Biological and Environmental Research - Earth and Environmental System Sciences
Earth and Environmental System Modeling
18 July 2016

The Use of Fractional Accumulated Precipitation for the Evaluation of the Annual Cycle of Monsoons

A better approach to evaluate the annual cycle of precipitation irrespective of mean-state bias


We have developed a new approach to assessing how well coupled-ocean atmosphere climate models simulate the annual cycle of precipitation. The new approach circumvents problems that arise when threshold-based approaches are used, since in may cases model dry or wet biases would provide an incorrect estimate of the time of monsoon onset and/or withdrawal. For various monsoon domains there are systematic, but distinct errors in the time of monsoon onset.


Our new methodology provides an objective approach for analyzing the annual cycle of precipitation, indicating that regional process studies are necessary to investigate the regionally specific causes of model systematic error. Compared to threshold-based approaches our new methodology is better-suited to validation of the annual cycle of precipitation during model development.


Previous threshold-based techniques for the evaluation of monsoon onset and withdrawal are not applicable to models with dry biases. Our new approach evaluates the fractional accumulation of pentad precipitation (for a given pentad this is the accumulation of rainfall up to and including that pentad divided by the total annual accumulation). This technique provides a uniform approach for evaluating the annual cycle irrespective of biases in total annual rainfall or the amplitude of the annual cycle. In the CMIP5 historical simulations the onset of monsoon rainfall is delayed over India, the Gulf of Guinea, and for the South American Monsoon, with early onset prevalent for the Sahel and for the North American Monsoon. This, in combination with the rate at which precipitation is accumulated during the monsoon indicate how well the models concentrate rainfall over the peak rainfall season, and the extent to which there is a phase error in the annual cycle. The lack of consistency in the phase error across the different monsoon domains suggests that a “global” approach to the study of monsoons may not be sufficient to rectify the regional differences. Rather, regional process studies are necessary for diagnosing the underlying causes of why the models are not responding properly to the prescribed annual cycle of solar forcing.

Kenneth R. Sperber
Lawrence Livermore National Laboratory (LLNL)

K. R. Sperber was supported by the Office of Science (BER), U.S. Department of Energy through Lawrence Livermore National Laboratory contract DE-AC52-07NA27344. H. Annamalai was supported by the Office of Science (BER) U.S. Department of Energy, Grant DEFG02-07ER6445, and also by three institutional grants (JAMSTEC, NOAA and NASA) of the International Pacific Research Center. We acknowledge the World Climate Research Programme's Working Group on Coupled Modelling, which is responsible for CMIP, and we thank the climate modeling groups (listed in Table 1 of this paper) for producing and making available their model output. For CMIP the U.S. Department of Energy's Program for Climate Model Diagnosis and Intercomparison provides coordinating support and led development of software infrastructure in partnership with the Global Organization for Earth System Science Portals.