings.
An important point to watch is that, if the telescope is even slightly out of focus,
then the objects will become impossible to see. If the eyepiece focusing mount has a
position scale, a note should be made of its reading when in focus. If, as is more
usual, the focusing mount is not calibrated, the telescope should be focused at night
and left undisturbed so that it is still in focus the next day. If this is not possible, then
the telescope can be focused on a terrestrial object on the horizon, provided that it is
at least several hundred m away, and the focus corrected to the position for an
infinite distance using the formula
1000f 2
Focal correction ¼ mm (8.7)
df
or for most purposes
1000f 2
Focal correction mm (8.8)
d
where the focal correction is the distance that the eyepiece must be moved towards
the objective, f is the focal length of the telescope in meters and d is the approximate
distance of the terrestrial object in meters.4
This last approach, however, is not very satisfactory because even for an object a
kilometer away, the focal correction is over 4 mm for a 0.2 m f10 telescope, and
therefore difficult to determine precisely in the absence of a position scale.

Q. Determine the correction needed to focus on Venus during the daytime using
an f6 0.5-m Newtonian telescope after it has been focused on a water-tower
whose distance from the observer is 850 m.
A. An f6 0.5-m telescope has a focal length (f) of 3 m (¼ 6 0.5)
From (8.7)
Focal correction ¼ 1,000 32/(850 3) ¼ 9,000/847 ¼ 10.6 mm
False Observations
For a variety of reasons, observations can be misleading, or even mistaken. This

comment applies to all types of observing, but is particularly true for visual
observations since there is then no hard copy that can be checked later or indepen-
dently. Problems are most likely to occur when the observer is trying to observe at
the limits of his or her equipment and his or her abilities. Sometimes the mistake
can result from preconceived ideas, leading to the observer seeing what he or she
expects to see, and not what is actually there. At other times, the false observation
can arise from the physiological nature of the human eye. Thus several objects on or
just below the resolution limit of the eye may be seen linked together to form a
linear feature. This is probably what happened with the Martian canals. They were
seen, not because the observers were poor, but because their eyesight was so acute
that they were almost able to see the volcanoes and craters that we now know
actually to be on the Martian surface.
4
NB “Google-Earth™” is very useful for finding such distances.
False Observations 179
Fig. 8.11 “Teardrop” effect

during a transit of Mercury or
Venus. The silhouette of the
planet still appears connected
to the limb of the Sun, even
when there should be a thin
thread of the photosphere
visible
Many of the rods and cones in the eye’s retina have cross-links, and this can lead
to problems when high levels of contrast occur in an image. The “tear-drop” effect
(Fig. 8.11), found when Mercury or Venus transits the Sun, is an example of this.
There is little an observer can do to counteract the possibility of false observations,
except to be aware that they can occur, and to be cautious about any results obtained
towards the limits of what may be possible.
Exercises
8.1. Calculate the energy per square meter at the surface of Earth from a star just
visible to the naked eye from a good observing site (ignore atmospheric
absorption). (Irradiance from a 0 magnitude star at the top of Earth’s atmo-
sphere: 2.48 108 W m2).
8.2. At what distance would a 20 W energy-saving light bulb just be visible to the
eye using the proposed E-ELT? (A star of magnitude 0 has an energy [irradi-
ance] of 2.48 108 W m2, at the top of Earth’s atmosphere. Assume that
this irradiance is halved by absorption in Earth’s atmosphere. You may find the
data in Table 8.1 and Example Calculation 8.2 useful).
8.3. Determine the correction needed to focus on Mercury during the daytime using
an f8 0.3-m Newtonian telescope after it has been focused on a tree whose
distance from the observer is 1,200 m.
8.4. Part of the rim of a lunar crater is observed to have a shadow with a length (L)
of 0.200 . The peak is 3.50 from the limb of the Moon (D), and the distance of the
terminator from the limb is 90 (P). What is the likely height of the crater rim?
(Take the lunar angular radius (R) to be 15.50 ).
Chapter 9
Detectors and Imaging
The Eye
The front line detectors for almost all astronomers are their own eyes. For many,
especially when using smaller telescopes, these are also the only detectors. The eye,
or more particularly vision, which is the result of the eye and brain acting in concert,
is, however, a very complex phenomenon, and some knowledge of its peculiarities
is essential for the observer. Thus reference has already been made in Chap. 8 to
averted vision, the effect of high contrasts (known as irradiation) and the combina-
tion of sub-resolution features (Martian canals). The structure of the whole eye
(Fig. 9.1) is well known from school, and need not be considered further here. It is
the structure of the eye’s detector, the retina, which is of importance.
The retina is a network of detecting cells connected by nerve fibers via the optic
nerve to the brain. The detecting cells are of four types: rods, and three types of
cones. The names of the cells come from their shapes. The three types of cones have
sensitivities that peak at about 430, 520, and 580 nm, and they provide us with color
vision. The overall sensitivity of the cones is, however, low. The rods are of only
one type, and have a response that peaks at 510 nm. There are about 108 rods and
6 106 cones but only 106 nerve fibers; hence many cells are linked to each nerve.
Cones are most abundant in the fovea centralis, and many there are singly
connected to nerve fibers. The fovea centralis is the point on the retina where the
light falls if we look directly at an object; the rods become commoner and the cones
fewer as distance increases away from this point.
The light-sensitive component of the rods is a molecule called rhodopsin, or
visual purple (from its colour). Under high light levels, the rhodopsin is mostly
inactivated and the rods have a low sensitivity. The cones then dominate vision and
we see things in color. At low levels of illumination, the rhodopsin regenerates
over a period of 20–30 min, restoring the rods to their full sensitivity. Vision is

182 9 Detectors and Imaging
Fig. 9.1 Structure of the

human eye (horizontal cross-
section)
then almost entirely via the rods. This effect has two consequences for
astronomers.
The first is the familiar effect of dark adaptation: immediately on leaving a
brightly lit area into the dark, very little can be distinguished. However, after a
few minutes, vision becomes much clearer, and after half an hour or so one can
often see quite easily even on a moonless night. This adaptation is partly a result
of the pupil of the eye increasing in size, but much more the effect of the
regeneration of the rhodopsin, which increases the retina’s sensitivity by a factor
of 100 or more. The rhodopsin is destroyed quickly by strong light, and so the
dark adaptation can easily be lost if the observer is careless with a torch or turns
on the main lights, etc. However, the sensitivity of rhodopsin ranges from about
380 to 600 nm. So deep red light, with a wavelength longer than 600 nm, should
not affect dark adaptation because the rhodopsin will not absorb that radiation.
For this reason, many observatories are illuminated with red lights to allow ease
of movement while preserving dark adaptation. It should be stressed, however,
that for this to work the light has to be very red indeed, with no short-wave
leakages.
The second effect of rod vision is that of lack of color. There is only the one type
of rod, and so at low light levels the eye cannot distinguish between different
wavelengths. Thus, as already mentioned (Chap. 8), most gaseous nebulae appear
milky white to the eye, instead of the multi-hued reds, greens and blues as seen on
many photographs.
The integrated sensitivity of the cones has a peak near 550 nm, while the
sensitivity of the rods peaks near 510 nm. If a bright star and a faint star are
simultaneously in the field of view of a telescope, the former may be seen with
color vision while the latter is detected by the rods. The two stars will thus be seen
in slightly different parts of the spectrum, and this can lead to incorrect estimates of
their relative magnitudes, with a faint blue star being seen as too bright, and a bright
red star as too faint, a phenomenon known as the Purkinje effect.
Charge Coupled Devices (CCDs) 183
Reference has been made elsewhere to other properties of the eye that affect
astronomical observation:
1. The eye’s logarithmic response to intensity, which results in the stellar magni-
tude scale (8.5).
2. Averted vision, which works because rods increase in abundance away from the
fovea centralis, and looking to the side of the object therefore causes its light to
fall onto the higher-sensitivity rods.
3. High contrast effects (the tear-drop effect, etc. – Fig. 8.11), which arise because
of the interconnection of many cells to a single nerve. If only a few of those cells
are stimulated, then the nerve does not trigger, and so the brain registers that part
of the image as dark, etc.
4. The resolution of the eye, which is 30 –50 , instead of the 2000 suggested by (2.4).
This is due to aberrations arising from the poor optical quality of the eye, the
granular nature of the retina because it is composed of individual rods and cones,
and the interconnection of many rods and cones to a single nerve.
Hence, these need not be considered further here.
Charge Coupled Devices (CCDs)
Many observers are content just to look through their telescopes, especially when
starting in astronomy. Sooner or later, however, most people want to obtain a
permanent record of what they can see through the telescope. This may, of course,
be in order to boast to friends and rivals in an astronomy society, or perhaps to
enable more rigorous measurements and serious work to be undertaken. There are
many types of detectors sensitive to light, but only two are likely to be of interest to
observers with small- to medium-sized telescopes; these are the charge coupled
device (CCD) and the photographic emulsion.1 They are each considered below.
The CCD has been in use on major telescopes since the early 1970s and has been
increasingly used by amateur astronomers from about the mid 1980s onwards. For
most imaging purposes the CCD has now completely replaced the photographic
emulsion and other types of detectors. Let us first therefore look at the basis of how
a CCD works.
Consider first an array of atoms, perhaps forming part of a crystal, and each with
one electron loosely attached (Fig. 9.2). The absence of an electron in such an array
(Fig. 9.3) leaves a hole that, because of the absence of the negative charge of the
electron, will then appear to be a positive charge. Applying a positive voltage from,
say, the right, will cause the electron on the left of the hole to hop into that hole,
leaving a new hole in its place one unit to the left (Fig. 9.4). The next left-most
electron then hops into that hole, and a new hole appears one unit further to the left,
1
The p-i-n photodiode is also a possibility, but only for photometry of point sources (Chap. 11).
Fig. 9.2 Array of atoms

and electrons
Fig. 9.3 Positive hole in

an array of atoms and
electrons
and so on. Thus effectively the hole behaves as though it is a positively charged
particle, and for most purposes we can treat it exactly as though it is such a particle.
The detection mechanism of a CCD is the production of electron–hole pairs in a
crystal of silicon. In the CCD, the absorption of a photon in p-type silicon2 “frees”
an electron from the grid, so that it can wander around within the solid (the
explanation for this in terms of energy bands within a solid is beyond the scope
of this book, but the interested reader is referred to any book on condensed matter
physics). A mobile positive hole, as above, is also produced by the absence of this
electron from its “normal” place. The electron and hole will move at random within
the crystal under the effects of the thermal motions of the atoms, and if they should
encounter each other, they will recombine and emit a photon. There will then have
2
Most of modern electronics is based upon silicon chips that have had very small quantities of
other elements added to the silicon – a process called “doping.” Some additives enable the
negative electrons to carry an electric current, producing n-type silicon; other additives cause
the electric current to be carried by the positive holes, producing p-type silicon. For further details
the interested reader is referred to physics and electronics texts.
Fig. 9.4 Motion of electrons

and a hole in an array of
atoms and electrons
been no net detection of the original photon. Thus the electron and the hole must
be kept apart.
In the CCD, the electron is attracted towards a positively charged electrode
placed on, but insulated from, the silicon. Simultaneously, the hole is repelled by
the same charge (Fig. 9.5). The electron is then stored beneath the electrode, where
it will subsequently be joined by other electrons produced in the same manner by
other photons. The hole is ejected into the body of the silicon and eventually lost.
Thus a negative charge builds up under the electrode, whose magnitude is propor-
tional to the intensity of the illuminating radiation.
Many such basic units can be produced in a two-dimensional array by standard
very large integrated circuit (VLSI) techniques (Fig. 9.6), to enable a two-
dimensional charge analog of the incoming photon image to be built up. The
image is read out from the silicon chip by moving (coupling) the charge under
the first electrode to an output electrode, from whence it may be detected.
The charge under the second electrode is then moved to the first, and thence to
the output. The charge under the third electrode is moved to the second and is then
moved to the first, and thence to the output, and so on.
Fig. 9.5 Basic unit of a CCD
Fig. 9.6 CCD array

(schematic); in practice,
CCDs can have up to
10 000 10 000 elements
The image is thus output as a series of charge packets, each of whose magnitudes
is proportional to the intensity of the image at the point in the two-dimensional
array from which the charge originated. This may then be displayed as a video
image if read at 50 or 60 Hz, or output to a computer, camera memory, etc., if longer
exposures are used.
The efficiency of detection of the CCD can reach 80% at wavelengths of about
700–800 nm, making it up to 100 times faster than photographic emulsion or the
human eye. The basic CCD has a spectral response extending from about 450 to
1,000 nm. This can be extended to shorter wavelengths by a suitable fluorescent
coating to convert the shorter-wavelength light into that detectable by the CCD.
After its efficiency, the other main advantage of the CCD is that its output can be
fed directly into a computer. The CCD’s disadvantages are its high cost and small
size compared with photographic emulsion.
The ubiquitous digital hand cameras are based upon CCDs and can sometimes
be used to obtain images through telescopes. Whether or not this is possible
depends upon mainly upon the camera’s focusing system. Apart from very cheap
cameras that have fixed foci, almost all modern cameras have automatic focusing
systems. The focusing system may be active or passive. Cameras with active
systems are unlikely to work on telescopes. The active system emits an ultrasound
or infrared beam and deflects the reflections from the objects to be imaged in order
to focus on them. The emitter and receiver of the focusing beam are likely to be to
one side of the camera’s lens. If the camera is held up to the eyepiece in order
to ‘look’ through the telescope, then the focusing beam will probably hit some part
of the telescope structure and focus on that, not on the image wanted by the
observer. It may be possible to alter the telescope’s focus until some sort of
image is achieved, but it will be difficult, and the results are likely to be poor.
Passive focusing systems utilize the image itself to focus and do so by
maximizing the contrast in the image. They will, therefore, usually work if the
camera is used on a telescope. To observe with such a camera that has a perma-
nently mounted lens, first focus the telescope by eye on the object of interest. Then
hold the camera up to the eyepiece so that the lens ‘looks’ through it. Be careful not
to let the outer surfaces of either the eyepiece lens or the camera lens scrape against
anything, or they may become scratched. By hand it will be difficult to hold the
camera up to the eyepiece without vibration or moving the telescope, so some sort
of mounting for the camera on the telescope may need to be devised.
With the camera in position on the telescope, the image is obtained by half-
depressing the exposure button as usual until the camera signals that it is ready and
then pressing completely to complete the exposure. In this way reasonable images
of the Moon may be achieved, but almost certainly not for other astronomical
objects – although if the camera has nighttime settings (giving longer exposures)
these may be worth trying.
Cameras with removable lenses (usually single-lens reflex or SLR cameras) may
have their lenses removed and be mounted using an adaptor directly onto the
telescope in place of the eyepiece. The primary lens or mirror of the telescope
then replaces the lens of the camera, and focusing is achieved using the telescope
focus mechanism, not that of the camera. One problem with this arrangement is that
the camera may try to focus continuously, thus rapidly draining its batteries and/or
causing vibrations.
However, none of the setups just described are likely to enable the observer to
image nebulae or galaxies, since these will require longer exposures than will be
available from the camera. Specialist astronomical CCD cameras will be needed for
this purpose.
A CCD camera designed for astronomical use has one main difference from
conventional digital cameras – the CCD chip and possibly other parts of the
camera as well are cooled. With the cameras designed for use on smallish
telescopes, the cooling is usually via a Peltier3 system, giving a temperature of
around 50 C below ambient. CCD arrays for major telescopes are likely to be
cooled with liquid nitrogen to 170 K (100 C). The reason for cooling is to reduce
the noise added to the image by the detector and to help prevent inter-pixel isolation
breakdown.
As we have seen above, the CCD relies upon the production of an
electron–hole pair to detect an incoming photon. Such electron–hole pairs,
though, can be produced by other processes, especially the thermal motions of
electrons within the silicon substrate of the device. Since the faint objects that are
usually of interest to astronomers require long exposures (seconds to hours), this
thermal noise (also called “dark noise” since it occurs even when no exposure is
being made) soon swamps the image. Lowering the temperature of the CCD by
any amount reduces the thermal noise, and it becomes almost zero at around
170 K. The effect of this is illustrated in Figs. 9.7, 9.8, and 9.9. Cooling the CCD
chip, of course, means that ice is likely to form onto it. To prevent this it has to be
contained within a sealed enclosure filled with a dry gas and with an optical-
quality window to allow the light beam to enter the enclosure. The window in turn
may need to be heated to around ambient temperature to prevent ice forming upon
it. Many of the commercially produced CCD cameras for astronomical use have
additional features such as a second CCD chip that is used to guide the telescope
automatically during the exposure and specialized software to allow for image
processing and data analysis.
Only in the last couple of decades or so have CCD cameras become cheap
enough to be considered for use outside of professional observatories. Even now,
“cheap” is a relative term, since a CCD will need a small computer, image
processing software and other accessories, and the observer can pay from about
half the cost of an entire 0.2 m Schmidt-Cassegrain telescope to several times that
amount for a working CCD package with around 1,000 1,000 pixels. The CCD,
however, is about 100 times as sensitive as either the eye or the photographic
emulsion, and so a 0.2-m telescope used with a CCD becomes the equivalent of a
2 m telescope used visually or photographically! The cost of the CCD considered in
this fashion becomes a huge bargain.
3
Briefly the Peltier effect is that if two suitable materials are joined in an electrical circuit with a
current flowing around it, one junction cools down while the other heats up. Suitable materials
include N- and P-doped bismuth telluride, lead telluride, antimony, bismuth, germanium and
silicon.
Fig. 9.7 Schematic

