Partitioning the variance between space and time

Here we decompose the space-time variance of near-surface air temperature using monthly observations for the global land surface (excluding Antarctica) from 1901-2000. To do that, we developed a new method for partitioning the total space-time variance, here called the grand variance, into separate spatial and temporal components. The temporal component is, in turn, further partitioned into the variance relating to different time periods and we use monthly data to decompose intra- and inter-annual components of the variance. The results show that the spatial and temporal components of the variance of near-surface air temperature have both, on average, decreased over time primarily because of reductions in the equator-to-pole (northern) temperature gradient, and because in cold regions, winter is generally warming faster than summer. We also found that in most regions, the inter-annual variance in near-surface air temperature has increased.