Studying vegetation distribution using ancillary and remote sensing data: A case study

A Landsat Thematic Mapper image (TM) was used, in combination with ancillary topographic and topoclimatic data, to study the distribution of vegetation classes in the Niwot Ridge-Colorado, U.S.A. A logical channel approach; i.e, spectral and ancillary data, for digital classification of remote sensing data was used. The analysis was performed using the SPSS statistical package. The vegetation class was dependent on nine selected topoclimatic and topographic data variables. These variables include: topographic slope, aspect, albedo with and without slope/aspect consideration, Normalized Difference (ND) with and without slope/aspect consideration, convexity, Potential Solar Insolation (PSI), and Slope Aspect Index (SAI). Three random sampling techniques, to select vegetation classes samples from the map, were used: random samples from all the regions in the study area, samples from hilly areas only, and samples using a strip area along the map profile. The results of this study showed that the combination of TM data with the topographic and topoclimatic data variables is an efficient way to study the distribution of wet and dry vegetation classes. Minor effects were found for samples locations on the discrimination analysis process.