STATISTICAL ANALYSIS OF THE SPATIAL RELASHIONSHIP OF CAATINGA AND PHYSIOGRAPHIC FACTORS THROUGH REMOTE DATA

Resumo

The caatinga is characteristic of the semi-arid region of Brazil, of significant coverage in the northeast of the country and, therefore, geoprocessing and statistical techniques can be important tools to analyze the environmental factors associated with its distribution. The present study had the objective of analyzing the relationship of: altimetry, slope, soil type and river proximity in the spatial distribution of the caatinga vegetation, using a database generated by geoprocessing techniques. NDVI was used, and statistical analysis was based on multiple regression and Pearson's correlation. The results show that the weather station is the only variable that presented a high determination coefficient with the caatinga vegetation (R2 = 0.956), while the other independent variables have weak or nonexistent correlation. With the techniques of geoprocessing was allowed to observe that the Caatinga has a generalist spatial distribution of occurrence, not having a defined pattern.