Login | Contact Us |
Bighorn Sheep Collaring
William B. Sloan, National Park Service
Widespread population declines and local extinction during the past century eliminated bighorn sheep (Ovis canadensis) from most of their historic range in the western United States. When the Maze District was added to Canyonlands National Park in 1971, the native bighorn sheep herd had been extirpated from the area. In January 1982, 23 bighorn were translocated from the Island in the Sky District's herd and released on the northeast side of the Maze District. Population numbers remain low, and impacts from increasing recreational use, long-term drought, and other environmental factors remain unknown.
The Maze District of Canyonlands National Park covers 73,000 acres to the west of the Green and Colorado Rivers and is the most remote district of the park. Monitoring the sheep population there on the ground has always been difficult due to the extreme remoteness of the area and to logistical constraints. In addition, monitoring from the air has been extremely difficult because of the rugged, convoluted nature of the terrain. For these reasons, the Maze population of bighorn has received less attention than the other populations in the park.
Sheep capture and collaring was accomplished by hiring a contract crew and helicopter. The collars, set to collect and record GPS location information four times per day, were attached to captured sheep in the Maze District in January 2008. Because data storage capacity for each collar was approximately 100 days, downloads were attempted once every four weeks for ten months. At the time of capture, blood draws and throat swabs were performed on each animal to facilitate testing for disease.
Attempts to download the data from overflights commenced in March. Persistent equipment failure triggered an adjustment in our methods. To improve data reception, we purchased the recommended three-element antenna, but that did little to help. The receiver continued to experience problems. However, for the purposes of VHF reception, the receiver was fully functional.
Although attempts to download the GPS data were unsuccessful from the overflights, each collared animal's VHF signal transmitted on a different frequency, allowing us to determine the location of each individual. This location information was layered with existing GIS themes to show seasonal movements, habitat utilization, overlap with recreational users, lambing areas, movement corridors, and potential barriers to movement.
At the end of this study, the collars were left on the sheep. It is anticipated that data from the collars will provide some valuable information for another two or three years, as determined by battery life. That data will be incorporated into the long-term monitoring program conducted by the National Park Service.
The interpretation of these results show the animals to be mostly disease free, with the exception of PI-3. Of particular interest is the fact that animal #250 had elevated PI-3 titer levels, indicating recent infection, and #250 died less than three months after being captured. PI-3 has been problematic in this, and many other, sheep populations in the past, eventually leading to pasteurellosis, pneumonia and an all-age die-off of many animals.
This project was needed to test the feasibility of using sophisticated GPS technology to provide essential data on bighorn sheep. To make that assessment, both GPS signal acquisition and multipath errors needed to be evaluated. A GPS satellite signal must first be received in order to record location data. For that data to be accurate, it cannot be distorted by bouncing off cliff walls or uneven terrain.
The second research question was determining the extent of multipath errors in this rugged, vertical environment. GPS signals bouncing off cliff walls can distort the timing of the signals creating erroneous location information. As noted above in the results, GPS collar #098 had less than 10% of its recorded locational data in the right place, and slightly less than 20% of collar #250's locations were accurate. Based upon this data from both collars, the conclusion can be drawn that the effect of multipath errors upon the proper location is extreme with these particular collars. However, on the positive side, approximately 23 and 53 locations, respectively, in three months were in the right location, giving more locations than either overflights or ground work would yield with traditional collars.
Despite the failure of the GPS collars and the touted high speed data transmissions, numerous positives have come from this project. The GPS data that was manually downloaded, valuable information needed in determining the feasibility of using this technology, did convincingly demonstrate the lack of GPS signal acquisition and an abundance of multipath errors, demonstrating the infeasibility of using this particular technology in the Maze District.
A greater understanding of the habitat and areas utilized by the Maze District sheep has been acquired. Overflights have shown movements and ranges of the animals. Ground observations have also shown this, as well as allowing the visual observation of the animals to determine demographics, lamb survival, habitat utilization and health.
If others of the GPS collars, when retrieved, have this percentage of correct GPS locations, much useful locational data will have been gleaned. It is anticipated, that these collars will continue to function for two or three more years, and additional direct, and limited GPS, data will be acquired and be of much value.
Based upon the negative answers to the research questions from these sophisticated GPS collars, efforts have already been, and will be made in the future, to educate other users to the failures and limitations of sophisticated GPS technology.
If the "state-of-the-art" GPS technology had worked as promised, a major step would have been made in eliminating the disadvantages and facilitating the tracking of wildlife in remote and rugged areas. However, GPS tracking technology is still not reliable, and the newer the technology, the less functional it currently seems. The more simple the GPS method, the more reliable the signal acquisition and locational accuracy. When it comes to the still developing and error-prone technology of GPS tracking of wildlife in dissected terrain, it appears simpler is still better.
Future Research Needs
Due to the extreme importance of the region's bighorn sheep population, ongoing monitoring and research is important to ensure the survival of this fragile species. Demographics, lamb survival, critical lambing and rutting areas, habitat utilization, lamb survival and herd recruitment, and overall herd health and related disease issues are a few of the important research and monitoring needs which will continue to need to be addressed on a continuing basis.
In the remote and dissected habitat utilized by bighorn sheep, monitoring efforts maximizing the effectiveness of traditional VHF collars, and secondarily, simpler but reliable GPS technology, will be necessary. With numerous herds located in the southeast Utah metapopulation, cyclic monitoring of the herds will need to continue.
As this project showed, reliable results are not yet to be expected from sophisticated GPS collars and receivers. The development of new GPS systems to monitor wildlife on a reliable basis has been slow, and new systems have been error-prone. New systems, when possible, should be used in the future when research proves them to be consistently reliable. Future research and monitoring, however, should rely upon time-tested methods until research such as this project produces positive results from GPS systems.