MULTI-SPATIOTEMPORAL SIMULATION OF EDGE EFFECT ON FOREST PATCHES IN THE BARRA SECA RIVER BASIN, ES

The Atlantic Forest is intensely fragmented and this fragmentation process has caused an expressive increase of forest remnants and, consequently, increased edge effect with different physical-biological intensities in the transition areas between the patch and the matrix. This study used landscape metrics to understand and analyze how different edge effect distances affect the structure of the forest landscape in the Barra Seca River basin (ES), in 1985, 1996, 2006 and 2016. Remote sensing images were processed and using the Bhattacharya algorithm with supervised classification, the forest patches of the study area were classified and isolated. Landscape ecology metrics were computed with Patch Analyst and V-Late 2 Beta extensions. The forest patches were divided into four size classes as follows smaller than 5 ha (C1); between 5 and 10 ha (C2); between 10 and 100 ha (C3); and over 100 ha (C4). The edge effect simulation using landscape metrics was performed using the edge effect distances of 20, 40, 60, 80, 100, 140, and 200 m. Forest fragmentation increased between 1985 and 2016 while the number of patches greater than 100 ha decreased. Currently, the basin landscape consists mainly of small patches, which have larger relative areas affected by edge effect while many patches smaller than 10 ha are completely dominated by edge effect for distances greater than 60 meters. The edge effect simulation for different distances allowed verifying the intensification of the edge effect on the forest patches of the Barra Seca River basin.