Assessing drought on agricultural productivity: a remote sensing approach to monitoring and adaptation

Droughts have various impacts depending on their severity and duration. In agriculture, drought can lead to crop failure and reduced yields. In this regard, it is imperative to conduct research on the effect of drought on productivity in agriculture. Data on crop yields, spanning from individual fields to global scales, is essential for farmers and policymakers. Existing sources of crop yield data, like regional agricultural statistics, frequently lack the necessary spatial and temporal detail. Vegetation indices (VIs) derived from remote sensing, such as NDVI (Normalized difference vegetation index), can effectively estimate crop yields through empirical modeling approaches. This study predicted crop yield by applying several indices by analyzing satellite images to estimate crop yield. The data to be used are open source Sentinel-2 imagery, python for regression analysis and the platform required to run is Google Earth Engine. Processing higher resolution images requires more computing resources than lower resolution images. Also, with the advent of cloud computing and open access computing portals such as Google Earth Engine, computing costs have decreased significantly. These technologies have made the processing of satellite images more economical. NDVI (measure of greenness of crops), NDMI (Normalized Difference Moisture Index) and SMI (Soil Moisture Index) data was calculated for several crop fields (area of Agsu region of Azerbaijan) for 3 years. A drought index was also applied to that area, and as a result, it was found that the productivity was low in the dry years. The aim here is to investigate the effect of climate change on crop productivity in Azerbaijan and study its effect on the economy