Just to rephrase, you are averaging sst over the dimensions for lat and lon within a certain range. Then create 12 plots for those averages changing year to year for a given month.
It looks like you are using times 26:386 instead of over all 431 timesteps in this data.
So, we can create a datetime array of those times.
ts = datetime(outtime_string(26:386,:));
Then loop through the months and create the plots.
for i=1:12
figure;
plot(ts.Year(ts.Month == i), sst_timeseries(ts.Month == i))
title(['Month ' num2str(i)])
end