illustration of a galaxy image
obtained in a 1-min exposure
using an uncooled CCD
camera
Fig. 9.8 Schematic illustration of a galaxy image obtained in a 5 min exposure using an uncooled
CCD camera
Fig. 9.9 Schematic

illustration of a galaxy image
obtained in a 10-min exposure
using a cooled CCD camera
For professional use, devices with up to about 10,000 10,000 pixels are
available. More commonly CCD arrays with sizes in the region of 1,000 1,000
pixels to 4,000 4,000 pixels are used, and several such arrays are stacked into a
mosaic in order to cover the whole of a telescope’s field of view (Fig. 9.10). The
electrical connections to the CCD arrays mean that there are gaps of the order of a
millimeter or two between them within such a mosaic. Several exposures will
therefore be needed fill in these gaps and to obtain continuous coverage of the
area of the sky being imaged. Apart from the additional time involved in repeating
images, this can cause problems when objects are undergoing rapid changes.
Furthermore processing such large images requires substantial computing power
and specialist software, well beyond that likely to be available outside of research
institutions.
Image Processing
One of the advantages of a CCD image is that it may be fed straight into a computer
for image processing, and normally this advantage is also a necessity. The CCD
image has to be corrected in various ways (see below), if optimum extraction of
information from it is to be achieved (all images, however obtained, may benefit
from or need image processing; it is mentioned here in connection with CCD
images only because they are already stored on a computer and can therefore easily
be processed).
Image Processing 191
Fig. 9.10 The CCDs within OmegaCam, the camera for ESO’s 2.6-m Very Large Survey
Telescope (VST). The camera images a square degree of the sky using a mosaic of 32 CCD
arrays, each of which is 2,024 4,048 pixels in size (Reproduced by kind permission of ESO)
Flat Fielding
The relative responses of all the elements may vary. An exposure of a uniform field
of view is therefore required and must be subtracted from the image to provide a
“flat field.”
Cosmic Rays
These are troublesome on long exposures. Each cosmic ray provides a bright
“spike” in the image, which may be mistaken for a feature. Long exposure images
must therefore be inspected carefully and such features removed; they are usually
so much brighter than the rest of the image that they are easily recognizable. If need
be, the affected pixels can be replaced by the average of their surroundings to
provide a “prettier” image – but the lost information is not regained.
Contrast Stretching
A CCD can resolve 100,000 or more gray levels, while a typical computer display
screen can only show 256 gray levels. A big improvement in usefulness may
therefore be obtained by mapping the main gray levels in the image to those
available on the display. Suppose, for example, that 90% of the data in an image
is between gray levels 1,200 and 3,760. By setting levels below 1,200 to display as
level 0 for the monitor, levels 1,200 and 1,209 to display as level 1 for the monitor,
levels 1,210 and 1,219 to display as level 2 for the monitor, levels 1,220 and 1,229
to display as level 3 for the monitor, and so on, then optimum use may be made of
the available information. Such a remapping of the image is called gray scaling4 or
contrast stretching.
Background and Dark Signal Subtraction
Thermal and other noise sources will result in electrons accumulating in the CCD
even when it is completely in the dark. The sky background, especially from urban
sites, can be quite bright. Both of these problems can be reduced by taking
additional exposures with the CCD, in the first case with the shutter closed, and
in the second case while looking at a featureless part of the sky near the object of
interest. These two exposures are then subtracted from the main exposure to reduce
the noise level and enhance the contrast.
Hot Spots
Some of the basic units (usually called pixels, the term being derived from picture
elements) may be faulty. This may mean that they always appear bright because of
high noise levels, or they may impede the flow of electrons from other units. In the
first case, the affected pixel can be corrected in a similar way to that for the cosmic
ray spikes. In the second case, correction of the affected pixels may be possible by
measuring the loss in the bad unit. Sometimes, however, the data in the pixels prior
to the bad one are lost, and a blank line will have to be accepted in the image.
4
The term “gray scaling” is also used for the process of converting a color image to a black and
white one.
Photography 193
Image Compression
Digital images generally require large amount of memory for their storage. Often
the amount of space needed can be significantly reduced, without losing any
information, by compressing the image. There are various algorithms for undertak-
ing this process, and they often work very well with astronomical-type images.
Thus an image of a star field will probably contain large areas of black sky. A row
within such an image that has a star towards its center is likely to be represented by
a sequence of numbers such as
0 0 0 0 0 0 0 0 0 0 0 0 10 20 50 25 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0
In a compressed version this would look like
0 12 10 20 50 25 5 0 14
where the run of identical values representing the black sky has been replaced by a
count of the number of those values. In this case the original information was
contained in 31 numbers. The same information in compressed form requires just
nine numbers – a compression by a factor of 3.4.
There are other approaches to image compression, such as storing the
differences between adjacent values rather than the values themselves (the
differences are likely to be smaller numbers than the values). In these and other
ways the memory required to store images can be reduced by a factor of ten or
more. If not all the information in an image is needed (for example, objects fainter
than a certain limit may be of no interest) then lossy compression may in some cases
reduce the storage space by a factor of a 1,000.
In general, image processing is undertaken for two purposes: first to correct
known deficiencies in the image (e.g., flat fielding) and second to optimize the
image to allow information to be retrieved by human beings (e.g., contrast
stretching). There are many other techniques to image processing such as removal
of the instrumental profile, use of false colors, removal of geometric distortion,
smoothing, image combination, edge enhancement, Lagrange transforms, Fourier
transforms, etc. In the second of the applications, image processing becomes much
more of an art than a science, and which technique to apply and when, in order to
obtain the information required, becomes a matter of judgment and trial-and-error.
These other techniques are beyond the scope of this book, and the interested reader
is referred to the specialist literature (see Appendix A of this book).
Photography
The use of photography for imaging by astronomers of all types has now largely
ceased. Photographic film, processing chemicals, printing paper, etc., are still
available if you belong to a camera club or via web sites such as e-Bay, but otherwise
are becoming difficult to obtain. Some amateur observers are likely to continue to
use photography while they can because it is still much cheaper than a CCD setup,
Fig. 9.11 Electron trap of a silver bromide grain in a photographic emulsion
but no major professional observatory has used photography for well over a decade
now. Nonetheless there are many records in archives that were obtained originally
by photography. (However, many of these will now have been digitized.) Photogra-
phy does have some idiosyncrasies, though, so it is still worth understanding a little
about the process in order to be able to interpret older material correctly.
The basic detection mechanism in photographic emulsion is the same as that of
the CCD – electron–hole pair production by an incident photon – but in silver
bromide rather than silicon. Thereafter, however, the process differs from that of
the CCD. The electron and hole are produced in a small crystal (grain) of the silver
bromide and are free to wander around within it under the effect of thermal motions.
As with the CCD, they must be kept separate to avoid recombination. In the
emulsion this is done chemically. The crystal structure of the silver bromide is
deliberately distorted (strained) by the addition of a small number of chlorine atoms
in solid solution (Fig. 9.11). The strain in the crystal lattice where such an atom
occurs results in a distortion of the electrical field, called a trap, which attracts the
electron and repels the hole.
The electron, immobilized at the trap, will neutralize a silver ion, leaving a silver
atom in the crystal structure. The presence of the silver atom then adds to the
effectiveness of the trap, so that the next electron produced by the absorption of a
photon is more likely to be held there than at other traps. The second electron
neutralizes a second silver ion, further enhancing the effectiveness of the trap, and
so on. Between 5 and 20 such silver atoms must accumulate before becoming stable
against dispersal by other processes, and such a cluster of atoms is then known as
the latent image. The hole must be eliminated, just as with a CCD. In the emulsion it
may be removed in two ways: either its random motions may take it to the surface
of the grain, where it will react with the gelatin within which the grains are
Other Detectors 195
suspended, or two holes may combine to neutralize a halogen ion, which in turn will
leak out of the grain to react with the gelatin.
Once the latent image has been produced, it must be processed. Processing is a
chemical procedure whereby the latent image is converted into a visible image. It
has at least two main stages: developing and fixing. Developers are reducing
chemicals that convert silver bromide into silver. Normally the reaction rate for
this conversion is very slow, but it is catalyzed by the presence of silver atoms in the
grains. The grains in the emulsion with latent images are thus converted completely
into silver much more quickly than those grains without a latent image and which
are still pure silver bromide.
Since a typical grain in the emulsion will contain some 1011 or so silver atoms,
developing amplifies the latent image by a factor of about 109. A developer, however,
will eventually convert all grains into silver, irrespective of the presence of a latent
image or not. Hence the developing times must be strictly controlled. After develop-
ing, the silver grains forming the image are still surrounded by the undeveloped silver
bromide grains. Fixing is a second chemical process that dissolves away the silver
bromide. Fixers will also dissolve the silver grains given enough time, so again the
process must only last for a strictly controlled reaction time.
The final image, after processing, consists of the silver grains suspended in
gelatin, the density of the grains being highest where the original light intensity
was greatest. On viewing such an image, the silver grains, with a size in the region
of 1 mm, absorb the light. The image therefore appears darkest where the original
intensity was highest, i.e., it is a negative image. Most photographic astronomical
work takes place directly on the negatives – something that appears strange at first,
but which quickly becomes familiar. If required, a photograph of the photograph
may be made (usually called a print), to produce a positive image.
Positional measurements from photographic images are straightforward. Essen-
tially a traveling microscope is centered on each image of interest in turn and the
coordinates read off from the position scales on its x and y axes. For spectroscopy
(Chap. 12) only movement along one axis is needed. The original light intensity
corresponding to the density on a photographic negative is more difficult to obtain.
Generally a calibration exposure is needed to convert density to intensity. The calibra-
tion exposure is a photograph of a number of sources of differing and known
brightnesses. Since the processing of a photographic emulsion affects the densities it
records, the calibration photograph must either be on a small portion of the film or plate
carrying the main image or, if separate, be taken from the same batch as that for the
main image and processed together with it. Even with these precautions, the accuracy
of the levels of brightness obtained via photography will be low by today’s standards.
Other Detectors
The eye, the CCD and (perhaps for a little while still) the photographic emulsion are
the main detectors used by astronomers for visual radiation (and a little way into the
infrared for the latter two). Other detectors, such as p-i-n photodiodes, avalanche
photodiodes, superconducting tunnel junction (STJ or Josephson junction)

detectors, and photomultipliers, may be encountered but are left for interested
readers to research for themselves (see Appendix A of this book). Outside the
visual region there are many types of detectors, some operating on similar lines to
the ones encountered already, others based upon quite different physical processes,
though are also outside the scope of this book.
Exercise
9.1. The following graph shows the intensity distribution of the pixels in the CCD
image of an isolated galaxy (the size of the galaxy is significantly smaller than
the area covered by the CCD). Suggest a suitable gray scaling (contrast stretch)
to display the galaxy to best advantage on a monitor with 256 distinguishable
gray levels. The CCD can detect a maximum intensity of 100,000.
Chapter 10
Data Processing
Introduction
Observing time on major telescopes is usually allocated to individual astronomers

or groups of astronomers a few nights at a time, and at most three or four such
sessions might normally be available over a year. This does not mean that the
astronomers concerned can relax for the other 50 weeks, but it is a measure of the
relative efforts required at a research level in obtaining the data in the first place
compared with the processing then needed. Usually many multiples of the time
spent observing have to be spent afterwards working on the data (photographs,
CCD images, spectra, photometric measurements, astrometric measurements, etc.)
in order to get it into the finally desired form.
Such an imbalance in the division of effort has been neither usual nor generally
appropriate among observers using smaller instruments. Many such observers use
their telescopes for the joy of gazing into the universe and, especially to begin with,
are not concerned with dissecting out the details; secondly, there has perhaps seemed
little that could be done with a small telescope by the way of astronomical research
that would not be done first or better on the larger telescopes; and thirdly data
processing often requires large amounts of computational effort that were unavail-
able outside of the large research institutions. A century or more ago, most obser-
vations were obtained with what would now be classed as small telescopes and were
studied in depth, but in the recent past little effort has normally gone into processing
the observations obtained with smaller telescopes.
In the last few years that situation has changed fundamentally. The advent of the
CCD means that the owner of a 0.2-m telescope can undertake observations that a few
decades ago would have required a 2-m telescope (Chap. 9), and for a quarter of the
price of a 0.2-m Schmidt-Cassegrain telescope, an observer can purchase a computer
with the memory and processing power of a mainframe computer of a few years ago.
The 2-m telescope, of course, also uses a CCD, and so can make observations
that would once have needed a 20-m telescope. Fortunately, the universe is large
enough to make a plethora of interesting and largely unstudied objects now available to

198 10 Data Processing
all sizes of telescope. Additionally there are now numerous archives that have open
access via the Internet, some of which have been put together with various software
packages to become “virtual observatories.” Also there are now many “citizen science”
projects in which amateurs and professional scientists combine their efforts to conduct
leading edge research, as discussed later in this chapter. Original research is thus open to
most astronomers who care to undertake it, and this means that data processing will
become much more widely used than in the past.
Data processing breaks down roughly into two main subprocesses – data reduction
and data analysis, and we consider each of these briefly below.
Data Reduction
This does not mean making your data smaller! It is the “mechanical” side of data
processing, that is, the process or processes required to correct known faults or
problems in data, and then to get the data into the most convenient form for the next
stage, which is data analysis. Parts or all of data reduction can often be done
automatically if the data are stored in the computer or available in computer-
readable form. The exact stages involved in data processing vary with the
observations and with the ultimate purpose for which they are to be used, but
some examples will demonstrate the type of operation involved:
• Removal of cosmic ray spikes from CCD images.
• Removal of the background, and the flat fielding of CCD images (Chap. 9).
• Contrast stretching, false color representation and other image processing
techniques (Chap. 9).
• Determination of the response (characteristic curve) of a photographic emulsion and
the conversion of the photographic density of the original image back into intensity.
• Correction of geometrical distortions in the image.
• Smoothing or other noise reduction procedures, such as adding together several
images. Generally a signal must be at least as strong as the noise level in order for
the object to be regarded as detected (a signal to noise ratio, or S/N, of 1).
In practice most work requires S/Ns of 10–1,000 or more. Images, especially
those from radio telescopes, may be processed in various ways to reduce the noise.
The main such procedures for this are known as CLEAN and maximum entropy
methods (MEM). However, the details of these are beyond the scope of this book
and the interested reader is referred to sources in Appendix A of this book.
• Correction for expansion or contraction or other temperature-induced defects.
• Reduction of the blurring effect (usually called the instrument profile or point spread
function) of the telescope and other ancillary instruments used to obtain the data.
• Removal of electrical supply interference or other cyclic defects.
• Conversion of intensity values into stellar magnitudes.
• Calibration of the wavelength along a spectrum.
• Correction for atmospheric absorption.
Errors and Uncertainties 199
• Correction for the effect of Earth’s velocity on wavelengths (Doppler shift) and
position in the sky (aberration).
• Correction for the effects of any filters used while obtaining the observations.
• Calculation of means, standard deviations and standard errors of the mean for a
set of measurements, etc.
Errors and Uncertainties
Space is not available to consider all these processes in detail here, but the last
mentioned is so fundamental that familiarity with it is essential to any observer
intending to progress beyond the simplest of observations. Whenever you make
measurements, you should always try to estimate or measure the errors/
uncertainties in those measurements. The most direct way of determining the
random errors is to repeat the measurements many times. The scatter in the results
then gives an idea of the errors. That scatter is quantified by the standard deviation:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uP
un
u ðX xi Þ2
ti¼1
s¼ (10.1)
n1
where s is the standard deviation, n is the number of measurements, xi is the ith

measurement and X is the mean value (average) of the measurements.1
The symbol S means that the quantity following it is summed over all possible
values of xi from i ¼ 1 to i ¼ n, i.e.
X
n
ðX xi Þ2 ¼ ðX x1 Þ2 þ ðX x2 Þ2 þ ðX x3 Þ2 þ ::: þ ðX xn Þ2 (10.2)
i¼1
The standard deviation gives a measure of how far from the true value a single
measurement is likely to lie; 68% of measurements should lie within 1s of the true value,
95% of measurements within 2s of the true value, and 99% of measurements within 2.5s
of the true value (see the discussion below, within the section about the correlation
coefficient, on the use of the terms “significant” and “highly significant” in this context).
The standard error of the mean, S, is a measure of how far from the true value the
mean or average of a number of measurements is likely to lie:
s
S ¼ pffiffiffi (10.3)
n
1
Many scientific calculators and computer spreadsheets have pre-programmed algorithms to calcu-
late standard deviations and also standard errors of the mean, correlation coefficients, t values and to
undertake linear regression etc. The handbooks or help sections of the calculators or computers will
need to be consulted for further details on how to use these facilities in specific cases.
As with the standard deviation the probability of a mean value being close to the
true value is that 68% of mean values are likely to be within 1S of the true value, 95%
within 2S of the true value, and 99% within 2.5S of the true value. Thus repeating
measurements does not make any individual measurement likely to be more correct
than any other, but the mean value of the measurements should (slowly) approach the
true value. The standard deviation and the standard error of the mean are themselves
uncertain. Unless you are using more than about a hundred measurements, both
quantities can only be relied upon to about one significant figure.

