Monitoramento de seca meteorológica com sensoriamento remoto na região médio norte de Mato Grosso

Abstract

Drought is characterized as one of the most complex and poorly understood of all natural disasters, impacting several areas of society, such as agriculture, industry, health, water distribution and energy generation. The main limitation of drought indexes used to monitor, identify and quantify precipitation anomaly is the lack of representative data of the area of occurrence. Measurements of hydro meteorological variables by satellites provide a good alternative in the absence of surface data. Thus, the present study intends to evaluate whether the use of the 3B43 V7 product of the Tropical Rainfall Measuring Mission (TRMM) multi satellite Precipitation Analysis (TMPA) is effective in generating monthly precipitation and drought maps from the monthly Standardized Precipitation Index in the northern region of Mato Grosso. Satellite-based data has been compared to rain gauge-based observation for the period between 1998 and 2017. Reliability of drought maps has been verified by evaluating the great drought that occurred during the 2015/16 harvest period. Results show product 3B43 V7 overestimates precipitation, although it can be used in the absence of surface data, since Nash-Sutcliffe coefficient (ENS) value was 0.75 and the Willmott concordance index (d) was 0.93. The correlation between SPI 3B43 V7 and SPI surface-based observation has been validated by ENS and d, whose values were 0.63 and 0.92, respectively. The drought maps confirmed the conditions reported in the bulletins of Instituto Mato-grossense de Economia Agropecuária (IMEA), which showed crop failure due to the lack of rain throughout critical stages of soybean and corn crops.