Page Banner

United States Department of Agriculture

Agricultural Research Service


item Norton, Jill
item Glenn, Nancy
item Weber, Keith
item Seefeldt, Steven - Steve
item Taylor, Joshua - Bret

Submitted to: Society of Range Management
Publication Type: Abstract Only
Publication Acceptance Date: 7/31/2005
Publication Date: 2/13/2006
Citation: Norton, J., Glenn, N., Weber, K., Seefeldt, S.S., Taylor, J.B. 2006. The use of remote sensing imagery to determine wildland burn severity in semiarid sagebrush-steppe rangelands. Society of Range Management.

Interpretive Summary: In order to facilitate the management of recently burned rangelands, knowledge of how much land burned and how severe the fire was is critical. It is quite possible that this information can be obtained using satellite imagery. Satellite imagery has been successfully used for this purpose in forests, but not for rangelands. Using a prescribed fire at the USDA-ARS U.S. Sheep Experiment Station near Dubois, Idaho, fire size and severity were measured on the ground and compared to satellite imagery analyses. The modifications of the methods used for forest fires improved the estimates of fire size and intensity. Use of satellite imagery shows promise as a tool to assist land managers, especially federal and state, in designing targeted management plans to improve vegetation recovery on burned rangelands.

Technical Abstract: The Normalized Burn Ratio (NBR) is a remote sensing-based index used to calculate the extent and severity of a fire. NBR functions well in forested ecosystems due to a high contrast of vegetation change before and after fire. Preliminary results indicate that an alternate burn severity algorithm is needed for semiarid rangelands to account for high soil reflectance and sparse vegetation cover. This study applies a burn severity assessment using Landsat and SPOT imagery following a fall 2005 prescribed fire at the USDA-ARS U.S. Sheep Experiment Station near Dubois, Idaho. Our burn severity assessment, applicable to semiarid environments, was developed using field and remote sensing data. This study applies an initial assessment NBR and a modified NBR algorithm to Landsat and SPOT imagery. The final model and technique will provide land managers with a tool to assist in accurately describing the characteristics (size, distribution, and severity) of burned rangeland areas. Thus, this remote sensing application will enhance fire management and recovery efforts across semiarid rangelands worldwide.

Last Modified: 06/27/2017
Footer Content Back to Top of Page