Data analysis using observations and numerical modeling to estimate carbon exchange in forest areas in Amazonia

• Grant number: 24/19065-0
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: February 01, 2025
• End date: December 31, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Luiz Augusto Toledo Machado
• Grantee: Angel Liduvino Vara Vela
• Host Institution: Instituto de Física (IF). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 22/07974-0 - Synergistic effects of climate change and land use on carbon source and sink of Amazon forest ecosystem, AP.PFPMCG.TEM

Abstract

The Amazon forest fulfills an essential role in the global balance of carbon and regional climate. The superficies and the atmosphere are intrinsically coupled by exchanges of energy, trace gases, and water vapor. On one side, the climate can influence the dynamics of land use changes, for example, favoring the occurrence of deforestation in drought events. On the other hand, deforestation can affect the climate through changes in the pattern of evapotranspiration, emissions from fires, and a decrease in the capacity to remove CO2 by the vegetation. The impacts of deforestation in the Amazon are not restricted to a local scale, and can affect several socio-economic sectors, from agricultural production in the Center West to water supply in the Southeast of Brazil. Therefore, it is essential to characterize the relationships between land use and atmospheric conditions. The objective of this postdoctoral project is to investigate the associations between land use, meteorological variables, and concentration of atmospheric pollutants in the Amazon, combining data from remote sensing and reanalysis models in an integrated and standardized database covering the period of 1985 to 2022. Data on meteorological variables on the surface from the ERA5-Land reanalysis, atmospheric pollutant concentration data monitored via satellite by the MODIS, AIRS, and TROPOMI sensors, and land use data from MapBiomas will be used. By standardizing all data to the same spatial grid and monthly temporal resolution, crosses will be made between the evolution of land use and variations in atmospheric conditions at different spatial scales, starting at a scale of 10 km. Factoring and grouping techniques such as EOF (empirical orthogonal functions) and cluster analysis will be applied to divide the study area into sub-regions with similar behavior. The aim is to carry out a longitudinal regression analysis, developing a linear mixed model in two stages. Based on the model parameters, it will be possible to assess the relative importance of atmospheric variables that contribute to deforestation and vice versa. This research will be developed as part of the project "Greenhouse Gas Emissions in Amazon and Data Analytics and Service System" (GHG) of the Research Center for Innovation in Gas (RCG2I, CPEFAPESP 2020/15230-5)