^{Originally posted by krikru September 27, 2023}
If I have two iris cubes, A and B, both on the same grid and with the same time points, how can I calculate the intra-seasonal (according the seasons that are added when calling add_season) Pearson correlations between A and B for each grid point? Can I do this somehow by only using Iris functions (i.e., not extracting the data property from both cubes and using NumPy), and without using the definition of the Pearson correlation?

Pearson correlation can usually be calculated with pearsonr, but this function doesn't seem to enable aggregation by an auxiliary variable, and I need to aggregate by season.