X-rays in the Solar System

Author's personal copy Provided for non-commercial research and educational use only. Not for reproduction, distribution or commercial use. This chapter was originally published in the book Encyclopedia of Solar system. The copy attached is provided by Elsevier for the author's benefit and for the benefit of the author's institution, for non-commercial research, and educational use. This includes without limitation use in instruction at your institution, distribution to specific colleagues, and providing a copy to your institution's administrator. All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing copies or access, or posting on open internet sites, your personal or institution’s website or repository, are prohibited. For exceptions, permission may be sought for such use through Elsevier’s permissions site at: http://www.elsevier.com/locate/permissionusematerial From Bhardwaj, A., Lisse, C. M., & Dennerl, K. (2014). X-rays in the Solar System. In T. Spohn, D. Breuer, & T. V. Johnson (Eds.), Encyclopedia of the Solar System, Elsevier (pp. 1019–1045). ISBN: 9780124158450 Copyright © 2014, 2007, 1999 Elsevier Inc. All rights reserved. Elsevier Author's personal copy Chapter 48 X-rays in the Solar System Anil Bhardwaj Space Physics Laboratory, Vikram Sarabhai Space Centre, Trivandrum, Kerala, India Carey M. Lisse Applied Physics Laboratory, Johns Hopkins University, Laurel, Maryland Konrad Dennerl Max-Planck-Institut für extraterrestrische Physik, Garching, Germany Chapter Outline 1. Introduction 2. Earth 2.1. Auroral Emissions 2.2. Nonauroral Emissions 2.3. Geocoronal Emissions 3. The Moon 4. Mercury 5. Venus 6. Mars 7. Jupiter 7.1. Auroral Emission 7.2. Nonauroral (Disk) Emission 8. Galilean Satellites 9. Io Plasma Torus 1019 1020 1020 1022 1022 1023 1025 1025 1027 1029 1029 1031 1031 1032 1. INTRODUCTION The usually defined range of X-ray photons spans w0.1e100 keV. Photons in the lower (<5 keV) end of this energy range are termed soft X-rays. In space, X-ray emission is generally associated with high-temperature phenomena, such as hot plasmas of 1 million to 100 million K and above in stellar coronae, accretion disks, and supernova shocks. However, in the solar system, X-rays have been observed from bodies that are much colder, T < 1000 K. This makes the field of planetary X-rays a very interesting discipline, where X-rays are produced from a wide variety of objects under a broad range of conditions. The first planetary X-rays detected were terrestrial X-rays, discovered in the 1950s. The first attempt to detect 10. Saturn 11. Rings of Saturn 12. Comets 12.1. Spatial Morphology 12.2. X-ray Luminosity 12.3. Temporal Variation 12.4. Energy Spectrum 12.5. Summary 13. Asteroids 14. Heliosphere 15. Summary Acknowledgments Bibliography 1032 1033 1034 1035 1037 1037 1038 1038 1039 1040 1041 1045 1045 X-rays from the Moon in 1962 failed, but it discovered the first extrasolar source, Scorpius X-1, which resulted in the birth of the field of X-ray astronomy. In the early 1970s, the Apollo 15 and 16 missions studied fluorescently scattered X-rays from the Moon. Such X-rays originate when energetic photons or particles remove an inner electron from atoms in the irradiated material. When the atom relaxes by filling the resulting gap with an outer shell electron, an X-ray photon with a characteristic energy is emitted. At low X-ray energies, this photon is usually produced by the transition of an n ¼ 2 to the n ¼ 1 shell electron, and is termed a Ka transition. A Kb transition would be for an n ¼ 3 to n ¼ 1 electron, while an La photon would be produced by an n ¼ 3 to n ¼ 2 transition, etc. Encyclopedia of the Solar System. http://dx.doi.org/10.1016/B978-0-12-415845-0.00048-7 Copyright Ó 2014 Elsevier Inc. All rights reserved. Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 1019 Author's personal copy 1020 PART | IX Exploring the Solar System Comet Hyakutake (C/1996 B2) ROSAT Wide Field Camera FIGURE 48.1 Chandra montage of solar system X-ray sources. Upper panel, from left to right: Venus, Mars, Comet Hyakutake (C/1996 B2), Jupiter (aurora þ reflected disk emissions). Bottom panel, from left to right: Saturn, Saturn rings, Earth (auroral emission), Moon (sunlit side emission). When Ka X-rays are produced in an oxygen atom, it is called oxygen Ka or O-Ka photon. Launch of the first X-ray satellite Uhuru in 1970 marked the beginning of satellite-based X-ray astronomy. The subsequently launched X-ray observatory Einstein discovered, after a long search, X-rays from Jupiter in 1979. Before 1990, the three objects known to emit X-rays (in addition to the Sun) were Earth, Moon, and Jupiter. In 1996, ROSAT (Röntgensatellit) made an important contribution to the field of solar system X-rays by discovering X-ray emissions from comets. This discovery revolutionized the field of solar system X-rays and highlighted the importance of the solar wind charge exchange (SWCX) mechanism in the production of X-rays in the solar system, which will be discussed in this chapter in various sections. Today, the field of solar system X-rays is very dynamic and in the forefront of new research. During the past one decade or so, our knowledge about the X-ray emission from bodies within the solar system has significantly improved. The advent of higher resolution X-ray spectroscopy with the Chandra and X-Ray Multi Mirror Mission (XMM)Newton X-ray observatories, followed by the nextgeneration SWIFT and Suzaku observatories has been of great benefit in advancing the field of planetary X-ray astronomy. Several new solar system objects are now known to shine in the X-ray (Figure 48.1). At Venus, Earth, the Moon, Mars, Jupiter, Saturn (including its rings), and asteroids, scattered solar X-rays have been observed. The first soft X-ray observation of Earth’s aurora by Chandra showed that it is highly variable, and the Jovian aurora is a fascinating puzzle that is just beginning to yield its secrets. The X-ray emission from comets, the exospheres of Venus, Earth, and Mars, and the heliosphere are all largely driven by charge exchange between highly charged minor (heavy) ions in the solar wind and gaseous neutral species. This chapter surveys the current understanding of X-ray emission from the solar system bodies. We start our survey locally, at the Earth, move to the Moon and the nearby terrestrial planets, and then venture out to the giant planets and their moons and rings. Next, we move to the small bodies, comets and asteroids, found between the planets, and finally we study the X-ray emission from the heliosphere surrounding the whole solar system and possibilities of X-rays from extrasolar systems. An overview is provided on the main source mechanisms of X-ray production from each object. For further details, readers are referred to the bibliography provided at the end of the chapter and references therein. 2. EARTH 2.1. Auroral Emissions Precipitation of energetic charged particles from the magnetosphere into Earth’s auroral upper atmosphere leads to ionization, excitation, dissociation, and heating of the neutral atmospheric gas. Deceleration, or braking of precipitating particles during their interaction with atom and molecules in the atmosphere, results in the production of a continuous spectrum of X-ray photons, called bremsstrahlung. This is the main X-ray production mechanism in the Earth’s auroral zones, for energies above w3 keV; therefore, the X-ray spectrum of the aurora has been found to be very useful in studying the characteristics Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy X-rays in the Solar System 1021 The PIXIE instrument aboard POLAR is the first X-ray detector that provides true two-dimensional global X-ray images at energies >3 keV. In Figure 48.2 two images taken by PIXIE in two different energy bands are presented. The auroral X-ray zone can be clearly seen. Data from the PIXIE camera have shown that the X-ray bremsstrahlung intensity statistically peaks at midnight, is significant in the morning sector, and has a minimum in the early dusk sector. During solar substorms, X-ray imaging shows that the energetic electron precipitation brightens up in the midnight sector and has a prolonged and delayed maximum in the morning sector due to the scattering of magnetic-drifting electrons and shows an evolution significantly different than that when viewed in the ultraviolet (UV) emissions. During the onset/expansion phase of a typical substorm, the electron energy deposition power is about 60e90 GW, which produces 10e30 MW of bremsstrahlung X-rays. By combining the results of PIXIE with the UV imager aboard POLAR, it has been possible to derive the energy distribution of precipitating electrons in the 0.1e100 keV range with a time resolution of about 5 min (see Figure 48.2). Because these energy spectra cover the entire energy range important for the electrodynamics of the ionosphere, important parameters like the Hall and Pedersen conductivities and the amount of Joule heating can be determined on a global scale with larger certainties than parameterized models can do. Electron energy deposition estimated from global X-ray imaging also gives valuable information on how the constituents of the upper atmosphere, like NOx, is modified by energetic electron precipitation. of energetic electron precipitation. In addition, particles precipitating into the Earth’s upper atmosphere give rise to discrete atomic emission lines in the X-ray range. The characteristic inner-shell line emissions for the main species of the Earth’s atmosphere are all in the low-energy range (nitrogen Ka at 0.393 keV, oxygen Ka at 0.524 keV, argon Ka at 2.958 keV, and Kb at 3.191 keV). Very few X-ray observations have been made at energies at which these lines are emitted. While charged particles spiral around and travel along the magnetic field lines of the Earth, the majority of the X-ray photons in Earth’s aurora are directed normal to the field, with a preferential direction toward the Earth at higher energies. Downward-propagating X-rays cause additional ionization and excitation in the atmosphere below the altitude where the precipitating particles have their peak energy deposition. The fraction of the X-ray emission that is moving away from the ground can be studied using satellite-based imagers (e.g. AXIS on Upper Atmosphere Research Satellite (UARS) and PIXIE on the POLAR spacecraft). Auroral X-ray bremsstrahlung has been observed from balloons and rockets since the 1960s and from spacecraft since the 1970s. Because of absorption of the low-energy X-rays propagating from the production altitude (w100 km) down to balloon altitudes (35e40 km), such measurements were limited to >20 keV X-rays. Nevertheless, these early omnidirectional measurements of X-rays revealed detailed information of temporal structures from slowly varying bay events to fast pulsations and microbursts. (b) UVI LBHL 0302:12 UT 2.5 107 2.5 108 20–21 MLT X-rays (1/keV s sr cm2) Electrons (1/keV s sr cm2) 0 (d) EOE = 9.89 keV 1000 100 10 0 20 10 E (keV) PIXIE 7.9–21.3 keV 0258 - 0308 UT X rays (1/s sr cm2) Photons (1/s sr cm2) 0 (c) PIXIE 2.8–9.9 keV 0300:30 - 0305:00 UT 10000 0 (e) 108 104 0.01 X rays (1/s sr cm2) 5000 X-rays (1/keV s sr cm2) (a) 1 100 E (keV) 2000 4000 21–22 MLT 1000 EOE = 3.18 keV EOE = 80.4 keV 100 10 0 10 20 E (keV) Electrons (1/keV s sr cm2) Chapter | 48 FIGURE 48.2 Earth’s aurora. POLAR satellite observation on July 31, 1987. (a) UVI and (b, c) PIXIE images in two different energy bands. (d) Left: Measured X-ray energy spectrum. An estimated X-ray spectrum produced by a single exponential electron spectrum with e-folding energy 9.89 keV is shown to be the best fit to the measurements. Right: The electron spectrum derived from UVI (thin line) and PIXIE (thick line). Both plots are averages within a box within 20e21 magnetic local time (MLT) and 64e70 magnetic latitude. (e) Same as (d) but within 21e22 MLT, where a double exponential electron spectrum is shown to be the best fit to the X-ray measurements. From Østgaard et al. (2001). 