QUANTITATIVE BIOMASS MEASUREMENT ON SALT DESERT SHRUB RANGELANDS UTILIZING LANDSAT COMPUTER COMPATIBLE TAPE DATA

Maintaining ongoing inventories of vast tracts of land using conventional ground survey techniques is difficult. Accurate and efficient techniques of remotely monitoring vegetation using Landsat digital data could result in substantial savings of both time and money. This is especially true in the sparsely vegetated Great Basin region.; There are five research objectives that have been addressed. First, a methodology for quickly estimating shrub biomass on Salt Desert rangelands has been developed. Secondly, the vegetation cover threshold at which Landsat can distinguish vegetation from soil within the Salt Desert ecosystem has been identified. Next, the ability of several common Landsat vegetation indices to estimate Salt Desert vegetation parameters has been addressed. Digital mapping of Salt Desert vegetation communities using Landsat data was completed. Finally, a biomass model using Landsat data has been developed and tested.; The field methodology utilized for collecting field biomass data is a modification of the reference unit technique. The modification utilizes regression techniques to allow the rapid development of the calibration equation that relates the number of reference unit equivalents on a transect or quadrat to biomass.; Landsat can distinguish vegetation from soil in areas with vegetation cover as low as seven percent. The vegetation index used to achieve discrimination of vegetation at a seven percent cover threshold was the Perpendicular Vegetation Index (PVI7).; The vegetation indices most suited for use in the low cover, shrub-dominated Salt Desert are the band ratio of MSS7/MSS5, and the PVI7. The PVI7 is slightly better at discriminating soil from vegetation, while the ratio of MSS7/MSS5 is more useful for identifying vegetation communities.; Digital vegetation maps of the Skull Valley, Utah field region have been produced using the PVI7 and MSS7/MSS5 vegetation indices. A vegetation map combining both of these vegetation indices has a statistically significant accuracy of 74.2 percent.; An additive regression model using digital numbers generated with a vegetation index was developed and tested as an estimator of range biomass. Due to difficulty in discriminating shrubs from high annual grasses, and other less significant problems, the model failed to estimate accurately Salt Desert range biomass.