Simulação de regimes microclimáticos utilizando imagens do sensor MODIS e dados meteorológicos em regiões de brejos de altitude ocorrentes em Pernambuco

Abstract

The semi-arid northeast Brazil specifically in the states of Pernambuco and Paraíba are the Mountain forests. In these locations, the exploitation of natural resources occurs gradually every years, because of the milder climate and greater water availability for the agricultural and livestock sectors. Studies related to the microclimatic conditions of mountain forests are fundamental for social and agricultural development, considering the climate as the most important factor for the emergence or disappearance of different plant or animal species, as well as the way of life of its population. The MTCLIM model was used in this research to estimate the global solar radiation, the air temperature (mean, minimum and maximum) and the vapor pressure deficit, with the objective to identify and analyze the spatialtemporal dynamics of meteorological elements as well as biophysical parameters to detect environmental changes through remote sensing techniques on satellite images for the Arapuá mountain, city of Carnaubeira da Penha-PE. Using sixteen days images MODIS: index of vegetation NDVI and superfice albedo available in 2006 and 2012. The NDVI were used in estimation Leaf Area Index (IAF) and Fraction of soil covered by vegetation (FC). It was used too surface data: average, minimum and maximum air temperatures (°C), precipitation (mm), the global solar radiation (W m-2) obtained in three automatic weather stations located in next study area belonging to the INMET and CPTEC. The thematic maps of meteorological elements: Tmax, Tmed, Tmin and Rg were processed by Demo Surfer 10.0 software while albedo, NDVI, FC and IAF were processed from SPRING software, version 5.1. It was accomplished the MTCLIM model calibration process to the automatic meteorological stations of the cities of Cabrobó and Triunfo. The estimated data were correlated with the observed Tmax, Tmed, Tmin, Rg and DPV using the regression analysis, considering Mean Error (EM), Mean Absolute Error (EMA) and WILLMOTT Correlation Index. The validation from MTCLIM model showed good results for the cities of Cabrobó and Triunfo with high values of correlation (r ) and determination index (d) for Tmed, Tmin, Tmax e Rg except for the vapor pressure deficit. The topography effect is very important for the modeling of Tmed, Tmin, Tmax and Rg in all scenarios, showing that the Arapuá mountain has a great influence on the spatial distribution of these meteorological elements. It follows that the main condition for the variation of vegetation conditions in this region was precipitation, which presented its highest values in the year of 2006.