108 104 0.01 1 100 E (keV) Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1022 PART | IX Limb scans of the nighttime Earth at low- to midlatitude by the X-ray astronomy satellite High Energy Astronomy Observatory-1 (HEAO-1) in 1977, in the energy range 0.15e3 keV, showed clear evidence of the Ka lines for nitrogen and oxygen sitting on top of the bremsstrahlung spectrum. The High-Resolution Camera (HRC-I) aboard the Chandra X-ray Observatory imaged the northern auroral regions of the Earth in the 0.1- to 10-keV X-ray range at 10 epochs (each w20 min duration) between December 2003 and April 2004. These first soft X-ray observations of Earth’s aurora (see Figure 48.3) showed that it is highly variable (intense arcs, multiple arcs, diffuse patches, and at times absent). Also, one of the observations showed an isolated blob of emission near the expected cusp location. Modeling of the observed soft X-ray emissions suggests that it is a combination of bremsstrahlung and characteristic K-shell line emissions of nitrogen and oxygen in the atmosphere produced by electrons. In the soft X-ray energy range of 0.1e2 keV, these line emissions are w5 times more intense than the X-ray bremsstrahlung. 2.2. Nonauroral Emissions The nonauroral X-ray emission above 2 keV from the Earth is almost completely negligible except for brief periods during major solar flares (Figure 48.4). However, at energies below 2 keV, soft X-rays from the sunlit Earth’s atmosphere have been observed even during quiet (nonflaring) Sun conditions. The two primary mechanisms for the production of X-rays from the sunlit atmosphere are: (1) Thomson (coherent) scattering of solar X-rays from the Exploring the Solar System electrons in the atomic and molecular constituents of the atmosphere and (2) the absorption of incident solar X-rays followed by the resonance fluorescence emission of characteristic K lines of nitrogen, oxygen, and argon. During flares, solar X-rays light up the sunlit side of the Earth by Thomson and fluorescent scattering (Figure 48.4); the X-ray brightness can be comparable to that of a moderate aurora. Around 1994, the Compton Gamma Ray Observatory (CGRO) satellite detected a new type of X-ray source from the Earth. These are very short-lived (1 ms) X-ray and g-ray bursts (w25 keVe1 MeV) from the atmosphere above thunderstorms, whose occurrence is also supported by the more recent Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI) observations. It has been suggested that these emissions are bremsstrahlung from upward-propagating, relativistic (megaelectronvolt) electrons generated in a runaway electron discharge process above thunderclouds by the transient electric field following a positive cloud-to-ground lightning event. 2.3. Geocoronal Emissions In the Earth’s exosphere (geocorona), SWCX with neutrals can produce X-rays. This process is now understood as a contribution to the soft X-ray background and to its longterm enhancements (LTEs) seen in the ROSAT all-sky survey. The LTEs in the ROSAT all-sky survey data are well correlated with the solar wind proton flux, suggesting that SWCX with H in the geocorona is the source of the LTEs. Chandra observations of the Moon are particularly interesting because they have cleanly separated FIGURE 48.3 Four Earth’s aurora X-ray images (shown on the same brightness scale) of the north polar regions obtained by Chandra HRC-I on different days (marked at the top of each image), showing large variability in soft (0.1e10 keV) X-ray emissions from Earth’s aurora. The bright arcs in these Chandra images show low-energy X-rays generated during auroral activity. The images are superimposed on a simulated image of the Earth. From Bhardwaj et al. (2007b). Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System 1023 FIGURE 48.4 (a) X-ray images of the Earth from the POLAR PIXIE instrument for the energy range 2.9e10.1 keV obtained on August 17, 1998, (left) and November 23, 1998, (right), showing the dayside X-rays during a solar X-ray flare. The grid in the picture is in corrected geomagnetic coordinates, and the numbers shown in red are magnetic local times. The terminator at the surface of the Earth is shown as a red dashed line. From Petrinec et al. (2000). (b) Left: a sudden increase and subsequent decrease in X-ray intensity were observed on August 17, 1998, shortly after 21 UT while the PIXIE instrument was observing the Earth’s northern hemisphere. This spike in X-ray intensity coincided with an X1 solar X-ray flare, as measured by GOES-10. Right: the occurrence of an M7 solar flare at w07 Universal Time (UT) resulted in a sudden increase and decrease in X-ray flux from the Earth on September 23, 1998. A later increase in the X-ray emissions from the Earth’s ionosphere (shortly before 12 UT) was due to increased auroral activity. From Bhardwaj et al. (2009). geocoronal from heliospheric charge exchange emissions and all other contributors to the soft X-ray background and have shown correlation of geocoronal X-rays with the flux of highly charged oxygen ions in the solar wind. XMM-Newton and Suzaku have observed correlations of the X-ray intensity from the Earth’s vicinity with the solar wind flux on timescales of about half a day, and also shortterm (w10 min) variations. The correlated variability of X-ray intensity with solar wind, and the lack of correlation with solar X-rays, suggests that the production of SWCX-induced X-rays is taking place in the Earth’s magnetosheath. 3. THE MOON X-ray emissions from the Earth’s nearest planetary body, the Moon (see also “The Moon”), have been studied in two ways: close-up from lunar orbiters (e.g. Apollo 15 and 16, Clementine, SMART-1, Kaguya, and Chandrayaan-1), and more distantly from Earth-orbiting X-ray astronomy telescopes (e.g. ROSAT and Chandra). Lunar X-rays result mainly from fluorescence of solar X-rays from the surface, in addition to a low level of scattered solar radiation. Thus, X-ray fluorescence studies provide an excellent way to determine the elemental composition of the lunar surface by remote sensing, since at X-ray wavelengths the optical properties of the surface are dominated by its elemental abundances. Elemental abundance maps produced by X-ray spectrometers (XRSs) on the Apollo 15 and 16 orbiters were limited to the equatorial regions but succeeded in finding geochemically interesting variations in the relative abundances of Al, Mg, and Si, such as the enhancement of Al/Si in the lunar highlands relative to the mare. The D-Compact X-ray Spectrometer (D-CIXS) instrument on Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1024 PART | IX SMART-1 has obtained abundances of Al, Si, Fe, and even Ca at 50-km resolution from a 300-km altitude orbit about the Moon. XRSs capable of higher spatial and spectral resolution have been flown on Kaguya and Chandrayaan-1. The X-ray charge-coupled devices on Kaguya suffered severe radiation damage en route to the Moon and could not perform well. C1XS on Chandrayaan-1 provided new data sets on composition of Mg, Al, Si, Ca, and Fe at scales of 15 10 5 0 0.001 0.10 Brigthness (R) (c) September 2001, S3 chip 4.0 Rate (counts/ks/keV/arcmin2) 1.00 0.50 1.0 0.25 500 1000 1500 2000 2500 Net observed 2.0 0.0 0.00 500 1000 1500 Energy (eV) 0.5 4 2000 2500 –0.25 1.0 1.5 Energy (keV) 6 8 Energy (keV) 2.0 10 2.5 12 Total observed Background 500 1000 1500 2000 2500 2000 2500 Net observed 0.50 0.25 0 0.0 September 2001, FI chips 0.00 0 0.75 1.0 –1.0 Au–M 0.75 2.0 0.0 0 3.0 2 1.00 Total observed Background (incl. flares) Background from soft flares 3.0 0 0–K Al–K Rate (counts/ks/keV/arcmin2) tEXP = 844 s tEXP = 1899 s 20 Mg–K March 9, 1991 Si–K (b) ROSAT Moon X-rays June 29, 1990 50 km, which indicates regions that may be richer in the sodic variety of plagioclase than previously thought. These recent experiments also suggest that particle-induced X-ray fluorescence could also contribute to the signal. Future X-ray experiments such as CLASS on Chandrayaan-2 aim at global elemental mapping with better spatial resolution (w12e25 km) and wider elemental overage (including direct detection of sodium) and will use refined Rate (counts/ks/keV/arcmin2) (a) Exploring the Solar System 0 500 1000 1500 Energy (eV) FIGURE 48.5 The Moon. (a) ROSAT soft X-ray (0.1e2 keV) images of the Moon at first (left side) and last (right side) quarter. The dayside lunar emissions are thought to be primarily reflected and fluoresced sunlight, while the faint nightside emissions are foreground due to charge exchange of solar wind heavy ions with H atoms in the Earth’s exosphere. The brightness scale in R assumes an average effective area of 100 cm2 for the ROSAT Position Sensitive Proportional Counter (PSPC) over the lunar spectrum. From Bhardwaj et al. (2007a). (b) Chandra spectrum of the bright side of the Moon. The green dotted curve is the detector background. K-shell fluorescence lines from O, Mg, Al, and Si are shifted up by 50 eV from their true values because of residual optical leak effects. Features at 2.2, 7.5, and 9.7 keV are intrinsic to the detector. From Wargelin et al. (2004). (c) Observed and backgroundsubtracted spectra from the September 2001 Chandra observation of the dark side of the Moon, with 29-eV binning. The left panel is from the back-illuminated Advanced CCD Imaging Spectrometer - Spectroscopic Array 3 (ACIS-S3) CCD, while the right panel shows the spectrum obtained with the front-illuminated (FI) ACIS-I Advanced CCD Imaging Spectrometer - Imaging Array (ACIS-I) CCDs, which have higher spectral resolution, but lower sensitivity. Oxygen emission from charge exchange is clearly seen in both spectra, and the energy resolution in the FI chips is sufficient that emission from M-L shell transitions of O7þ ions is largely resolved from L-K transitions of O6þ ions. Emission lines from transitions of what is likely Mg-Ka around 1340 eV, Al-Ka at about 1550 keV, and Si-Ka at approximately 1780 keV are also apparent in the FI spectrum. See also Ewing et al. (2013). Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System methodologies to take into account signals from PIXE, enabling observations even during lunar night. These global maps will be complemented by rovers on the planned missions Selene-2 as well as Chandrayaan-2 which are expected to carry alpha particle spectrometers to study local surface chemistry through in situ quantitative analysis. Early lunar X-ray observations from Earth orbit were made with ROSAT. A marginal detection by the Advanced Satellite for Cosmology and Astrophysics (ASCA) is also reported. Figure 48.5(a) shows the ROSAT images of the Moon; the right image is data from a lunar occultation of the bright X-ray source GX5-1. The power of the reflected and fluoresced X-rays observed by ROSAT in the 0.1e2 keV range coming from the sunlit surface was determined to be only 73 kW. The faint but distinct lunar nightside emissions (100 times less bright than the dayside emissions) were originally interpreted as being produced by bremsstrahlung of solar wind electrons of several hundred electronvolts impacting the nightside of the Moon on its leading hemisphere. However, this was before the GX5-1 data were acquired, which clearly show lunar nightside X-rays from the early trailing hemisphere as well. A new, much better and accepted explanation is that the heavy ions in the solar wind charge exchanges (SCWX) with geocoronal and interstellar H atoms that lie between the Earth and Moon result in foreground X-ray emissions between ROSAT and the Moon’s dark side. This was confirmed by Chandra Advanced CCD Imaging Spectrometer (ACIS) observations in 2001 (see Figure 48.5(c)). The July 2001 Chandra observations also provide the first remote measurements that clearly resolve discrete K-shell fluorescence lines of O, Mg, Al, and Si on the sunlit side of the Moon (see Figure 48.5(b)). The observed oxygen Ka (OeKa for short; which means K-shell emissions from oxygen (O) atoms) line photons correspond to a flux of 3.8  10 5 photons/s/cm2/arcmin2 (3.2  10 14 erg/s/ cm2/arcmin2). The MgeKa, AleKa, and SieKa lines each had roughly 10% as many counts and 3% as much flux as OeKa line, but statistics were inadequate to draw any conclusions regarding differences in element abundance ratios between highlands and maria. Later Chandra observations of the Moon used the photon counting, high spatial resolution HRC-I imager to look for albedo variations due to elemental composition differences between highlands and maria. The observed albedo contrast was noticeable, but very slight, making remote elemental mapping difficult. 