Predictability and the relationship between subseasonal and interannual variability during the Asian summer monsoon

Abstract

The relationship between subseasonal and interannual variability of the Asian summer monsoon has been investigated through analysis of the dominant modes of variability in the 40‐year NCEP/NCAR Re‐analysis, with complementary satellite and surface‐based precipitation data. The hypothesis that the characteristics of monsoon subseasonal variability (i.e. weather regimes) are modulated on interannual time‐scales in a systematic and therefore predictable manner has been tested. The null hypothesis is that predictability of the seasonal mean monsoon behaviour requires that the effects of the slowly varying components of the climate system be correctly simulated. An interannual mode of monsoon variability has been identified which is closely related to the observed seasonal mean all‐India Rainfall (AIR). A counterpart of this mode has also been identified at subseasonal time‐scales which projects strongly on to the daily AIR, confirming that a common mode of monsoon variability exists on sub‐seasonal and interannual time‐scales. It has been shown that the temporal behaviour of this subseasonal mode, as described by the probability distribution function (PDF) of the principal component time series, does not show any evidence of bimodality, the shape of the PDF being Gaussian. Further, it has been shown that the mean of the PDF is systematically and significantly perturbed towards negative (positive) values in weak (strong) monsoon years as categorized in terms of the seasonal mean AIR. This translation in the mean of the PDF, rather than a change in shape of the PDF, suggests that anomalous monsoons are not associated with changes in weather regimes. Further analysis has confirmed that low‐frequency modulation of the basic state is primarily responsible for these shifts in the subseasonal PDFs, supporting the null hypothesis that predictability of the seasonal mean monsoon requires that the effects of the slowly varying components of the climate system be correctly simulated. Thus, model improvements to reduce systematic errors in the mean simulation and the response to low‐frequency boundary forcing may improve the prospects for dynamical seasonal prediction. However, the results indicate that only a subset of the subseasonal modes are systematically perturbed either by the El Nino Southern Oscillation or in weak vs. strong monsoon years, suggesting that predictability is likely to be limited by the chaotic, internal variability of the monsoon system.