Implementation of Remote Sensing-Based Methods to Map Evapotranpiration and Monitor Drought Conditions in Python

One of the key surface variables for hydrological applications, monitoring of natural and anthropogenic water consumption and closing energy balance and water budgets is evapotranspiration (ET). There is currently a strong need for high temporal and spatial resolution ET products for climate and hydrological modelers. In the same way, drought is a costly weather phenomenon since it significantly devastates agriculture, subsequently economy. It is crucial to monitor and predict drought for proactive decision-making purposes to mitigate its impacts. Remote sensing data and methods have been heavily utilized to monitor agricultural drought and map evapotranspiration rates on the Earth's surface. For these purposes, a fully-automated ET model was developed in Python programming language which can be driven by any satellite products such as the Landsat, the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Visible Infrared Imaging Radiometer Suite (VIIRS). Similarly, a set of independent tools were implemented in Python to download necessary satellite data, compute intermediate results and produce final drought maps. Later, these independent tools were streamlined into a large Python script to efficiently automate drought map production from the MODIS satellite products. Moreover, generic python functions were developed to form time series by extracting model results from hundreds of satellite images which will be subsequently used in the validation of ET and drought monitoring results against ground truth. Implementation details of these models will be discussed, and their results will be presented.