The role of initial condition on sub-seasonal to seasonal drought forecast for South and Southeast Asia

A significant portion of the economy in South and Southeast Asia comes from agriculture, which is strongly influenced by hydroclimate extremes, such as droughts. In principle, a hydrological monitoring and forecasting system is critical to provide valuable information to help local governments and decision-makers monitor and forecast the progress of those extreme hydroclimate events. The South and Southeast Asia Hydrological Monitoring and Subseasonal to Seasonal Forecasting System (SAHFS-S2S) is established using Noah-MultiParamiterization (NoahMP) land surface model within the NASA Land Information System (LIS) Framework for such purpose. The monitoring system is driven by Climate Hazards Group InfradRed Precipitation with Stations (CHIRPs) precipitation and Global Data Assimilation System (GDAS) datasets, while the forecasting system is driven by NASA's Goddard Earth Observing System Model - Sub-seasonal to seasonal (GEOS-S2S) dataset downscaled by NCAR General Analog Regression Downscaling (GARD) algorithm.

In these Asian regions, however, the hydrological cycle is strongly influenced by human activity, such as irrigation, which causes groundwater depletion in areas heavily dependent on groundwater irrigation. In this study, we include an irrigation scheme in the SAHFS-S2S and GRACE data assimilation in the monitoring system to better estimate the hydrological cycles. A better monitoring system will provide the forecasting system with more accurate initial conditions, which are critical for a more accurate hydrological forecast in drought-prone areas. The evaluation of the system focuses on the soil moisture forecast with different initial conditions provided by the monitoring system with and without irrigation presence and GRACE data assimilation.