The 2008 Joint Meeting of the Society for Range Management and the America Forage and Grassland Council. |
Tuesday, January 29, 2008 - 2:00 PM
Accurate Modeling of Bare-Earth Exposure Using Remote Sensing and Geographic Information Systems
Jacob T. Tibbitts1, Nancy Glenn2, and Keith T. Weber1. (1) GIS Training and Research Center, Idaho State University, 921 South 8th Avenue Stop 8104, Pocatello, ID 83209, (2) Idaho State University, Boise Center Aerospace Laboratory, Idaho Water Center, 322 East Front Street, Suite 240, Boise, ID 83702
A major concern of both public and private land managers is the loss of rangeland resources. Regardless of the cause of rangeland resource loss (urbanization, drought, invasive weeds, or fire), the degradation of healthy rangelands –leading to decreased productivity and sustainability--and subsequent desertification pose problems for ranchers, land management agencies, and the economy of many western states. Within semi-arid ecosystems, the amount of ground cover is a key indicator of rangeland health. As ground cover decreases, bare earth exposure increases which can lead to a decrease in soil stability, productivity, and sustainability. Thus, accurate and reliable detection and modeling of bare earth exposure is important for extensive, efficient, and reliable assessments of rangeland health and monitoring. Ground based measurements of ground cover can be prohibitive due to high cost, inaccessibility, or other logistic constraints. Nonetheless, remote sensing technologies and the use of geographic information systems allows for extensive, economical, and objective monitoring of bare earth. This study, located at the O’Neal Ecological Reserve in southeastern Idaho, develops a bare earth exposure model through classification of remotely sensed satellite imagery correlated with field data. Accuracy assessments were performed to ensure the model could reliably detect bare earth exposure within acceptable error rates. This work shows potential for early detection of bare earth exposure for monitoring trends over time related to various management decisions or environmental phenomena.