Run a pre-existing recipe

CSET works by running recipes that describe how to transform and and visualise data. It comes with a collection of pre-written recipes to get you started. In this tutorial you will use CSET to plot the mean surface air temperature of a forecast.

We will create a basic spatial plot of the mean surface air temperature. There is a pre-existing recipe for this that can be retrieved with the CSET cookbook command.

Try the following:

cset cookbook -o recipes mean_surface_air_temperature_spatial_plot.yaml

This will write out a recipes folder containing recipe .yaml file to your current directory. We will use the mean_surface_air_temperature_spatial_plot.yaml recipe.

Now you need to find some data to process. You can download an example file here, or with the following command.

curl -LO https://github.com/MetOffice/CSET/raw/main/tests/test_data/air_temp.nc

Now we are ready to run our recipe. This is where we use the cset bake command. This takes the input data file, an output path and the recipe file. The output should be a directory, but it will be created if it does not exist.

cset bake -i air_temp.nc -o output/ -r recipes/mean_surface_air_temperature_spatial_plot.yaml

This will run the recipe and leave its output in the specified output directory.

The most interesting output will be the plot, which you can look at with xdg-open output/plot.png.

You’ve now successfully run CSET with a pre-existing recipe. In the next tutorial we will see what is going on inside.