Observed and predicted trends in tropical Pacific rainfall

The island nations of the tropical Pacific are particularly vulnerable to climate change. With their reliance on rainfall for nearly all of their fresh water, changes in rainfall patterns are of particular importance to Pacific Islanders. Rainfall is also a key component of atmosphere-ocean heat exchange here, and is thus an important driver of the weather in a region that in turn influences the global climate system.

Rain gauge observations have always been a primary source of rainfall data. The Pacific Rainfall Program (PACRAIN) rainfall database aggregates gauge data from a variety of sources into a homogenous format. Some of the observation sites have data going back to the 19th Century. While these long-term records are useful for climate studies, the use of any in situ data in this region is hampered by the large distances between sites. This is especially true for rainfall due to its highly localized and discrete nature.

Numerical forecast and climate models can be used to fill in the gaps in sparsely instrumented regions like the Pacific. Reanalysis products, such as those produced by NCEP/NCAR, use a forecast model to assimilate historical data and create objective analyses that are spatially and temporally homogeneous. This homogeneity makes them very useful for trend analysis. Unlike reanalysis products, climate models do not attempt to faithfully recreate or forecast the exact state of the atmosphere at a given time. However, they should be climatologically consistent and thus also useful for determining long-term trends.

PACRAIN gauge data and numerical model products will be used to diagnose ongoing changes in tropical Pacific rainfall with the aim of predicting future trends. These predictions can then be compared to climate model forecasts.