Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

An Advanced, Low-Cost, GPS-Based Animal Tracking System: Rangeland Ecology & Management May 2006

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/43254578

An Advanced, Low-Cost, GPS-Based Animal Tracking System

Article  in  Rangeland Ecology & Management · May 2006


DOI: 10.2111/05-162R.1 · Source: OAI

CITATIONS READS

81 9,865

9 authors, including:

Patrick E. Clark
United States Department of Agriculture
92 PUBLICATIONS   1,756 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Monitoring herd's mobility using animal GPS collars View project

Prescribed Fire Project View project

All content following this page was uploaded by Patrick E. Clark on 02 January 2018.

The user has requested enhancement of the downloaded file.


Rangeland Ecol Manage 59:334–340 | May 2006

Technical Note
An Advanced, Low-Cost, GPS-Based Animal Tracking System
Patrick E. Clark,1 Douglas E. Johnson,2 Mark A. Kniep,3 Phillip Jermann,4 Brad Huttash,3
Andrew Wood,5 Michael Johnson,6 Craig McGillivan,7 and Kevin Titus8
Authors are 1Range Scientist and 4Biological Science Technician, US Department of Agriculture–Agricultural Research Service, Boise,
ID 83712; 2Professor and 6Software Programmer, Department of Rangeland Ecology and Management, Oregon State University,
Corvallis, OR 97331; 3Electrical Engineers, Micron Technology, Inc, Boise, ID 83716; 5Electrical Engineer, Hewlett-Packard Company,
Boise, ID 83714; and 7current and 8former electrical engineering students, Boise State University, Boise, ID 83725.

Abstract
An improved global positioning system (GPS)–based animal tracking system is needed to meet quickly evolving demands of
ecological research, range livestock production, and natural resource management. Commercially available tracking systems lack
the data storage capacity needed to frequently collect animal location data (e.g., 15-minute intervals or less) over long-term
deployment periods (e.g., 1 year or more). Some commercial systems have remote data–download capabilities, reducing the need
to recapture tagged animals for data retrieval, but these systems download data via satellite (Argos), global system for mobile
communications (GSM) cellular telephone, or telemetry radio frequencies. Satellite systems are excessively expensive, and GSM
cellular coverage is extremely limited within the United States. Radio-based systems use narrow-band very-high– or ultra-high
frequencies requiring the user to obtain frequency allocations. None of these existing systems were designed to provide continual,
real-time data access. The Clark GPS Animal Tracking System (Clark ATS) was developed to meet the evolving demands of
animal ethologists, ecologists, natural resource managers, and livestock producers. The Clark ATS uses memory-card technology
for expandable data storage from 16 megabytes to 8 gigabytes. Remote data downloading and program uploading is
accomplished using spread-spectrum radio transceivers, which do not require narrow-band radio frequency allocations. These
radios also transmit, at a user-defined time interval, a real-time, GPS-location beacon to any Clark ATS base station within range
(about 24 km or 15 miles line of sight). Advances incorporated into the Clark ATS make it possible to evaluate animal behavior at
very fine spatial- and temporal-resolution over long periods of time. The real-time monitoring provided by this system enables
researchers to accurately examine animal distribution and activity responses to acute, short-term disturbances relative to longer-
term behavioral patterns. The Clark ATS also provides a huge time- and cost-savings to researchers and natural resource
managers attempting to relocate a tagged animal in the field for direct observation or other operations.