4. MERCURY Being too close to the Sun, Mercury (see also “Mercury”) cannot be observed by any X-ray observatory orbiting 1025 around the Earth. Launched on August 3, 2004, the MESSENGER (MErcury Surface, Space ENvironment, GEochemistry, and Ranging) spacecraft conducted its first flyby of Mercury on January 14, 2008, followed by two subsequent encounters on October 6, 2008, and September 29, 2009, prior to Mercury orbit insertion. On March 18, 2011, it became the first probe to orbit the planet Mercury. The XRS onboard MESSENGER has observed X-ray fluorescence emission from Mercurydthereby providing important information on the elemental composition of its surface. The XRS spectra have revealed Mercury’s surface to differ in composition from those of other terrestrial planets and the Moon. The XRS has also observed electron-induced X-ray emission from the dark side of Mercury. These X-ray fluorescence emissions are produced by interaction of w1- to 10-keV electrons with Mercury’s surface and the abundance results derived from this technique for Mg, Si, and Al are found to be consistent with those derived from solar-induced X-ray fluorescence. In fact, electron-induced X-ray fluorescence has been used to derive the spectrum of electrons precipitating onto the Mercury surface, which is in agreement with the MESSENGER’s Energetic Particle Spectrometer (EPS)-measured spectrum. This showed that both the dayside and the nightside of an airless planetary body can be mapped by X-ray fluorescencedwhile on the dayside solar photons produce fluorescence and on the nightside its electrons. The XRS has also been used to study astrophysical objects during MESSENGER’s flybys of Mercury. In future, the Mercury Imaging X-ray Spectrometer (MIXS) aboard the European Space AgencyeJapan Aerospace Exploration Agency BepiColombo mission will help further investigate Mercury’s X-rays and look for even solar wind-induced charge exchange emission, since the strength of solar wind flux is very high at Mercury. 5. VENUS The first X-ray observation of Venus was obtained by Chandra in January 2001. It was expected that Venus would be an X-ray source due to two processes: (1) charge exchange interactions between highly charged ions in the solar wind and the Venusian exosphere and (2) scattering of solar X-rays in the Venusian atmosphere. The predicted X-ray luminosities were w0.1e1.5 MW for the first process, and w35 MW for the second one, with an uncertainty factor of about 2. The Chandra observation of 2001 consisted of two parts: grating spectroscopy with Low Energy Transmission Grating (LETG)/Advanced CCD Imaging Spectrometer - Spectroscopic Array (ACIS-S) and direct imaging with ACIS-I. This combination yielded data of high spatial, spectral, and temporal resolution. Venus was clearly detected as a half-lit crescent, exhibiting Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1026 PART | IX Exploring the Solar System FIGURE 48.6 Venus. (a) First X-ray image of Venus, obtained with Chandra ACIS-I on January 13, 2001. The X-rays result mainly from fluorescent scattering of solar X-rays on C and O in the upper Venus atmosphere, at heights of 120e140 km. In contrast to the Moon, the X-ray image of Venus shows evidence for brightening on the sunward limb. This is caused by the fact that scattering takes place on an atmosphere and not on a solid surface. (b) Spectral scan. Scales are given in kiloelectronvolts and angstroms. The observed C, N, and O fluorescent emission lines are enclosed by dashed lines; the width of these intervals matches the size of the Venus crescent (22.800 ). From Dennerl, Burwitz, Englhauser, Lisse, and Wolk (2002). (c) Spatial and spectral distribution of X-ray photons from Venus, obtained on March 27, 2006 with Chandra. Left: Distribution of the X-ray photons in the energy range 0.3e0.8 keV. Photons in the fluorescence band FL are marked with bright diamonds, and those in the charge exchange bands CX1 and CX2 with dark circles. The extraction regions for the limb and disk spectra are superimposed on light and dark gray, respectively; the circle indicates the geometric size of Venus. Right: X-ray spectra for the limb (top) and the disk region (bottom), with the energy bands CX1, FL, and CX2 marked. The spectrum of the limb region is dominated by two emission lines in the CX1 and CX2 bands. These lines are almost absent in the disk spectrum, which is dominated by emission in the FL band. The energies of characteristic lines expected for charge exchange and fluorescence are marked by vertical lines. From Dennerl (2008). Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System considerable brightening on the sunward limb (Figure 48.6); the LETG/ACIS-S data showed that the spectrum was dominated by K-shell fluorescence of oxygen and carbon (OeKa and CeKa), and both instruments indicated temporal variability of the X-ray flux. An average luminosity of 55 MW was found, which agreed well with the theoretical predictions for scattered solar X-rays. In addition to the CeKa and OeKa emission at 0.28 and 0.53 keV, respectively, the LETG/ACIS-S spectrum also showed evidence for NeKa emission at 0.40 keV. An additional emission line was indicated at 0.29 keV, which might be the signature of the C 1s / p* transition in CO2. The observational results are consistent with fluorescent scattering of solar X-rays by the majority species in the Venusian atmosphere, and no evidence of the 30 times weaker charge exchange interactions was found. Simulations showed that fluorescent scattering of solar X-rays is most efficient in the Venusian upper atmosphere at heights of w120 km, where an optical depth of 1 is reached for incident X-rays with energy 0.2e0.9 keV. The second Chandra observation of Venus in March 2006 showed clear signature of charge-exchanged X-rays from the exosphere (Figure 48.6), which was marginally detected again in the October 2007 Chandra observation. The bright emission feature is O6þ emission near 565 eV, which is also an important feature in X-ray spectra of comets and the Martian exosphere. The appearance of Venus is different in optical light and in X-rays. The reason for this is that the optical light is reflected from clouds at a height of 50e70 km, while scattering of X-rays takes place at higher regions extending into the tenuous, optically thin parts of the thermosphere and exosphere. As a result, the Venusian sunlit hemisphere appears surrounded by an almost transparent luminous shell in X-rays, and Venus looks brightest at the limb because more luminous material is there. Because X-ray brightening depends sensitively on the density and chemical composition of the Venusian atmosphere, its precise measurement will provide direct information about the atmospheric structure in the thermosphere and exosphere. This opens up the possibility of using X-ray observations for monitoring the properties of these regions that are difficult to investigate by other means, as well as their response to solar activity. 6. MARS The first X-rays from Mars were detected on July 4, 2001, with the ACIS-I detector onboard Chandra. In the Chandra observation, Mars showed up as an almost fully illuminated disk (Figure 48.7). An indication of limb brightening on the sunward side, accompanied by some fading on the opposite side, was observed. The observed morphology and X-ray luminosity of w4 MW, about 10 times less than at Venus, 1027 was consistent with fluorescent scattering of solar X-rays in the upper Mars atmosphere. A single narrow emission line caused by K-shell fluorescence emission of oxygen dominated the X-ray spectrum. Simulations suggest that scattering of solar X-rays is most efficient between 110 km (along the subsolar direction) and 136 km (along the terminator) above the Martian surface. This behavior is similar to that seen on Venus. No evidence for temporal variability or dust-related emission was found, which is in agreement with fluorescent scattering of solar X-rays as the dominant process responsible for Martian X-rays. A gradual decrease in the X-ray surface brightness between 1 and w3 Mars radii is observed (see Figure 48.7). Within the limited statistical quality of the low flux observations, the spectrum of this region (halo) resembled that of comets, suggesting that they are caused by charge exchange interactions between highly charged heavy ions in the solar wind and neutrals in the Martian exosphere (corona). For the X-ray halo observed within 3 Mars radii, excluding Mars itself, the Chandra observation yielded a flux of about 1  10 14 erg/cm2/s in the energy range 0.5e1.2 keV, corresponding to a luminosity of 0.5  0.2 MW for isotropic emission, which agrees well with that expected theoretically for the SWCX mechanism. The first XMM-Newton observation of Mars in November 2003 confirmed the presence of the Martian X-ray halo and allowed a detailed analysis of its spectral, spatial, and temporal properties to be made. Highresolution spectroscopy of the halo with XMM-Newton Reflection Grating Spectrometer (RGS) revealed the presence of numerous (w12) emission lines at the positions expected for deexcitation of highly ionized C, N, O, and Ne atoms (Figure 48.8). The three Ka lines of O VII emission were resolved and found to be dominated by a long-lived forbidden triple to singlet transition, as expected for charge exchange but not fluorescence or collisionally induced X-ray emission. This was the first definite detection of charge exchange-induced X-ray emission from the exosphere of another planet. The XMM-Newton observation confirmed that the fluorescent scattering of solar X-rays from the Martian disk is clearly concentrated on the planet, and is directly correlated with the solar X-ray flux levels. On the other hand, the Martian X-ray halo was found to extend out to w8 Mars radii, with pronounced morphological differences between individual ions and ionization states (Figure 48.8). The halo emission exhibited pronounced variability, but, as expected for solar wind interactions, the variability of the halo did not show any correlation with the solar X-ray flux. Mars was found to be dimmer in observations made during solar minimum, suggesting direct correlation with solar activity. Suzaku could not even detect Mars in X-rays in an observation in April 2008. Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1028 PART | IX Exploring the Solar System FIGURE 48.7 Mars. (a) First X-ray image of Mars, obtained with Chandra ACIS-I. The X-rays result mainly from fluorescent scattering of solar X-rays on C and O in the upper Mars atmosphere, at heights of 110e130 km, similar to Venus. The X-ray glow of the Martian exosphere is too faint to be directly visible in this image. From Dennerl (2002). (b) Spatial distribution of the photons around Mars in the soft (E ¼ 0.2e1.5 keV) and hard (E ¼ 1.5e10.0 keV) energy range, in terms of surface brightness along radial rings around Mars, separately for the dayside (offset along projected solar direction >0) and the nightside (offset <0); note, however, that the phase angle was only 18.2 . For better clarity, the nightside histograms were shifted by one decade downward. The bin size was adaptively determined so that each bin contains at least 28 counts. The thick vertical lines enclose the region between 1 and 3 Mars radii. (c) X-ray spectra of Mars (top) and its X-ray halo (bottom). Crosses with 1s error bars show the observed values; the model spectra, convolved with the detector response, are indicated by gray curves (unbinned) and by histograms (binned as the observed spectra). The spectrum of Mars itself is characterized by a single narrow emission line (this is most likely the O-Ka fluorescence line at 0.53 keV; the apparent displacement of the line energy is due to optical loading). At higher energies, the presence of an additional spectral component is indicated. The spectral shape of this component can be well modeled by the same 0.2-keV thermal bremsstrahlung emission which can be used as a “technical” proxy for characterizing the basic spectral properties of the X-ray halo in an instrument-independent way. From Dennerl (2002). Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System 1029 Energy (eV) 580 570 560 550 540 530 520 510 8 CO2 f 4 1s 1s 1s i 3σμ 4σg 1π μ Counts/ks/A r 1s 6 1π g O6+ 2 Residuals 0 2 0 –2 22 22 23 Wavelength (Å) 23 24 24 Cross dispersion angle (arcmin) –2 –1 0 1 2 204 206 208 210 212 Dispersion angle (arcmin) 214 FIGURE 48.8 Mars. Imaging spectroscopy of Mars with XMM-Newton’s RGS in November 2003 showing images of Mars and its halo in the individual emission lines of ionized oxygen and fluorescence of CO2. (Top) RGS spectrum obtained from a 10000 -wide area along the cross-dispersion direction, showing the region around the CO2 doublet and the O VII multiplet. (Bottom) dispersed images in the same wavelength range as above. The apparent diameter of Mars during this observation was 12.200 . (Right) X-ray image of Mars in November 2003 with XMM-Newton/RGS in the emission lines of charge exchange (green-blue) and fluorescence of solar X-rays (orange). The black circle indicates the size of planet. From Dennerl et al. (2006). 7. JUPITER 7.1. Auroral Emission Like the Earth, Jupiter emits X-rays both from its aurora and its sunlit disk. Jupiter’s UV auroral emissions were first observed by the International Ultraviolet Explorer (IUE) and soon confirmed by the Voyager 1 ultraviolet spectrometer as it flew through the Jupiter system in 1979 (see Bhardwaj and Gladstone (2000) for review). The first detection of the X-ray emission from Jupiter was also made in 1979; the Einstein satellite detected X-rays in the 0.2e3.0 keV energy range from both poles of Jupiter, due to the aurora. Analogous to the processes on Earth, it was expected that Jupiter’s X-rays might originate as bremsstrahlung by precipitating electrons. However, the power requirement for producing the observed emission with this mechanism (1015e1016 W) is more than two orders of magnitude larger than the input auroral power available as derived from Voyager and IUE observations of the UV aurora. (The strong Jovian magnetic field excludes the bulk of the solar wind from penetrating close to Jupiter, and the solar wind at Jupiter at 5.2 AU is 27 times less dense than at the Earth at 1 AU.) Precipitating energetic sulfur and oxygen ions from the inner magnetosphere, with energies in the 0.3e4.0 MeV/nucleon range, were suggested as the source mechanism responsible for the production of X-rays on Jupiter. The heavy ions are thought to start as neutral SO and SO2 molecules emitted by the volcanoes on Io into the Jovian magnetosphere, where they are dissociated and ionized by solar UV radiation, and then swept up into the huge dynamo created by Jupiter’s rotating magnetic field (see also “Jupiter” and “Planetary Magnetospheres”). Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1030 PART | IX The ions eventually become channeled onto magnetic field lines terminating at Jupiter’s poles, where they emit X-rays by first charge stripping to a highly ionized state, followed by charge exchange and excitation through collisions with H2. ROSAT ’s observations of Jupiter X-ray emissions supported this suggestion. The spatial resolution of these early observations was not adequate to distinguish whether the emissions were linked to source regions near the Io torus of Jupiter’s magnetosphere (inner magnetosphere) or at larger radial distances from the planet. The advent of the Chandra and XMM-Newton X-ray observatories revolutionized our understanding of Jupiter’s X-ray aurora. High-spatialresolution (<1 arcsec) observations of Jupiter with Chandra in December 2000 (see Figure 48.9) revealed that most of Jupiter’s northern auroral X-rays come from a “hot spot” located significantly poleward of the UV auroral zones (20e30 Jupiter’s radius, RJ), and not at latitudes connected to the Io plasma torus (IPT) (inner magnetosphere). The hot spot is fixed at 60 e70 magnetic latitude and 160 e180 longitude (in “system III” coordinates rotating with the Jovian magnetosphere) and occurs in a region where anomalous infrared and UV emissions (the so-called flares) have also been observed. On the other hand, auroral X-rays from 70 to 80 S latitude spread almost halfway across the planet (starting at w300 and spreading through 0 out to 120 longitude). The location of the auroral X-rays connects along magnetic field lines to regions in the Jovian magnetosphere well in excess of 30 Jovian radii from the planet, a region where there are insufficient S and O ions to account for the X-ray emission. Acceleration of energetic ions was invoked to increase the phase space distribution, but now the question was whether the acceleration involved outer magnetosphere heavy ions or solar wind heavy ions. Surprisingly, Chandra observations also showed that X-rays from the Jovian aurora pulsate with a periodicity that is quite systematic (approximately 45-min period) at times (in December 2000) and irregular (20e70 min range) at other times (in February 2003). The 45-min periodicity is highly reminiscent of a class of Jupiter high-latitude radio emissions known as quasiperiodic radio bursts, which had been observed by Ulysses in conjunction with energetic electron acceleration in Jupiter’s outer magnetosphere. During the 2003 Chandra observation of Jupiter, the Ulysses radio data did not show any strong 45-min quasiperiodic oscillations, although variability on timescales similar to that in X-rays was present. Chandra also found that X-rays from the northern and southern auroral regions are neither in phase nor in antiphase, but that the peaks in the south are shifted from those in the north by about 120 (i.e. one-third of a planetary rotation). The Chandra and XMM-Newton spectral and spatial observations have now established that X-rays from Jupiter’s aurora are basically of two types: (1) soft (w0.1e2 keV) and (2) hard (>2 keV) X-rays (Figure 48.9). (b) 10–4 10–3 Black: North aurora Red: South aurora Green: Disk 10–5 Normalized counts/s/keV 0.01 (a) Exploring the Solar System 0 2.0 4.0 Brightness (R) 6.0 0.2 0.5 1 2 Channel energy (keV) 5 10 FIGURE 48.9 Jupiter. (a) Detailed X-ray morphology first obtained with Chandra HRC-I on December 18, 2000, showing bright X-ray emission from the polar “auroral” regions, indicating the high-latitude position of the emissions, and a uniform distribution from the low-latitude “disk” regions. From Gladstone et al. (2002). (b) Combined XMM-Newton European Photon Imaging Camera (EPIC) spectra from the November 2003 observation of Jupiter. Data points for the north and south aurorae are in black and red, respectively. In green is the spectrum of the low-latitude disk emission. Differences in spectral shape between auroral and disk spectra are clear. The presence of a high-energy component in the spectra of the aurorae is very evident, with a substantial excess relative to the disk emission extending to 7 keV. The horizontal blue line shows the estimated level of the EPIC particle background. From Branduardi-Raymont et al. (2007). Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System The soft X-rays are line emissions, which are consistent with high-charge states of precipitating heavy O and S ions from the Jovian magnetospheric that are accelerated to attain energies of >1 MeV/nucleon before impacting the Jovian upper atmosphere. There is no evidence for the role of solar wind ions due to the absence of signatures of carbon ions in the X-ray spectrum. Modeling studies which include a source of gaseous neutral atoms (Io’s SiOx and SOx for Jupiter, Enceladus’ H2O for Saturn), solar UV photolysis and ionization (to Sþ, Siþ, and Oþ for Jupiter and Hþ and Oþ for Saturn), followed by pickup, acceleration, and polar precipitation plus charge exchange of the ions at 0e2 keV energies also supports this. The higher (>2 keV) energy X-rays are basically bremsstrahlung from precipitating magnetospheric energetic electrons (see Figure 48.9) and originate from locations that spatially coincide with the Far Ultra Violet (FUV) auroral oval, suggesting that the source of both, X-rays and UV emissions, is the same. The variability on timescales of days suggests a link to changes in the energy distribution of the precipitating magnetospheric electrons and may be related to the solar activity at the time of observation. 7.2. Nonauroral (Disk) Emission The existence of low-latitude “disk” X-ray emission from Jupiter was first recognized in ROSAT observations made in 1994. As for the inner planets, it was suggested that elastic scattering of solar X-rays by atmospheric neutrals (H2 and He for Jupiter) and fluorescent scattering by the dominant multielectron atom, carbon, via carbon K-shell X-rays (mainly from methane (CH4) molecules located below the Jovian homopause) were the sources of the disk X-rays. A general decrease in the overall X-ray brightness of Jupiter observed by ROSAT over the years 1994e1996 was found to be coincident with a similar decay in solar activity index (solar 10.7 cm flux). A similar trend is seen in the data obtained by Chandra in 2000 and 2003; Jupiter’s disk was about 50% dimmer in 2003 compared to that in 2000, which is consistent with variation in the solar activity. A Chandra observation in early 2008 during the New Horizon flyby also showed a dimmer Jupiter, consistent with a decrease in solar activity. First direct evidence for temporal correlation between Jovian disk X-rays and solar X-rays was provided by XMM-Newton observations in November 2003, which demonstrated that day-to-day variations in disk X-rays of Jupiter are synchronized with variations in the solar X-ray flux, including a solar flare that had a matching feature in the Jovian disk X-ray light curve. Chandra observations of December 2000 and February 2003 also support this association between light curves of solar and planetary X-rays. However, there is an indication of higher X-ray 1031 counts from regions of low surface magnetic field in the Chandra data, suggesting the presence of some particle precipitation. The higher spatial resolution observation by Chandra has shown that nonauroral disk X-rays are relatively more spatially uniform than the auroral X-rays (Figure 48.9). Unlike the w40  20-min quasiperiodic oscillations seen in auroral X-ray emission, the disk emission does not show any systematic pulsations. There is a clear difference between the X-ray spectra from the disk and from the auroral region on Jupiter; the disk spectrum peaks at higher energies (0.7e0.8 keV) than that of the aurora emission (0.5e0.6 keV) and lacks the high-energy component (above w3 keV) present in the latter (see Figure 48.9). 8. GALILEAN SATELLITES The Jovian Chandra observations on November 25e26, 1999, and December 18, 2000, discovered the X-ray emission from the Galilean satellites (Figure 48.10). These satellites are very faint when observed from the Earth’s orbit (by Chandra), and the detections of Io and Europa, although statistically very significant, were based on w10 photons each! The energies of the detected X-ray events ranged between 300 and 1890 eV and appeared to show a clustering between 500 and 700 eV, suggestive of oxygen K-shell fluorescent emission. The estimated power of the X-ray emission was 2 MW for Io and 3 MW for Europa. There were also indications of X-ray emission from Ganymede. X-ray emission from Callisto seems likely at levels not too far below the Chandra sensitivity limit because the heavy ion fluxes of the Jovian magnetosphere are an order of magnitude lower than at Ganymede and Europa, respectively. Emissions from Io were also seen in a February 2003 Chandra observation, although at weaker levels. The most plausible proposed emission mechanism is inner (K-shell) ionization of surface and near-surface atoms by incoming magnetospheric ions followed by prompt X-ray emission. Oxygen should be the dominant emitting atom either on a SiOx (silicate) or SOx (sulfur oxides) surface (Io) or on an icy one (the outer Galilean satellites). It is also the most common heavy ion in the Jovian magnetosphere. The extremely tenuous atmospheres of the satellites are transparent to X-ray photons with these energies, as well as to much of the energy range of the incoming ions. However, oxygen absorption in the soft X-ray is strong enough that the X-rays must originate within the top 10 mm of the surface in order to escape. Simple estimates suggest that excitation by incoming ions dominates over electrons and that the X-ray flux produced is within a factor of three of the measured flux. The detection of X-ray emission from the Galilean satellites thus provides a direct measure of the interactions of the Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1032 PART | IX Io Europa 15 15 10 10 5 5 0 0 –5 –5 –10 –10 –15 –15 –15 –10 –5 0 0.02 0.04 Exploring the Solar System 5 0.06 10 15 0.08 –15 –10 –5 0.02 0.04 0 5 0.06 0.08 10 15 FIGURE 48.10 Galilean Moons. Chandra X-ray images of Io and Europa (0.25 keV < E < 2.0 keV) from November, 1999 observations. The axes are labeled in arcsec (1 arcsec w 3000 km) and the scale bar is in units of smoothed counts per image pixel (0.492 by 0.492 arcsec). The solid circle shows the size of the satellite (the radii of Io and Europa are 1821 km and 1560 km, respectively), and the dotted circle shows the size of the detect cell. From Elsner et al. (2002). magnetosphere of Jupiter with the satellite surfaces. An intriguing possibility is placement of an imaging XRS onboard a mission to the Jupiter system. 