Proposed herein is an effort to establish a regional S2S prediction system for the state of texas that is capable of producing forecasts of precipitation and streamflow anomalies for weeks 3-6. The…

The University of Texas at Arlington (UTA) received a $638,275 research grant award from a civilian federal agency with an ultimate completion date of February 14, 2027. The award supports development of a regional Subseasonal-to-Seasonal (S2S) prediction system for Texas designed to forecast precipitation and streamflow anomalies for weeks 3-6. The technical approach couples a regional implementation of the Weather Research and Forecasting Advanced Research WRF-ARW model with the National Water Model, nested within the National Weather Service Global Ensemble Forecast System (GEFS)-V12 grid, with domain coverage extending from the Gulf of Mexico to Oklahoma/Arkansas and from eastern Mexico to Alabama. The system will be reinitialized using remotely sensed data including SMAP soil moisture, MODIS skin temperature and vegetation indices, and MODIS sea surface temperature data.

The project will execute two sets of forecasts using GEFS-V12 control runs with and without remotely sensed land condition data assimilation to quantify data infusion impacts. Hindcast experiments will focus on historical extreme events including the 2015 flood, 2017 Hurricane Harvey, and the 2019 flash drought. Model forecasts will be validated against surface observations and compared to National Weather Service S2S streamflow and soil moisture products to assess skill improvements from enhanced process representations and data assimilation techniques. The contract carries no set-aside designation and will be performed at UTA's Arlington, Texas location under the Office of Research Administration Division.

Proposed herein is an effort to establish a regional S2S prediction system for the state of texas that is capable of producing forecasts of precipitation and streamflow anomalies for weeks 3-6. The system will be based on a coupling of a regional implementation of WRF-ARW with the national water model (NWM). In the proposed undertaking, the WRF-ARW will be set up for a domain that extends from the gulf of mexico to the oklahoma/arkansas, and from the eastern portion of mexico to alabama IT will be nested within the NWS global ensemble forecast system (gefs)-V12 grid. The coupled system will be reinitialized using remotely sensed soil moisture from smap, modis skin temperature, dynamic vegetation, and modis sea surface temperature. Two sets of forecasts will be produced running the coupled modeling system using gefs-V12 control run as lateral boundary conditions. The first/second control run will be done without/with any infusion of remotely sensed land conditions to identify the impacts of data assimilation. The hindcast experiment will focus on windows surrounding the flood of 2015, hurricane harvey of 2017, and the flash drought of 2019. Forecasts of precipitation and other variables from the coupled model will be validated against surface observations and analyses. These forecasts will also be gauged relative to NWS S2S streamflow and soil moisture products to determine skills gained through improved representations of processes and data infusions.