Resumen
Se necesita un sistema mejorado de rastreo de animales basado en GPS para satisfacer las crecientes demandas de investigación
ecológica, producción de ganado en pastizales y el manejo de los recursos naturales. Los sistemas de rastreo comerciales
disponibles carecen de la capacidad de almacenaje de datos necesaria para colectar frecuentemente la localización del animal
(por ejemplo, a intervalos de 15 minutos o menos) en un perı́odo largo de tiempo (un año o mas). Algunos sistemas comerciales
tienen capacidad de descargar datos a larga distancia, reduciendo la necesidad de recapturar los animales marcados para
recuperar los datos, pero estos sistemas descargan los datos vı́a satélite (Argos), a través de telefonı́a celular GSM o de
radiofrecuencias de telemetrı́a. Los sistemas de satélite son excesivamente caros y la cobertura de la telefonı́a celular GSM es
extremadamente limitada dentro de Estados Unidos de América. Los sistemas basados en radio usan bandas estrechas de
frecuencia VHF o UHF, requiriendo que el usuario obtenga asignaciones de frecuencia. Ninguno de los sistemas existentes
fueron diseñados para proveer un acceso continuo en tiempo real. El Sistema de Rastreo de Animales Clark GPS (Clark ATS) fue
desarrollado para satisfacer las demandas de los etólogos animal, ecólogos, manejadores de recursos naturales y productores de
ganado. El Clark ATS utiliza tecnologı́a de tarjeta de memoria para expandir la capacidad de almacenaje de datos de 16
megabytes a 8 gigabytes. La descarga remota de datos y la carga del programa se logra usando radio transcriptores de espectro
amplio, que no requieren la asignación de frecuencias de radio de banda angosta. Estos radios también transmiten, a un
intervalo de tiempo definido por el usuario, en tiempo real, la localización de la baliza de GPS a cualquier base de Clark ATS
dentro del rango ( aproximadamente 24 km o 15 millas en lı́nea recta). Los avances incorporados al Clark ATS hacen posible
evaluar el comportamiento animal a una resolución espacial y temporal muy fina por largos perı́odos de tiempo. El monitoreo en
tiempo real suministrado por este sistema permite a los investigadores examinar acertadamente la distribución de los animales y
las actividades en respuesta a disturbios severos a corto plazo en relación a los patrones de comportamiento a largo plazo. El
Clark ATS también proporciona grandes ahorros de tiempo y costos a los investigadores y manejadores de recursos naturales
que intentan relocalizar en el campo a los animales marcados para realizar observaciones directas u otras operaciones.
Key Words: activity budgets, animal behavior, global positioning system, habitat use, real-time, telemetry tracking
This research was funded by the US Department of Agriculture–Agricultural Research Service.
Mention of manufacturer, trademark name, or proprietary product does not constitute endorsement by the US Department of Agriculture, Oregon State University, or Boise State
University and does not imply their approval to the exclusion of other products that may also be suitable.
Correspondence: Patrick E. Clark, Northwest Watershed Research Center, USDA–ARS, 800 Park Boulevard, Suite 105, Boise, ID 83712-7716. Email: pclark@nwrc.ars.usda.gov
Manuscript received 26 September 2005; manuscript accepted 12 March 2006.

