Gross primary productivity in areas of different land cover in the western Brazilian Amazon

Abstract Uncertainty regarding gross carbon sequestration at local, regional, and global scales can be reduced by monitoring the land surface processes at high spatial and temporal resolutions. In this sense, the objective of this study was to estimate gross primary productivity (GPP) in a region of the western Brazilian Amazon using Landsat 8 OLI/TIRS images, and to evaluate possible changes in estimated productivity among areas of different land use and in different seasonal conditions (August 04, 2013, representing the dry season, and October 10, 2014, representing the rainy season). The images were subjected to atmospheric and radiometric corrections. As the basic input for the model, the balance of radiation and other components of the energy balance were calculated using the Surface Energy Balance Algorithm for Land. The GPP estimated through the OLI/TIRS sensor was compared with the MOD17A2 product of the MODIS sensor. The comparison of GPP by the OLI/TIRS sensors and the MOD17A2H product were based on the mean error (ME), simple linear regression (r2) and correlation coefficient (r). The estimated GPP indicated distinctions among land use types, however, similarities with MOD17A2H were detected only with the image from the rainy season (r = 0.53), with a slight underestimation for all land uses. The dry season image showed r2 = 0.11, r = 0.33 and ME = −0.48 gCm−2day−1 for extractivism, while the land use types were overestimated by the model. The OLI/TIRS sensor estimates need to be validated with data from flow towers. The focus on carbon sequestration by forest ecosystems and the reduction of CO2 emissions is the foundation for mitigating climate change damage and consequences at regional and global levels. Currently deforestation of the Amazon rainforest is growing absurdly. Policies contrary to the control of deforestation will only cause accelerated climate change, which will endanger the lives of future generations.