Improving Hydrological Representation in the Community Noah Land Surface Model for Intra-seasonal to Interannual Prediction Studies (GC07-075)

Project Summary:

Accurately predicting precipitation on intra-seasonal to interannual timescales in North America is critical for effective management of agriculture, water resources, and other societal needs, which is one of the primary objectives of the NOAA CPPA FY2007 Priorities.

Despite significant progress in understanding physical processes and mechanisms controlling predictability of precipitation, the prediction skill is still limited. This limited skill may be due to an inadequate data record for surface boundary conditions (e.g., soil moisture), deficiencies in model parameterizations (e.g., surface hydrology), and improper initialization techniques for the prediction system.

The proposed project will improve the prediction skill by

1. Improving land-surface hydrologic parameterizations controlling soil moisture memory,
2. developing high-resolution land surface assimilation datasets consistent with the improved hydrologic parameterizations and the coupled land-climate model system, and
3. documenting the impacts of the improved realism in these parameterizations on intra-seasonal to interannual precipitation prediction.

We propose to use the latest version of the unified Noah LSM, jointly maintained at NCEP and NCAR, to investigate how its augmentation with additional land memory processes (e.g. groundwater and dynamic vegetation) influences its soil moisture memory. We will apply the enhanced Noah in the continental USA to develop high-resolution datasets of land surface state variables (e.g., soil moisture) for a period of 2000–2004 in conjunction with NCAR’s high-resolution land data assimilation system (HRLDAS). We will then perform ensembles of simulations using the Weather Research and Forecasting (WRF) meso-scale model coupled to the Noah LSM. Initialized with HRLDAS land data, the simulations will illustrate the role of soil moisture, groundwater, vegetation, frozen soil, and snow in predicting precipitation at intra-seasonal to interannual timescales.

We selected Noah because of its long heritage and extensive applications. Through previous funding, we have already tested different versions of Noah with the first four of the following submodels:

1. a simple groundwater scheme,
2. a topography-based runoff production scheme,
3. a lateral routing scheme for overland flow,
4. a short-term dynamic vegetation scheme, and
5. a multi-layer physically based snowpack scheme.

The main model enhancement work in this project is to integrate these submodels, in a consistent fashion, into the officially-released offline version of the unified Noah LSM and to test the enhanced Noah with IHOP-02 and other datasets. One of the key deliverables is to transition the improved Noah to regional and global climate prediction models in NCEP pre-implementation test beds to demonstrate sufficient hindcast performance to warrant operational implementation. The research tasks will be closely coordinated between the investigators and scientists at the University of Texas at Austin, NCEP, and NCAR.

The proposed research is directly relevant to the CPPA FY2007 Priorities in “Predictability and Prediction Studies”. It is also relevant to other priorities including “Understanding and Predicting of Monsoon Systems in Americas”, “Improving Ensemble Hydrologic Prediction” and “CPPA Synthesis Projects”.