Agriculture drought assessment based on remote sensing, cloud computing, multi-temporal analysis. A case study: the Mostiștea Plain (Romania)
Abstract: Agricultural drought is one of the most important natural hazards worldwide, affecting a significant proportion of the global population. Earth Observation multi-spectral imagery satellites can provide a comprehensive picture of all land and sea areas of the Earth. Free of charge and open access imagery from missions such as Sentinel-2 provides high quality imagery with rapid high revisit period. Earth Engine© developed by Google Inc. provides the possibility to view and analyse petabytes of remote sensing data in archives that include more than thirty years of satellite imagery and scientific datasets. This paper proposes a cloud- based computation approach and analysis of multi-temporal, high resolution Sentinel-2 imagery on the Mostiștea Plain (Romania) in order to evaluate the agriculture drought. Custom javascript code was created in the Code Editor for calculating and analyzing remote sensing-based indices between 2017 and 2019. The results were classified into six classes: Water, No drought, Light drought, Moderate drought, Heavy drought, Severe drought. According to the classification, the southern half of Mostiștea Plain was the most affected area by a heavy agricultural drought during 2017-2019 period.