Lesson S3 - Regridding NLDAS forcing data to the WRF-Hydro domain

Overview

In this lesson, we cover regridding NLDAS2 forcing data to a WRF-Hydro domain using the Python-based WRF-Hydro Forcing Engine

Required software and datasets

Meteorological Forcing Engine (MFE) software

The WRF-Hydro Meteorological Forcing Engine (MFE) software can be downloaded from https://github.com/NCAR/WrfHydroForcing

Data sources

For this training, data has already been included for the simulation time period.

NOTE: These data are for instructional purposes only and do not include forcing data used for model spinup

If you would like to acquire other data, NLDAS2 forcing data can be retrieved from the NASA data server at https://hydro1.gesdisc.eosdis.nasa.gov/data/NLDAS/NLDAS_FORA0125_H.002/.

See the following site for information on how to access data: https://disc.gsfc.nasa.gov/data-access

See the following instructions for how to download data files using wget: https://disc.gsfc.nasa.gov/information/howto?title=How%20to%20Download%20Data%20Files%20from%20HTTP%20Service%20with%20wget

Step 1: Extract the NLDAS2 forcing data provided with this tutorial

Step 2: View the contents of the input_files directory

Let's view the data in the NLDAS2 inputs:

Step 3: View the domain metadata files

The Forcing Engine requires two additional static input files; a WRF geogrid file (for grid information), and an optional Spatial Metadata template file to copy attributes from.

We have the necessary files, now we can begin the regridding process.

Editing the forcing engine configuration file

The WRF-Hydro Forcing Engine requires a configuration file to specify the input types, forecast time periods, and other options needed for regridding and pre-processing the forcing data.

Step 1: Examine the supplied nldas_fe.config configuration file and verify the times and forcing input types

Regridding the forcing data using the forcing engine

In this section we will regrid all our forcing data using the configuration shown above

Step 1: Run the Forcing Engine driver (genForcing.py)

genForcing.py takes 3 arguments: the path to the configuration file, a version number, and a configuration name. The version and name are arbitrary and used for logging and are inserted as netCDF Global Attributes in the LDASIN output files.

The Forcing Engine sends most of its output to a log file specified in the configuration file. Only severe errors are printed to stdout so a successful run will produce no output.

Step 2: View the Forcing Engine Log:

Step 3: View the contents of the output_files directory

The Forcing Engine will generate output files in a directory heirarchy rooted in output_files. Lets view the contents of the directory:

Step 4: View the contents of the LDASIN files

Now, let's look at the actual regridded data in the output files. Try changing the variable from "RAINRATE" to others in the file to see the other fields.

© UCAR 2020