Developing a digital soil survey update protocol at the Golden Spike National Historic Site (Utah).

摘要

There are few guidelines for soil survey update procedures. A soil survey update protocol using geographic information systems (GIS) and remote sensing was developed at Golden Spike National Historic Site (GOSP). Slope derived from digital elevation models (DEMs) using GIS was compared to in-field slope measurements for selected map unit polygons. The slopes of most polygons were accurately predicted using DEMs, except polygons belonging to steep-sided, short-sloped drainages. Limestone rock outcrop was identified using Landsat 7 imagery. Bands 5 and 2 and threshold values composed a normalized spectral index model. Slope and fractional vegetation index layers were added to the model yielding 87% overall accuracy for predicting rock outcrop. Ecological sites were refined using Landsat 7 imagery. Unsupervised classification produced 12 classes that were clumped into 3 classes. Percent class composition within polygons determined ecological site. Ecological site classification using remote sensing could facilitate monitoring of rangeland state and transition.