Collaboration between scientists at UMBC–GEST and NASA–GSFC, the NCAR Research Applications Laboratory (RAL), and Baron Advanced Meteorological Services (BAMS), has produced a modeling framework for the application of traditional land surface models (LSMs) in a distributed hydrologic system that can be used for analysis and prediction of routed stream discharge hydrographs. This collaboration is oriented on near-term system implementation across Romania for flood and flash-flood diagnosis and forecasting as part of the World Bank-funded Destructive Waters Abatement (DESWAT) program. Meteorological forcing from surface observations, model analyses and numerical forecasts are employed in the NASA–GSFC Land Information System (LIS; Kumar et al. 2006) to drive the Unified Noah LSM with Noah-Distributed (Gochis and Chen, 2003) components, stream network delineation and routing schemes original to this work.

The Unified Noah LSM is the product of a joint modeling effort between several research partners including NCAR, the NOAA National Center for Environmental Prediction (NCEP), and the Air Force Weather Agency (AFWA). At NCAR, extensions to the Unified Noah LSM have been developed for LSM applications in a distributed domain in order to address the lateral redistribution of soil moisture by surface and subsurface flow processes. These advancements have been integrated into LIS and coupled with original formulations for hydraulic channel network definition and specification, linkages with the Noah-Distributed overland and subsurface flow framework, distributed cell-to-cell hydraulic routing, and a parametric baseflow accounting model.

We begin with an overview of the system components, including their formulations and organization, and then address results of the first U.S. case study performed with this system for a precipitation event over a headwater basin in the southern Appalachian Mountains. This event occurred in October 2005 following the landfall of Tropical Storm Tammy in South Carolina, following a long dry period in the region. Though these conditions supported the demonstration of watershed response to strong precipitation forcing under nearly ideal and easily-specified initial conditions, comparisons with simulations that included a spin-up period are also addressed. In the latter case, two years of NASA–NOAA North American Land Data Assimilation System (NLDAS) forcing data were employed for conditioning of soil moisture in the test watershed prior to the studied precipitation event. The results presented here will compare simulated and observed stream discharge conditions at various locations in the watershed, including both unregulated and modified stream segments.

In the test watershed, two significant precipitation episodes over two days are represented with 1-km NEXRAD and MM5 analysis-based input datasets. Results in the spatial distribution of soil moisture and at identified stream gauge locations demonstrate expected physical responses to the precipitation input, including valley-bottom surface saturation in the watershed and peaks in the simulated stream hydrographs that are well-correlated with the spatio-temporal distribution of precipitation. Issues in distributed hydrological modeling related to pre-processing of the topographic datasets, the representation of baseflow and general groundwater processes in the course of simulation, and accounting for regulated stream segments (e.g. withdrawals, reservoirs) are also discussed.

References

Gochis and Chen, 2003: Hydrological Enhancements to the Community Noah Land Surface Model, NCAR/TN-454+STR, 68 pp.

Kumar, S.V., C.D. Peters-Lidard, Y. Tian, P.R. Houser, J. Geiger, S. Olden, L. Lighty, J.L. Eastman, B. Doty, P. Dirmeyer, J. Adams, K. Mitchell, E.F. Wood, and J. Sheffield, 2006: Land information system: An interoperable framework for high resolution land surface modeling. Environmental Modeling & Software, v. 21, pp. 1402-1415.