Q. A spacecraft has made repeated measurements of the number density of atoms
in interplanetary space. If the measurements were made over a short period of
time (so that the spacecraft has not moved very far, nor would it be likely that the
density of the atoms has really altered) determine the average value of the
density, its standard deviation and the standard error of the mean.
Data:
Measured number of atoms per cubic meter (in millions)
5.5 6.1 3.0 5.4 4.9 4.8 5.2 6.2
A. The average value of these measurements is

X ¼ (5.5 + 6.1 + 3.0 + 5.4 + 4.9 + 4.8 + 5.2 + 6.2)/8
¼ 41.1/8 ¼ 5.1375
Thus
Average density ¼ 5.1375 106 m3
and so
S (X xi)2 ¼ [(5.1375 5.5)2 + (5.1375 6.1)2
+ (5.1375 3.0)2 + (5.1375 5.4)2 + (5.1375 4.9)2
+ (5.1375 4.8)2 + (5.1375 5.2)2 + (5.1375 6.2)2]
¼ 0.1314 + 0.9264 + 4.5689 + 0.0689 + 0.0564
+ 0.1139 + 0.0039 + 1.1289 ¼ 6.9989
and so from (10.1) we have
s ¼ [6.9989/(8 1)]0.5 ¼ [0.9998]0.5 ¼ 0.9999
Thus
Standard deviation ¼ 0.9999 106 m3
From (10.3) we have
S ¼ 0.9999/(8)0.5 ¼ 0.9999/2.8284 ¼ 0.3535
Thus
Standard error of the mean ¼ 0.3535 106 m3
Data Analysis 201
If you cannot repeat the measurements, you should still try realistically to
estimate the errors, for example from the known precision of your measuring
instrument, and the degree of “fuzziness” of the point you are measuring, etc.
Numerical results should be presented in the form
XS (10.4)
Do not give the mean value to more significant places than are justified by the
error estimate. Thus 12.5678 0.0002 is acceptable, but 12.55678 0.1345
should be given at the final stage (more figures can be retained during the calcula-
tion, and remember that only one significant figure is normally justified for the
standard error of the mean) as 12.6 0.1.

Q. Express the results from Example Calculation 10.1 correctly
A. From Example Calculation 10.1
X ¼ 5.1375 106 m3
s ¼ 0.9999 106 m3
S ¼ 0.3535 106 m3
Thus the correct expression of the result is
Number density of atoms ¼ 5.1 0.4 106 m3
Data Analysis
Data analysis is the conversion of the data after reduction into an astronomically
interesting form. Often it will more appropriately be called data synthesis, and this is
where the astronomer usually makes his or her most significant contribution to
the whole observing process. Data analysis varies even more than data reduction
in the stages involved. At its simplest, it may be just the comparison of a new
measurement of the magnitude of a star with one made previously to see if it has
varied in brightness. At its most complex, it may involve the detailed computer
modeling of a stellar or planetary atmosphere or interior, interstellar nebula, galaxy,
cluster, etc., in order to determine rates of nucleosynthesis, element abundances,
velocity, temperature or density structures, sources of energy, and so on, and it may
need to include data of many different types, from many different sources, require
complex theories, abstruse mathematics, and advanced concepts from physics and
other sciences.
Data analysis thus depends both upon the nature of the observations and the
purpose for which they are intended. It is not possible to give any sort of general
guide to what may be involved. Some of the processes are discussed in Chaps. 8, 11,
and 12, but the observer will need to seek a more specialist guidance when he or she
gets to the stage of undertaking original research.
There are numerous statistical procedures that can be applied within data
analysis, and many of these require advanced knowledge of statistics and/or have
only specialist applications. Three statistical techniques, however, are designed to
answer basic questions that crop up very frequently in experimental work:
• What equation best fits the data that I have obtained?
• Do my measurements of quantity A show that it is related to quantity B?
• Has the object that I am measuring changed between the two occasions when I
studied it?
These three questions are addressed by the statistical techniques of linear
regression, the correlation coefficient, and the Student’s t test. The practical use
of these techniques is outlined below; their theoretical background, however, is left
for the interested reader to pursue in more specialist texts on statistics and data
analysis.
Linear Regression
This technique is also known as linear least squares curve fitting. It finds the linear
equation that is the “best” fit to a set of measurements of pairs of (possibly) related
quantities, such as the periods and luminosities for Cepheid stars, sunspot numbers
and terrestrial magnetic activity, radial velocities of galaxies and their distances,
etc. That is to say, for a set of measurements of x and y, it finds values of a and b that
give the “best” fit to the data in an equation of the type
y ¼ ax þ b (10.5)
The technique has to be used with some caution. First, the word “best” has been
put here in quotes because it is defined in an arbitrary manner. Linear regression
produces the line for which the sum of all the squares of the distances of the
measured points in the y direction from the line is minimized (Fig. 10.1). It can
therefore give excessive weightings to data points well away from the average. Other
definitions of the difference between the line and the data points could be used, such
as the modulus of the distance from the line or the logarithm of the distance.
However, these are mathematically more difficult to manipulate, and so the “least
squares” solution is generally adopted. For what it is worth, if you give a graph with a
scatter of points on it to a number of people and ask each one to fit a straight line to
the points by eye, then the results are usually quite close to the least squares fit.
Data Analysis 203
Fig. 10.1 The least squares sum
The second problem with linear regression is that it will always give an answer,
even when the quantities are connected in a non-linear fashion such as an exponential
or sine relationship,2 or when there is no relationship at all. It must therefore be used
with care and with some physical insight into the processes being studied. The
significance of the relationship determined by linear regression can also be assessed
using the correlation coefficient (see below).
The third problem is that linear regression assumes that one set of data has a
much higher level of precision than the other. This is often the case, for example, in
the light curve of a variable star; the times of the observations will be very much
more accurately known than the magnitude of the star. The formula then assigns the
x variable to this high precision data set. (The data can simply be relabeled if it is the
y set that is more accurate.) However, if both measurements have significant
uncertainties, then linear regression may not give the best fitting curve.
With the above cautions, the use of the linear regression method just requires the
use of two slightly forbidding, but nonetheless straightforward, equations. Using
the same notation as for (10.1), the values of a and b in (10.5) are given by:
2
You may still be able to fit a least squares linear equation to a set of data points even when the
physics of the phenomenon suggests that the relationship is not linear. For example, if theory
suggests a quadratic equation (i.e., y ¼ a x2 etc.) for a relationship, then y should have a linear
relationship with √x and the linear regression formulae will give the linear least squares best fit for
it. You should note, however, that this will not usually be quite the same as the quadratic least
squares best fit to the same data, but it is much simpler to calculate and may be good enough to get
the analysis started.
P
n
ðX xi ÞðY yi Þ
i¼1
a¼ P
n (10.6)
ðX xi Þ2
i¼1
b ¼ Y aX (10.7)

Q. We might well expect that the luminosity, l (in units of the solar luminosity,
4 1026 W), of a globular cluster would be linearly related to the number of
stars, n, in the cluster. Determine a best fit formula for that relation using the
following set of measurements of the luminosities and estimates of star numbers
of some globular clusters:
Data:
Cluster M3 M5 M4 M13 M92 M22 M15
l 302,000 209,000 21,000 159,000 159,000 48,000 331,000
n 210,000 60,000 60,000 300,000 140,000 300,000 450,000
A. We have (with the luminosity substituting for x, and the number of stars for y)
Average luminosity ¼ L ¼ 175,600
Average star numbers ¼ N ¼ 217,100
So we get (some of the information is for Example Calculation 10.4)
li (L–li) (L–li)2 ni (N–ni) (N–ni)2 (L–li) (N–ni)

302,000 126,400 1.5977 1010 210,000 7,100 5.0410 107 8.9744 108
209,000 33,400 1.1156 109 60,000 157,100 2.4680 1010 5.2471 109
21,000 154,600 2.3901 1010 60,000 157,100 2.4680 1010 2.4288 1010
159,000 16,600 2.7556 108 300,000 82,900 6.8724 109 1.3761 109
159,000 16,600 2.7556 108 140,000 77,100 5.9444 109 1.2799 109
48,000 127,600 1.6282 1010 300,000 82,900 6.8724 109 1.0578 1010
331,000 155,400 2.4180 1010 450,000 232,900 5.4242 1010 3.6193 1010
S 8.1976 1010 1.2334 1011 4.3661 1010
From (10.6) therefore:
P
n
ðL li ÞðN ni Þ
i¼1 4:3661x1010
a¼ P
n ¼ ¼ 0:5326
ðL li Þ2 8:1976x1010
i¼1
(continued)
Data Analysis 205
(continued)
and
b ¼ N aL ¼ 217100 0:5326 x 175600 ¼ 123600
and so we have the linear formula (taking reasonable account of the number of
significant figures in the original data) relating the number of stars in a globular
cluster to its luminosity:
Star number ¼ 0:53 x luminosity þ 120000
That this is not an unreasonable formula may be seen from Figure Example 10.3.
Figure Example 10.3 Globular clusters’ luminosities and star numbers – observed data and a
linear regression fit to them
However a more thorough analysis of the significance of the formula is

contained in Example Calculation 10.4.
Correlation Coefficient
Since linear regression produces a formula whether the two quantities are related or
not, it is usually necessary to check how significant the correlation of the measure-
ments may be, before using the formula for other purposes. Calculation of the
correlation coefficient performs this task.
However, the question that the use of the correlation coefficient actually answers
is, “What is the probability that the measurements are not related?” A highly
probable result from the use of the correlation coefficient therefore implies that
the quantities measured are not related, while a low probability result implies that
they are related. This approach to assessing probabilities is quite common within
statistics (see also Student’s t test below), but it can cause confusion when first
Fig. 10.2 The correlation coefficient
encountered. However, the test can never tell you that two quantities are definitely
related; it only gives the probability of that being the case.
By convention, a correlation is called significant if its probability is 5% (i.e.,
there is a 95% chance that the quantities are related) and highly significant if its
probability is 1% (i.e., there is a 99% chance that the quantities are related).
A research paper would not normally be accepted for publication if a claimed
correlation was poorer than significant (the result of the test was a probability
higher than 5%, i.e., the chance that the quantities are related was less than 95%).
This conventional use of the terms significant and highly significant also applies in
other circumstances such as Student’s t test (below), standard deviation and the
standard error of the mean (above).
The calculation of the value of the correlation coefficient uses a formula (10.8)
that is similar to that involved in linear regression. (Note that if you are using both
techniques, some quantities are duplicated between the formulae, and so do not
have to be calculated twice.) But the interpretation of the result requires the use of a
graph (Fig. 10.2) or a set of statistical tables, and introduces a new quantity – the
number of degrees of freedom. This is a statistical concept whose theoretical
background is beyond the scope of this book. The practical realization of the
number of degrees of freedom for the correlation coefficient however is simple.
It is just one less than the total number of pairs of data points.
The value of the correlation coefficient, r, is determined via the formula
P
n
ðX xi ÞðY yi Þ
i¼1
r ¼ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (10.8)
P n P n
ðX xi Þ2 ðY yi Þ2 :
i¼1 i¼1
Data Analysis 207
It is then interpreted from Fig. 10.2 (r has a value of +1 when the two quantities
are perfectly correlated; of 1, when the two quantities are perfectly anti-correlated;
and a value of 0 when they are uncorrelated).

Q. Determine the significance level of the correlation of the data in Example
Calculation 10.3 and comment on the usefulness of the formula found there (the
data and the preliminary calculations may be found in the example).
A. From (10.8) and the figures in Example Calculation 10.3, we have:
P
n
ðL li ÞðN ni Þ
i¼1 4:3661 1010
r ¼ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P n P n 8:1976 1010 1:2334 1011
ðL li Þ2 ðN ni Þ2
i¼1 i¼1
4:3361 1010
¼ ¼ 0:4320
1:0014 1011
There are seven pairs of measurements, so the number of degrees of freedom

is six. From the graph in Fig. 10.2 we may therefore see that the result of the
correlation coefficient test is a probability much higher than 5%. Since this is
the probability that the quantities being measured are not related, we may state
(on the basis of this data at least) that there is no significant correlation between
a globular cluster’s luminosity and the number of stars that it contains.
The formula found in Example Calculation 10.3 is therefore not a useful way
of estimating the number of stars in a globular cluster.
NB: Using more extensive tables than the data shown in Fig. 10.2, we find that
r 0.4 corresponds to a significance level of 8–9% – i.e., the probability that
there is a relationship between a globular cluster’s luminosity and the number of
stars that it contains is over 90% – and that, most people would say, is good
betting odds.
So what this analysis really says is that there may well be a useful relation-
ship between a globular cluster’s luminosity and the number of stars that it
contains, but the data obtained so far is insufficient to get a claim of such a
relationship published in a refereed professional journal. Thus, what is needed is
more research in order to produce better and/or more data that may show that
there is a significant relationship – or may demonstrate the lack of a relationship
– more convincingly.
Student’s t Test
This test tries to answer the question “Do two sets of measurements of the same
object or quantity at different times (or wavelengths, positions, polarizations, etc.)
show that it has changed?3”
Like the correlation coefficient, the test is based on the assumption that the two sets
of data do not differ from each other, and determines the probability that any observed
difference is due to random errors only. The lower the probability value given by
the test, therefore, the higher the probability that the two sets of measurements do
show that the object has changed. Thus a result evaluated at 1% by Student’s t test
gives a 99% chance of the measured quantity having changed.
Theoretically any value of the probability between 0 and 1 (0–100%) can be
found from applying Student’s t test, though in practice it will never be possible to
state with certainty that a quantity has or has not changed. As with the correlation
coefficient, the values 5% and 1% are chosen arbitrarily as decision points, and sets
of measurements evaluated at these two levels are termed significant and highly
significant, respectively.
Student’s t parameter is found from the equation

X X
XA XB A B

t ¼ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (10.9)
2 S2 þ S2
s s 2
B
A
þ B A
NA NB
where A and B are the two sets of measurements, X is the average value for the set, s
its standard deviation and N the number of measurements in the set.
For Student’s t test the number of degrees of freedom is the number of
measurements in each set minus 2 (i.e., NA + NB2). The significance of the result
of the test may then be read off the graph in Fig. 10.3.

