Useful bits of code that I want access to from all projects.
I use cartoplot for rapidly visualising arrayed data for the Arctic. It uses cartopy v0.19 and goes like this:
from cartoplot import cartoplot
cartoplot(lon,lat,data)
Where all inputs are arrays and data is one-shorter in both dimensions than the coordinates. You can specify some stuff with the keywords.
Allows masking into different Arctic regions using the MASIE grid.
Will get to cleaning up and documenting this soon hopefully, but it relies on mask.nc file
Converts lon/lat coordinates to xy coords with units of meters specified in the NSIDC ease grid (Lambert equal area). Super useful for calculating the distance between points. This code also uses the most recent version of pyproj which has the Transformer syntax, which lots of people haven't yet switched to (so may be a useful template). Also has the "inverse" keyword which goes the other way, xy to lon/lat.
I use this all the time, it just regrids data. Eg. polar stereographic to EASE. It takes the lon/lat coordinates of your original data (plus the data itself), and the target lon/lats, and converts. To do this it converts all the points to xy, because otherwise the warped space of lon/lat coordinates screws things up. Just a wrapper for scipy.griddata really, but handles the lon/lat issue conveniently.
This code also contains regrid_fast which regrids large 3D arrays by just constructing the triangular mesh once and using the same mapping over and over. I used this for my 2021 Comms E&E paper a lot. This is not just a wrapper for scipy.griddata.