Estimativa da produção primária bruta em áreas com diferentes tipos de cobertura da terra

Abstract

Carbon sequestration by terrestrial biomes plays a significant role on the global carbon cycle to mitigate the atmospheric CO2 increase and its consequences on climate changing. Gross Primary Production (GPP) is a biophysical parameter of any ecosystem and can assist understanding the spatial and temporal dynamics of carbon flows. The use of orbital remote sensing techniques has been a valuable tool because it allows a panoramic view, at different scales, with high precision and lower operating cost when compared to the traditional techniques of monitoring land use and carbon cycle cover. Although several researches have been successful in using remote orbital sensors to monitor land use and land cover and quantify GPP, the main challenge is developing a systematic and able methodology that makes it easy to identify land use, land cover and to obtain GPP values. An alternative to solve these demands is the Google Earth Engine (GEE) platform, a geographic spatial data repository that enables large-scale environmental data analysis with broad computational capabilities as it uses Google servers for processing and storage. Thus, paper 1 aims at developing a systematic methodology to classify different kinds of land use and land cover with Landsat 8 images on GEE platform, while paper 2 aims at evaluating the methodology to estimate GPP by remote sensing techniques, based on Landsat 8 sensor images for different kinds of land use and land cover. This methodology is based on the Estimating Absorbed Photosynthetically Active Radiation model - A PAR - by vegetation, associated with the light use efficiency model and was implemented on GEE platform.