Q. The spacecraft mentioned in Example Calculation 10.1 makes a second set of
measurements 24 h after the first set. The results are
Number of measurements ¼ NB ¼ 10
Average density ¼ XB ¼ 5.3817 106 m3
Standard deviation ¼ sB ¼ 1.0981 106 m3
Standard error of the mean ¼ SB ¼ 0.3472 106 m3
(continued)
3
This is not a simplified test suitable for students that is to be superseded by a better test for higher
levels of work. The name comes from the statistician who devised the test, William Gosset. He
wrote popular mathematics articles under the pen-name “Student.”
Data Analysis 209
Fig. 10.3 Student’s t test
(continued)
Is there a significant difference between the space densities of atoms at the
two sample sites?
A. From Example Calculation 10.1 the first set of results are
Number of measurements ¼ NA ¼ 8
Average density ¼ XA ¼ 5.1375 106 m3
Standard deviation ¼ sA ¼ 0.9999 106 m3
Standard error of the mean ¼ SA ¼ 0.3535 106 m3
Thus from (10.9)

XA XB 5:1375 10 6
5:3817 10 6
t ¼ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
¼ vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
!
u
sA s2B u 0:9999 106 2
1:0981 106
2

þ t þ
NA NB 8 10

0:2442 10 6 0:2442 106

¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ ¼ 0:4928
1:2498 1011 þ 1:2058 1011
4:9554 105
From Fig. 10.3, with 16 degrees of freedom (¼ 8 + 10 2), we may see that a
value of t of about 0.5 is well above the 5% significance level (actually at about
50%) and so the difference between the two values obtained for the space density
of atoms is not significant.
Archives, Data Mining, Virtual Observatories

and Citizen Science Projects
Modern major astronomical instruments – whatever wavelength region they may

cover and whether ground- or space-based – generate vast quantities of data. For
example, the proposed Square Kilometer Array (SKA) radio telescope is expected
to produce around one exabyte4 of information every day or so. To put this in
perspective, currently personal computers have hard disks with memories of a
terabyte or so. A million such computers would thus be needed to store the results
just 1 day’s observing by the SKA.
Astronomers have traditionally archived their data for possible future use. Even
when this meant just stacking photographic plates in a suitable room somewhere at
the observatory, the archive was invaluable for seeing if a star had changed
brightness, to measure proper motions, to search for pre-discovery images of a
new comet, asteroid or nova and so on.
Now, many of those photographic archives have been digitized, and all new data
is archived in that form so that the information is readily available. This has led to a
new way of undertaking astronomical research, called data mining. Essentially this
just means looking through archive material to see if there is something new to be
inferred from it. Since most observations are made for a specific purpose, the data
that is generated may contain information that the original researchers did not use.
Alternatively different data sets may be combined to produce new insights or data
processing and analysis techniques may improve with time and allow for better
interpretation of the data. For example, in 2003 ESO’s VLT was used to obtain
images of the star b Pic in order to try and detect a suspected exoplanet. None was
then found, but the planet was detected in 2008 within that same set of data using a
more refined analysis technique.
It might seem that once an archive has been digitized, then it could be accessed
via the Internet by anyone based anywhere in the world. To some extent this is true –
the archives at the Centre de Données astronomiques de Strasbourg (http://cdsweb.
u-strasbg.fr/), for example, are open to anyone without restriction, as are the archives
of the Hipparcos spacecraft (http://apm5.ast.cam.ac.uk/hipp/hipparcos.html). How-
ever other archives have their access restricted to recognized scientists (although in
many such cases anyone, including amateur astronomers, who can demonstrate a
viable research program and who has suitable computing facilities may given access
if the archive manager is contacted). The volume of data there may be in the archive
and the time therefore that it takes to work through it are likely to be of more
importance in restricting access. At a broadband transmission rate of 10 megabits per
second, for example, a terabyte of data would take nearly 10 days to download.
4
1018 bytes. See Appendix D in this book for a list of the standard prefixes for very large and very
small numbers.
Archives, Data Mining, Virtual Observatories and Citizen Science Projects 211
In the last decade or so therefore a number of virtual observatories have been

developed. The virtual observatory comprises one or more archives of data together
with software to process that data to a greater or lesser extent. The researcher will
thus only have to download the results of such processing, not the raw data itself.
Generally the processed data will require far less memory than the raw data. Thus in
Example Calculation 10.1, the raw data comprises eight numbers, the processed
data three (number of measurements, average density, and standard deviation – or
four numbers if the standard error of the mean is included). Usually the reduction in
the volume of data will be by a far larger amount than this – compressing an image
of a star field (Chap. 9) may easily reduce the data by a factor of 10 or a 100.
The National Virtual Observatory (NVO – http://www.us-vo.org/) based in the
United States and the Euro-Virtual Observatory (Euro-VO – http://www.euro-vo.
org/pub/) are examples of virtual observatories that are open to anyone to use.
Using a virtual observatory, especially for genuine research projects, is likely to
require the person concerned to have a good deal of expertise both in astronomy and
computing. Anyone with a computer and an Internet connection, though, can take
part in real research via various citizen science schemes. Some of these schemes
merely require the use of a computer’s spare capacity at times when it is not
otherwise required, but without much additional input from the owner. Software
is downloaded onto the computer to process small portions of data taken from a
much larger database. The results are then sent back to the central organizers for
collation and further analysis. Other schemes are similar but additionally require
involvement of the reader in the analysis to a greater or lesser extent.
One of the longest running astronomy-related citizen science projects is
SETI@home (http://setiathome.berkeley.edu/). This started in 1999 and uses up
to three million home computers to analyze data from radio telescopes in a search
for extra-terrestrial intelligences. Other schemes include Galaxy Zoo (http://www.
galaxyzoo.org/), which aims to classify the hundreds of thousands of galaxies in the
HST archive and MoonWatch (http://astro.ukho.gov.uk/moonwatch/index.html),
whose purpose is to find the youngest new Moon visible to observers in mid-
latitudes. An Internet search for ‘citizen science’ will soon find those projects that
are still active at any particular time.
Exercises
10.1. Calculate the mean, standard deviation and standard error of the mean of the
following set of measurements of the Sun’s angular diameter in seconds of arc:
959.6 959.3 959.5 959.8 959.75 959.0 960.1 959.7 959.65 959.4
10.2. Express the following numbers correctly:

12.556 78 0.1345
3.59 0.01
5002 103
0.001 2345 0.000 6789
3.09 106 105

3.318 1016 2.7 1016
10.3. Find a linear relationship between the apparent magnitudes and numbers of
asteroids in each one-magnitude range.
Apparent magnitude range Number of asteroids

4.5 0.5 2
5.5 0.5 1
6.5 0.5 6
7.5 0.5 25
8.5 0.5 80
9.5 0.5 160
10.5 0.5 320
11.5 0.5 800
Hence predict the number of asteroids in the range 15m to 16m. How realistic
is this answer likely to be?
10.4. A connection is suspected between the number of astronomy students at an
observing session and the temperature. Is there a real correlation?
Number of students 20 10 15 21 20 5 25 25 25 28
Temperature ( C) 5 7 8 0 5 9 9 5 15 10
10.5. The separation of a close double star has been measured on two occasions.
Has the separation changed over time?
First set of measurements in seconds of arc (A)

1.29 1.33 1.34 1.35 1.27
Second set of measurements in seconds of arc (B)
1.27 1.30 1.26 1.28 1.25 1.27
10.6. Has the brightness of a star changed between two observing sessions?
First set of brightness measurements (A)

8.302 8.324 8.306 8.294 8.289 8.295
8.310 8.296 8.322 8.284 8.285 8.321
8.294 8.309 8.347
Second set of brightness measurements (B)
8.284 8.315 8.289 8.271 8.263 8.293
8.173 8.313 8.257 8.258 8.299 8.312
Chapter 11
Photometry
Introduction
Photometry is the process of determining intensity or magnitude of or within the

original object. It most commonly involves measurements of point sources or near-
point sources like stars and planets, but it can also involve measuring the integrated
intensities or magnitudes of more extended objects. Any of the detectors discussed in
Chap. 9 can be used for photometry. The CCD and p-i-n photodiode are, however,
the most straightforward because they have linear responses, and their outputs can be
converted directly into magnitudes using (8.5). The eye and the photographic
emulsion are non-linear detectors, and their responses are therefore more complex
to convert into magnitudes. The visual estimation of magnitude having already been
discussed in Chap. 8, we shall consider CCD photometry below. Photographic
photometry can still be undertaken, but it is rather pointless to do so since the
practical side is complex; measuring the photographic images requires expensive
and sophisticated machinery, and the final results are of relatively low accuracy.
Readers interested in the technique will, though, find it discussed in earlier editions
of this book.
CCD Photometry
This is straightforward since the CCD response is linear. Scintillation, however,

will normally spread the images, even of stars, over several pixels. The actual
intensity must therefore be found by adding the intensities of those pixels together.
Fortunately most manufacturers of small astronomical CCDs supply software that
enables this to be done and in some cases to give the stars’ magnitudes directly.
For this, however, there will need to be some standard stars (stars with known
magnitudes) within the image as well as the star(s) of interest. The large CCD
arrays used on major telescopes will have their own customized software written for

214 11 Photometry
them, and anyone using them will probably be formally trained in their use or at
least have access to comprehensive handbooks and computer-based help systems.
Filters will normally need to be used to restrict the range of wavelengths being
detected (see the section on UBV photometry below). If you wish to compare your
magnitudes with those measured elsewhere, then the CCD-filter combination will
need to give the same overall response as the standard detector-filter combinations.
Since the wavelength sensitivity of the CCD differs markedly from that of the eye,
photomultiplier and p-i-n photodiode, filters sold for use with those detectors will
not give the correct response when used with a CCD. Most manufacturers of
astronomical CCDs, however, will also supply suitable filters for use with them.
The dynamic range of CCDs is large (100,000 or more – equivalent to 12.5
stellar magnitudes). However, it is possible to saturate the image, and the
magnitudes then obtained will be incorrect. If the magnitudes of a very bright star
and a very faint star are both needed, it may be necessary to take two images, one
short enough not to saturate the image of the bright star, and the other long enough
to give a good image of the faint star. Some CCDs have extra electrodes that drain
off electrons as pixels approach saturation. This process is known as anti-blooming.
The images will not then be saturated, but the effect is to make the CCD response
non-linear at high intensities. You will need to check with the manufacturer of your
CCD to find out if your device operates in this way. If it does, then you will only be
able to use it for photometry on images that are well away from being saturated.
Absolute Magnitude
The magnitudes so far considered are all as the objects appear in the sky. They are
therefore termed apparent magnitudes, and are denoted by the lower case m. The
relationship between energy and apparent magnitude was discussed in Chap. 8 (8.5).
Clearly, apparent magnitude tells us nothing about the actual energy or luminosity of
the object. Thus the Moon (Table 8.2) has an apparent magnitude of 12.7, and
Betelgeuse one of 0.73, but the Moon is not actually 60,000 times brighter than
Betelgeuse – we just see it so because the Moon is so much closer to us.
To provide a quantity that is related to the actual luminosity of the object, we use
the absolute magnitude (denoted by the upper case M). The absolute magnitude is
defined as the apparent magnitude that the object would have if its distance were
10 pc. Differences in absolute magnitudes thus reflect real differences in the levels
of brightness of the objects. The absolute and apparent magnitudes are related by
M ¼ m þ 5 5log10 D (11.1)
where D is the distance of the star in parsecs.

The absolute magnitudes of some objects are listed in Table 11.1, from which we
may see that the true relationship between the Moon and Betelgeuse is that the latter
is 1015 times brighter than the former (however, see also the section on bolometric
magnitude, below).
Absolute Magnitude 215
Table 11.1 Absolute Object M

magnitudes
Sun +4.83
Full Moon +32
Venus (maximum) +29
Jupiter (maximum) +26
Sirius A +1.4
Betelgeuse 6.05
Polaris 3.63 (variable)
Type 1a supernova 19
Faintest white dwarfs +16
Faintest brown dwarfs +20

Q. Determine which of the two stars, Castor and Pollux (the “heavenly twins” in
Gemini – a Gem and b Gem), is actually the brighter star.
Data:
Castor: m ¼ +1.58m, Distance ¼ 15.3 pc
Pollux: m ¼ +1.14m, Distance ¼ 10.4 pc
A. From (11.1)
MCastor ¼ mCastor þ 5 5log10 DCastor ¼ 1:58 þ 5 5log10 15:3

¼ 6:58 5:92 ¼ 0:66m
MPollux ¼ mPollux þ 5 5log10 DPollux ¼ 1:14 þ 5 5log10 10:4
¼ 6:14 5:09 ¼ 1:05m
Thus Castor, whose apparent magnitude makes it fainter than Pollux (0.67)
in the sky, is actually the brighter star by 0.39m (¼ 1.43).
Additional note: Castor is in fact a visual binary system whose combined
luminosities give the above apparent magnitude of +1.58m. The two components
of the binary have individual apparent magnitudes of +2.0m and +2.9m. Their
absolute magnitudes are thus +1.08m and +1.98m, making both stars actually
fainter than Pollux (by factors of 0.97 and 0.42), although Castor A is almost
as bright as Pollux.
However, both Castor A and Castor B are spectroscopic binary stars, and there
is a faint eclipsing binary star system some 7200 away that is probably also gravita-
tionally linked to the other stars. The two spectroscopic binaries have primary
components with spectral class A (Chap. 12) and secondary components of
spectral class M. The secondary stars will therefore be much fainter than the
primary stars (by around a factor of 0.01) and so the absolute magnitudes of
the primary stars will still be close to +1.08m and +1.98m. Pollux is a single star but
does have a three Jupiter-mass planets in orbit around it.
216 11 Photometry
Wavelength Dependence
Stars and many other astronomical objects have spectra that are approximately
black-body1 in overall shape (Fig. 11.1). A hotter star will therefore be relatively
brighter in the blue parts of the spectrum compared with a cooler star. Conversely,
the latter will be relatively brighter at longer wavelengths. The value obtained for
the magnitude therefore depends upon the wavelength at which it is measured.
Magnitudes estimated by eye are called visual magnitudes (mv or Mv) and
correspond roughly to the relative energies of the objects at 510 nm. Magnitudes
Temperature = 20,000 K
x1.0
Relative Intensity
x25
(arbitrary units)
x600
x40,000
x700,000
Temperature = 500 K
x20,000,000
Temperature = 300 K
x500,000,000
Temperature = 100 K
x10,000,000,000
0 2 4 6 8 10
Wavelength (microns)
Gamma rays,
Visible Infrared
X-rays,
Ultra-violet
Fig. 11.1 Black-body curves from 100 to 20,000 K. Note that in reality the curves do not cross each
other. The curve for a lower temperature is always below that for a higher temperature over the
whole of the wavelength range. However in order to illustrate the results for a range of temperatures,
in this plot, apart from the first curve (20,000 K), the remaining curves have been expanded vertically
by the factor shown in each case – leading to the apparent crossing of the curves
1
A black-body is not black (except at a temperature of 0 K). It is defined as an object that perfectly
absorbs electromagnetic radiation of all wavelengths. There is no such thing as a true black-body
in nature, but a small hole in the side of a closed box comes close. When a black-body is heated it
emits radiation with a characteristic spectrum (Fig. 11.1) that peaks at some wavelength and then
tails off towards zero intensity at longer and shorter wavelengths. The peak occurs at shorter and
shorter wavelengths as the temperature rises, an effect known as Wien’s law. The Sun, for
example, with a temperature of 5,780 K has its emission peaking in the green (at a wavelength
of 502 nm).
Bolometric Magnitude 217
obtained from blue-sensitive photographs are called photographic magnitudes

(mp or Mp) and correspond to the energies around 440 nm. Magnitudes obtained
from orthochromatic photographs, which mimic the spectral sensitivity of the eye,
are called photovisual (mpv or Mpv). CCDs and other detectors such as photomul-
tipliers have spectral sensitivities that differ from each other, and from those of the
eye and the various types of photographic emulsion.
It is therefore usual to use filters to isolate the wave band over which a magnitude
is to be measured. There are innumerable such filter systems currently in use by
astronomers, many of which are designed for highly specific and restricted purposes,
such as monitoring the strength of a particular spectrum line, and have little merit
outside that application. The most commonly used system of filters is, however, in
very widespread use, and is called the UBV system. It is the only one that we shall
consider here.
The commonest of the photometric filter systems in the visual is called the UBV
system. It uses three filters, centered on 365 nm (Ultraviolet), 440 nm (Blue)
and 550 nm (Visual), each with a bandwidth of about 100 nm. The magnitudes
obtained through the filters are designated U, B and V. The V magnitude thus
corresponds roughly to that which would be estimated by eye. The B magni-
tude corresponds roughly to that obtained using blue-sensitive photographic
emulsion. The U magnitude does not have a visual or photographic analog and is
badly affected by atmospheric absorption, so is of little use except when obtained
from high altitude sites.
One of the main reasons for undertaking photometry is that it can provide a
simple method of determining the temperature of a star. In the UBV system, the
temperature is obtained from the difference between the B and V magnitudes
(usually called the B-V color index) using the semi-empirical relationship
8540
T K (11.2)
ðB VÞ þ 0:865
Bolometric Magnitude
The true luminosity of a star is given by the absolute bolometric magnitude, MBol.
The bolometric magnitude takes account of the energy emitted at all wavelengths
from gamma rays to radio waves. The bolometric magnitude may be estimated from
the star’s absolute magnitude in the blue-green region (V or MV) and its spectral
class (Chap. 12). The bolometric correction (BC – Fig. 11.2) is the amount that must
be subtracted2 from the visual absolute magnitude to give the bolometric absolute
magnitude, i.e.,
2
In some sources the bolometric correction may be listed as a negative number that must then be
added to the visual magnitude to give the bolometric magnitude.
218 11 Photometry
Fig. 11.2 Bolometric correction (This is the curve for main sequence stars. The relationships for
giant and supergiant stars are slightly different from this curve, but only by small amounts.)
MBol ¼ MV BC (11.3)
The total energy emitted by the star (i.e., its luminosity) is then
Luminosity ¼ 3 x 1028 x 102 x MBol =5 W (11.4)

Q. Determine the absolute bolometric magnitudes of Pollux and the Sun (using
the result from Example Calculation 11.1 and the data in Table 11.1 and
Fig. 11.2) and hence their luminosities. Which is the brighter star and by what
factor?
Data:
Spectral type of Pollux: K0
Spectral type of the Sun: G2
A. From Example Calculation 11.1 and Table 11.1
MPollux ¼ +1.05m
MSun ¼ +4.83m
From Fig. 11.2 we get
BCPollux ¼ +0.25m
BCSun ¼ +0.10m
From (11.3)
(continued)
Photometers 219
(continued)
Pollux: MBol ¼ MV BC ¼ 1:05 0:25 ¼ 0:8m
Sun MBol ¼ MV BC ¼ 4:83 0:1 ¼ 4:73m
From (11.4)
LuminosityPollux ¼ 3 x 1028 x 102 x MBol = 5 ¼ 3 x 1028 x 102 x 0:8 = 5