334 RANGELAND ECOLOGY & MANAGEMENT 59(3) May 2006


INTRODUCTION spectrum radio to a hand-held base station capable of receiving
and displaying collar locations on a digital map; 4) collar units
Study of animal ecology using telemetry tracking systems began having a large (up to 8 gigabyte), user-expandable, on-board
in the late 1950s and early 1960s (Le Munyan et al. 1959; data storage capacity; and 5) collar components having very
Eliassen 1960; Marshall et al. 1962; Cochran and Lord 1963; low power demand (mean consumption , 100 mW) and
Mech et al. 1965) using collars or tags emitting very-high fre- batteries with very high capacity (19 AH D-cells). The relatively
quency (VHF) radio-signal pulses. Intensive monitoring of widely low cost of the Clark GPS Animal Tracking System (Clark
roaming animals with VHF systems, however, was costly, time- ATS) would help the user to economically deploy the system on
consuming, and often posed risks to personnel safety. With the an adequate sample size of animals, which may not have been
launch of the Nimbus 3 satellite (Kenward 1987) and, later, the possible using a more expensive, commercial tracking system.
Argos system (Fancy et al. 1988), it became possible to auto- The capabilities of the Clark ATS would also allow the user to
matically collect and transmit location data from widely roaming deploy the system for up to 3 weeks at a data-capture rate of once
or migrating animals (e.g., polar bear and caribou) using sat- every minute without the need to recollect and service the collar.
ellite communication technology (for examples, see White and For longer-term deployments, the user could configure the
Garrott 1990). The positional accuracy of these location data, system to acquire data at 15-minute intervals for up to 1 year
however, was quite coarse (6 300 m) (Britten et al. 1999), thus without service.
negating their use for habitat-selection studies. Deployment of
the NAVSTAR (Navigation Geographic Positioning System
[GPS]), declared fully operational in 1995, enabled develop- MATERIALS AND METHODS
ment of animal tracking systems with unprecedented positional
accuracy (6 5 m) (e.g., Rodgers et al. 1996). These GPS-based Clark GPS Animal Tracking System
The Clark ATS consists of a GPS tracking collar (Fig. 1) and
tracking systems allowed evaluation of animal movement and
a hand-held, mobile base station (Fig. 2). The tracking collar
habitat selection at very fine spatial resolution.
collects and stores GPS-fix information including collar loca-
Despite these technological advances, however, telemetry
tion (latitude and longitude), date and time (Greenwich mean),
tracking systems have not kept pace with the evolving demands
and parameters indicating fix quality (e.g., dilution of precision
of ecological research. Costs of commercial GPS tracking
and number satellites used) on a removable memory card
collars severely limit the sample size (i.e., individual animals)
(CompactFlash) contained within the collar. Raw satellite data
and statistical power that researchers have available for animal
(e.g., carrier phase, pseudorange, and Doppler measurements)
ethology and ecology studies. Commercial GPS collars also have
acquired and used by the GPS receiver to calculate a GPS fix are
data-storage constraints that hinder collecting location data
also stored on the memory card allowing postdifferential
with high-temporal frequency (e.g., every 15 minutes or less)
correction of the GPS locations for improved spatial accuracy.
over long deployment periods (up to 1 year or more). Con-
Use of the removable memory card for data storage allows the
sequently, intensive investigations of habitat selection, short- and
user to quickly remove data from the collar while still in the
long-range movements, and other animal behaviors cannot be field. Changing the data storage capacity of the collars for
conducted over multiple seasons or years without frequently different applications is as simple as inserting a memory card
downloading and erasing data from the collar memory. Some with a different capacity.
commercial systems have remote data–download capabilities, The collar and hand-held base-station units are wirelessly
reducing the need to frequently recapture collared animals for linked by radio transceivers. These radio transceivers operate in
data retrieval. These systems, however, download data via the spread-spectrum radio bandwidth (902–928 MHz). Spread-
satellite (e.g., Schwartz and Arthur 1999), global system for spectrum radio transceivers can frequency-hop throughout this
mobile communications (GSM) cellular telephones, or teleme- wide bandwidth looking for an unused frequency for data
try radio frequencies (Rodgers et al. 1996). Satellite communi- transmission (Ziemer et al. 1995). Frequency-hopping helps
cation is very expensive. The GSM cellular coverage in the maximize successful and timely data transmissions and elimi-
wildlands of North America is extremely limited. Radio-based nates conflicts, even when many different users are in the same
systems use narrow-band VHF or ultra-high frequency (UHF) coverage area. Besides collecting and storing raw GPS-fix in-
requiring the user to obtain frequency allocations, which dictate formation described above, the collar also collects GPS-fix data
where and when these systems can be used. None of these that have been differentially corrected, in real-time, using the
existing systems were designed to provide the continuous, real- wide-area augmentation system (WAAS). The collar then
time data access that is often desired by ecologists, animal transmits these WAAS-corrected data to a Clark ATS base
ethologists, and other researchers. station via the radio link. In this way, the collar is transmitting
The objective of this research was to develop a robust GPS- a differentially corrected GPS location beacon to any base
based, real-time animal tracking system with the following station within range, allowing real-time tracking of the collar
attributes: 1) individual collars and mobile base-station units location. Collar location data received by the base station are
costing less than $1 000 each (US dollars in 2006); 2) spread- stored in an electronic database on the base station for later use.
spectrum radio frequency communication between collar and These location data may also be plotted in real-time on digital
base station allowing remote uploading of programming and topographic maps and orthophotographic images using base-
downloading of postdifferentially correctable GPS data; 3) real- station software. The radio transceivers also make it possible to
time collar tracking capabilities where GPS data describing the remotely download data stored on the collars and to remotely
current location of the collar would be transmitted via spread- upload new programming to the collar from the base station.

