Paper Reviewed
Anandhi, A. and Nanjundiah, R.S. 2015. Performance evaluation of AR4 Climate Models in simulating daily precipitation over the Indian region using skill scores. Theoretical and Applied Climatology 119: 551-566.

In a paper published in Theoretical and Applied Climatology, Anandhi and Nanjundiah (2015) studied several Coupled General Circulation Models (CGCMs) that were utilized in the development of the Intergovernmental Panel on Climate Change's fourth assessment report (IPCC AR4). More specifically, they analyzed the several models' abilities to simulate daily precipitation over India on a 2.5° x 2.5° latitude by longitude basis throughout the 20th century. This they did by comparing the models' outputs with a similar but observational dataset for India that had been prepared by the National Climate Centre of the Indian Meteorological Department, based on measurements obtained at 2,140 rain gauge stations over the period 1951-2004.

And what did the two researchers thereby learn? Very simply, they found that (1) "no single model performs best for all the categories and zones considered," that (2) "most models underestimated the daily precipitation rate in the 0-1 mm/day range," and that they (3) "overestimated it in the 1-15 mm/day range," all of which suggests that if current climate models cannot accurately "predict the past," why should we believe they can accurately predict the future?