¼ 3 x 1028 x 0:48 ¼ 1:4 x 1028 W
LuminositySun ¼ 3 x 1028 x 102 x MBol = 5 ¼ 3 x 1028 x 102 x 4:73 = 5
¼ 3 x 1028 x 0:013 ¼ 3:9 x 1026 W
Thus Pollux is brighter than the Sun by a factor of 36.
Spatial Information
Much of photometry is concerned only with the brightnesses of points or small sources.
Sometimes, however, the variation of brightness with position is important. Thus in
spectroscopy (Chap. 12) the variation of intensity with wavelength is needed; in the
image of an extended object the precise variation of intensity across it may be required,
or its integrated magnitude may be wanted. Such information may be obtained by
repeated or scanned observations with a point detector, but is generally more easily
found from an image. With a CCD image, each pixel is essentially a point photometer,
and the magnitude of its charge is directly related to the intensity at that point.
Photometers
Magnitudes may be estimated by eye or found from CCD or photographic images as

just discussed, but for the highest precision work, instruments specifically designed
to obtain brightness measurements, known as photometers, must be used. These
vary widely in their details, but most designs are based upon the basic photometer
(Fig. 11.3). This has a small entrance aperture to reduce background noise, a finding
and guiding system, a means of interchanging filters rapidly, and a detector. The
latter, for visual work, is usually an avalanche photodiode or a p-i-n photodiode.
Often the detector needs to be cooled to reduce thermal noise. A Fabry lens may
also be used; this focuses the telescope objective onto the detector and provides a
uniform image that does not move over the detector’s surface with the scintillation
motion of the source or with tracking errors. Small photometers based upon p-i-n
photodiodes are produced commercially, but a CCD can be used in their place much
of the time.
220 11 Photometry
Fig. 11.3 Basic photometer
Observing Techniques
Photometry using a photometer requires precise observing techniques if it is to give

accurate results. A standard star for comparison is needed and this should be close
to the star of interest in the sky, of similar brightness, and of similar spectral type
(Chap. 12). The observing sequence is then to measure first the comparison star,
then the star of interest, and then the comparison star again to ensure that the
observing conditions or the instrument settings, etc., have not changed during the
time taken for the observations. For each filter that is to be used, first the sky
background is measured, then the star, then the background again, and this
sequence is repeated four or five times. This procedure is also to ensure that
conditions are not changing and to enable a measure of the uncertainties (Chap.
10) to be obtained. The output, which now will usually be to a computer, but in the
past would have been to a chart recorder, of such an observing run is shown
schematically in Fig. 11.4. Magnitudes can be obtained from the data using (8.5)
and then color indices, temperatures, absolute magnitudes, etc., calculated.
Although magnitudes can be obtained from single CCD images, it is better to use
several such images, so that uncertainties, changing sky conditions, etc. can simi-
larly be estimated. Otherwise the principles of the process are essentially similar to
that for the basic photometer. Spacecraft such as Kepler using arrays of CCDs can
achieve photometric measurements of stars to within 0.002 % – a far better
degree of precision than most ground-based equipment. To achieve this, Kepler
takes an image every 6 s and is deliberately slightly de-focused so that the star
images are spread over 20–30 pixels.
Observing Techniques 221
Fig. 11.4 Schematic output

from an observing run with
a basic photometer: (a) first
set of readings for the
comparison star; (b) set
of readings for the star
of interest; (c) second
set of readings for
the comparison star
Exercises
11.1. A standard star, of apparent V magnitude 7.31, gives an output from a
photometer of 0.362 mV. What are the apparent V magnitudes of two nearby
stars for which the outputs from the same photometer are 4.079 mV and
209 mV?
11.2. From what distance could we just detect Jupiter, if observing in the visible
with the best of present-day techniques, but from well outside the Solar
System (ignore contrast problems with the Sun)? Suggest whether or not
we should be able to detect comparable planets around other stars by this
method.
11.3. Determine the absolute bolometric magnitude and the total energy emitted by
Betelgeuse using the information in Table 11.1 and Fig. 11.2, given also that
its spectral class is M2 I. The Sun emits 4 1026 W, so how much brighter is
Betelgeuse than the Sun?
Chapter 12
Spectroscopy
Introduction
Spectroscopy is the study of the way that the brightness of an object varies with
wavelength. UBV photometry gives some information of this type (Chap. 11). But,
by common consent, spectroscopy normally has a spectral resolution of 1% or
better, photometry a spectral resolution of 1% or worse. Although it is the most
fruitful technique available to astronomers, capable of yielding information on
temperatures, compositions, luminosities, pressures, magnetic fields, levels of
excitation and ionization, surface structure, line of sight velocities, turbulent
velocities, rotational velocities, expansion/contraction, binarity, and, less directly,
distances, masses and ages, spectroscopy has generally found little favor among
users of smaller telescopes. This is almost certainly because of the long exposures
normally required, even on large telescopes, in order to obtain a spectrum.
With a CCD detector, however, spectroscopy becomes eminently possible for
telescopes of 0.2 or 0.3 m aperture and above, and small spectroscopes are now
available commercially. This chapter is therefore written more in the hope of
encouraging wider interest in the topic among observers with such instruments
than as a guide to current practice.
Spectroscopes
Spectral resolution and dispersion are the two primary criteria governing the
performance of spectroscopes. They are defined by
l
Spectral resolution ¼ R ¼ (12.1)
Dl

224 12 Spectroscopy
dl
Dispersion ¼ (12.2)
dx
where l is the operating wavelength, Dl is the smallest wavelength interval that can
be distinguished and dl is the change in wavelength over a linear distance dx along
the spectrum.
Spectral resolutions used in astronomy range from 10 to 100,000 (Dl ¼ 50 to
0.005 nm for visual work), and dispersions from 200 to 0.01 nm mml. The lower
resolutions and poorer dispersions mostly arise, however, in searches and survey
work, and so minimum useful values are more normally about 300–500 for spectral
resolution and 10–20 nm mml for dispersion. Such levels of performance are
achievable by quite a simple spectroscope using a small prism or diffraction grating.
There is room here only to outline what is involved in designing a spectroscope, and
the interested reader is referred to more specialist texts (Appendix A of this book) for
further information.
The basic spectroscope contains six major elements (Fig. 12.1): an entrance slit to
give a pure spectrum and to limit background noise, a widener, a collimator, a
dispersing element, a focusing element and a detector. The final spectrum is com-
posed of overlapping monochromatic images of the entrance slit, and this is why
features in the spectrum are normally seen as lines. The slit should thus be as narrow
as possible, without excluding too much light from the object being observed, and its
sides should be accurately parallel to each other.
If guiding is to be undertaken, then this is usually accomplished by having the
outer faces of the slit jaws polished and then observing the overspill of light from
the object at the sides of the slit. The widener is to broaden the spectrum at right
angles to its length; otherwise stellar spectra would be too narrow to use.
An oscillating parallel-sided block is often used, or the observer can use the telescope
drives to trail the image up and down the slit during the exposure. The collimator
and focusing element are conventional optical components and can be either lenses
or mirrors.
The dispersing element separates the light into its component wavelengths.
It can be either a small prism or a diffraction grating. In the latter case it should
be blazed (its individual apertures angled to direct the light into the desired spectral
order), or it will waste most of the light. The detector can be a photographic
emulsion or a CCD, but in the former case the observer will need to be prepared
for very long exposures. As a very rough guide, a widened spectrum with a
dispersion of 10 nm mm1 for a magnitude 4 star obtained on a 0.3-m telescope
would require an exposure of between 10 min to an hour with a CCD detector.
The spectroscopes used on large telescopes can be very complex and cost an
appreciable fraction of the value of the telescope itself. This is because so much
information is obtainable from a spectrum. The spectroscope is thus designed to
operate at a range of dispersions, resolutions and wavelength ranges, and to be as
efficient as possible. In the last decade or so, the efficiency has been improved
dramatically by obtaining the spectra of many objects simultaneously. The spectro-
scope is fed by a number of fiber optic cables (“light pipes”). The other ends of the
Spectroscopy 225
Fig. 12.1 Basic spectroscope (Shown with a diffraction grating as the dispersing element,
although this can be replaced by a prism)
Fig. 12.2 Use of a direct vision spectroscope for visual spectroscopy
cables are positioned, under computer control, at the focal plane of the telescope in
such a way as to coincide with the images of the objects of interest. For example,
the Visible Multi-Object Spectrograph (VIMOS) on the 8.3-m Melipal instrument
(which forms a part of ESO’s VLT) covers from 360 to 1,000 nm in wavelength at a
spectral resolution of up to 2,500 for up to 200 objects in a single exposure.
For visual work, a much simpler approach can be used that will enable the main
spectrum lines to be seen, as well as the differences between spectral types (see
below) and stellar and nebular spectra. A small direct vision spectroscope of
the type sold for use in chemistry labs for element identification in flame spectra
can be attached to the end of a conventional eyepiece (Fig. 12.2) and will enable
low-dispersion spectra of the brighter stars to be seen directly.
Spectroscopy
Reference has already been made to the wide range of information derivable from
spectra. An indication of how that information is obtained is summarized below, but
the interested reader will need to consult more specialized texts (Appendix A of this
book) before undertaking serious work in any of these areas.
226 12 Spectroscopy
Spectral Type
One of the quickest ways of extracting information is via the spectral type of a star.
This is a classification system based upon eye estimates of the features visible on
medium-dispersion spectrograms. With experience, an observer can find a great deal
of information about the star for a small investment of his or her time and effort.
Early in the history of stellar spectroscopy, it was seen that stellar spectra
showed certain recurring themes that enabled them to be grouped together. One
of these early classification systems was based upon the relative intensities of
the hydrogen lines with respect to the other lines in the spectrum. Spectra with
the strongest hydrogen lines were group A, those with slightly weaker lines, group
B, slightly weaker still, group C, and so on. The differences between spectra were
initially attributed to evolution, with the spectra changing in the sense of decreasing
hydrogen line intensity and of increasing complexity, as the star became older. The
least complex spectra, types A, B, etc., were thus thought to come from young (or
early) stars, while the more complex spectra came from older (or later) stars.
Stellar spectra do, of course, change as the star ages, but not in a simple,
straightforward manner. The main underlying reason for the differing appearances
of stellar spectra is variation in their surface temperatures. A more useful stellar
spectral classification would thus be based upon a temperature sequence of the
spectra, and that is how the present system is arranged.
Unfortunately, the change from the earlier classification method to that of the
present day was accomplished by rearranging and adapting the older system, not by
starting afresh. The present spectral classification system is thus now rather untidy
and unnecessarily difficult to use. It seems unlikely to be scrapped, however, so the
student must become familiar with it as it stands.
The modern system of classification was originally codified in Harvard’s Henry
Draper star catalog, and so is sometimes known as the Harvard classification
system. The Harvard system was then further developed by Morgan, Keenan and
Kellman at Yerkes and is now more commonly known as the MKK system. There
are 16 groups of spectra, of which 10, labeled by the letters,
OBAFGKMLTY
form the core. The O-type stars are the hottest normal stars (50,000 K), and the
L-type the coolest (2,000 K). The hottest brown dwarfs1 are mixed with the coolest
stars in class L, with classes T and Y comprising just brown dwarfs and with
temperatures down to 500 K. A useful mnemonic for the order of the classes is
1
Brown dwarfs are not brown in color – the name arose for historical reasons. Their actual color to
the eye would be very deep red, but most of their emission is in the infrared. They are stars whose
energy comes from low temperature nuclear fusion reactions such as the conversion of deuterium
and lithium to helium, not the conversion of ordinary hydrogen to helium, which is the power
source of the Sun and normal stars.
Spectroscopy 227
Oh Be A Fine Girl/Guy Kiss Me – Love To You
Each of the major classes is subdivided into ten, with Arabic numerals denoting
the subdivisions. The Sun, for example, is of spectral class G2, while Polaris is F7
and Sirius A2. The first three groups are often called early-type stars, while the last
two are called late-type stars. This, however, is now just a convenient convention,
and no longer has the evolutionary significance once attributed to it.
Recently the original decimal subdivision of the core classes was adapted to
provide a smoother variation with temperature by adding and deleting some of the
classes. The full range of spectral types in current use is thus (the subdivisions of
the last three classes are still somewhat under debate):
Core Subdivisions
O 4 5 6 7 8 9 9.5
B 0 0.5 1 2 3 5 7 8 9.5
A 0 2 3 5 7
F 0 2 3 5 7 8 9
G 0 2 5 8
K 0 2 3 4 5
M 0 1 2 3 4 7 8
The missing numbers, though, are likely still to be encountered, especially in

older sources.
The other six letters
RNSWPQ
are much more rarely used and denote unusual groups of stars such as the carbon stars
(R and N, sometimes also called C-type stars), and the Wolf–Rayet stars (W), etc.
The main spectral features of the core spectral types are:
O Few lines present; those actually visible are mostly due to highly ionized
silicon and nitrogen, etc. Hydrogen Balmer and ionized helium lines visible.
B Balmer lines strengthening, neutral helium and lower stages of ionization of
silicon and nitrogen, etc., now producing lines. Towards the lower temperature
end of the group, the neutral helium lines disappear.
A Balmer lines peak in intensity at A0; singly ionized calcium lines appear.
Thereafter the Balmer lines weaken and numerous lines due to singly ionized
metals start to appear.
F Lines in the spectra become more numerous, Balmer and ionized metal lines
continue to weaken, neutral metal lines strengthen.
G Ca II, H and K lines peak in their intensities, Balmer lines continue to weaken,
and neutral metal lines continue to increase in intensity.
K The Balmer lines are still just visible, many lines due to neutral metals are now
present, and a few molecular bands (TiO) appear.
M The TiO bands now dominate the spectrum.
228 12 Spectroscopy
L Lines and bands due to metal hydrides and the alkali metals dominate the
spectrum.
T Methane features are prominent in the spectrum.
Y Brown dwarfs belonging to this class have yet to be unequivocally identified.
They are likely to have spectra dominated by ammonia, water and methane.
The precise determination of the spectral class of a star from its spectrum is via
the intensity ratios of pairs of lines that happen to be especially sensitive to
temperature or luminosity. The required line pairs vary with spectral class, some
only being used to distinguish one or two subclasses, others being of wider use.
The student is referred to more specialized texts on the topic (see Appendix A of
this book) for further details of the techniques of spectral classification. Once the
spectral class of a star is determined, then its temperature and many other properties
may be found from tables (see Appendix A of this book).
Luminosity Class
The widths of spectral lines increase as the gas pressure in the regions producing
them increases. Now, the masses of normal stars vary by only about a factor of
1,000, while their diameters vary by a factor of about 10,000, and their volumes by
a factor of 1012. The physically largest stars must therefore have much lower
densities than the physically smallest stars, and hence the larger stars have the
lower surface pressures. Thus the spectral lines originating from large, and hence
very luminous, stars generally have smaller widths than those from smaller stars.
The overall effect of differing pressures is to change the intensity ratios between
some pairs of lines in the spectra from stars that have identical temperature-based
spectral classes, but differing luminosities. The luminosity class is added as a
Roman numeral after the temperature spectral class. Classes I–IV are the giant
stars (supergiants, bright giants, giants and subgiants respectively). The majority
of stars, including the Sun, fall into class V and are called main sequence or dwarf
stars. Classes VI and VII are the subdwarfs and white dwarfs. A more complete
classification for the earlier examples is therefore: the Sun G2V, Polaris F7 I or F7
II and Sirius Al V.
Radial Velocity
After spectral classification, the velocity of an object along the line of sight, usually
known as its radial velocity, is the commonest parameter to be obtained from a
spectrum. For this purpose a comparison spectrum is usually required. This is an
emission-line spectrum from an artificial source. The comparison spectrum is
Spectroscopy 229
Fig. 12.3 Wavelength comparison spectrum such as might be used for a basic spectrograph
placed either side of the main spectrum2 (Fig. 12.3). The wavelengths of the lines in
the comparison spectrum are known, and so those of lines in the main spectrum may
be found by interpolation. These will normally be different from the rest of the
wavelengths of those same lines because of the Doppler shift, and so the object’s
velocity can be found from the Doppler formula:
cDn cDl
Radial velocity ¼ v ¼ ¼ (12.3)
n l
where v is the velocity of the object along the line of sight, c is the velocity of light
(Chap. 7), Dl and Dn are the wavelength or frequency shifts: Dl ¼ lOlL, Dn ¼
nLnO, where subscript “O” denotes an observed value and subscript “L” denotes a
laboratory (unshifted) value.
The convention is used that the radial velocity is positive when directed away
from Earth, and negative when directed towards Earth.
Spectrophotometry
The greatest return of information from a spectrum comes from the detailed study
of spectrum line strengths and of their shapes (profiles). This is known as spectro-
photometry. The process usually requires extensive computer modeling of the
2
Techniques developed in the last decade or so to enable the detection of exoplanets from the very
small velocity changes that the orbiting planet induces in its host star include the use of absorption
cells that superimpose comparison lines directly onto the star’s spectrum. Such details go beyond
the scope of this book and the interested reader will need to consult other sources (see Appendix A
of this book).
230 12 Spectroscopy
region producing the spectrum. The models are used to produce predicted spectra,
and these are then compared with the observations. The model is adapted until as
close a fit as possible to the observations is obtained and the properties of the
observed object inferred from the best-fit model. The process can become
extremely complex, and any further discussion is best left to more advanced texts.
Exercise
12.1. Calculate the line-of-sight velocity with respect to Earth of a star when the
Balmer H-a line in its spectrum is observed to be at a wavelength of
655.2 nm. (Balmer H-a line rest wavelength: 656.2868 nm; speed of light
in a vacuum: 2.998 108 m s1).
Appendix A: For Further Reading
Journals
Only the major and relatively widely available journals are listed. There are
numerous more specialized research-level journals available in academic libraries.
Popular
Astronomy
Astronomy Now
Ciel et Espace
Journal of the British Astronomical Association
New Scientist
Practical Astronomy
Publications of the Astronomical Society of the Pacific
Scientific American
Sky & Telescope
Research
Astronomical Journal
Astronomy and Astrophysics
Astrophysical Journal
Monthly Notices of the Royal Astronomical Association
Nature
Science

