Estimativa de erosão a partir do índice de brilho de imagens orbitais: ferramenta aplicada ao diagnósticos das pastagens degradadas

Abstract

Inadequate soil management and the absence of conservation practices favor the degradation of pastures and can trigger adverse environmental changes and damage under the terms of Federal Law nº. 6,938 / 1981. Based on this premise, this work aimed to estimate soil losses caused by water erosion in pasture areas using the brightness index (BI) of the annual Landsat 8 image series in different geological formations. A specifically prepared Google Earth Engine (GEE) script automatically extracts the BI from the images. The study took place in the Environmental Protection Area (APA) of the Uberaba river basin (Minas Gerais, Brazil). To achieve the objective, sampling was carried out at 37 collection points, with 5 simple soil samples extracted 20 cm from the surface and random probing with the digital penetrograph at 60 cm depth, in a radius of 20 meters around each sampling point, and then analyzed for USLE (Universal Soil Loss Equation) and BI zonal soil loss statistics in a Geographic Information System. The BI versus USLE soil loss regression models allowed estimating BI soil losses over APA pastures. The model's accessories were remarkable. The soil loss maps in the APA were validated in pasture phytophysiognomies by surveying resistance to penetration in 37 randomly selected locations. The results were satisfactory, especially those based on BI. BI losses increased towards higher resistances. Increased losses identified also occurred in regions exposed to environmental land use conflicts (actual uses that deviate from land capacity or natural use). Overall, the BI approach proved efficient for accurately tracking soil losses and pasture degradation across large areas, with the advantage of being in a single parameter easily accessed through sensed data. From an environmental point of view, this is an important result, as the accurate diagnosis and prognosis of degraded pastures is essential to implement mitigation measures after the “polluter pays principle”, even more so in Brazil where the areas occupied by degraded pastures are huge.