Separate out tropics and extratropics

[apendergrass]

Calculate precip mean and std change, and huss change, for three regions (all land):

• tropics (< |30° lat|)
• NH XT (>30° lat)
• SH XT (<-30° lat)
  And for two seasons:
• DJF
• JJA

For the plots, average the NH XT, DJF with SH XT, JJA and vice versa to get XT land in winter and XT land in summer

[apendergrass]

Looking back at original code, the way this happened was like this (matlab code):

cmip5prstd(itime,iind,:)=sqrt(regionseason(prdjfstd.^2,prjjastd.^2,lat,cella,landmask));

where prdjfstd and prjjastd are maps of the std of pr for DJF and JJA for some time period.

Then the relevant parts of regionseason are

function [regsea,names]=regionseason(djf,jja,lat,cella,landmask)

summer=jja;
winter=djf;
summer(:,lat<0)=djf(:,lat<0);
winter(:,lat<0)=jja(:,lat<0);

nm = @(field) field./nansum(nansum(field));
ss = @(field) nansum(nansum(field,1),2);

laf=cella.*landmask;
lmmax=max(max(landmask));

nlon=size(cella,1); 
latmat=repmat(lat',[nlon 1]);
trop=abs(latmat)<=30; 
xt=abs(latmat)>30; 
nhxt=latmat>30; 
shxt=latmat<-30;

regsea=NaN(32,1); 
names=cell(32,1);

...

m=nm(trop.*laf);
regsea(3)=ss(djf.*m);
regsea(4)=ss(jja.*m);
names{3}='Trop land DJF';
names{4}='Trop land JJA';

...

m=nm(xt.*laf);
regsea(7)=ss(winter.*m);
regsea(8)=ss(summer.*m);
names{7}='XT land Winter';
names{8}='XT land Summer';
...

Then the changes are calculated afterward, from those values for each time period.

nt=size(cmip5tas,1);
ntm1=nt-1;

dtascmip5=cmip5tas(2:end,:)-repmat(cmip5tas(1,:),[ntm1 1]);

dcmip5prstd=cmip5prstd(2:end,:,:)./repmat(cmip5prstd(1,:,:),[ntm1 1 1]);
dcmip5prmean=cmip5prmean(2:end,:,:)./repmat(cmip5prmean(1,:,:),[ntm1 1 1]);
dcmip5qmean=cmip5qmean(2:end,:,:)./repmat(cmip5qmean(1,:,:),[ntm1 1 1]);

keepmodels=find(~isnan(cmip5qmean(1,:,1)));

mmmdcmip5prmean=mean(dcmip5prmean(:,keepmodels,:),2);
mmmdcmip5prstd=mean(dcmip5prstd(:,keepmodels,:),2);
mmmdcmip5qmean=mean(dcmip5qmean(:,keepmodels,:),2);
mmmdtascmip5=mean(dtascmip5(:,keepmodels),2);

So to recap, the maps for DJF and JJA for each time period, then combined with the landmask and areacella to get tropical DJF+JJA, XT winter+summer land mean values for that time period, and then all the changes are calculated relative to the base period.

This is for std, but I used the same function (+approach) for mean pr and mean huss, without the RMS part.