This video outlines four main methods to remap or interpolate gridded datasets in netcdf or grib format:
nearest neighbour interpolation
bilinear interpolation
bicubic interpolation
conservative remapping

We consider both the downscaling and upscaling of data and I warn about the potential danger of accidentally subsampling a field using some of the methods when interpolating to a coarse grid.

I then show how to quickly and easily carry out interpolation and remapping using climate data operators (CDO) giving an example using temperature data from ERA5. I introduce three ways to specify the new grid
simple global regular latlon grid
define a region and resolution in a text grid descriptor file
map directly one field to the grid of another

By the end of this video you should be handle the remapping of any gridded datasets in netcdf or grib format and thus be able to compare different climate or numerical weather prediction (NWP) models of different resolutions either with each other or with gridded satellite observational data.