Modelos de índice de umidade do solo obtido por sensoriamento remoto no sertão pernambucano

Abstract

In semi-arid regions, growth and productivity of agricultural crops are directly linked to the amount of water available from rainfall and / or irrigation. The accurate assessment of soil moisture is difficult to be performed by means of field measurements because the typical field methods are complex and expensive. This work aimed to develop models of multiple regressions of the soil surface moisture content using images from the MODIS sensor and meteorological data (rainfall (mm), mean air temperature (C °), maximum temperature and minimum air (° C ), sunshine (hours) Evaporation (mm), relative humidity (%) wind speed (km / h) Ouricuri and Arcoverde belonging to the middle region of Pernambuco Hinterland for the years 2006 and 2012. it also held -if the zoning of the aridity index to identify the cities where there are clusters of aridity and correlate the soil moisture content with weather and biophysical variables. the inverse interpolation technique of the square of the distance was applied to spatialise the aridity index using data of precipitation and air temperature from 1911 to 1990 measured at 82 weather stations. a total of 44 vegetation index images and 88 temperature of the obtained surface of MODIS / Terra were used to obtain the moisture content from soil. The normalized difference vegetation index images were classified to quantify and assess the areas of land use and land cover. The results showed that the high vegetation protection class disappeared from 2006 to 2012, followed by an increase in areas of low vegetation cover. They observed also trends of aridity core training in the cities of Arcoverde, Sertânia, Iguaraci, Afogados da Ingazeira, Ingazeira, Carnaíba and Tabira, as presented aridity index at the boundary of semi-arid regions. It was concluded that the soil moisture content was related linearly very well with NDVI and Ts obtained from MODIS products for the city of Arcoverde - PE. The relationship between soil moisture content by evaporation and wind speed was moderate, but reversed. The models of multiple regressions can be used to estimate the moisture content of the soil surface from meteorological data from the municipalities of Ouricuri - PE and Arcoverde - PE for the years 2006 and 2012.