DOI 10.1007/978-1-4614-4891-4, # Springer Science+Business Media New York 2013
232 Appendix A: For Further Reading
Other Books by the Author
Exoplanets – Finding, Exploring and Understanding Alien Worlds, Chris Kitchin,

Springer, 2011.
Astrophysical Techniques 5th Edition, Taylor & Francis 2009.
Galaxies in Turmoil, Springer, 2007.
Illustrated Dictionary of Practical Astronomy, Springer, 2002.
Solar Observing Techniques, Springer-Verlag, 2001.
Photo-Guide to the Constellations, Springer-Verlag, 1997.
Seeing Stars, co-author R. Forrest, Springer-Verlag, 1997.
Optical Astronomical Spectroscopy, IOP Publishing, 1995.
Journeys to the Ends of the Universe, co-author Adam Hilger, 1990.
Stars, Nebulae and the Interstellar Medium, co-author Adam Hilger, 1987.
Early Emission Line Stars, co-author Adam Hilger, 1982.
Star and Other Catalogs, Atlases, and Reference Books
Astronomical Almanac, HMSO/US Government Printing Office (published

annually).
Allen’s Astrophysical Quantities, A. N. Cox (Ed), Springer, 2001.
Cambridge Encyclopaedia of the Sun, K. R. Lang, Cambridge University Press,
2001.
Cambridge Star Atlas, W. Tirion, Cambridge University Press, 2011.
Deep Sky Observer’s Year, G Privett and P. Parsons, Springer-Verlag, 2001.
Deep Sky Observing, S. R. Coe, Springer-Verlag, 2000.
Encyclopaedia of Astronomy and Astrophysics, edited by P. Murdin, Nature and
IoP Publishing, 2001.
Encyclopaedia of Planetary Sciences, edited by J. H. Shirley and R. W. Fairbridge,
Kluwer Academic Publishers, 2000.
Field Guide to the Deep Sky Objects, M. Inglis, Springer-Verlag, 2001.
Handbook of the British Astronomical Association, British Astronomical Associa-
tion (published annually).
Handbook of Double Stars: For the Use of Amateurs, E. Crossley, J. Gledhill, J. M.
Wilson, Cambridge University Press, 2011.
Norton’s Star Atlas, I. Ridpath, Addison Wesley, 2003.
Observing the Caldwell Objects, D. Ratledge, Springer-Verlag, 2000.
Planetary Nebulae: A Practical Guide and Handbook for Amateur Astronomers,
S. J. Hynes, Willmann-Bell, 1991.
Sky & Telescope’s Pocket Sky Atlas, R. W. Sinnott, Sky Publishing Corporation,
2006.
Sky Atlas 2000.0, W. Tirion, Sky Publishing Corporation, 2000.
Appendix A: For Further Reading 233
Sky Catalogue 2000, Volumes 1 and 2, A. Hirshfield and R. W. Sinnott, Cambridge

University Press, 1992.
3,000 Deep Sky Objects: An Annotated Catalogue, E. Aranda, Springer, 2011.
Yearbook of Astronomy, Macmillan, published annually.
Practical Astronomy Books
Astronomical Spectroscopy for Amateurs, K. M. Harrison, Springer, 2011.

Astronomy with Small Telescopes, S. F. Tonkin, Springer-Verlag, 2001.
Backyard Astronomer’s Guide, T. Dickinson, Firefly Books, 2007.
CCD Astrophotography: High Quality Imaging from the Suburbs, A, Stuart,
Springer, 2006.
Choosing and Using a Dobsonian Telescope, N. English, Springer, 2011.
Choosing and Using a Schmidt-Cassegrain Telescope, R. Mollise, Springer-Verlag,
2001.
Choosing and Using a Refracting telescope, N. English, Springer, 2011.
Complete Guide to the Herschel Objects: Sir William Herschel’s Star Clusters,
Nebulae and Galaxies, M. Bratton, Cambridge University Press, 2011.
Deep Sky Companions: The Secret Deep, S. J. O’Meara, Cambridge University
Press, 2011.
Deep Sky Wonders: A Tour of the Universe, S. French, Firefly Books, 2011.
Digital SLR Astrophotography, M. A. Covington, Cambridge University Press,
2007.
How to Photograph the Moon and Planets with Your Digital Camera, T. Buick,
P. Pugh, Springer, 2011.
Observational Astronomy: Techniques and Instrumentation, E. C. Sutton,
Cambridge University Press, 2011.
Observing the Messier Objects with a Small Telescope: In the Footsteps of a Great
Observer, P. Pugh, Springer, 2011.
Observing the Moon, P. Wlasuk, Springer-Verlag, 2000.
Using the Meade ETX, M. Weasner, Springer-Verlag, 2002.
Introductory Books
AstroFAQs, S. F. Tonkin, Springer-Verlag, 2000.

Astronomy on the Personal Computer, O. Montenbruck, T. Pfleger, Springer-
Verlag, 2000.
Astrophysics is Easy! – An Introduction for the Amateur Astronomer, M. D. Inglis,
Springer, 2007.
Big Questions: The Universe, S. Clark, Quercus, 2010.
234 Appendix A: For Further Reading
Dark Side of the Universe: Dark Matter, Dark Energy, and the Fate of the Universe,
I. Nicolson, Canopus, 2007.
Deep Space, S. Clark, Quercus, 2008.
Discovering the Solar System, B.W. Jones, Wiley-Blackwell, 2007.
DK Illustrated Encyclopaedia of the Universe, Dorling Kyndersley, 2011.
Galactic Encounters: Our Majestic and Evolving Star-System, From the Big Bang
to Time’s End, W. Sheehan, C. J. Conselice, J. Baum, Springer 2012.
Guidebook to the Constellations: Telescopic Sights, Tales and Myths, P. Simpson,
Springer, 2012.
Illustrated Guide to Astronomical Wonders: From Novice to Master Observer,
R. B. Thompson, B. F. Thompson, O’Reilly Media, 2007.
Universe, RA Freedman and WJ Kaufmann III, WH Freeman, 2010.
Universe: Solar System, Stars and Galaxies, M. Seeds, D. Backman, Cengage
Learning Custom Publishing, 2011.
Appendix B: Constellations
Constellation Abbreviation Constellation Abbreviation

Andromeda And Leo Leo
Antlia Ant Leo Minor LMi
Apus Aps Lepus Lep
Aquarius Aqr Libra Lib
Aquila Aql Lupus Lup
Ara Ara Lynx Lyn
Aries Ari Lyra Lyr
Auriga Aur Mensa Men
Boötes Boo Microscopium Mic
Caelum Cae Monoceros Mon
Camelopardalis Cam Musca Mus
Cancer Cnc Norma Nor
Canes Venatici CVn Octans Oct
Canis Major CMa Ophiuchus Oph
Canis Minor CMi Orion Ori
Capricornus Cap Pavo Pav
Carina Car Pegasus Peg
Cassiopeia Cas Perseus Per
Centaurus Cen Phoenix Phe
Cepheus Cep Pictor Pic
Cetus Cet Pisces Psc
Chamaeleon Cha Piscis Austrinus PsA
Circinus Cir Puppis Pup
Columba Col Pyxis Pyx
Coma Berenices Com Reticulum Ret
Corona Australis CrA Sagitta Sge
Corona Borealis CrB Sagittarius Sgr
Corvus Crv Scorpius Sco
Crater Crt Sculptor Scl
Crux Cru Scutum Sct
Cygnus Cyg Serpens Ser
(continued)

236 Appendix B: Constellations
Constellation Abbreviation Constellation Abbreviation

Delphinus Del Sextans Sex
Dorado Dor Taurus Tau
Draco Dra Telescopium Tel
Equuleus Equ Triangulum Tri
Eridanus Eri Triangulum Australe TrA
Fornax For Tucana Tuc
Gemini Gem Ursa Major UMa
Grus Gru Ursa Minor UMi
Hercules Her Vela Vel
Horologium Hor Virgo Vir
Hydra Hya Volans Vol
Hydrus Hyi Vulpecula Vul
Indus Ind
Lacerta Lac
Appendix C: Answers to Exercises
2.1. (a) x8,148

(b) 19,051
2.2. (a) x21, 35.7 mm
(b) x195, 3.8 mm
(c) 0.7500 , 2.7 m, 5.55 m
2.3. 391 m
4.1. 2 h 53 min 9.5 s
1 h 4 min 0 s
6 h 11 min 50s
19 h 26 min 16 s
4.2. 46 310 1500
221 300 000
272 220 000
318 180 1500
4.3. 6 h 31 min 4 s
4.4. 4 h 42 min 55 s
4.5. 23 h 13 min 35 s
4.6. Sirius: 16 h 28 min 26 s
Betelgeuse: 17 h 18 min 25 s
Neither star would be visible
4.7. Altitude ¼ +70 200 5100 ; azimuth ¼ 107 070 5300 West
5.1. 50 160 5100 East
5.2. 8 370
5.3. 6 h 46 min 14 s
16 410 0800

238 Appendix C: Answers to Exercises
5.4. Venus: 1.597 years

Jupiter: 1.092 years
Pluto: 1.004 years
5.5. (a) No; lunar eclipses can only occur at full Moon.
(b) Yes; the Moon’s orbit rotates in space once every 18.7 years, and so a solar
eclipse can occur sometime anywhere around Earth’s orbit.
6.1. You should get times near to 11 h 42 min p.m. and 4 h 42 min a.m.
8.1. Sixth magnitude star: 9.87 1011 W m2
8.2. Distance : 2.1 106 km
8.3. Distance: 4.8 mm
8.4. Height: 183 m (likely height in the range 150–200 m)
9.1. The galaxy extends from intensities of about 36,000–48,000. One (of many)
suitable gray scalings would thus be:
CCD intensity Monitor level

0–35 999 0
36 000–36 049 1
36 050–36 099 2
36 100–36 149 3
36 150–36 199 4
and so on up to
48 600–48 649 253
48 650–48 699 254
48 700–100 000 255
10.1. Mean ¼ 959.58

s ¼ 0.302
S ¼ 0.0955
Thus the final answer should be 959.6 0.1
10.2. 12.6 0.1
3.59 0.01
5000 100
0.012 0.0007 or 0.001 0.0007
3.1 106 105
3 1016 3 1016
10.3. Number of asteroids ¼ 91.64 m – 558.9. (Number of asteroids ¼ 90 m – 560
is more in line with the number of significant figures in the original data.)
Hence the number of asteroids in the range 15–16m (m ¼ 15.5) is 835. This
predicted number is likely to be far too small since the original data show a
much steeper than linear relationship (probably exponential) in the number of
asteroids with decreasing brightness.
Appendix C: Answers to Exercises 239
10.4. r ¼ 0.85. With nine degrees of freedom, we find from Fig. 10.3 that the
significance level < 1 %. The correlation of student numbers with tempera-
ture is therefore highly significant.
10.5. t ¼ 2.59. The number of degrees of freedom is nine. So, from the graph in
Fig. 10.3, we may see that the result of Student’s t test is a probability in the
region between 5 % and 1 % (actually just slightly less than 5 %). Therefore,
the result is significant and there is greater than 95 % chance that the stars’
separation has changed with time.
10.6.
NA ¼ 15 XA ¼ 8.305 sA ¼ 0.017
NB ¼ 12 XB ¼ 8.285 sB ¼ 0.021
Giving
t ¼ 2.67
The number of degrees of freedom is 25. So from the graph in Fig. 10.3,
we may see that the result of Student’s t test is a probability slightly less
than 1 %. Therefore the result is highly significant, and there is greater than
99 % chance that the star is a variable.
11.1. Magnitudes: 4.67, 7.91
11.2. 27 pc (for Jupiter to have an apparent magnitude of +28 – see Table 8.2); thus
Jovian-sized planets could be detectable out to a few tens of parsecs.
11.3. For Betelgeuse, MV ¼ 6.05, and BC ¼ 1.7 so from Equation (11.3),
MBol ¼ 7.75
So from Equation (11.4)
Luminosity ¼ 3.8 1031 W.
Betelgeuse is thus 94,000 times brighter than the Sun.
12.1. 496 km s1 towards Earth.
Appendix D: SI and Other Units
SI Prefixes
Prefix Multiplier Symbol

atto 1018 a
femto 1015 f
pico 1012 p
nano 109 n
micro 106 m
milli 103 m
centi 102 c (not recommended)
deci 101 d (not recommended)
deca 101 da (not recommended)
hecto 102 h (not recommended)
kilo 103 k
mega 106 M
giga 109 G
tera 1012 T
peta 1015 P
exa 1018 E
SI Units
Physical quantity Unit Symbol

Angle radian rad
Capacitance farad F (s4 A2 m2 kg1)
Electric charge coulomb C (A s)
Electric current ampere A
(continued)

242 Appendix D: SI and Other Units
Physical quantity Unit Symbol

Electrical resistance ohm O (m2 kg s3 A2)
Energy joule J (m2 kg s2)
Force newton N (kg m s1)
Frequency hertz Hz (s1)
Length meter m
Luminous intensity candela cd
Magnetic flux density tesla T (kg s2 A1)
Mass kilogram kg
Power watt W (m2 kg s3)
Pressure pascal Pa (kg m1 s2)
Solid angle steradian sr
Temperature kelvin K
Time second s
Voltage volt V (m2 kg s3 A1)
Other Units in Common Use in Astronomy
Unit Symbol Equivalent

Ångstrom Å 1010 m
Astronomical unit AU 1.495 978 70 1011 m
Atmosphere atm 1.013 25 105 Pa
Bar bar 105 Pa
Dyne dyn 105 N
Electron volt eV 1.6022 1019 J
Erg erg 107 J
Gauss G 104 T
Jansky jy 1026 W m2 Hz1
Light year ly 9.4605 1015 m
Micron m, mm 106 m
Parsec pc 3.0857 1016 m
Solar luminosity L 3.8478 1026 W
Solar mass M 1.9891 1030 kg
Solar radius R 6.960 108 m
Appendix E: The Greek Alphabet
Letter Lower case Upper case

Alpha a B
Beta b Β
Gamma g G
Delta d D
Epsilon e Ε
Zeta z Ζ
Eta Z Η
Theta y Y
Iota i I
Kappa k K
Lambda l L
Mu m M
Nu n Ν
Xi x X
Omicron o Ο
Pi p P
Rho r Ρ
Sigma s S
Tau t Τ
Upsilon u Υ
Phi f F
Chi w Χ
Psi c C
Omega o O

Index
A Aperture synthesis. See also Interferometer

Aberration (optical) e-MERLIN, 30
astigmatism, 12, 19, 48–51 filled aperture, 30, 53
chromatic, 6–8, 12–14, 19, 48, 49 unfilled aperture, 30
coma, 12, 19, 48–50 very large array, 29
distortion, 14, 48–51 Aphelion, 115, 122, 126
field curvature, 19, 48, 49 Apocynthion, 122
spherical, 6–9, 12, 19, 22, 48, 49 Apogee, 117, 122
Aberration (stellar), 6–9, 12–14, 19, 22, 40, 42, Apojove, 122
48–50, 102, 115–116, 181, 197 Apparent magnitude, 167, 168, 210, 212,
Absolute magnitude, 169, 212–213, 215, 218 213, 237
Absorption cell, 227 Archives, 192, 196, 208–210
Achromat, 12, 14 Arctic circle, 103
condition for achromatism, 48 Arecibo radio telescope. See Radio telescope,
Achromatic doublet. See Lens, achromatic Arecibo
Active supports. See Mirror, active supports Artificial guide star. See Atmospheric
Adaptive optics. See Atmospheric compensation
compensation, adaptive optics Astigmatism. See Aberration (optical)
Airy disc, 33 Astrolabe, 104–106
Alidade, 105, 106 Astronomical Almanac, 84, 86, 87, 91, 151,
ALMA. See Atacama Large Millimeter 161, 230
Array (ALMA) Astronomical refractor, 5
Almagest, 109 Yerkes, 14
Alt-alt mounting. See Mounting (telescope) Atacama Large Millimeter Array
Alt-az mounting. See Mounting (telescope) (ALMA), 26, 29. See also Aperture
Altitude, 8, 22, 56, 75–80, 93, 98, 104, 105, synthesis
135, 136, 149, 154, 215, 236 Atmosphere, 15, 22–25, 34, 41, 56, 57, 99,
Analemma, 81, 83, 85 102–104, 122, 123, 138, 146, 151,
Andromeda Galaxy, 37, 174, 175 154, 158, 159, 167, 175, 199, 204
Angular measure, 23, 78, 79 Atmospheric compensation
Antarctic circle, 103 adaptive optics, 25
Anti-blooming CCD, 212 artificial guide star, 25
Anti-reflection coating, 42 correcting mirror, 24, 25
Apastron, 122 guide star, 24, 25

