Future changes in the frequency and intensity of drought and flood events will prove critical to the resilience of many communities and ecosystems. Hence, quantifying and understanding the large uncertainty in projected changes in local precipitation is a key research issue in climate science. This study aims to map, and begin to understand, the relative roles of uncertain modeling and natural variability, a topic that has received little attention to date. A number of questions remain, as follows. What is the split between these uncertainty sources at spatial scales closer to the size of the societies and ecosystems threatened by changing precipitation? How important is the size of the ensemble for this type of analysis? What can it teach us about the uncertain processes and mechanisms that produce such a range of precipitation responses, and how do these mechanisms vary spatially?

Data from four ensembles are used here. Three are perturbed physics ensembles founded on one of the Met Office Hadley Centre models, and the fourth is a multi-model ensemble, utilizing data from the CMIP3 archive. The largest – 280-member – ensemble illustrates a rich pattern in the varying contribution of modeling uncertainty, with similar features found using the CMIP3 ensemble (despite its limited sample size, which restricts it value in this context).

The contribution of modeling uncertainty to the total uncertainty in local precipitation change is found to be highest in the deep tropics, particularly over South America, Africa, the east and central Pacific, and the Atlantic. In the moist maritime tropics, the highly uncertain modeling of sea-surface temperature changes is transmitted to a large uncertain modeling of local rainfall changes. Over tropical land and summer mid-latitude continents (and to a lesser extent, the tropical oceans), uncertain modeling of atmospheric processes, land surface processes and the terrestrial carbon cycle all appear to play an additional substantial role in driving the uncertainty of local rainfall changes. In polar regions, inter-model variability of anomalous sea ice drives an uncertain precipitation response, particularly in winter. In all these regions, there is therefore the potential to reduce the uncertainty of local precipitation changes through targeted model improvements and observational constraints.

In contrast, over much of the arid subtropical and mid-latitude oceans, over Australia, and over the Sahara in winter, internal atmospheric variability dominates projected precipitation changes. Here, model improvements and observational constraints will have little impact on the uncertainty.