9. IO PLASMA TORUS The IPT is known to emit at extreme ultraviolet (EUV) energies and below, but it was a surprise when Chandra discovered that it was also a soft X-ray source. The 1999 Jovian Chandra observations detected a faint diffuse source of soft X-rays from the region of the IPT. The 2000 Chandra image, obtained with the HRC-I camera (Figure 48.11), exhibited a dawn-to-dusk asymmetry similar to that seen in the EUV. Figure 48.11 shows the background-subtracted Chandra/ACIS-S IPT spectrum for November 25e26, 1999. This spectrum shows evidence for line emission centered on 574 eV (very near a strong O VII line), together with a very steep continuum spectrum at the softest X-ray energies. Although formed from the same source, the spectrum is different from that of the Jovian aurora because the energies, charge states, and velocities of the ions in the torus are much lowerdthe bulk ions have not yet been highly accelerated. There could be contributions from other charge states because current plasma torus models consist mostly of ions with low charge states, consistent with photoionization and ion-neutral charge exchange in a low-density plasma and neutral gas environment. The 250- to 1000-eV energy flux at the telescope aperture was 2.4  10 14 erg/cm2/s, corresponding to a luminosity of 0.12 GW. Although bremsstrahlung from nonthermal electrons might account for a significant fraction of the continuum X-rays, the physical origin of the observed IPT X-ray emission is not yet fully understood. The 2003 Jovian Chandra observations also detected X-ray emission from the IPT, although at a fainter level than in 1999 or 2000. The morphology exhibited the familiar dawn-to-dusk asymmetry. 10. SATURN The production of X-rays at Saturn (see also “Saturn”) was expected because, like the Earth and Jupiter, Saturn was known to possess a magnetosphere with energetic electrons and ions within it; however, early attempts to detect X-ray emission from Saturn with Einstein in December 1979 and with ROSAT in April 1992 were negative and marginal, respectively. Saturnian X-rays were unambiguously observed by XMM-Newton in October 2002 and by the Chandra X-ray Observatory in April 2003. In January 2004, Saturn was again observed with Chandra ACIS-S in two exposures, one on 20 January and the other on 26e27 January, with each observation lasting for about one full Saturn rotation. The X-ray power emitted from Saturn’s disk is roughly one-fourth of that from Jupiter’s disk, which is consistent with Saturn being twice as far from the Sun and the Earth as Jupiter. Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System 1033 Callisto (a) Io 5 0 –5 N –15 –10 Europa 0 –5 Normalized counts/s keV 10 Ganymede 10 (b) 5 12 14 16 18 15 E FIGURE 48.11 Plasma Torus. (a) Chandra/HRC-I image of the IPT (December 18, 2000). The axes are labeled in units of Jupiter’s radius, RJ, and the scale bar is in units of smoothed counts per image pixel. The paths traces by Io, Europa, and Ganymede are marked on the image. Callisto is off the image to the dawn side. The regions bounded by rectangles were used to determine background. The regions bounded by dashed circles or solid ellipses were defined as source regions. (b) Chandra/ACIS-S spectrum for the IPT from November 1999. The solid line presents a model fit for the sum of a power law spectrum and a Gaussian line, while the dashed line represents just a pure power law spectrum. The line is consistent with K-shell fluorescent emission from oxygen ions. From Elsner et al. (2002). 0.10 0.08 0.06 0.04 0.02 0.00 0.2 0.4 0.6 0.8 Channel energy (keV) The January 2004, Chandra observation showed (Figure 48.12) that X-rays from Saturn are highly variabled a factor of twoefour variability in brightness over 1 week. The bright X-rays from Saturn’s south polar cap on January 20 (see Figure 48.12, left panel), which are not evident in the January 26 observation (see Figure 48.12, right panel) and in earlier Chandra observations are an extension of the disk X-ray emission of Saturn. No evidence of auroral X-rays from Saturn has been found so far, which could be due to the limited sensitivity of current X-ray observatories. As is the case for Jupiter’s disk, X-ray emission from Saturn is due to the scattering of the incident solar X-ray flux. An X-ray flare has been detected from the nonauroral disk of Saturn during the Chandra observation on January 20, 2004, which, taking light travel time into account, coincided with an M6-class flare emanating from a sunspot that was clearly visible from both Saturn and Earth. This was the first direct evidence suggesting that Saturn’s disk X-ray emission is principally controlled by processes happening on the Sun. Further, a good correlation 1.0 has been observed between Saturn X-rays and F10.7 solar activity (see also “The Sun”) index, suggesting a solar connection. The spectrum of X-rays from Saturn’s disk is very similar to that from Jupiter’s disk (Figure 48.12). 11. RINGS OF SATURN The rings of Saturn (see also “Planetary Rings”), known to be made of mostly water (H2O) ice, are one of the most fascinating objects in our solar system. The discovery of X-rays from the rings of Saturn was made with Chandra ACIS-S observations in January 2004 and April 2003. X-rays from the rings are dominated by emission in a narrow (w130 eV wide) energy band of 0.49e0.62 keV (Figure 48.13). This band is centered on the oxygen Ka fluorescence line at 0.53 keV, suggesting that fluorescent scattering of solar X-rays from oxygen atoms in the surface of H2O icy ring material is the likely source mechanism for ring X-rays. The X-ray power emitted by the rings in the 0.49- to 0.62-keV band on January 20, Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1034 FIGURE 48.12 Saturn. (a) Chandra ACIS X-ray 0.24- to 2.0-keV images of Saturn on January 20 and 26, 2004. Each continuous observation lasted for about one full Saturn rotation. The horizontal and vertical axes are in units of Saturn’s equatorial radius. The white scale bar in the upper left of each panel represents 10 arcsec. The two images, taken a week apart and shown on the same color scale, indicate substantial variability in Saturn’s X-ray emission. From Bhardwaj et al. (2005a). (b) Disk X-ray spectrum of Saturn (red curve) and Jupiter (blue curve). Values for the Saturn spectrum are plotted after multiplying by a factor of 5. From Bhardwaj (2006). PART | IX January 20, 2004 (a) January 26, 2004 1. 1. RS 0 RS 0 –1. –1. –2. –1. 1. 0 RS –2. 2. 0 (b) Exploring the Solar System 0.02 Brightness (R) –1. 0 RS 1. 2. 0.04 Comparison of Chandra ACIS-S X-ray spectrum of Saturn and Jupiter disk 0.05 Saturn, January 20, 2004 Jupiter, February 24, 2003 Counts/s/keV 0.04 0.03 0.02 0.01 0.00 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 Energy (keV) 2004, is 84 MW, which is about one-third of that emitted from the Saturn disk in the 0.24- to 2.0-keV band. The projected rings have about half the surface area of the Saturn disk, consistent with this ratio. The X-ray power emitted by the rings in the 0.49- to 0.62-keV band could vary from 30 to 150 MW depending on the observation period. Figure 48.13 shows the X-ray image of the Saturnian system in January 2004 in the 0.49- to 0.62-keV band, the energy range where X-rays from the rings were unambiguously detected. The observations of January 2004 also suggested that, similar to Saturn’s X-ray emission, the ring X-rays are highly variableda factor of two to three variability in brightness over 1 week. There is an apparent asymmetry in X-ray emission from the east (morning) and west (evening) ansae (the apparent extremities of the rings, looking like two handles; see Figure 48.13(a)). However, when the Chandra ACIS-S data sets of January 2004 and April 2003 are combined, the evidence for asymmetry is not that strong. Recent study by XMM-Newton suggests no direct relationship between Saturn disk X-rays and ring X-rays: while disk X-rays follow solar activity, the X-rays from rings do not, suggesting the role of other processes in their production, like meteoric impact-induced spokes or lightning-induced electron beams. 12. COMETS The ROSAT discovery of X-ray emission in 1996 from C/1996 B2 (Hyakutake) created a new class of cold, 102e103 K X-ray-emitting objects. Observations since 1996 have shown that the very soft (E < 1 keV) emission is due to an interaction between the solar wind and the comet’s atmosphere (see also “Physics and Chemistry of Comets”), and that X-ray emission is a fundamental property of comets. Theoretical and observational work in the two decades since the discovery has demonstrated that charge exchange collision of highly charged heavy solar wind ions with cometary neutral species is the best explanation for the emission. In fact, of the solar system bodies with associated X-ray emission, comets are the best example of a nearly pure charge exchange emitting system, as their gravitationally unbound atmospheres (or comae), are tenuous and highly extended (typically 105e106 km in radius), unable to scatter Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System 1035 January 20, 2004 (a) January 26, 2004 1. 1. RS RS 0 –1. –1. –2. –1. 0 RS 1. 0 2. –2. –1. 1. 2. 0.016 January 20 January 26 0.014 0.012 0.010 10 9 January 20 8 7 6 5 4 3 2 1 0.008 Counts/s/keV 0 RS 0.01 0.02 0.03 Brightness (R) Counts/s/keV (x0.001) (b) 0 0 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Energy (keV) 0.006 0.004 0.002 0.000 –0.002 –0.004 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Energy (keV) FIGURE 48.13 Saturn’s rings. (a) Chandra ACIS X-ray images of the Saturnian system in the 0.49- to 0.62-keV band on January 20 and 26e27, 2004. The X-ray emission from the rings is clearly present in these restricted energy band images; the emission from the planet is relatively weak in this band (see Figure 48.12 (a) for an X-ray image of the Saturnian system in the 0.24- to 2.0-keV band). (b) Background-subtracted Chandra ACIS-S3-observed X-ray energy spectrum for Saturn’s rings in the 0.2e2.0 keV range on January 20 and 26e27, 2004. The cluster of X-ray photons in the w0.49- to 0.62-keV band suggests the presence of the oxygen Ka line emission at 0.53 keV in the X-ray emission from the rings. The inset shows a Gaussian fit (peak energy ¼ 0.55 keV, s ¼ 140 eV), indicated by the dashed line, to the ACIS-observed rings’ spectrum on January 20 suggesting that X-ray emissions from the rings are predominantly oxygen Ka photons. From Bhardwaj et al. (2005b). many solar X-rays but highly capable of intercepting a large amount of solar wind ions as they stream away from the Sun. Recently, X-ray emission has also been detected from a comet (C/2011 W3) flying through the lower solar corona, with the solar X-ray imager X-Ray Telescope (XRT) onboard Hinode. This emission is thought to arise from the ionization of cometary material in the corona. The observed characteristics of the emission can be organized into the following four categories: (1) spatial morphology, (2) X-ray luminosity, (3) temporal variation, and (4) energy spectrum. Each of the observed characteristics depends on the nature of the comet’s coma and the solar wind it interacts with. We discuss the typical nature of each of these characteristics next. 12.1. Spatial Morphology X-ray and EUV images of C/1996 B2 (Hyakutake) made by the ROSAT High Resolution Imager (HRI) and Wide Field Camera (WFC) look very similar (Figure 48.14). Except for C/1990 N1, 2P/Encke, 73P/SW-3B, and C/2011 W3 (see below), all EUV and X-ray images of comets have exhibited similar spatial morphologies. The emission is largely confined to the sunward side of the cometary Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1036 PART | IX Exploring the Solar System (a) x 1010 1.8 1.6 1 0 1.2 1 –1 0.8 –2 0.6 –3 –4 0.4 0.2 –5 x 