Estimativa de escoamento superficial no Brasil utilizando o método NRCS-CN por computação em nuvem (google earth engine)

Abstract

The present study focuses on the management of surface runoff, essential for the conservation of water resources and sustainable use of Brazilian biomes. The objective is to estimate and evaluate the surface runoff across the Brazilian territory in a distributed manner, employing the NRCS-CN method on the Google Earth Engine (GEE) platform. The methodology involves calculating runoff depth pixel by pixel, using remote sensing data and the JavaScript language. Data from precipitation (Climate Hazards Group InfraRed Precipitation with Station Data – CHIRPS), land use and land cover (Land Cover Type 1 MODIS), and soil type (OpenLandMap Soil Texture Class USDA System) were applied. Based on these, Curve Number (CN) values were classified, adjustable according to the Antecedent Moisture Conditions (AMC) of the soil. The processed data were exported (in TIF and CSV formats) for map production in GIS software (ArcGis and Qgis), application of descriptive statistics (Excel), and significance testing (Fisher's Exact Test - MATLAB). The relationship between runoff and precipitation, considering the characteristics of Soil Type and Land Use, was investigated through Fisher's Exact Test. The results indicate high variability in the spatiotemporal behavior of surface runoff in Brazil. It is noteworthy that the Pampa Biome showed the highest average annual runoff of 569 mm (2002), while the Caatinga had the lowest, at 40 mm (2012), in a historical series from 2001-2020. In this period, the driest year had a runoff of 7.02% compared to the year with the most intense runoff (529 mm difference). A total of 120 mosaics were generated with the results of the precipitation and runoff relationships, covering all combinations of soil type and land use/occupation. This study provides valuable knowledge for the development of climate change adaptation plans and the mitigation of water-related disaster risks.