This project will develop and demonstrate a scalable, rapidly deployable modeling system that couples high-resolution distributed hydrological models with convection-permitting atmospheric models to improve weather and water forecasting at installation scales. Convective precipitation remains the single largest gap in atmospheric understanding, as current models parameterize rather than resolve convection. Soil moisture heterogeneity drives convection initiation, yet this process is poorly represented in existing models because land-surface models are simplified and unable to represent lateral water movement or groundwater storage. By integrating a high-resolution distributed groundwater-surface water hydrological model with a convection-permitting atmospheric model, this work will capture the soil moisture-precipitation feedback mechanisms that drive convection and improve predictive skill across temporal scales from sub-daily to seasonal. Testing at the Jornada Experimental Range in New Mexico will validate the coupled system using the Army Research Laboratory's Meteorological Sensor Array and establish a proof-of-concept for future two-way coupling and multi-decadal forecasting.

The project will develop a one-way coupled groundwater-surface water-meteorological modeling system using high-resolution distributed hydrological models integrated with a convection-permitting Weather Research and Forecasting atmospheric model. The distributed hydrological model will simulate lateral surface and subsurface flow dynamics at sub-kilometer resolution, providing realistic soil moisture distributions that account for topography and groundwater storage. The convection-permitting atmospheric model will resolve primary circulation features of convective systems at 1-4 kilometer resolution, capturing both convective and orographic precipitation. Validation will leverage observations from the dense Meteorological Sensor Array at the Jornada Experimental Range, comparing model outputs against measured precipitation, wind, temperature, and humidity profiles. Sensitivity experiments will quantify uncertainties related to physiographic heterogeneity and evaluate model performance across sub-daily, seasonal, and decadal timescales.

This research supports installation operations by improving weather and water forecasting capabilities needed for resilience planning, land management, and operational decision-making. Current atmospheric and hydrological models operate independently, limiting their ability to represent soil moisture-precipitation feedbacks that drive convection and affect water availability. By demonstrating a coupled modeling system that captures these interactions, this capability will enhance predictive skill for flash droughts and convectively driven flash floods at subseasonal to seasonal timescales. The system will provide installation managers with actionable information across multiple temporal scales, from sub-daily operational decisions to seasonal resource planning. Results will also support conservation strategies for species dependent on specific moisture and temperature conditions. The proof-of-concept will establish a foundation for future two-way coupling and multi-decadal forecasting, enabling the Department of Defense to better understand and predict the complex interactions between land, water, and atmosphere that affect installation operations and natural resources. (Anticipated Project Completion - 2026)