Study of land uses in the Sorocabuçu River watershed, Ibiúna-SP, (Brazil) for the evaluation of eligible areas in MDL projects

Abstract

The constant population growth along with the demand for natural resources put pressure on forests' capacity to capture CO2, among other benefits, thus arising the need to implement monitoring and conservation measures. Considering the above, the objective of this study was to identify degraded areas eligible for the development of CDM projects in the Sorocabuçu River watershed, Brazil. To achieve this, land use was classified for the years 2000 and 2020 using Multi-layer Perceptron neural networks, which were evaluated using a confusion matrix and Kappa index. These classifications allowed the determination of transition potential using the LCM tool, which, along with NDVI calculation, enabled the identification of areas eligible for CDM project implementation through a Fuzzy system. In the land use classification, a loss of natural vegetation areas was identified, while the NDVI indicated the absence of areas with vegetation in the extremely healthy category. As for transition potential, it was determined that the study area mostly exhibits low potential; however, there are areas with high potential. Finally, it was found that the study area comprises 1.38% degraded areas with high eligibility for CDM project implementation.