Sensoriamento remoto aplicado à análise temporal da relação uso da terra/temperatura e albedo de superfície na bacia do Rio Vieira no Norte de Minas Gerais

Abstract

This study aimed to analyze with the help of TM data - Landsat 5, the relationship between the changes in land use and the variation of temperature and surface albedo, during period winter in basin River Vieira for the years 1985, 1995 and 2010. The choice of river basin Vieira is justified because this area, located in the North of Minas Gerais, was the scene of intense and rapid population growth. So this fact led to several modifications to the systems of river basin Vieira land use, such as: rapid urban expansion, rehabilitation of degraded areas and human disturbance of natural areas. Allowing this context, the development of this work in an environment with appropriate changes in land use systems necessary for the research. The methodological procedures needed to analyze the relationship between changes in land use and the variation of temperature and surface albedo in the Rio Vieira were the following: the first procedure considered the treatment of satellite images in goal to map land use; this mapping, in turn, followed the hierarchical scheme proposed by Heymann Legend (1994). The second procedure considered the application of the algorithm semi empirical Surface Energy Balance Algorithms for Land - SEBAL, which describes the full swing of radiation on the Earth's surface from the spectral bands of visible, near infrared and thermal infrared. From an operational standpoint, the runtime of the methodology applied to the mapping of land use, since the acquisition of orbital products to validate the mappings, was relatively short. A fact that deserves attention, especially because it led to the development of a wide range of temporal information of Vieira River basin. With emphasis on the recovery of 54.79 km ² of natural vegetation in 25 years. From the viewpoint of the thermal mapping, the coefficient of variation showed a correlation between the temperature obtained by the SEBAL model and the air temperature, measured at about 1.5 m high, ranging from 0.61 to 11.38 % and an overall accuracy r ² = 0.961. This validation was highly effective in order that the data Landsat TM 5 are collected at an altitude of 705 km above the surface. The statistical validation shows that the estimate of the infrared emissivity obtained by this algorithm satisfy the need for obtaining the thermal modeling of river basin Vieira from the spectral channels of Landsat 5 TM sensor. It is recommended the SEBAL algorithm for studies seek to that analyze the spatial distribution of surface temperature data. Considering that the temperature of the surface is of great value to the understanding of various climatic phenomena.