Análise da distribuição espacial de áreas queimadas através da função K de Ripley

Abstract

The aim of this study was to evaluate the use of Ripley’s K function to determine the spatial pattern of burned areas, which occurred in six Conservation Units from September 3 to October 5, 2008. The study area is located in the north of Minas Gerais state, Brazil, and measures a total of 936,901 ha. A cartogra - phic data base was used for the spatial analysis; with burned polygons mapped through the Landsat5 TM image processing techniques. The burned areas were separated into size classes in order to highlight the spatial patterns. For the spatial pattern analysis Ripley’s K function was calculated for all burned areas as well as for each size class. 194 burned areas were mapped, which corresponds to a total of 19,538 ha and represents 2.09% of the study area. For all burned areas and for those in class sizes smaller than 100 ha the hypothesis of Complete Spatial Randomness was rejected. In other words, these areas had some aggregated distribution pattern along the scale of analysis, indicating that anthropogenic, biotic or abiotic aspects could have contributed to these patterns. For burned areas larger than 100 ha the hypothesis of Complete Spatial Randomness was attended, which means that these areas showed a random spatial pattern. In these areas, fire was used carelessly, ignoring the prevention techniques and then widening into large proportions. This kind of behavior is common throughout the area, making it difficult to predict the spatial pattern of future burned areas.