subtract first 50 year historical mean rather than piC mean to get deltas for hist+ssp
From Norman: "the piC data not matching the hist+scenario levels, which creates problems when subtracting piC to obtain delta values (e.g. MRI-ESM2-0 for SSP126, which is about 10°C off for tas, pr looks also all over the place for various models). This is done in the notebook, under 'Application to new scenarios', in the second cell where 'td_ssp' is assigned. The plots are done 3/4 cells below that. As an intermediate solution, I have used the first 50 years of the historical for subtraction, which gives better results, and might even be a better solution overall. You should be able to reconstruct the issues from there. ."
subtract first 50 year historical mean rather than piC mean to get deltas for hist+ssp
From Norman: "the piC data not matching the hist+scenario levels, which creates problems when subtracting piC to obtain delta values (e.g. MRI-ESM2-0 for SSP126, which is about 10°C off for tas, pr looks also all over the place for various models). This is done in the notebook, under 'Application to new scenarios', in the second cell where 'td_ssp' is assigned. The plots are done 3/4 cells below that. As an intermediate solution, I have used the first 50 years of the historical for subtraction, which gives better results, and might even be a better solution overall. You should be able to reconstruct the issues from there. ."