Evaluation of Water and Carbon Estimation Models in the Caatinga Biome Based on Remote Sensing

The study of energy, water, and carbon exchanges between ecosystems the atmosphere is important in understanding role vegetation regional microclimates. However, they are still relatively scarce when it comes to Caatinga vegetation. This aims identify differences dynamics critical environmental variables such as net radiation (Rn), evapotranspiration (ET), fluxes (gross primary production, GPP) contrasting recovered (dense Caatinga, DC) degraded (sparse SC) state Paraíba, northeastern Brazil. Estimates were performed using Surface Energy Balance Algorithm for Land (SEBAL), comparisons estimated measured data conducted based on coefficient determination (R2). Eddy Covariance (EC) method comparison with same derived from Moderate Resolution Imaging Spectroradiometer (MODIS) aboard Terra satellite. estimates showed higher Rn values DC, indicating that this area should have greater energy availability physical, biological, chemical processes. R2 daily observations was 0.93. ET SEBAL relation observed values; however, presented better spatial discrimination surface features. MOD16A2 algorithm, closer agreed seasonality Enhanced Vegetation Index (EVI). DC generally than SC, while MODIS (GPP MOD17A2H) a temporal behavior observations. difference two areas more evident rainy season. GPP MOD17A2H 0.76 0.65 respectively. In addition, Observed modeled lower, i.e., 0.28 0.12 capture CO2 considering whole year, SC showing notable increase absorption only predominant underestimation but evidenced effects land use cover changes over all seasons.