I haven't been putting much energy into the Global Warming changing climate debate lately. You can disappear from the lack of debate for a year or two and pretty much not miss anything important. The same old people are trying to use simple models to combat simple models and there are still "Global Average Surface Temperature Anomaly" food fights.
So imagine if you will, that the only observational information you had was Sea Surface Temperature from 30S to 30N and CO2 concentration change. From that you need to model the rest of the world "surface" temperature and estimate the CO2 radiant forcing impact on "global climate".
Now if you have your pet "ideal" simple model you can use that or you can sneak a peak at some of the other data if you like and use correlations in a statistical model of you choice. However, you cannot assume any land use impacts other than any you can support with your model, climate driven desertification or such.
30S-30N SST only by the way is 52% of the ocean area and 37% of the global area. Energy wise, it is in the ballpark of about 70% of the global energy.
KNMI. Now see how well you can "project" what happens in the rest of the world.