Evaluating the TRMM 3B43 monthly precipitation product using gridded raingauge data over Australia

Abstract

This work compares two monthly rainfall datasets for Australia for the period between January 1998 and December 2010: The TRMM (Tropical Rainfall Measuring Mission) and Other Sources Rainfall Product (3B43) and the Australian Bureau of Meteorology (BoM) monthly gridded values based on raingauge data. The spatial variability in the cross correlation between each dataset over time reveals values > 0.90 for areas with denser ground-based networks of raingauges and higher rainfall, falling to around 0.20 for more arid areas with fewer raingauges. There appears to be some variation in the cross correlation over time, with values on average decreasing slightly after 2004, accompanied by increased scatter. The application of principal component analysis confirms the high correlation between these rainfall products, demonstrated by very similar spatial and temporal variability patterns. Histograms reveal that both rainfall products follow an almost identical probability distribution. The austral summer appears to show the strongest correlation between datasets, with the other seasons being slightly less correlated, but very similar to each other, although the differences between all seasons are minor with cross correlation values > 0.80 for Australia as a whole. Regarding the correlation between datasets as a function of climate, the steppe and temperate climate zones show a higher correlation than the tropical and desert zones. It is therefore proposed that the TRMM 3B43 monthly product would be appropriate for use in climate studies in Australia, keeping in mind that some discrepancies between the datasets for individual months arise. It is also potentially of value for the areas of Australia with fewer raingauge stations