The application of Landsat image data to rangeland assessment and monitoring: the development and demonstration of a land image-based resource information system (LIBRIS)

Abstract

A land image-based resource information system (LIBRIS) has been developed to support rangeland resource management on both regional and local scales. The study area is centred on the mining town of Broken Hill (lat. 32°S, long. 141°30’E) which lies within the semi-arid rangelands of southern Australia. LIBRIS is a geographic information system developed to integrate Landsat spectral data with other relevant spatial data sets (tenure, rangeland type, elevation and slope) not only to improve the interpretation of Landsat data for rangeland applications, but also for report generation and the modelling and derivation of other landscape variables such as susceptibility to erosion. A simple vegetation–Landsat response model is used within LIBRIS. The model, based on field reflectance data, has two indices representing the proportions of ‘cover’ and ‘greenness’ (vigour) of the sparse rangeland vegetation. The index of ‘cover’ is used extensively in this report. Classifications of cover classes derived from Landsat are integrated with land tenure boundaries to present reports of land cover classes and changes in these classes over time (1972–80) for pastoral properties and for the various rangeland types. Finally, the land cover information has been incorporated with slope to derive a map and report of susceptibility to erosion; an index which is of fundamental importance to arid land management. The utility of LIBRIS in integrating Landsat data with collateral data for arid land management is thus demonstrated.