246 Index
Atmospheric compensation (cont.) Celestial longitude, 90, 96, 97

Hartmann sensor, 24 Celestial pole, 56, 71, 95, 128
multi-conjugate adaptive optics, 25 Celestial sphere, 71–76, 93, 99, 101, 112
tip-tilt mirror, 25 Centre de Données astronomiques de
Atmospheric refraction, 99, 103, 138 Strasbourg, 166, 208
Atomic clock, 82 Čerenkov radiation, 144
Autoguider, 138 Chandler wobble, 83
Autumnal equinox, 77, 84 Charge coupled devices (CCDs), 138,
Avalanche photodiode, 217 181–189, 191, 211, 212, 215, 218
Averted vision, 38, 173, 179, 181 anti-blooming, 212
Azimuth, 22, 56, 75–80, 93, 98, 100, 126, 135, array, 47, 48, 50, 184, 186, 188, 189
136, 149, 236 charge coupling, 183
detection efficiency, 185
detection mechanism, 182, 192
B electron-hole pair, 182, 186, 192
Background signal subtraction, 190 mosaic, 47, 48, 50, 188, 189
Bacon, F.R., 5 Chaucer, G., 106
Barlow lens, 35, 45 Chromatic aberration. See Aberration
Bayer, J., 165 (optical)
Bearing, 75, 128, 131 Circle of least confusion, 49
Bell metal, 10 Circular polarisation, 145
Binary star, 170, 171, 213 Circumpolar object, 101–103
visual, 91, 171 declination, 101, 103
Binoculars Citizen science, 196, 208–210
gyro-stabilised, 65 Civil time, 80–85, 87, 88, 100, 134, 152
night vision, 37, 65 Coating, 15, 42, 62, 150, 185
Black body, 214 anti-reflection, 42, 160
Bolometric correction, 215, 216 Coelostat, 135, 163
Bolometric magnitude, 215–217, 219 Collimator, 222
Brown dwarf, 213, 224, 226 Colour index, 215
Colour vision, 169, 172, 180
Coma. See Aberration (optical)
C Comparison spectrum, 226, 227
c. See Velocity of light (c) Compass point, 72
Caldwell catalogue, 175 Conjunction
Camera inferior, 116
digital, 185, 186 superior, 125, 126
Schmidt (see Schmidt camera) Constellation, 25, 56, 74, 75, 102, 104, 110,
SLR, 185 111, 137, 151, 158, 164, 165, 230,
Camera obscura, 5 232–234
Cardinal point, 72 Contrast stretching, 190, 196
Cassegrain, L., 11 Copernican model. See Solar system
Cassegrain telescope, 11 Copernicus, N., 109
Cassini, G., 8 Correcting lens
Catadioptric, 19 Maksutov telescope, 20
Catoptric, 19 prime focus, 16
CCD camera Schmidt camera, 19
astronomical, 139, 186, 211, 212 Schmidt-Cassegrain telescope, 20
astronomical cooling, 186 Correcting mirror. See Atmospheric
CCD photometry, 211–212 compensation, correcting mirror
CCDs. See Charge coupled Correlation coefficient, 197–201,
devices (CCDs) 203–206
Celestial latitude, 90, 96, 97 Cosine rule. See Spherical trigonometry
Index 247
Cosmic rays effects, 143, 189, 196 E

Coudé focus, 16, 46 Earth
Coudé system, 16, 17 orbit, 75, 81, 91, 115, 117, 122, 155, 236
Coudé telescope, 16, 46 perihelion, 103, 115, 122
Cross wire eyepiece, 135, 138 rotation, 16, 28, 54, 56, 71, 76, 77, 82,
83, 89, 99, 102, 107, 112, 117, 127,
128, 131, 133
D Eclipse
Dark adaptation, 37, 170, 180 annular solar, 121
Dark noise, 186 limits on occurrence, 122
Dark signal subtraction, 190 lunar, 107, 118, 122, 123, 126
Data analysis, 186, 196, 199–207 partial solar, 121
Data mining, 208–210 solar, 117–121, 126
Data processing, 51, 53, 195–210 total solar, 117, 120
Data reduction, 196–197, 199 Ecliptic, 74, 75, 80, 81, 84, 90, 96, 97, 102,
Day-time observing, 175–176 104, 112–115, 117, 126
Dec. See Declination E-ELT. See European Extremely Large
Declination, 54, 55, 72, 77–81, 83, 85, 88–90, Telescope (E-ELT)
99–101, 103, 111, 113, 128, 133, Effective focal length, 45–48, 57
134, 136, 137, 149, 151, 152, 165, 166 Electro-magnetic radiation
axis, 54, 90, 128, 149 amplitude, 144–146
Deferent, 109 circular polarisation, 145
Degree, 23, 60, 69, 76, 79, 83, 90, 95, 98, 99, electric vector, 144, 145
103, 109, 114, 117, 136, 147, 151, 153, elliptical polarisation, 147
189, 199, 204–206, 218, 237 frequency, 143, 146
Degree of freedom, 204–207, 237 intensity, 144
della Porta, G., 5 interactions with matter, 145
Detector, 19, 26, 27, 31, 36, 40, 46, 50, 56, 115, linear polarisation, 146, 147
135, 138, 143–145, 149, 172, 179–194, phase, 144, 146
211, 215, 217, 221, 222 photon, 144–145
photon counting, 47 polarisation, 145
Dewing-up, 150 refractive index, 143
Diffraction, 19, 23, 24, 33–35, 38, 41, 47, 57, velocity, 143, 144
61, 151, 160, 171, 222, 223 wavelength, 143–146
airy disc, 33 Electron-hole pair, 182, 186, 187
Diffraction limit, 23, 24, 33–35, 38, 41, 47, 57, Electron volt (eV), 144, 145, 240
61, 151, 159, 171 Ellipsoid. See Telescope mirror,
Digges, L., 3, 4 secondary-ellipsoidal
Digital camera Elliptical polarisation, 147
active focusing, 185 e-MERLIN. See Aperture synthesis, e-MERLIN
passive focusing, 185 English mounting. See Mounting (telescope)
Dioptric, 19 Epicycle, 109
Direct vision, 223 Epoch, 113, 114
Dispersion, 12, 221–224 Equation of time, 82, 83, 84, 102
Distortion. See Aberration (optical) Equator, 69, 71, 72, 78, 80, 81, 84, 88–90, 97,
Diurnal motion, 99–100 102, 104, 112, 136
Dobsonian mounting. See Mounting Equatorial mounting. See Mounting (telescope)
(telescope) Equatorial platform, 64, 65, 131
Dollond, J., 12 Equilux, 103
Dome, 11, 14, 34, 154 Equinox, 77, 81, 83, 102, 103
Double star, 41, 120, 158, 169–172, 174, Errors (in data), 138, 157, 197–199, 206, 217
210, 230 ESO. See European Southern Observatory’s
Dove prism, 132 (ESO)
248 Index
European Extremely Large Telescope Filled aperture. See Aperture synthesis,

(E-ELT), 22, 23, 32, 168, 169, 177 filled aperture
European Southern Observatory’s (ESO), 18, Filter
22, 23, 26, 98, 120, 123, 128, 131, comet, 46
173, 189, 208, 223 full aperture, Sun, 45, 161, 162, 164
eV. See Electron volt (eV) H-a, 46
Exit pupil, 34, 35 H-a cut-off, 46
Exoplanet, 57, 120, 208, 227, 230 light pollution rejection, 46
Extended source Nebula, 46
image brightness, 31, 36, 168 U, B, V, 215
surface brightness, 36, 37, 172 Finder chart, 152, 153
Eye Finding, 6, 24, 64, 89, 135, 136, 149,
colour vision, 169, 179, 180 151–153, 156, 161, 162, 171,
cone cells, 31 173, 175, 217, 230
dark adapted (see Dark adaptation) First point of Aries (FPA), 84, 88–90,
fovea centralis, 37, 179, 181 102, 112
pupil, 31, 34, 35, 37, 180 Five parts rule. See Spherical trigonometry
resolution, 34, 35, 57, 176, 179, 181 Flamsteed, J., 165
retina, 37, 38, 144, 162, 176, 179–181 Flat fielding, 181, 191, 196
rod cells, 31, 180 Focal correction, 175, 176
sensitivity, 6, 37, 172, 179, 180, 181, 215 Focal length, 3, 5, 8, 11, 32, 35, 41, 42,
structure, 37, 179, 185, 221 45–48, 56, 57, 175, 176
Eyepiece effective, 45–48, 57
crosswire, 135, 138, 155 Focal plane, 22, 46, 48, 50, 223
Erfle, 42 Focal point, 11
guiding, 41, 137–139 Focal ratio, 8, 14, 32, 162, 163
Huyghenian, 42 Focus, 3, 5, 6, 9–12, 15, 16, 18, 19, 21, 22,
Kellner, 42 26, 38, 46–50, 57, 65, 132, 163, 175,
maximum focal length, 41 177, 185
micrometer, 41, 155, 169–171 Focusing, 6, 63, 175, 185, 222
multi-lens, 14 Ford, D., 106
Nagler, 42 Fork mounting. See Mounting (telescope)
orthoscopic, 42 Foucault test, 39
parfocal, 42, 44 Four parts rule. See Spherical trigonometry
Plössl, 42 FPA. See First point of Aries (FPA)
projection, 45, 163 f-ratio. See Focal ratio
Ramsden, 42 Fringe, 52
wide angle, 41, 42 interferometer, 51, 52
Eye relief, 34, 35, 42 Full aperture filter, 162
F G
False observations, 176–177 Gaia spacecraft, 110
Feed. See Radio telescope Galactic latitude, 90
Fibre optic, 22, 30, 46, 222 Galactic longitude, 90
Field curvature. See Aberration (optical) Galilean refractor, 3–5
Field of sharp focus, 15, 18, 19. See also Galilean satellites, 116–118, 160
Field of view Galileo, 3, 5, 7
Field of view. See also Field of sharp focus Gamma ray, 166, 214, 215
eyepiece, 35, 36, 127, 152 Gemini telescope, 25
telescope, 35, 36, 47, 138, 151, 152, 180, 188 German mounting. See Mounting (telescope)
Field rotation. See Mounting (telescope), Giant Magellan Telescope (GMT), 22, 81,
alt-az-field rotation 82, 84, 98
Index 249
Glass I
crown, 12, 49 Image brightness. See also Light grasp
flint, 12, 49 extended source, 31, 36, 37, 47, 51
GMT. See Giant Magellan Telescope; point source, 31–36, 49
Greenwich Mean Time Image compression, 191
Gran Telescopio Canarias (GTC), 22 Image de-rotator, 133
Grating, 222, 223 Image processing
Great circle, 72, 76, 77, 93, 95, 111 background signal subtraction, 190
Greatest elongation, 116, 158 contrast stretching, 190
Greek alphabet, 241 cosmic rays effects, 189, 191, 196
Greenwich mean time (GMT), 22, 82, 84, 98 dark signal subtraction, 190
Gregorian telescope, 9, 11 flat fielding, 189
Gregory, J., 9, 10 hot spots, 190
Grossetest, B.R., 5 image compression, 191
GTC. See Gran Telescopio Image scale (IS), 46–47
Canarias (GTC) Infra-red, 18, 22, 25, 28, 57, 145, 146, 159, 162,
Guide star, 24, 25, 165 185, 193, 224
artificial (see Atmospheric compensation, Interference fringes, 33
artificial guide star) Interferometer. See also
Guiding Aperture synthesis
autoguider, 138 base-line, 27, 110
off-axis, 138 fringe, 52, 53
operation, 53
path difference, 27, 52, 53
H resolution, 25, 27, 30, 51–53
HA. See Hour angle (HA) two-element, 52
Hadley, J., 12 very-long base-line, 27, 110
Half-wave dipole. See Radio telescope Intermediate frequency. See Radio telescope
Hall, C.M., 12 International Atomic Time
Hartmann sensor. See Atmospheric (TAI), 82, 83
compensation Irradiation, 179
Heliocentric model, 109
Heliocentric time, 91
Herschelian telescope, 12, 16, 57 J
Herschel, J., 166, 174 James Webb Space Telescope
Herschel, W., 12, 14, 57, 166, 174 (JWST), 22–24
1.2-metre reflector, 14 Jansky, K., 26, 240
HET. See Hobby Eberly Telescope (HET) Julian date
Hevelius, J., 8 heliocentric, 91
Hipparchus, 112, 166, 167 modified, 91
Hipparchus’ star catalogue, 166 Julian day, 91
Hobby Eberly Telescope (HET), 22 Julian day number, 91
Horizon, 72, 73, 75, 76, 99–101, 103, 104, Jupiter, 5, 25, 116, 117, 120, 122, 158, 160,
123, 175 168, 213, 236, 237
Horseshoe mounting. See Mounting satellites, 5, 116, 160
(telescope) JWST. See James Webb Space Telescope
Hot spots, 190 (JWST)
Hour angle (HA), 54, 55, 77–80, 84, 88–90, 95,
98, 99128, 133, 134, 137, 151
HST. See Hubble Space Telescope (HST) K
Hubble Space Telescope (HST), 11, 25, 38, Keck telescope, 32, 39
39, 62, 165 Kepler, J., 109
Huygens, C., 6, 8 Kepler spacecraft, 48, 120, 124, 218
250 Index
L Meridian
Large Binocular Telescope (LBT), 22 Greenwich, 69, 78, 79, 84
Latitude, 69–71, 77, 78, 81, 87, 89, 96–101, right ascension, 136
103–106, 132–134, 149, 163 Messier, C., 174
LBT. See Large Binocular Telescope (LBT) Messier object, 152, 174
Least squares equation, 201 Metius, J., 5
Lens Microwave, 12, 145–146
achromatic, 12, 48 Minute-of-arc, 15, 18, 23, 90, 136
apochromatic, 12, 13 Mirror. See also Telescope mirror
converging, 3, 5, 11, 12 active supports, 40, 54
correcting (see Correcting lens) correcting (see Atmospheric compensation,
diverging, 3, 5, 12, 45 correcting mirror)
relay, 8 figuring, 39, 61
simple, 6, 7, 12, 13, 35 flat, 10, 15, 62, 135
Libration, 108, 155 grinding, 38, 61, 135
Light honeycomb, 40
photon (see Photon) lightweight, 16
wave nature, 32 metal-on-glass, 15
Light gathering power, 22. See also multi (see Multi-mirror telescope)
Light grasp off-axis, 12, 22, 39
Light grasp, 31, 32, 36, 47, 64. polishing, 38, 39, 61, 135
See also Light gathering power surface accuracy-1/8 wavelength
Linear polarisation, 146, 147, 175 requirement, 38, 61, 160
Linear regression, 197, 200–204 tip-tilt (see Atmospheric compensation,
Lippershey, H., 3, 5 tip-tilt mirror)
Local sidereal time (LST), 84–90, Mirror coating
98, 133, 134 aluminium, 62
Longest day, 102 silicon dioxide, 62
Longitude, 69, 70, 77–82, 84, 87, 90, silver, 62
96–98, 137 MKK system, 224
LST. See Local sidereal time (LST) Modified English mounting.
Luminosity type, 226 See Mounting (telescope)
Moon
libration, 155
M motion, 107
M31, 37, 90, 174 orbital motion, 107, 116, 155
M104, 172, 173 phase, 107, 116, 118, 154
Magnetic declination, 72 rotation, 107
Magnetic deviation, 72 Moon maps, 153, 154
Magnetic variation, 72, 74 orientation, 153–154
Magnification Moore, P., 175
maximum, 151 Mount Hopkins telescope, 21
minimum, 34–37, 41, 44 Mounting (Camera)
Magnitude barn door, 56
bolometric, 212, 215–217 Haig, 56
limiting, 169, 173 Scotch, 56
photographic, 215 Mounting (telescope)
photo-visual, 215 alignment, 135–138
visual, 169, 211, 214, 215 alt-alt, 135
Maksutov telescope, 20, 160 alt-az, 16, 18, 54, 56, 59, 64, 75, 89,
Martian canals, 176, 179 128–135, 140, 149, 152
Mean sun, 80, 81, 84 alt-az-field rotation, 56, 132, 133, 149
Medicean stars, 5 alt-az-zero field rotation, 56, 133, 134
Index 251
basic requirements, 127 Parsec, 110, 114, 212

