Overview

Obtain and build the EMOSLIB, ecCodes, and FFTW software required to convert the ECMWF Nature Run GRIB files to NetCDF files containing data that has been converted to a latitude/longitude grid. More information about the Nature Run data can be found on the CIRA website.

Obtaining the code

Clone this repository to your local machine using git clone https://github.com/CSU-CIRA/ECO1280_environment.git

Run instructions

Environment set up

The conversion tools are intended to run in an environment described by the Dockerfile in this repository. To minimize the size of the final image, the Dockerfile is set up as a multistage build, with the first stage downloading and building the EMOSLIB, ecCodes, and FFTW software and the second stage shedding the build tools and artifacts and installing the conversion tools. To build and use the Docker image, your machine will need Docker installed.

Building the image

Build the image using the command docker build -t eco1280_convert:1.0 . Although the tag (-t) is optional, it will assign a name and version to the resulting image which will make it easily locatable and also provide a mechanism to have multiple versions available for testing and using upgrades. Note the . assumes you are working in the directory that contains the Dockerfile. To provide an alternate path, trade the period out for -f /path/to/the/Dockerfile

Creating a container from the image

Although the default Docker behavior will create a container that runs as root, it is best practice to declare your own user when you create the container to avoid messy security issues. For compatability with the host machine, you can use your user and preferred group ids in the container, or if compatability is not an issue, you may simply create a new user:group identification for the container. To get your user and group id on the host (provided that you are working in a linux-like environment), use id -u [user name] and id -g [group name].

To create and run a container from the image, use docker run -it --name eco1280_convert -u uid:gid -v /path/to/data/on/host:/path/to/data/in/the/container eco1280_convert:1.0, adding -v flags before the image name as necessary to provide any data mounts needed for the code to be able to access the necessary input data. Again, the --name is optional but makes the container easily locatable.

Entering the container

The -it flag in the docker build command will start an interactive session within the container, where you will be dropped in the /ECO1280_int_scripts directory where the conversion scripts are located. If the container is stopped (you can check by running docker ps), start it by running docker start eco1280_convert and then enter it using docker exec -it eco1280_convert /bin/bash.

Running the lat/lon converter

The main script for interpolation and conversion is convert2ll. It takes input file from standard in and its output location as a command line argument. A sample call looks like: echo /path/to/input/gxuz_ml_reducgg_0.grb | ./convert2ll -O /path/for/output/netcdf