Auxilliary Workflows

Regionalization

The full regionalization workflow generates formulation and parameter assignments through nwm-region-mgr. RTE provides run_region.sh, run_regionalization.py, and sbatch_run_region.sh as execution wrappers for that workflow.

For workflow configuration, outputs, and Slurm guidance, see the documentation of the nwm-region-mgr repository.

Standalone Regionalized Forecasts

run_region_standalone.sh and run_regionalization_standalone.py do not generate regionalization parameters. They run an ngen forecast for a gage or VPU using formulation assignments, catchment groups, and parameters produced by a previously completed regionalization workflow.

Caveats for Regionalization Workflows

At runtime, host disk mounts occur within run_region.sh. These include various data directory mounts for inputs, static forcing data files, intermediary outputs, and output realization files.

Regardless of build instruction, forcing configuration files are sourced from the container path /ngen-app/ngen-forcing/NextGen_Forcings_Engine_BMI/BMI_NextGen_Configs/config_templates/.

Coastal Forcing

make_coastal_forcing.py runs the Forcing Engine in gridded mode to create atmospheric forcing inputs for coastal models. Example calls are provided by make_coastal_forcing.sh. See the nwm-coastal repository for complete workflow and input-data setup documentation.

Data Assimilation

run_output_postprocess.py and run_output_mosaic.py expose output-production and mosaicing workflows supplied by nwm-data-assimilation. See the nwm-data-assimilation repository for complete workflow and configuration documentation.