59(3) May 2006 335


Figure 1. The Clark global positioning system Animal Tracking System
collar, which can be sized to fit most livestock and large, terrestrial
wildlife species, including A, Cattle and elk (990 g), and B, domestic or
wild sheep, goats, deer, and wolves (720 g). The sheep/wolf-sized
Figure 2. The Clark global positioning system Animal Tracking System
collars use shorter belting lengths and a smaller enclosure (box) to
hand-held base station including personal digital assistant, spread-
reduce bulk and weight.
spectrum radio transceiver, and waterproof case.

Details concerning the Clark ATS are also presented at the receiver, an active GPS patch antenna, a spread-spectrum radio
website associated with this paper: http://clark.nwrc.ars.usda. transceiver, and an omnidirectional radio antenna (Table 1).
gov/collars/index.php. Plans and instructions for constructing The collar is powered by either 1 or 2 high-capacity, D-cell
the Clark ATS are available from the corresponding author batteries, depending on the planned length of deployment and
upon request. weight considerations. Other collar components include a printed
circuit board (PCB), a back-up battery, a regulator, capacitors,
Collar Components resistors, and various electrical connectors. A bill of materials
Major components of the GPS telemetry collar include a single- list for the components needed to construct the collar is avail-
board computer with a CompactFlash memory-card slot, a GPS able at the Clark ATS website mentioned above.

336 Rangeland Ecology & Management


Hand-Held Base-Station Components Table 1. Electronic, environmental, and physical parameter specifica-
Major components of the hand-held, mobile base station tions for the major components included in the Clark global positioning
include a Pocket-PC personal digital assistant (PDA), spread- system (GPS) Animal Tracking System (Clark ATS) collars and base
spectrum radio transceiver, omnidirectional antenna, and stations.1,2
waterproof PDA case (Fig. 2; Table 1). The radio transceiver
Component Parameter Specification Notes
and PDA components of the base station were each powered by
separate rechargeable battery packs carried outside the PDA Computer Voltage, power supply 3.3 V Single-board
case in a belt pack or backpack. Other base-station components Power consumption, max. 412 mW CompactFlash
included a level-shifter, regulator, rocker switch, antenna- write
adapter cable, PDA synchronization/serial-adaptor cable, ca- Power consumption, 20 lW Suspend mode
pacitors, and resistors. As with the collar, a bill of materials for suspend
the base station is listed at the Clark ATS website. Temperature range 408 to 858C —
Interface, RS-232 1 channel 115 200 baud
max.
Collar Construction Interface, UART serial 7 channels 115 200 baud
What follows is a very general sequence of steps involved in max.
assembling the electronic components and constructing hard- Memory, nonvolatile 1 MB Flash memory
ware components of the collar. A very detailed, step-by-step set Size, physical dimensions 35.5 3 63.5 3 W, L, and H
of instructions and illustrations for use in constructing the
17.8 mm
tracking collar can be accessed at the Clark ATS website.
GPS receiver Voltage, power supply 3.0 V 16-channel
Nearly all of the electronic components of the collar are
receiver
mounted and electrically connected via the PCB (Fig. 3). The
Power consumption, max. 375 mW —
location where each component should be mounted is marked
Temperature range 408 to 858C —
with white silk-screening on the PCB. The soldering involved in
mounting and connecting components is simple and can be Interface, UART serial 2 channels 115 200 baud
accomplished by a person with little or no previous experience. max.