1011 x 104 5 4 7 3 6 V(km) 2 5 1 0 4 –1 –2 3 2 0 –3 –4 –5 –5 –4 –3 –2 –1 0 1 U(km) (c) 1.4 2 V(km) 3 2 3 x 104 1 5 4 Counts/s/keV (b) 1 2 3 4 5 –5 –4 –3 –2 –1 x 104 0 1 U(km) 2 3 4 0.2 5 0.4 x 104 0.6 0.8 Channel energy (keV) 1 (d) Comet X-ray Signal (counts/s) 0.5 Comet 0.0 SW Btotal –0.5 –1.0 Solar X-rays –1.5 SW proton flux –2.0 3 4 5 6 7 Time (UT Day of July 1997) 8 9 FIGURE 48.14 The rich behavior of X-ray emission seen from comets. (a) Cometary X-ray emission morphology. Images of C/1996 B2 (Hyakutake) on March 26e28, 1996 UT: ROSAT HRI 0.1e2.0 keV X-ray, ROSAT WFC 0.09e0.2 keV EUV, and visible light, showing a coma and tail, with the X-ray emission contours superimposed. The Sun is toward the right, the plus signs mark the position of the nucleus, and the orbital motion of the comet is toward the lower left in each image. From Lisse et al. (1996). (b) Morphology as a function of comet gas production rate (given in terms of molecules/s in the lower right of each panel). Note the decreasing concentration of model source function and the increasing importance of diffuse halo emission in the extended coma as the gas production rate increases. From Lisse et al. (2005). (c) Chandra ACIS spectra of the X-ray emission from three comets. All curves show ACIS-S3 measurements of the 0.2- to 1.0-keV pulse height spectrum, with 1s error bars and the best-fit emission line þ thermal bremsstrahlung model convolved with the ACIS-S instrument response as a histogram. Pronounced emission due to O7þ and O6þ is evident at 560 and 660 eV, and for C5þ, C4þ, and N5þ emission lines at 200e500 eV. Best-fit model lines at 284, 380, 466, 552, 590, 648, 796, and 985 eV are close to those predicted for charge exchange between solar wind C5þ, C6þ, C6þ/N6þ, O7þ, O8þ, and Ne9þ ions and neutral gases in the comet’s coma. (Black) ACIS spectra of C/LINEAR 1999 S4 (circles), from Lisse et al. (2001). (Red) Comet McNaught-Hartley spectra (squares), after Krasnopolsky et al. (2003). Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System coma; almost no emission is found in the extended tails of dust or plasma. The X-ray brightness gradually decreases with increasing cometocentric distance r with a dependence of about r 1. The emission morphology and range depend on the amount of neutral gas pouring out of the comet’s nucleus. The brightness merges with the soft X-ray background emission at distances as small as 104 km for weakly active comets, but the most actively outgassing comets can be X-ray luminous out to 106 km. For the least active comets, the X-ray emission tracks the regions of densest coma gas, usually at the nucleus or along jets and shells; for the highly outgassing comets, the coma is collisionally thick to charge exchange and the region of peak emission is crescent shaped with a brightness peak displaced toward the Sun from the nucleus (Figure 48.14). The distance of this peak from the nucleus appears to increase with increasing gas production rate; for Hyakutake, it was located at w2  104 km. Numerical simulations of the solar wind interaction with Hyakutake including charge exchange have been used to generate X-ray images. A global magnetohydrodynamic model and a hydrodynamic model were used to predict solar wind speeds and densities in addition to the X-ray emission around a comet. The simulated X-ray images are similar to the observed images. In the collisionally thick case, the gas production rate can be directly determined from the observed X-ray morphology, as it was demonstrated for four comets observed by ROSAT and XMM-Newton. It is also possible to deduce the location of the cometary bow shock by a tomographic analysis of the X-ray morphology. This technique was successfully applied to XMM-Newton data of comet C/2000 WM1 (LINEAR). A completely different morphologyda tail-like structuredwas observed from Comet Lovejoy (C/2011 W3) with XRT on Hinode 30 min after perihelion, when the comet was only w320,000 km above the solar photosphere. The X-ray emission was similar to the EUV emission, but offset along the local magnetic field lines. This emission, however, is unlikely to result from charge exchange interactions but rather from direct ionization of the cometary material. 1037 12.2. X-ray Luminosity The observed X-ray luminosity, Lx, of comets is mainly determined by the gas production rate and the heavy ion flux in the solar wind, and also by instrumental parameters like the energy bandpass and the observational aperture. Typical values of Lx range from 0.01 to 1 GW. This is of the order of 10 4 of the total luminosity of a comet. For weakly active comets, with Lx < 0.1 GW, a roughly linear correlation between optical and X-ray luminosities is observed. For the brightest X-ray comets, a plateau or asymptote in the X-ray production is seen at a maximum value of w1016 erg/s. Particularly dusty comets, like C/1995 O1 (HaleeBopp), 103P/Hartley 2, or 73P/SchwassmannWachmann 3 appear to have less X-ray emission than would be expected from their overall optical luminosity Lopt. This is most likely a consequence of the fact that the optical luminosity of a comet is dominated by the amount of dust, while the X-ray luminosity is controlled by the amount of gas. The peak X-ray surface brightness decreases with the inverse square of the heliocentric distance r, independent of the gas production rate. 12.3. Temporal Variation Photometric light curves of the X-ray and EUV emission typically show a long-term baseline level with superimposed impulsive spikes of a few hours’ duration, with positive excursions typically three to four times that of the baseline emission level. Figure 48.14 demonstrates the strong correlation found between the time histories of the solar wind proton flux (a proxy for the solar wind minor ion flux), the solar wind magnetic field intensity, and a comet’s X-ray emission, for the case of comet 2P/Encke observed almost continuously over the course of 2 weeks in 1997 by ROSAT and Extreme Ultraviolet Explorer (EUVE) (Figure 48.14(d)). Another long-term study was conducted for comet 9P/Tempel 1 in 2005, supporting the Deep Impact mission. Comparison of comet luminosities with time histories of the solar wind proton flux, oxygen ion flux, and solar X-ray flux shows a strong correlation between the cometary emission and the = (Green) 2P/Encke spectrum taken on November 24, 2003, multiplied by a factor of 2. The C/1999 S4 (LINEAR) and C/McNaught-Hartley 2001 observations had an average count rate on the order 20 times as large, even though Encke was closer to Chandra and the Earth when the observations were being made. Note the 560-eV complex to 400-eV complex ratio of 2e3 in the two bright, highly active comets, and the ratio of approximately 1 for the faint, low-activity comet Encke. From Lisse et al., op. cit (2005). (d) Temporal trends of the cometary X-ray emission. Light curve, solar wind magnetic field strength, solar wind proton flux, and solar X-ray emission for 2P/Encke 1997 on July 4e9, 1997, UT. All error bars are 1s. D, ROSAT HRI light curve, July 4e8, 1997. >, EUVE scanner Lexan B light curve July 6e8, 1997, UT, taken contemporaneously with the HRI observations, and scaled by a factor of 1.2. Also plotted are the Comprehensive Solar Wind Laboratory for Long-Term Solar Wind Measurements (WIND) total magnetic field Btotal (*), the Solar and Heliospheric Observatory (SOHO) Charge, Element, and Isotope Analysis System (CELIAS)/Solar Extreme Ultraviolet Monitor (SEM) 1.0e500 Å solar X-ray flux (>), and the SOHO CELIAS solar wind proton flux (boxes). There is a strong correlation between the solar wind magnetic field/density and the comet’s emission. There is no direct correlation between outbursts of solar X-rays and the comet’s outbursts. From Lisse et al. (1997). Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1038 PART | IX solar wind oxygen ion flux, a good correlation between the comet’s emission and the solar wind proton flux, but no correlation between the cometary emission and the solar X-ray flux. Up until 2005, the temporal variation of the solar wind dominated the observed behavior on all but the longest timescales of weeks to months. A “new” form of temporal variation was demonstrated in the Chandra observations of comet 2P/Encke 2003, wherein the observed X-ray emission is modulated at the 11.1-h period of the nucleus rotation. Rotational modulation of the signal should be possible only in collisionally thin (to SWCX) comae with weak cometary activity, where a change in the coma neutral gas density can directly affect the cometary X-ray flux. Imaging of the X-ray emission of comet 103P/Hartley 2, compared to optical groundbased images of the comet obtained during the Deep Impact Extended mission flyby of the comet, also seem to show correlated rotational modulation of the comet’s X-ray emission. 12.4. Energy Spectrum Until 2001, all published cometary X-ray spectra had very low spectral energy resolution (DE/E w 1 at 300e600 eV), and the best spectra were those obtained by ROSAT for C/1990 K1 (Levy) and C/1990 N1 (TsuchiyaKiuchi), and by BeppoSAX for C/1995 O1 (HaleeBopp). These observations were capable of showing that the spectrum was very soft. However, due to the limited spectral resolution, continuum emission could not be distinguished from a multiline spectrum, as it would result from the SWCX mechanism. It was found that thermal bremsstrahlung was a good “technical” proxy for characterizing the basic spectral properties in an instrumentindependent way, given the limited spectral resolution and the lack of a more realistic model spectrum. All these spectra were consistent with bremsstrahlung temperatures kT between 0.2 and 0.3 keV. Nondetections of comets C/Hyakutake, C/Tabur, C/HaleeBopp, and 55P/TempeleTuttle using the X-Ray Timing Explorer Proportional Counter Array (XTE PCA) (2e30 keV) and ASCA Solidstate Imaging Spectrometers (SIS) (0.6e4 keV) imaging spectrometers were consistent with an extremely soft spectrum. In 2001, the first high-resolution cometary X-ray spectrum was obtained, using Chandra X-ray observatory measurements of the emission from comet C/1999 S4 (LINEAR) as it passed close by the Earth. Discrete line emission signatures due to highly ionized oxygen and carbon were immediately apparent. Higher resolution spectra of cometary X-ray emission are now common. Eight comets were studied with the Chandra X-ray observatory spectroscopy in the period 2000 to 2006, covering Exploring the Solar System the transition from solar maximum to solar minimum. Figure 48.15(a) shows the (background-subtracted raw) spectra for eight of the comets, and Figure 48.15(b) and (c) show at which ecliptic latitude and phase in the solar cycle the spectra were observed. It is immediately obvious that there are spectral differences. In Figure 48.15(a), three spectral bands are indicated, dominated by emission from (1) C V, C VI, N VI (“C þ N”); (2) by O VII; and (3) by O VIII ions, and the spectra are arranged so that, from top to bottom, flux is systematically shifted from lower to higher energy bands. The quantitative results of the spectral fits clearly show that the flux in the C þ N band is anticorrelated to that in the O VIII band (Figure 48.15(d)), indicating that the comets were exposed to different solar wind conditions. As can be seen in Figure 48.15(b) and (c), all the comets which were observed at high latitudes happened to be there during solar maximum, when the equatorial solar wind had expanded into these regions. This implies that, until 2006, Chandra had not observed any comet exposed to the polar wind. This situation changed in October 2007, during solar minimum, when the nucleus of comet 17P/Holmes experienced a spectacular outburst, which increased its dust and gas outflow and optical brightness by almost a million times within hours, from under 17 mag to 3 mag, making it by far the optically brightest comet observable by Chandra since its launch. At the time, comet 17P/Holmes was located at a sufficiently high heliographic latitude (19 ) to be exposed to the polar wind at solar minimum. It was thus expected that this comet would exhibit considerably different X-ray properties, and in fact this was observed: 17P/Holmes became the first comet where Chandra did not detect any significant X-ray emission at all. The most likely explanation for this dramatic X-ray faintness is that the polar wind was so diluted