H111-0033
A Meteorological Forcing Engine for Hydrology: Development and Applications

Friday, 11 December 2020
Poster
Yongxin Zhang1, Ryan Cabell2, Joe Grim2, Logan Karsten3, David Gochis2, Brian A. Cosgrove4, Cham Pham5, Zhengtao Cui6 and Ridwan Siddique1, (1)National Center for Atmospheric Research, Boulder, CO, United States, (2)NCAR, Boulder, CO, United States, (3)ClimaCell, Boulder, United States, (4)NOAA/NWS/OHD, Silver Spring, MD, United States, (5)NOAA/NWS/NWC, Silver Spring, MD, United States, (6)NOAA, National Weather Service, Silver Spring, MD, United States
Abstract:
A Python-based Meteorological Forcing Engine (MFE) has been developed at the National Center for Atmospheric Research (NCAR) for preparing meteorological input forcing variables for the WRF-Hydro model and its operational instantiation as the NOAA National Water Model. This Python-based MFE was built to replace a prior system for pre-preprocessing meteorological inputs built from an ad-hoc set of scripts in the NCAR Command Language (NCL).

The new MFE was developed in a way to easily customize and scale out complex interpolation, bias correction and downscaling processes using robust community software such as the Earth System Modeling Framework (ESMF) for regridding and the Message Passing Interface (MPI) for parallel processing. The MFE currently offers the NWM and WRF-Hydro users a highly-configurable package to easily prepare meteorological inputs to their modeling studies, and it can also be conveniently adapted to other hydrological modeling frameworks. The community is invited to contribute to this package as new forcing products are generated in the future, or additional products become desirable.

Driven by simple, user-specified configuration files, the WRF-Hydro MFE can handle multiple input forcing products, generate interpolation weight files between different model domains, run spatial interpolation from the input grids to the model grids using different approaches, and apply optional downscaling and bias corrections to any or all selected forcing variables. Some of the input data operators support ingest of historical regional and global reanalysis datasets that can be used for retrospective hydrological studies. Other input data operators support operational real-time numerical weather prediction (NWP) model products for real-time forecasting and nowcasting applications. The MFE is also designed to use supplemental precipitation products that offer spatially distributed quantitative precipitation estimates from sensors such as Doppler radar and satellites to enhance NWP estimated or reanalyzed precipitation where available.

In the presentation, the development and applications of the MFE over the CONUS will be introduced.

<< Previous Abstract | Next Abstract >>