The first step is to mount the GPS receiver, GPS antenna Fix acquisition time , 3.5 s Hot-start mode
connector, and other components required to supply and Accuracy, horizontal3 6.3 m 95% CEP4
regulate power and to communicate with the GPS. Power is Size, physical dimensions 25.4 3 25.4 3 W, L, and H
then applied to the PCB to confirm the GPS is operating 3.0 mm
properly. The GPS functions, such as power-saving modes, may GPS antenna Voltage, power supply 3.0 V Active antenna
be configured at this time as well. Next, the single-board Power consumption, max. 63 mW —
computer and other components providing backup power and Amplifier gain 27 dB —
serial communication to the computer are soldered to the PCB. Size, physical dimensions 40.0 3 48.0 3 W, L, and H
A memory card containing the collar test program and 13.0 mm
parameter file (see Collar Programming section below) is Radio Voltage, power supply 3.3 V Spread spectrum
then inserted in the Compact Flash card-slot of the single- transceiver Power consumption, max. 743 mW —
board computer and power is applied to the PCB once again. Temperature range 408 to 808C —
Assuming the GPS antenna is connected and has a clear view of Interface, UART serial 1 channel 115 200 baud
the sky, the GPS should collect GPS-fix information and max.
transmit it to the single-board computer where the data are Transmission range , 24 km Line of sight
stored on the memory card. With power disconnected, the card Size, physical dimensions 41.9 3 48.1 3 W, L, and H
can then be removed and its contents reviewed using a memory
5.1 mm
card reader to confirm proper function of both the GPS and
Radio, Gain 2.14 dBi —
single-board computer. Finally, the radio transceiver is
antenna Size, physical dimensions 20.0 3 137.0 3 W, L, and H
soldered to the PCB. If it becomes necessary to remove the
1.0 mm
computer from the PCB (e.g., due to computer malfunction or
Battery Voltage 3.6 V D-cell
damage), this operation requires desoldering of both the radio
transceiver and the computer. Consequently, delaying mounting Capacity5 19 AH —
of the radio until the GPS and computer have been fully tested Size, physical dimensions 35.0 3 62.5 mm Diam and H
1
can save a lot of frustration. Once mounted, the radio is Bill of materials lists and detailed instructions for constructing the Clark ATS collars and
base stations can be acquired from the corresponding author or can be downloaded from
configured and tested for successful communication with
the following website: http://clark.nwrc.ars.usda.gov/collars/.
a radio configured to act as a base station (see Base-Station 2
Max indicates maximum; UART, universal asynchronous receiver/transmitter; MB, mega-
Construction section below). byte; W, width; L, length; H, height; CEP, circular error probability; dB, decibel; dBi, gain in
The electronic components of the collar are highly suscep- decibels referenced to an isotropic radiator; AH, ampere hours; and diam, diameter.
3
Calculated using uncorrected (non–wide-area augmentation system) location data collected
tible to damage by moisture and impact and must be housed in by GPS receivers contained within 5 deployment-ready Clark ATS collars.
a tough, waterproof enclosure. A collar made of conveyor 4
CEP is the radius of a circle (horizontal) that is centered at the GPS antenna’s true position,
belting material is used to attach the enclosure and its contents and contains 95% of the GPS locations.
5
to an animal during deployment. The following is a general list Actual battery capacity depends on the resistance load, current demand, and minimum
voltage-level requirements of the application.
of steps for constructing and assembling the collar belting and

