Estimating droughts in Central Asia based on several climate indices and climate factors

Droughts are one of the most severe natural hazards, causing extensive economical losses and often regional conflicts due to water scarcity. Central Asia is especially sensitive to droughts because the region heavily depends on water for hydropower and irrigation. The drought warning system is needed for water-related political discussions and decision-makers in the region to mitigate potential water governance and make action plans for agriculture depending on the water availability.

Central Asia has limited ground observations and additionally a lot of outdated measurement stations from the Soviet era, therefore the usage of remotely sensed data is beneficial in this region.

Here we use several drought indices to analyse historical droughts and possibly predict future droughts. Already developed indices such as Drought Severity Index (DSI) based on GRACE and GRACE-FO total water storage anomaly (TWSA) data and a widely used Standard Precipitation Index (SPI), based only on precipitation, were calculated. Additionally, other climate factors were investigated and their statistical relationship with DSI and SPI, such as groundwater and particularly, snow cover and snow water equivalent (SWE). We applied near real-time monthly TWSA data from GRACE/GRACE-FO and MSWEP (Multi-Source Weighted-Ensemble Precipitation). Daily snow cover data from MODIS was used and Copernicus products for groundwater and SWE.

In Central Asia, most of the water resources come from snow and glacier melt, coming from the Pamir, Tian Shan and Hindukush mountains, and the study was focused on assessing the snow cover and SWE in the winter months, particularly before the drought period. The goal was to understand and quantify the relationship between these climate factors and historical droughts based on DSI and SPI. Ultimately, the established relationship between climate factors and drought indices could provide an early warning for the upcoming arid and drought period. The obtained results show that drought conditions can be well identified using the remote sensing information in Central Asia.  The results also show that the region is experiencing more frequent drought conditions in the last 5 years.