Page Banner

United States Department of Agriculture

Agricultural Research Service

Research Project: MANAGEMENT TECHNOLOGIES FOR ARID RANGELANDS

Location: Range Management Research

Title: Modeling erosion in a southern New Mexico watershed using agwa: Sensitivity to variations of input precision and scale

Authors
item Burnett, Sarah -
item Karl, Jason
item Abbott, Laurie -

Submitted to: Society for Range Management Meeting Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: December 15, 2011
Publication Date: January 29, 2012
Citation: Burnett, S., Karl, J.W., Abbott, L. 2012. Modeling erosion in a southern New Mexico watershed using agwa: Sensitivity to variations of input precision and scale [abstract]. 65th Annual Meeting of the Society for Range Management, January 29-February 3, 2012, Spokane, Washington. p. 0217.

Technical Abstract: Rangeland environments are particularly susceptible to erosion due to extreme rainfall events and low vegetation cover. Landowners and managers need access to reliable erosion evaluation methods in order to protect productivity and hydrologic integrity of their rangelands and make resource allocation decisions. Predicting erosion and its impacts, however, is a challenge due to the spatially variable nature of erosion and the difficulty of measuring it. One erosion model, the automated geospatial watershed analysis (AGWA), uses broad-scale GIS layer inputs to model erosion over entire watersheds to predict relative erosion rates and identify areas where erosion is high. In this research we conducted a sensitivity analysis of the AGWA model, to determine how varying the precision and scale of model inputs affects the magnitude and spatial distribution of erosion predictions. We compared AGWA outputs from three different scales and precisions of input data from a private ranch in Grant County, New Mexico: detailed vegetation and soils input layers derived from remote sensing and field measurements, ecological-site state mapping, and AGWA national-scale vegetation and soils layers. Additionally, erosion hotspots identified by AGWA from each input set were compared to locations within the ranch using rangeland health indicators to determine how AGWA predictions were sensitive to scale and precision in predicting areas susceptible to extreme erosion. The results of this study demonstrate how the AGWA model can be used to identify locations vulnerable to erosion and determine what scale of input is necessary to make useful predictions.

Last Modified: 9/20/2014
Footer Content Back to Top of Page