johnrobertlawson / WEM

WRF Ensemble Management (USE EVAC)
30 stars 16 forks source link

=== WEM

NOTE - WEM has been refactored. Use evac (Ensemble Verification and Creation) here: https://github.com/johnrobertlawson/evac.git

WRF Ensemble Management can help you create WRF ensembles from GEFS reanalysis data via submodule lazyWRF. It can also post-process the data (such as computing ensemble means, creating postage-stamp plots of all members, etc) via submodule postWRF.

As a sidenote, I encourage all users of Python to cite packages and acknowledge fellow users in publications and presentations. There is often little incentive for researchers to open source their efforts, yet it is a process that needs to be encouraged.

Where do I start?

Documentation is here (incomplete): http://johnrobertlawson.github.io/WEM/

./lazyWRF/ contains scripts that form the basis of automating your WRF ensemble runs. ./postWRF/bin/ contains examples of post-processing that you may like to perform with the module. The other essential file you will need to personalise for post-processing is /bin/settings.py. The class therein contains all the settings for loading data, saving output, etc. Almost all settings can be left as default (by not specifying a setting), other than essentials like the path to your WRF data, the path to output figures, etc.

To run lazyWRF, the top-level script must be in your WPS folder to allow WPS executables to see the namelist.wps. So you might need to soft-link from your WPS directory to where you keep your top-level lazyWRF/WEM controlling scripts (e.g., ln -sf /path/to/WEM/scripts/ in your WPS folder). At least, I can't find a way around this.

Contributors & Attributions

Some files or methods contain attributions to other programmers whose code has been refactored to fit this project (or is/will become a prerequisite). In summary, thanks to:

SHARPpy

HootPy/PulsatrixWx project

URL: http://www.atmos.washington.edu/~lmadaus/pyscripts.html

PyWRF project (Monash)

URL: http://code.google.com/p/pywrf/

URL: https://github.com/scaine1/pyWRF/

PyWRFPlot project

URL: https://code.google.com/p/pywrfplot/