59(3) May 2006 337


Figure 3. Schematic illustrating the placement and electrical connections for all the electronic components of the Clark global positioning system
Animal Tracking System collar.

enclosure. First, 2 equal lengths of conveyor belting are lam- configured to operate as a base station (i.e., a server). The
inated together with a double-row of stitching around the edge radio and other components and wiring used to power the
of the belting. Lengths of the belting will differ depending on system and translate the serial output of the radio are mounted
the neck circumference of the animal species of interest (Figs. on a small PCB. The PCB and its wiring harness are then
1A and 1B). Next, neck size–adjustment slots or holes are cut secured inside the waterproof PDA case. Finally, the PDA is
into the belting. The GPS and radio antennas and cables are then secured inside the PDA case, connected to the PCB via the
threaded between the 2 layers of belting. The belting is attached wiring harness, and the case is then closed and sealed. During
to the sides of the enclosure using mounting brackets. The GPS deployment, the radio antenna and the power supply cables for
and radio antenna cables ends are then inserted through a cable the radio and PDA are connected to the base-station PCB via
gland in the wall of the enclosure and plugged into their respec- waterproof, bulkhead connectors passing through the wall of
tive connectors on the PCB. Finally, a battery pack consisting of the PDA case (Fig. 2).
2 D-cell batteries and associated capacitors is secured to the
inside of the enclosure lid with silicon adhesive. Before testing
Collar Programming
or deploying the collar, the power cable from the battery pack
The duty cycle or timing of operations for all major compo-
must be connected to the PCB, and the enclosure is then closed
nents of the tracking collar is controlled by a program placed in
and sealed. A bill of materials and detailed instructions needed
the nonvolatile memory (1 megabyte [MB]) of the single-board
for constructing the collar belting and enclosure are available
computer. This program uses a text file of control parameters,
on the system website.
which is edited by the user and loaded on the collar’s memory
card before deployment. Parameters in this text file specify the
Base-Station Construction date/time when the program should initiate data collection thus
As with the collar construction, a detailed set of instructions allowing collar deployments with a delayed start. The text file
for constructing the hand-held base station is available at the also directs the program in how often GPS data should be
Clark ATS website. The following is a general description of collected and how frequently information for the most recent
how the electronics and hardware of the base station are GPS fix (i.e., a GPS-location beacon) should be transmitted by
constructed and assembled. First, the radio transceiver is the radio transceiver to the base station. Data included in