and its ionization so low that only very little X-ray flux was generated by charge exchange at energies above w300 eV. An instrumental effect, i.e. a loss of sensitivity, can definitively be ruled out, because only two months later, another comet, 8P/Tuttle, was observed with Chandra, and this comet, at low latitude (3 ), was clearly detected in X-rays. 12.5. Summary Driven by the solar wind, cometary X-rays provide an observable link between the solar corona, where the solar wind originates, and the solar wind where the comet resides. They are the cleanest example of charge exchangedriven X-ray emission, and should prove to be quite valuable in understanding other astrophysical charge exchange systems found wherever cold neutral and hot ionized gases meetde.g. in the entire heliosphere, in stellar winds, in massive star-forming regions, in the expanding Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System 1039 C + N OVII OVIII (a) (b) D 73P/2006 100 H A B E 0 H Ecliptic plane 200 F E 100 1 AU 0 C F 0 9P/2005 100 G 0 C/2000 WM1 100 C (c) Monthly number of sunspots C/2001 Q4 200 Counts G Sun 2P/2003 200 180 H 120 100 80 60 40 20 1996 1998 2000 2002 2004 2006 Year C/1999 S4 A 250 (d) 50 H 500 C/1999 T1 B 250 0 4000 D 153P/2002 C + N/OVII 0 Cold, fast E Warm, slow F 10 A G 5 0.4 0.6 0.8 Energy (keV) 1.0 0.1 B D C 2000 0 EF G 140 0 0 500 AB C D 160 Hot, fast, disturbed 0.5 OVIII/OVII 1.0 FIGURE 48.15 Summary of the spectral results obtained with Chandra for all the comets (denoted by AeH) which were observed from 2000 to 2006. (a) The 0.3- to 1.0-keV pulse height distributions, (b) the ecliptic latitudes, (c) phases in the solar cycle of the observed comets, and (d) the deduced information about the solar wind heavy ion content. Two comets were observed interacting with low ionization temperature but fast winds arising from the bottom of the solar corona (E, H); at least two with high ionization temperature but slow winds arising from the top of the Sun’s corona (F, G, and possibly A and C); and two comets (B & D) appeared to have interacted with disturbed solar winds found during flares, coronal mass ejections, or solar sector boundary crossings. Adapted from Bodewits et al. (2007). shells of supernova remnants, in active galaxies, or in clusters of galaxies. In our own solar system, once we have understood the SWCX mechanism’s behavior in cometary comae in sufficient detail, we will be able to use comets as probes to measure the solar wind throughout the inner heliosphere. This will be especially useful in monitoring the solar wind in places hard to reach with spacecraftdsuch as over the solar poles, at large distances above and below the ecliptic plane, and at heliocentric distances greater than a few astronomical units. For example, about one-third of the observed soft X-ray emission is found in the 530- to 700-eV oxygen O7þ and O6þ lines; observing photons of this energy will allow studies of the oxygen ion charge ratio of the solar wind, which is predicted to vary significantly between the slow and fast solar winds at low and high solar latitudes, respectively. 13. ASTEROIDS X-rays from asteroids have been studied by experiments on two in situ missions, the X-ray/gamma-ray spectrometer (XGRS) on the Near Earth Asteroid Rendezvous (NEAR)eShoemaker mission to asteroid 433 Eros, and the XRS on the Hayabusa mission to asteroid 25143 Itokawa Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1040 PART | IX (see also “Near Earth Asteroids”). The only attempt to detect X-rays from an asteroid remotely was a 10-ks, observation by Chandra on December 11, 2001, of 1998 WT24, but it was unsuccessful. The results of the in situ observations show X-ray emission due to fluorescence and scattering of incident solar X-rays, similar to the emission seen from the surface of the airless Moon. In fact, the best measurements were obtained during a strong solar flare, when the incident solar X-rays were highly amplified. As for the Moon, X-ray spectroscopy of resonantly scattered solar X-rays can be used to map the elemental composition of the surface. NEAR-Shoemaker entered an Eros orbit on February 14, 2000, and completed a 1-year-long mission around it. Eros at 33  13  13 km in size is the second largest near-Earth asteroid, and its “day” is 5.27 h long. Eros exhibits a heavily cratered surface with one side dominated by a huge, scallop-rimmed gouge; a conspicuous sharp, raised rimmed crater occupies the other side. The XRS part of the XGRS detected X-rays in the 1e10 keV energy range to determine the major elemental composition of Eros’ surface. The XRS observed the asteroid in low orbit (<50 km) during May 2, 2000, to August 12, 2000, and again during December 12, 2000 to February 2, 2001. These observations suggest that elemental ratios for Mg/Si, Al/Si, Ca/Si, and Fe/Si on Eros are most consistent with a primitive chondrite and give no evidence of global differentiation. The S/Si ratio is considerably lower than that for a chondrite and is most likely due to surface volatilization (“space weathering”). The overall conclusion is that Eros is broadly “primitive” in its chemical composition and has not experienced global differentiation into a core, mantle, and crust, and that surface effects cause the observed departures from chondritic S/Si and Fe/Si. Hayabusa reached the asteroid 25143 Itokawa on September 12, 2005. The first touchdown occurred on November 19, 2005. The observations made during the touchdown, a period of relatively enhanced solar X-ray flux, returned an average elemental mass ratio of Mg/Si ¼ 0.78  0.07 and Al/Si ¼ 0.07  0.03. These early results suggest that, like Eros, asteroid Itokawa’s composition can be described as an ordinary chondrite, although occurrence of some differentiation cannot be ruled out. The composition and structure of the rocks and minerals in asteroids provide critical clues to their origin and evolution and are a fundamental line of inquiry in understanding the asteroids, of which more than 20,000 have been detected and cataloged. It is interesting to note that for both Eros and Itokawa the compositions derived by remote X-ray observations using spacecraft in close proximity to the asteroid seem consistent with those found using Earth-based optical and infrared spectroscopy. Exploring the Solar System 14. HELIOSPHERE The solar wind (see also “The Solar Wind”) flow starts out slowly in the corona but becomes supersonic at a distance of few solar radii. The gas cools as it expands, falling from w106 K down to about 105 K at 1 AU. The average properties of the solar wind at 1 AU are proton number density w7/cm3, speed w 450 km/s, temperature w 105 K, magnetic field strength w5 nT, and Mach number w8. However, the composition and charge state distribution far from the Sun are “frozen-in” at coronal values due to the low collision frequency outside the corona. The solar wind contains structure, such as slow (300 km/s) and fast (700 km/s) streams, which can be mapped back to the Sun. The solar wind “terminates” in a shock called the heliopause, where the ram pressure of the streaming solar wind has fallen to that of the interstellar medium (ISM) gas. The region of space containing plasma of solar origin, from the corona to the heliopause at w100 AU, is called the heliosphere. A very small part of the solar wind interacts with the planets and comets; the bulk of the wind interacts with neutral ISM gas in the heliosphere and neutral and ionized ISM at the heliopause. X-ray emission from the heliosphere has also been predicted from the interaction of the solar wind with the interstellar neutral gas (mainly HI and HeI) that streams into the solar system. It has been demonstrated that roughly half of the observed 0.25-keV X-ray diffuse background can be attributed to this process (see Figure 48.16). Solar and Heliospheric Observatory (SOHO) observations of neutral hydrogen Lyman-alpha emission show a clear asymmetry in the ISM flow direction, with a clear deficit of neutral hydrogen in the downstream direction of the incoming neutral ISM gas, most likely created by SWCX ionization of the ISM. The analogous process applied to other stars has been suggested as a means of detecting stellar winds. Also a strong correlation between the solar wind flux density and the ROSAT “LTEs,” systematic variations in the soft X-ray background of the ROSAT X-ray detectors, has been shown. Photometric imaging observations of the lunar nightside by Chandra made in September 2001 does not show any lunar nightside emission above an SWCX background. The soft X-ray emission detected from the dark side of the Moon, using ROSAT, would appear to be attributable not to electrons spiraling from the sunward to the dark hemisphere, as proposed earlier, but to SWCX in the geocorona and the column of heliosphere between the Earth and the Moon (Section 3). Just as charge exchange-driven X-rays are emitted throughout the heliosphere, similar emission must occur within the astrospheres of other stars with highly ionized stellar winds that are located within interstellar gas clouds that are at least partially neutral. Although very weak, in Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy Chapter | 48 X-rays in the Solar System 1041 FIGURE 48.16 Heliosphere. (Upper panel) ROSAT All-Sky Survey map of the cosmic X-ray background at 1/4 keV. The data are displayed using an Aitoff projection in galactic coordinates centered on the galactic center with longitude increasing to the left and latitude increasing upward. Low intensity is indicated by purple and blue while red indicates higher intensity. (Lower panel) same as above except the contaminating LTEs (SWCX emission) were not removed. The striping is due to the survey geometry where great circles on the sky crossing at the ecliptic poles were scanned precessing at w1 /day. From Snowden et al. (1997). principle, this emission offers the opportunity to measure mass loss rates and directly image the winds and astrospheres of other main sequence late-type stars. Imaging would provide information on the geometry of the stellar wind, such as whether outflows are primarily polar, azimuthal, or isotropic and whether or not other stars have analogs of the slow (more ionized) and fast (less ionized) solar wind streams. 15. SUMMARY Table 48.1 summarizes our current knowledge of the X-ray emissions from the planetary bodies that have been observed to produce soft X-rays. Several other solar system bodies, including Titan, Uranus, Neptune, and inner icy satellites of Saturn, are also expected to be X-ray sources, but they are yet to be detected. X-rays are expected from these bodies due to scattering of solar X-rays as well as SWCX and/or magnetospheric ion precipitation and electron bremsstrahlung. However, due to larger distance from Sun, and hence much reduced solar radiation and solar wind flux, the X-rays produced at these objects would be at level much lower than the detection capability of current X-ray observatories. X-rays would also be produced in extrasolar planetsdthrough processes similar to those in our solar system. However, detecting X-rays from exoplanets would be a challenging task since the flux would be very weak due to large distances involved. Table 48.2 lists the spacecraft missions and satellitebased observatories mentioned in the text, which have contributed to the growth of planetary X-rays and our current understanding of the processes of X-ray production. Upcoming X-ray observatories, like Astro-H, eROSITA/SRG, and Athena, and planetary missions (BepiColombo, Selene-2, Chandrayaan-2) carrying experiments are also listed, which would have better sensitivity, effective area, and resolution, thus providing better tools to significantly advance the field of solar system X-rays. Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Author's personal copy 1042 TABLE 48.1 Summary of the Characteristics of Soft X-ray Emission from Solar System Bodies Object Emitting Region Power Emitted1 Special Characteristics Possible Production Mechanism Earth Auroral atmosphere 10e30 MW Correlated with magnetic storm and substorm activity EB þ characteristic line emission from atmospheric neutrals due to electron impact Nonauroral atmosphere 40 MW Correlated with solar X-ray flux FS by atmosphere Auroral atmosphere 0.4e1 GW Pulsating (w20e60 min) X-ray hot spot in north polar region Energetic ion precipitation from magnetosphere and/or solar wind þ EB Nonauroral atmosphere 0.5e2 GW Relatively uniform over disk RS þ possible ion precipitation from radiation belts Dayside surface 0.07 MW Correlated with solar X-rays FS by the surface elements on dayside Jupiter Moon Comets Sunward-side coma Comets in the solar corona Coma plus tail Venus Sunlit atmosphere Nightside emissions are w1% of the dayside 0.2e1 GW 50 MW Exosphere Mars 5 SWCX with geocorona 6 Intensity peaks in sunward direction, w10 e10 km ahead of cometary nucleus and is correlated with solar wind parameters SWCX with cometary neutrals Offset from the EUV emission along magnetic field Ionization of cometary material Emissions from w120 to 140 km above the surface FS by C and O atoms in the atmosphere Emissions from region 1.2 times Venus radius SWCX with Venus exospheric neutrals Sunlit atmosphere 1e4 MW Emissions from upper atmosphere at heights of 110 e130 km FS by C and O atoms in the upper atmosphere Exosphere 1e10 MW Emissions extend out to w8 Mars radii SWCX with Martian corona Io Surface 2 MW Emissions from upper few micrometers of the surface Energetic Jovian magnetospheric ions impact on the surface Europa Surface 3 MW Emissions from upper few micrometers of the surface Energetic Jovian magnetospheric ions impact on the surface Plasma torus 0.1 GW Dawn-dusk asymmetry observed EB þ ? Sunlit disk 0.1e0.4 GW Varies with solar X-rays RS þ FS by atmosphere þ EB Rings of Saturn Surface 80 MW Emissions confined to a narrow energy band around at 0.53 keV. FS by atomic oxygen in H2O ice þ ? Asteroid Sunlit surface Emissions vary with solar X-ray flux FS by elements on the surface Mercury Dayside Emissions vary with solar flux FS of solar X-rays by elements on the surface Depends on precipitating electron spectrum Electron-induced FS by elements on the surface Emissions vary with solar wind SWCX with heliospheric neutrals Nightside Heliosphere Entire heliosphere 16 10 W SWCX, solar wind charge exchange is the charge exchange of heavy, highly ionized solar wind ions with neutrals; FS, fluorescent scattering of solar X-rays; RS, resonant scattering of solar X-rays by atmospheric constituents; EB, bremsstrahlung from precipitating energetic electrons. The question mark (?) refer to some other process(es) at work not clearly known as of now. 1 The values quoted are values at the time of observation. X-rays from all bodies are expected to vary with time. For comparison, the total X-ray luminosity from the Sun is 1020 W. Exploring the Solar System IPT Saturn PART | IX Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Nightside (geocoronal) Author's personal copy Chapter | 48 TABLE 48.2 Satellites and Spacecraft Mentioned in the Text, Chronologically Sorted According to Launch Date Category Date Agency, Country Uhuru First X-ray satellite X-ray Astronomy 1970e1973 NASA (USA) Apollo 15 Fourth manned lunar landing Moon JulyeAugust 1971 NASA (USA) Apollo 16 Fifth manned lunar landing Moon April 1972 NASA (USA) HEAO-1 High Energy Astronomy Observatory-1 X-ray Astronomy 1977e1979 NASA (USA) UV spectrometer Voyager 1 Outer solar system Jupiter, Saturn 1977epresent NASA (USA) IUE International Ultraviolet Explorer UV Astronomy 1978e1996 NASA, ESA, Science and Engineering Research Council (SERC) Einstein (HEAO-2) High Energy Astronomy Observatory 2 X-ray Astronomy 1978e1981 NASA (USA) PSPC, HRI, WFC ROSAT Röntgensatellit X-ray Astronomy 1990e1999 Germany, USA, UK Ulysses Solar Wind Observatory Solar Wind 1990e2009 ESA, NASA CGRO Compton Gamma-Ray Observatory Gamma-ray Astronomy 1991e2000 NASA (USA) AXIS UARS Upper Atmosphere Research Satellite Earth 1991e2011 NASA (USA) Lexan B EUVE Extreme Ultraviolet Explorer EUV Astronomy 1992e2001 NASA (USA) SIS ASCA Advanced Satellite for Cosmology and Astrophysics X-ray Astronomy 1993e2001 ISAS (Japan) Clementine Lunar Mission Moon 1994 Strategic Defense Initiative Organization (SDIO), NASA (USA) WIND Solar Wind Observatory Solar Wind 1994epresent NASA, ESA, ISAS CELIAS/SEM SOHO Solar and Heliospheric Observatory Sun 1995epresent ESA, NASA PCA RXTE Rossi X-ray Timing Explorer X-ray Astronomy 1995e2012 NASA (USA) NEAR-Shoemaker Near Earth Asteroid Rendezvous Asteroid 433 Eros 1996e2001 NASA (USA) PIXIE POLAR Earth Magnetosphere Earth 1996e2008 NASA (USA) BeppoSAX Giuseppe Occhialini (Beppo) Satellite per Astronomia a raggi X X-ray Astronomy 1996e2003 Italy, Netherlands GOES-10 Weather satellite Earth 1997e2009 NOAA/NASA (USA) ACIS, HRC-I Chandra Chandrasekhar X-ray Observatory X-ray Astronomy 1999epresent NASA (USA) EPIC, RGS XMM-Newton X-ray Multi Mirror Mission X-ray Astronomy 1999epresent ESA (Continued) 1043 Description X-rays in the Solar System Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Instrument, Mission Author's personal copy 1044 TABLE 48.2 Satellites and Spacecraft Mentioned in the Text, Chronologically Sorted According to Launch Datedcont’d Category Date Agency, Country RHESSI Reuven Ramaty High Energy Solar Spectroscopic Imager Sun 2002epresent NASA (USA) XRS Hayabusa Asteroid Sample Return Mission Asteroid 25143 Itokawa 2003e2010 JAXA (Japan) D-CIXS SMART-1 Lunar Mission Moon 2003e2006 ESA EPS, XRS MESSENGER Mercury Surface, Space Environment, Geochemistry and Ranging Mercury 2004epresent NASA (USA) SWIFT Gamma-Ray Burst Mission Gamma-ray Astronomy 2004epresent NASA, with international participation Deep Impact (Extended) Comet Mission Comets Tempel 1 and Hartley 2 2005e2007, 2008e2013 NASA (USA) Suzaku Astro E2 X-ray Astronomy 2005epresent ISAS/JAXA (Japan) New Horizon Outer Solar System Pluto 2006epresent NASA (USA) XRT Hinode Solar Observations Sun 2006epresent JAXA (Japan), with international Participation X-Ray CCDs Kaguya Lunar Mission Moon 2007e2009 JAXA (Japan) X-Ray CCDs Chandrayaan-1 Lunar Mission Moon 2008e2009 ISRO (India) Astro-H Planned X-ray Astronomy w2014 JAXA, NASA MIXS BepiColombo Planned Mercury w2015 ESA/JAXA eROSITA SRG Planned X-ray Astronomy w2016 Germany, Russia Rover Selene-2 Planned Moon w2017 JAXA (Japan) CLASS, orbiter Chandrayaan-2 Planned Moon w2016e2017 ISRO (India) Athena Planned X-ray Astronomy w2028 ESA CCD, charge-coupled device; ESA, European Space Agency; ISAS; ISRO, Indian Space Research Organisation; JAXA, Japan Aerospace Exploration Agency; NASA, National Aeronautics and Space Administration. Exploring the Solar System Description PART | IX Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045 Instrument, Mission Author's personal copy Chapter | 48 X-rays in the Solar System ACKNOWLEDGMENTS A large part of this chapter is based on the review article by Bhardwaj et al. (2007a), which is a collective effort of several authors, and we deeply acknowledge all the authors of that paper. We also thank the entire solar system X-ray community whose works have led to this review. BIBLIOGRAPHY Bhardwaj, A. (2006). X-ray emission from Jupiter, Saturn, and Earth: a short review. In Advances in Geosciences (Vol. 3); (pp. 215e230). Singapore: World Scientific. Bhardwaj, A., Elsner, R. F., Waite, J. H., Jr., Gladstone, G. R., Cravens, T. E., & Ford, P. G. (2005a). Chandra Observation of an X-ray Flare at Saturn: Evidence for Direct Solar Control on Saturn’s Disk X-ray Emissions. Astrophysical Journal Letters, 624, L121eL124. Bhardwaj, A., Elsner, R. F., Waite, J. H., Jr., Gladstone, G. R., Cravens, T. E., & Ford, P. G. (2005b). The Discovery of Oxygen Ka X-ray Emission from the Rings of Saturn. Astrophysical Journal Letters, 627, L73eL76. Bhardwaj, A., Elsner, R. F., Gladstone, G. R., Cravens, T. E., Lisse, C. M., Dennerl, K., et al. (2007a). X-rays from solar system objects. Planetary and Space Science, 55, 1135e1189. Bhardwaj, A., Gladstone, G. R., Elsner, R. F., Østgaard, N., Waite, J. H., Jr., Cravens, T. E., Chang, S.-W., Majeed, T., & Metzger, A. E. (2007b). First Terrestrial Soft X-ray Auroral Observation by the Chandra X-ray Observatory. Journal of Atmospheric and Solar Terrestrial Physics, 67, 179. http://chandra.harvard.edu/ press/05_releases/press_122805.html. Bhardwaj, A., Elsner, R. F., Gladstone, G. R., Branduardi-Raymont, G., Dennerl, K., Lisse, C. M., Cravens, T. E., et al. (2009). X-ray Emission from Planets and comets: Relationship with Solar X-rays and Solar Wind. Advances in Geosciences, 15, 229e244. Bhardwaj, A., & Gladstone, G. R. (2000). Auroral emissions of the giant planets. Reviews of Geophysics, 38, 295e353. Bodewits, D., et al. (2007). Spectral Analysis of the Chandra Comet Survey. Astronomy and Astrophysics, 469, 1183e1195. Branduardi-Raymont, G., Bhardwaj, A., Elsner, R. F., Gladstone, G. R., Ramsay, G., Rodriguez, P., Soria, R., Waite, J. H., Jr., & Cravens, T. E. (2007). Astronomy and Astrophysics, 463, 761e774. Branduardi-Raymont, G., Bhardwaj, A., Elsner, R. F., & Rodriguez, P. (2010). X-rays from Saturn: a study with XMM-Newton and Chandra over the years 2002e05. Astronomy and Astrophysics, 510, A73. Christian, D. J., Bodewits, D., Lisse, C. M., Dennerl, K., Wolk, S. J., Hsieh, H., et al. (2010). Chandra observations of comets 8P/Tuttle and 17P/Holmes during solar minimum. The Astrophysical Journal Supplement Series, 187, 447e459. Dennerl, K. (2002). Astronomy and Astrophysics, 394, 1119e1128. Dennerl, K. (2008). Planetary Space Science, 56, 1414. Dennerl, K. (2010). Charge transfer reactions. Space Science Reviews, 157, 57e91. 1045 Dennerl, K., Burwitz, V., Englhauser, J., Lisse, C., & Wolk, S. (2002). Discovery of X-rays from Venus with Chandra. Astronomy and Astrophysics, 386, 319e330. Dennerl, K., Lisse, C. M., Bhardwaj, A., Burwitz, V., Englhauser, J., Gunell, H., et al. (2006). First observation of Mars with XMM-Newton. High resolution X-ray spectroscopy with RGS. Astronomy and Astrophysics, 451(2), 709e722. Dennerl, K., Lisse, C. M., Bhardwaj, A., Christian, D. J., Wolk, S. J., Bodewits, D., et al. (2012). Solar system X-rays from charge exchange processes. Astronomische Nachrichten, 333(4), 324e334. Elsner, R. F., Gladstone, G. R., Waite, J. H., Crary, F. J., Howell, R. R., Johnson, R. E., et al. (2002). Astrophysical Journal, 572, 1077e1082. Ewing, et al.. (2013). Astrophysical Journal, 763. article id. 66. Gladstone, G. R., Waite, J. H., Grodent, D., Lewis, W. S., Crary, F. J., Elsner, R. F., et al. (2002). Nature, 415, 1000. Kharchenko, V., Bhardwaj, A., Dalgarno, A., Schultz, D., & Stancil, P. (2008). Modeling spectra of the North and South Jovian X-ray auroras. Journal of Geophysical Research, 113, A08229. Krasnopolsky, V. A., Greenwood, J. B., & Stancil, P. C. (2004). X-ray and extreme ultraviolet emission from comets. Space Science Review, 113, 271e374. Lisse, C. M., et al. (1996). Science, 292, 1343e1348. Lisse, C. M., et al. (1997). Icarus, 141, 316e330. Lisse, C. M., et al. (2005). Chandra observations of Comet 2P/Encke 2003: first detection of a collisionally thin, fast solar wind charge exchange system. Astrophysical Journal, 635, 1329e1347. Lisse, C. M., et al. (2007). Chandra observations of Comet 9P/Tempel 1 during the deep impact campaign. Icarus, 190, 391e405. Lisse, C. M., Cravens, T. E., & Dennerl, K. (2004). X-ray and extreme ultraviolet emission from comets. In M. C. Festou, H. U. Keller, & H. A. Weaver (Eds.), Comet II (pp. 631e643). Tucson: Univ. Arizona Press. McCauley, P. I., Saar, S. H., Raymond, J. C., Ko, Y.-K., & Saint-Hilaire, P. (2013). Extreme-ultraviolet and X-ray observations of comet Lovejoy (C/2011 W3) in the lower corona. Astrophysical Journal, 768, 161e172. Østgaard, N., Stadsnes, J., Bjordal, J., Germany, G. A., Vondrak, R. R., Parks, G. K., et al. (2001). Journal of Geophysical Research, 106, 26081. Petrinec, S. M., McKenzie, D. L., Imhof, W. L., Mobilia, J., Chenette, D. L., et al. (2000). Journal of Atmospheric and Solar Terrestrial Physics, 62, 875e888. Snowden, S. L., Egger, R., Freyberg, M. J., McCammon, D., Plucinsky, P. P., Sanders, W. T., et al. (1997). Astrophysical Journal, 485, 125e135. Wargelin, B. J., Markevitch, M., Juda, M., Kharchenko, V., Edgar, R., Dalgarno, A., et al. (2004). Astrophysical Journal, 607, 596e610. Wegmann, R., & Dennerl, K. (2005). X-ray tomography of a cometary bow shock. Astronomy and Astrophysics, 430, L33eL36. Wolk, S. J., Lisse, C. M., Bodewits, D., Christian, D. J., & Dennerl, K. (2009). Chandra’s close encounter with the disintegrating comets 73P/2006 (Schwassmann-Wachmann 3) fragment B and C/1999 S4 (LINEAR). Astrophysical Journal, 694, 1293e1308. Encyclopedia of the Solar System, Third Edition, 2014, 1019e1045