Evaluation of Landsat MSS data for classifying and characterizing natural vegetation on a regional basis

Vita.Landsat multispectral scanner imagery and digital data were evaluated for use in classifying and characterizing natural vegetation in northcentral Texas and southcentral Oklahoma. Landsat imagery was used to map and classify natural vegetation throughout the region. Computer compatible Landsat digital data were analyzed to determine the relationship between spectral reflectance and different vegetation parameters measured on the ground. The study reports on three meaningful contributions in remote sensing of natural vegetation from satellite data products. They are (1) a natural vegetation classification suitable for remote sensing use, (2) the evaluation of different dates of imagery for relative information content, and (3) information relating Landsat multispectral scanner digital data to quantitative vegetation measurements. The natural vegetation classification was developed for a 10,000 square mile regional area covering parts of northcentral Texas and southcentral Oklahoma. Fifteen vegetation types were identified on Landsat imagery and were ecologically determined from association tables prepared from supportive ground data. The classification is most suitable for generalized levels of land use and natural resource planning