Skip to content

A scripting framework for running the WRF model, mainly designed to support operational forecasts.

License

Notifications You must be signed in to change notification settings

samwisehawkins/wrftools

Repository files navigation

wrftools

The master branch has changed significantly and now depends on a scheduler such as the Sun Grid Engine. This makes it much easier to run large historical simulations. A branch corresponding to the old master is available as the synchronous branch.

What is it?

A framework for running WRF simulations.

It is designed to be flexible and extendable. There are some tools out there which run WRF, but they are not easily modified. This is designed to provide a framework which is easily customised and modified.

Not yet python 3 compatible.

Install

It relies on two other custom python packages. Grab these and install as you usually would. (If you are working on Maestro, these are already installed)

$>git clone https://github.com/samwisehawkins/confighelper.git
$>python confighelper/setup.py install

$>git clone https://github.com/samwisehawkins/loghelper.git
$>python loghelper/setup.py install

Get the main repository

$>git clone https://github.com/samwisehawkins/wrftools.git

Quick start

There are three main scripts:

  • init.py intialises a base directory
  • prepare.py prepare simulations in seperate directories
  • submit.py submit simulations them to the scheduler

Assuming you have wrftools repository and the dependencies installed.

mkdir myforecast
cp /some/path/namelist.wps ./
cp /some/path/namelist.input ./
cp wrftools/config/init.yaml ./
python wrftools/init.py --config=init.yaml
python prepare.py --config=prepare.yaml
ls -l      # ooh, look at all those simulations waiting to go!
python submit.py --config=submit.yaml

Documentation

See doc/documentation.md for more details


Author Sam Hawkins samwisehawkins@gmail.com

About

A scripting framework for running the WRF model, mainly designed to support operational forecasts.

Resources

License

Stars

Watchers

Forks

Packages

No packages published