Impacto da mudança de cobertura do solo na dinâmica da GPP, ET e WUE utilizando produtos derivados do sensor MODIS em Mato Grosso

Abstract

Despite having ecological importance, the state of Mato Grosso has experienced intense change in its land cover in recent decades and the effects of these changes on climatic variables need to be studied. Thus, the objective of this work was to evaluate the impact of land cover change on the dynamics of energy and mass exchanges, using products derived from the MODIS sensor in Mato Grosso, Brazil. The study was divided into three stages: i) Validation and analysis of land cover products MCD12Q1 and CUT-IBGE; ii) Validation and analysis of the MODIS Evapotranspiration product MOD16A2iii) Analysis of the spatiotemporal dynamics of biophysical variables in Forest, Cerrado, Pasture and Agricultural environments in Mato Grosso. In step i, 25 control points were randomly created in each class to obtain the confusion matrix. From the confusion matrix, 7 indexes were generated for individual performance evaluation and 2 indexes for general performance analysis of the evaluated classes. In step ii, the MOD16A2 evapotranspiration data were validated using Evapotranspiration data calculated by the Bowen Ratio method obtained at four experimental sites in Mato Grosso. In step iii, the annual and seasonal averages of the Precipitation, EVI, Ts, GPP, ET and WUE variables were calculated from values extracted from 100 points in each of the coverages, derived from the intersection between the product MCD12Q1 and CUT-IBGE. The performance of the CUT-IBGE product was superior to MCD12Q1. The spatial resolution and the similarity between some phytophysiognomies were limiting factors in the classification of land cover in Mato Grosso. The loss of natural cover of the Forest and Cerrado between 2000 and 2016 was mostly caused by the increase in Agricultural and Pasture areas, due to the agricultural profile of the region. The MOD16A2 product proved to be an important alternative in Evapotranspiration studies at a local and regional scale. The results obtained in this study demonstrated the good ability of the product MOD16A2 to estimate the ET in transitional environments Amazon ForestCerrado, Cerrado and Pantanal. The Cerrado, Pasture and Agricultural Area environments have similar patterns of energy and mass exchange throughout the year. This suggests that the proper management of pasture and agricultural areas associated with the preservation of legally protected Cerrado areas can coexist, allowing to meet the growing demand for food in the world. On the other hand, the replacement of forest areas by other covers reduces the EVI, ET, GPP and WUE, thus significantly modifying the interactions between the soil-plantatmosphere system in these locations.