338 Rangeland Ecology & Management


these radio transmissions are labeled with a collar identifica- mature cows that were using a high-elevation (, 2 438 m),
tion value defined by the user as a parameter in the text file. sagebrush-steppe rangeland in central Idaho during July–August
Finally, when directed by radio queries from the base station, 2005. The collars were recovered in September 2005, data were
the collar program allows the user to remotely download data downloaded, and the battery life for each collar was determined
from the collar and remotely upload new programming. The by contrasting the start date-time and end date-time (i.e., bat-
collar program, source code, and an example of the parameter tery exhaustion date-time) values within the data record.
text file are available for downloading at the Clark ATS
website.
RESULTS AND DISCUSSION
Base-Station Programming The Clark GPS Animal Tracking System includes GPS-based
There are 3 software programs used by the hand-held base sta- tracking collars and a hand-held, mobile base station. Major
tion. The first program receives incoming GPS-location beacon components of the tracking collars include a single-board com-
transmissions and stores these data in a database on the PDA of puter with a memory-card slot (CompactFlash), a 16-channel
the base station. Incoming GPS-location beacon data are plotted GPS receiver, and a spread-spectrum radio transceiver. All these
in real-time on digital topographic maps or orthophotographic components are mounted and electrically connected on a single
images using the second program. The third program allows the PCB. These collar electronics are then contained within a water-
base station to send queries to collars requesting their stored GPS tight housing, which is mounted on a collar that can be sized
data. This program allows the user to remotely install new for different livestock or wildlife species. The tracking collar
programming on a collar’s computer. All 3 base-station pro- collects and stores time-stamped data for the geographic loca-
grams, their source code, and instructions for their installation tion of the collar, which can be downloaded to a distant base
and use are available for downloading at the system website. station using the radio transceiver.
At this writing, the cost for a Clark ATS collar was about
Collar Testing $840, excluding costs for about 2 hours of labor required for
The horizontal error or circular error probability (CEP) of GPS construction. Commercial GPS tracking collars with remote
location data collected by the Clark ATS collars was evaluated data–access capabilities typically cost about 3 to 4 times more
using 4 collars placed, equidistant from each other, at 1 m in per collar than the Clark ATS collar. This cost-savings allows
height, around the circumference of a circle that had a radius of the user to substantially increase the sample size and statistical
5 m and was centered over a geodetic survey point (point power available in the experimental design of animal ethology
identifier ¼ OH1352; NGS 1992). The collars were configured and ecology studies.
to collect data at 1-minute intervals and were deployed for The software programming capabilities available within
21 days, thus yielding more than 29 000 locations per collar. A the single-board computer of the Clark ATS collar allow the
95% CEP value for each collar was determined by finding the user to implement complex duty-cycle/power-management
radius of a circle that was centered over the true location of the schedules (such as altering the data-collection rate in response
collar (i.e., as determined by a 5-m offset from the geodetic to animal-activity levels). The computer also facilitates future
survey point location) and which contained 95% of the integration and control of other sensor technology (e.g.,
locations. A mean 95% CEP was calculated based on the accelerometers, digital cameras, microphones, or pulse oxim-
CEP values from all 4 collars. eters and other animal physiology sensors) within the Clark
Range-testing of the radio transceivers in the Clark ATS was ATS collar.
conducted using 4 collars and 2 base stations. Of the 4 test The 16-channel GPS receiver allows the Clark ATS collar to
collars, 2 collars were built with the radio antenna entirely take full advantage of all GPS satellites in view at any one time
enclosed between the layers of collar belting, whereas the re- (i.e., 1 to 12 satellites), while still leaving at least 4 channels
maining 2 collars were built with the antenna fully extended open for other functions, such as acquiring WAAS data for real-
outside of the collar belting. These 2-antenna configurations time, differential-correction of collar location data. Field testing
allowed testing of collar belt effects on radio-transmission range. indicated the mean 95% CEP for uncorrected GPS location
During testing, the base stations were established on an open data collected with Clark ATS collars was 6.3 m (Table 1). Clark
ridgetop where a clear, line-of-sight distance of 24 km was ATS collars collecting WAAS-corrected data yielded a mean
available. Only 1 base station was operated at any one time to 95% CEP of 2.7 m.
avoid interference problems. The test collars were trans- The spread-spectrum radio transceiver in the Clark ATS
ported away from the base station in a vehicle and were collar has several advantages over track systems using GSM
periodically removed from the vehicle and allowed to at- cellular telephone, satellite, or narrow-band telemetry radio fre-
tempt to transmit their GPS location data to the base station quencies for telemetering data between collars and base stations.
from increasingly longer distances. Once a maximum distance Currently, GSM cellular telephone coverage is very limited in the
for successful data transmission was reached using 1 base station, wildlands of North America and other regions of the world.
the second base station was employed to confirm these findings. Tracking systems relying on GSM service, consequently, have
Battery life under intensive, short-duration collar deploy- very limited utility for rangelands and other wildland applica-
ments was tested using 5 Clark ATS collars deployed on beef tions. Use of commercial satellite communications is expensive
cattle. The collars were configured to collect and store GPS data and may have very restrictive data-throughput levels (e.g., 32
at 1-minute intervals and to attempt to transmit the most current bytes per transmission). A tracking system relying on Service
location data every 15 minutes. The collars were deployed on Argos for data retrieval would incur a cost of about $140 per