DIY construction, 131 Parsons, W., 14, 15
Dobsonian, 64, 65, 132 1.8-metre reflector, 14
English, 128, 129 Peltier cooling, 186
equatorial, 16, 54, 55, 64, 89, 128, 129, Periastron, 122
131, 133, 135, 136, 149 Pericynthion, 122
fixed, 16, 22, 135, 136, 163 Perigee, 122, 157
fork, 128, 129, 131 Perihelion, 103, 115, 122, 126, 166
German, 129, 139 Perijove, 122
horseshoe, 128, 131 Phase, 5, 107, 116, 118, 119, 144, 146, 154
modified English, 128, 129 Photographic, 12, 19, 45, 144, 181, 185, 186,
yoke, 129 191–193, 196, 208, 211, 215, 217, 222
Multi-conjugate adaptive optics. Photographic magnitude, 215
See Atmospheric compensation, Photographic photometry, 45, 211, 215,
multi-conjugate adaptive optics 217, 222
Multi-mirror telescope, 21 Photography
detection mechanism, 182, 192
developing, 193
N electron-hole pairs, 186, 192
Nadir, 72 fixing, 193
Nasmyth focus, 16, 18 latent image, 192, 193
Nasmyth system, 17 negative image, 193
New general catalogue (NGC), 174 print, 193
Newton, I., 10, 38 silver bromide, 192, 193
Newtonian telescope, 10, 19, 32, 61, 62, 64, Photometer, 15, 40, 46, 127, 138, 157, 217–219
160, 176 Photometry
Night vision glasses, 65 CCD, 211–212
Noise observing procedure, 218
dark, 186 Photomultiplier, 144, 194, 212
thermal, 186, 190, 217 Photon, 47, 144–145, 173, 182, 183, 186, 192
Photon counting, 47
Photo-visual magnitude, 215
O p-i-n photodiode, 181, 193–194, 211, 212, 217
Objective. See Telescope objective Pixel, 47, 186, 188–190, 194, 211, 212,
Observatory, 18, 19, 25, 56–57, 69, 80, 81, 86, 217, 218
87, 90, 101, 150, 163, 165, 180, 186, Planck curve, 145
192, 207–209 Planet
Observing site, 24, 56–57, 150, 151, 153, 158, direct motion, 108
164, 169 motion, 102, 106, 108, 109
requirements, 56 retrograde motion, 108
Occultation, 95, 96, 116, 117, 120, 158, 160 Planisphere, 104–106, 134, 151
Old Moon in the new Moon’s arms, 154, 155 Pleiades, 174
Opera glass, 3 Pogson, N., 167
Opposition, 116, 123, 158, 166 Pogson’s equation, 167
Orbital period, 123, 126, 170. See also Point source, 5, 10, 27, 31–36, 48–53, 169,
Sidereal period 181, 211
Polar axis, 54, 89, 128, 135–137, 149
Polarimetry, 147
P Polaris, 101, 112, 113, 136, 168, 213, 225, 226
Paraboloid. See Telescope mirror, Polarisation, 145–147, 175, 206
primary-paraboloidal Positions in the sky, 52, 69–97, 113
Parallax, 102, 114–115 Precession, 75, 84, 102, 112–114, 152
Parallax angle, 114 Prime focus telescope, 16, 22, 26, 46, 47
252 Index
Prime meridian, 77–79, 133 Ritchey-Chretién telescope, 15, 18,

Prism, 65, 132, 222, 223 38, 64
Prominence spectroscope, 164 Rómer, O., 118
Proper motion, 109–111, 113, 114, 208
Ptolemy, 109
Purkinje effect, 180 S
Pyrex, 40, 61 Safety, 150–151, 162
SALT. See South African Large
Telescope (SALT)
Q Saturn
Quadrature, 117 Cassini division, 8
rings, 8, 65, 160
satellites, 8, 106, 160
R Schmidt, B., 18
Radial velocity, 226–227 Schmidt camera, 18–20, 38, 47, 49
Radian, 23, 33, 47, 52, 275 Schmidt-Cassegrain telescope, 20, 40, 41,
Radio telescope 59–60, 150, 186, 195
Arecibo, 27, 135 Schmidt-Newtonian telescope, 19
dish, 25, 26, 29, 135 Scintillation, 22, 24, 211, 217
feed, 22, 26, 46 Seasons, 102–104
half-wave dipole, 26 Second-of-arc, 23–25, 34, 47, 48, 78, 79, 90,
intermediate frequency, 27 113, 114, 128, 155–157
RATAN-600 (see RATAN-600 radio Seeing, 24, 44, 149–151, 158, 159,
telescope) 176, 208
receiver, 27, 60, 140, 185 Seidel aberration. See Aberration (optical)
resolution, 25–27 Serrurier truss, 53, 54
super-heterodyne, 27 Setting, 90, 99–101, 103, 104, 135–137, 152,
Yagi, 26 167, 175, 185, 190, 218
RATAN-600 radio telescope, 27 Setting circle, 90, 128, 135–137, 149, 151, 152,
Rayleigh limit, 34 158, 161, 173, 175
Receiver. See Radio telescope Shortest day, 102
Reflection, 8, 9, 24, 44, 63, 185 Short, J., 11
law of, 6 Sidereal clock, 77
Reflector, 10, 12, 14, 15, 18–20, 53, 54, 61, Sidereal day, 54, 76–77, 83, 86, 89, 99, 128
64, 160 Sidereal period, 123–126. See also
Refraction, 8, 12, 99, 103, 132 Orbital period
atmospheric, 138 Sidereal time
Refractive index, 49, 143 approximate, 87, 88
Refractor calculation, 100
astronomical (see Astronomical refractor) definition, 77
Galilean (see Galilean refractor) local, 82, 83, 87–88, 100, 134, 137
Resolution relation with HA and RA, 95
airy disc, 33 Signal-to-noise ratio, 46, 47, 196
diffraction-limited, 23, 171 Significance level, 205, 207
interferometer, 25, 27–30, 50–53 Silver mirror test, 62
Rayleigh limit, 34 Sine rule. See Spherical trigonometry
spectral, 221–223 SI units and prefixes, 207, 239–240
Rete, 105, 106 Slit, 147, 164, 222
Rhodopsin, 37, 179, 180 SLR camera, 185
Right ascension (RA), 55, 88–91, 93, 95–97, Solar clock, 77
100, 108, 110–111, 113, 114, 128, 133, Solar day, 77, 86, 91
136–139, 151, 165, 166 mean, 76, 83
Rising, 99–101, 103, 104 Solar diagonal, 44–45
Index 253
Solar system Stellar magnitude, 167, 169, 181, 196, 212.

Copernican model, 7 (see also See also Absolute magnitude; Apparent
Heliocentric model) magnitude; Magnitude
heliocentric model, 109 (see also Stellar nomenclature
Copernican model) Bayer system, 165
Solar time, 81–82, 84, 86, 134 catalogue number, 165
mean, 77, 80–82, 84 Flamsteed number, 165
Solstice, 77, 85, 100, 102 pulsars, 166
Sombrero galaxy, 172, 173 variable stars, 165
South African Large Telescope Student’s t test, 200, 203–207
(SALT), 22, 135 Subaru telescope, 47
Space telescope, 11, 20–25, 38, 39, 62, 165 Summer solstice, 77, 85, 100
Spectral resolution, 221–223 Summer triangle, 102
Spectral type, 216, 218, 223–226 Sun, 5, 25, 44, 45, 71, 74–76, 80, 81, 83–85,
MKK system, 224 91, 99, 100, 102–104, 106, 107, 109,
Spectrograph, 223, 227 115–120, 122–124, 140, 146, 151, 154,
Spectrophotometry, 227–228 158, 160–164, 167, 168, 175, 177, 213,
Spectroscope 214, 216, 217, 224–226
absorption cell, 227 motion, 75, 81
basic, 222, 223 Sundial, 81, 82
collimator, 222 Superconducting tunnel junction, 27, 194
comparison spectrum, 226–227 Super-heterodyne. See Radio telescope
direct vision, 223 Surface brightness, 36, 37, 172
dispersion, 222–224 Synodic period, 123–126
grating, 222, 223
prism, 222, 223
radial velocity, 226–227 T
resolution, 221–223 Tear-drop effect, 177, 181
slit, 222 Telecompressor, 45
spectrophotometry, 227–228 Telephoto property, 11, 18
Spectroscopy, 147, 149, 193, 217, Telescope. See also Catadioptric; Catoptric;
221–228 Dioptric; Reflector; Refractor
Speculum metal, 10, 14, 15 aerial, 8, 26
Spherical aberration. See Aberration (optical) alignment, 21, 53, 60, 105, 135–137, 140
Spherical polar coordinates, 69–71, 75, 89 alignment-completely automatic, 60,
Spherical triangle. See Spherical trigonometry 140, 169
Spherical trigonometry alignment-data base, 60, 171
cosine rule, 93, 94, 96 alignment onto object, 53, 60, 100, 117,
five parts rule, 95 122, 135–137, 140
four parts rule, 94, 96 alignment-semi-automatic, 60
sine rule, 92, 94 cassegrain (see Cassegrain telescope)
spherical triangle, 94, 95 collimation, 63, 64, 127
Spider, 62, 64 commercial, 105, 135, 139
Spring equinox, 77 Coudé (see Coudé telescope)
Square kilometer array (SKA) radio DIY construction, 131
telescope, 208 field of view (see Field of view, telescope)
Standard deviation, 197, 198, 204, Gregorian (see Gregorian telescope)
206, 207, 208 Herschelian (see Herschelian telescope)
Standard error of the mean, 197–199, 204, large, 11, 12, 15, 18, 21, 22, 24, 46, 47, 132,
206, 207, 209 147, 221, 222
Standard star, 167, 211, 218 major, 15, 16, 46–48, 56, 127–128, 131,
Star colour, 179 181, 186, 195, 211–212
Star diagonal, 42–44 Maksutov (see Maksutov telescope)
254 Index
Telescope. (cont.) Twilight, 158

mounting (see Mounting (telescope)) astronomical, 102–103
multi-mirror (see Multi-mirror telescope) civil, 103
Newtonian (see Newtonian telescope) nautical, 103
objective, 14, 38–40, 128, 160, 217 Twinkling, 24, 151
objective lens, 32, 65
objective mirror, 9, 32, 38
prime focus (see Prime focus telescope) U
radio (see Radio telescope) U, B, V filters, 215
Ritchey-Chretién (see Ritchey-Chretién Ultra low expansion (ULE), 40, 61
telescope) Ultra-violet, 22, 146, 159, 215
Schmidt-Cassegrain (see Schmidt- Uncertainties (in data), 197–199, 201, 218
Cassegrain telescope) Unfilled aperture. See Aperture synthesis
Schmidt-Newtonian (see Schmidt- Universal time (UT), 82–83
Newtonian telescope) Ursa Major, 111
second hand, 60, 63 change of shape with time, 111
small, 10, 15, 19, 38, 40, 42–44, 56, 59–65,
127, 128, 132, 157, 158, 172, 174, 179,
195, 221 V
space (see Space telescope) Velocity of light (c), 118, 143, 144, 227
terrestrial (see Terrestrial telescope) Venus, 5, 25, 32, 116–119, 158, 160, 168, 176,
tube, 9, 53, 62, 127, 128, 150, 160, 161, 163 177, 213
tube-Serrurier truss, 53–54 phase, 5, 116, 119
Telescope mirror Vernal equinox, 77, 81, 84, 102
primary, 9, 11, 12, 15, 16, 19, 21–23, 38, Very large array (VLA). See Aperture synthesis
53, 54, 61–63, 135, 162, 185 Very large survey telescope (VLST), 189
primary-hyperboloidal, 11, 15, 21 Very large telescope (VLT)
primary-paraboloidal, 9, 12, 21, 61 Antu, 18, 173
primary-spherical, 9, 10, 12, 19, 22, 48 ISAAC, 10, 18
secondary, 9, 11, 12, 15, 19–21, 46, 53, 62, Melipal, 223
64, 159, 160 Very-long base-line (VLBI).
secondary-ellipsoidal, 9 See Interferometer
secondary-hyperboloidal, 11, 15, 21 Virtual observatory, 196, 208–209
Telescope objective, 14, 38–40, 160, 217 Vision, 37, 38, 65, 169, 172, 174, 179–181,
Telescope tube, 9, 53, 62, 127, 128, 150, 195, 223
160–163 averted, 38, 174, 179, 181
Terrestrial telescope, 8 Visual binary star, 91, 171
Thermal noise, 186, 217 Visual magnitude, 169, 214, 215
Thirty meter telescope (TMT), 22, 25 Visual observing
Tide, 107, 117 day-time observing, 175–176
Time, 3, 10–12, 22, 24, 25, 28, 40, 41, 59–62, false observations, 176–177
71, 72, 76–91, 99, 100, 102, 104, 105, finding, 149, 151–153, 156, 161, 162, 172,
107–111, 113, 116–118, 123, 125–128, 174, 176
131, 133–137, 140, 144, 145, 149–155, galaxies, 151, 170, 172–175
157, 158, 162–164, 173, 188, 193, 195, moon, 46, 153–158, 160, 164, 168, 175
198, 208, 209, 217, 218, 224 moon-occultations, 158, 160
heliocentric, 91 moon-TLPs, 157
Time zone, 81, 82, 86–88, 133, 134 nebulae, 46, 152, 172–175
Timocharis, 112, 166 planets, 151, 158–160, 172, 174, 175, 177
Tip-tilt mirror. See Atmospheric compensation problems, 45, 132, 150–152, 154,
Tracking, 6, 54, 56, 128, 131, 138, 217 161, 176
Transit, 116–118, 120, 136, 160, 177 safety, 150–151, 162
exoplanet detection, 120, 124, 208, 227 star hopping, 152, 158, 174
Index 255
stars, 45, 151–153, 157–159, 164–172, Wien’s law, 214

173–175 Winter solstice, 77, 85, 102
stars-magnitude estimation, 171
sun, 44, 151, 154, 158, 160–164, 167, 168,
175, 177 X
sun-eyepiece projection, X-ray, 144–146, 160
45, 163
sun-full aperture filter, 45,
162, 164 Y
sun-H-a filter, 164 Yagi. See Radio telescope
sun-prominence spectroscope, 164 Yoke mounting. See Mounting (telescope)
sun-warnings and precautions, 150, 158,
161, 175, 193
sun-Zurich sunspot number, 163 Z
Visual purple, 37, 179 Zenith, 42–43, 62, 72, 75–77, 99, 103,
von Fraunhofer, J., 14 131, 135
Zenith distance, 62, 75–76, 103
Zerodur, 40, 61
W Zodiac, 74–75
Wave front, 24, 25 Zodiacal constellations, 74, 75
Weber-Fechner law, 166 Zurich sunspot number, 163

Hci 634 G

Uploaded by

Copyright:

Available Formats

Hci 634 G

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hci 634 G

Uploaded by

Copyright:

Available Formats

Undergraduate Lecture Notes in Physics

For further volumes:

ULNP titles must provide at least one of the following:

ULNP especially encourages new, original, and idiosyncratic approaches to physics

Neil Ashby, Professor, Professor Emeritus, University of Colorado Boulder, CO,

Telescopes and Techniques

ISSN 2192-4791 ISSN 2192-4805 (electronic)

# Springer Science+Business Media New York 1995, 2003, 2013

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Chris Kitchin is Professor Emeritus of Astronomy at the University of Hertfordshire

Preface to the First Edition .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. . . .. . .. vii

Part II Positions and Motions

4 Positions in the Sky . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Eclipses and Related Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Part III Observing

7 Electromagnetic Radiation . . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. 145

9 Detectors and Imaging .. . .. . .. . .. .. . .. . .. .. . .. . .. .. . .. . .. . .. .. . .. . .. .. . .. . .. 181

Appendix A For Further Reading . .. . . . . . .. . . . . . .. . . . . .. . . . . . .. . . . . . .. . . . . .. . 231

Appendix B Constellations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

Appendix C Answers to Exercises . . . . .. . . . . . . . .. . . . . . . . . .. . . . . . . . .. . . . . . . . . .. 237

Appendix D SI and Other Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Appendix E The Greek Alphabet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

C. R. Kitchin, Telescopes and Techniques, Undergraduate Lecture Notes in Physics, 3

Fig. 1.1 Optics of the Galilean refractor

Fig. 1.2 Suggested optical system of Digges’ possible telescope

Fig. 1.3 Optics of the

Fig. 1.4 Spherical aberration

Fig. 1.5 Spherical aberration in a simple lens

Fig. 1.7 Simple terrestrial

Fig. 1.8 Ray paths for

Fig. 1.9 Gregorian telescope

Fig. 1.10 Newtonian

Fig. 1.11 Cassegrain telescope

Cassegrain.1 This optical design, or variations on it, is still almost universally

Fig. 1.12 Achromatic doublet

Fig. 1.14 Modern

Fig. 1.15 Telescope used

Fig. 1.16 The Coudé system

Fig. 1.17 The Nasmyth

The Schmidt Camera

Fig. 1.19 Optical layout

camera, is called a catadioptric instrument (reflectors can also be called catoptric

Fig. 1.20 Optical layout

Fig. 1.21 A 180-mm

Multi-Mirror and Space Telescopes

The history of telescopes shows a recurring theme:

Fig. 1.22 Optical layout of

resolution (2.3) of 0.0500 2 has enabled it to make some outstanding discoveries.

360 ¼ a full circle

For some purposes, though, radians need to be used, and

Fig. 1.25 An interferometer and its output

C. R. Kitchin, Telescopes and Techniques, Undergraduate Lecture Notes in Physics, 31

Fig. 2.1 Optical paths in

where D is the diameter of the objective in meters. A typical amateur astronomer’s

Example Calculation 2.1

f o ¼ Focal ratio D ¼ 4 0:5 ¼ 2 metres

Thus from (2.1)

From (2.2) we have