The National Digital Forecast Database (NDFD) is stored in GRIB2 format on the NDFD Server, which keeps the NDFD GRIB2 files up-to-date and accessible. In many cases, you don't work with GRIB2 files directly--instead, we convert them to NetCDF4 files (which are slightly more friendly). The conversion process can be a headache, though.
This set of functions makes working with GRIB2 files a little easier. It provides a way to get a list of available NDFD areas along with the forecast time ranges and variables available for each area. From these available parameters, we can download data directly from the NDFD servers and load individual variables into Pandas Dataframes. Variables are stored locally to make everything work a little faster.
This project works mainly with NDFD Grid Data, but NDFD Point Access Data is available through a series of XML calls. I'm not exactly sure what you need for your project, but I'd be happy to create an example using the Point Access Data portal.
To download the necessary requirements, git clone this repository and run this command:
pip install requirements.txtThis should work in a Conda environment to, but I'm not certain of that.
The cfgrib library, which we use to load GRIB2 files, may require some additional tools for setup. If you are running MacOS, you can install this extra stuff using this command:
brew install eccodesOr, if you use Conda for package management:
conda install -c conda-forge eccodes