59(3) May 2006 339


month per collar transmitter for a typical deployment (Service with spread-spectrum radio frequency communication between
Argos 2005). This service package would, however, only include collar and base station, real-time collar tracking capabilities,
2 data-transmission time slots per day with a maximum of 32 large (up to 8 GB) user-expandable on-board data storage, and
bytes of data per transmission. Tracking systems using narrow- very low power demand (mean consumption , 100 mW) and
band radios operate on frequencies reserved by the Federal with very high battery capacity (19 AH D-cells) was successfully
Communications Commission (FCC) for wildlife telemetry. met. These attributes and capabilities of the Clark ATS should
Users of these tracking systems must obtain a narrow-band allow the user to economically deploy the system on an adequate
frequency allocation for each radio before deployment. These sample of animals for up to 3 weeks at a data capture rate of once
frequency allocations are often time-consuming to acquire and every minute without the need to re-collect and service the collar.
limit radio use to specific study areas with finite study duration. For longer-term deployments, the user could configure the system
The FCC has authorized spread-spectrum radio transceivers to to acquire data at 15-minute intervals for up to 1 year without
operate throughout the 902–928 MHz bandwidth; thus fre- service. The real-time tracking capabilities of the Clark ATS
quency allocations are not required. With spread-spectrum should enable researchers to accurately examine animal distri-
technology, a base-station radio transceiver (Server) and a collar bution and activity responses to acute, short-term disturbances
radio transceiver (Client) must correctly identify each other and relative to longer-term behavioral patterns. Real-time GPS
coordinate communication while hopping between frequencies. tracking may also provide time- and cost-savings to researchers
These requirements not only ensure successful data transmission and natural resource managers attempting to relocate a tagged
despite heavy radio traffic and electronically noisy environments animal in the field for direct observation or other operations.
but also essentially encrypt and protect the transmitted data
from unauthorized access.
Range-testing of spread-spectrum radio transceivers con-
ACKNOWLEDGMENTS
tained in the Clark ATS revealed that collars built with the
The authors wish to thank S. Hardegree and several anonymous reviewers
radio antenna fully extended outside of the collar belting were
for their critical review and comments on drafts of this manuscript.
capable of successfully transmitting GPS location data packets
to a base station over 24 km away (line-of-sight). Collars with
the antenna fully enclosed between the 2 layers of the collar LITERATURE CITED
belting were capable of successful transmissions out to 8 km. In
cases where it is critical to enclose the antenna within the collar BRITTEN, M. W., P. L. KENNEDY, AND S. AMBROSE. 1999. Performance and accuracy
belting to prevent antenna damage, such as in predator-tracking evaluation of small satellite transmitters. Journal of Wildlife Management
applications, the user should expect a similar reduction in trans- 63:1349–1358.
mission range. COCHRAN, W. W., AND R. D. LORD. 1963. A radio-tracking system for wild animals.
The memory-card data storage used in the Clark ATS collars Journal of Wildlife Management 27:9–24.
ELIASSEN, E. 1960. A method for measuring the heart rate and stroke/pulse
provides a large increase in capacity over commercially avail-
pressures of birds in normal flight. Arbok Universitet Bergen, Matematisk
able GPS-tracking collars, which typically can store no more
Naturvitenskapelig 12:1–22.
than 10 000 differentially correctable collar locations. At this FANCY, S. G., L. F. PANK, D. C. DOUGLAS, C. H. CURBY, G. W. GARNER, S. C. AMSTRUP, AND
writing, the capacity of CompactFlash cards ranged from W. L. REGELIN. 1988. Satellite telemetry: a new tool for wildlife research and
16 MB to 8 gigabytes (GB). In the Clark ATS, a 64-MB card management. Washington, DC: US Fish and Wildlife Service. Resource
is capable of storing about 124 872 GPS locations (postdiffer- Publication 172. 54 p.
entially correctable) or the equivalent of about 88 days of data KENWARD, R. E. 1987. Wildlife radio tagging: Equipment, field technique and data
collected at 1-minute intervals. About 2 years of 30-second data analysis. London, UK: Academic Press. 222 p.
could be collected using a 1-GB card. LE MUNYAN, C. D., W. WHITE, E. NYBERT, AND J. J. CHRISTIAN. 1959. Design of
The Clark ATS collars have a relatively high battery-life ex- a miniature radio transmitter for use in animal studies. Journal of Wildlife
pectancy because they are constructed using components with Management 23:561–565.
MARSHALL, W. H., G. W. GULLION, AND R. G. SCHWAB. 1962. Early summer activities of
very low power requirements (see Table 1) and are powered
porcupines as determined by radio-positioning techniques. Journal of Wildlife
with high-capacity batteries. Battery life under intensive collar Management 26:75–79.
deployments was tested using Clark ATS collars deployed on MECH, L. D., V. B. KEUCHLE, D. W. WARNER, AND J. R. TESTER. 1965. A collar for
beef cattle. Collars configured to collect and store GPS loca- attaching radio transmitters on rabbits, hares, and raccoons. Journal of
tion data at 1-minute intervals exhausted the useable capacity Wildlife Management 29:898–902.
(11 842 mAH) of the battery (3.6 V) powering the single-board [NGS] NATIONAL GEODETIC SURVEY. 1992. NGS data sheet. Designation: chalky, PID:
computer and GPS receiver after about 24 days (563.4 hours), OH1352, marker: triangulation station disk. Silver Spring, MD: National
thus yielding a mean, overall power consumption of 76.5 mW. Oceanic and Atmospheric Administration. 3 p.
Despite this relatively low power consumption, however, battery- RODGERS, A. R., R. S. REMPEL, AND K. F. ABRAHAM. 1996. A GPS-based telemetry
life expectancy tends to be the most limiting factor affecting system. Wildlife Society Bulletin 24:559–566.
SCHWARTZ, C. C., AND S. M. ARTHUR. 1999. Radio-tracking large wilderness animals:
how long the Clark ATS can be successfully deployed.
integration of GPS and Argos technology. Ursus 11:261–274.
SERVICE ARGOS. 2005. Service Argos catalog 2005 (USA JTA). Largo, MD: Service
Argos, Inc. 7 p.
MANAGEMENT IMPLICATIONS WHITE, G. C., AND R. A. GARROT. 1990. Analysis of wildlife radio-tracking data. San
Diego, CA: Academic Press. 383 p.
The objective of this research to develop a GPS-based animal ZIEMER, R. E., R. L. PETERSON, AND D. E. BORTH. 1995. Introduction to spread
tracking system (Clark ATS), costing less than $1 000 per unit, spectrum communications. Upper Saddle River, NJ: Prentice Hall. 700 p.

340 Rangeland Ecology & Management

View publication stats

You might also like