The seasonal predictability of the Asian summer monsoon in a two-tiered forecast system

Abstract

An extensive set of boreal summer seasonal hindcasts from a two tier system is compared with corresponding seasonal hindcasts from two other coupled ocean–atmosphere models for their seasonal prediction skill (for precipitation and surface temperature) of the Asian summer monsoon. The unique aspect of the two-tier system is that it is at relatively high resolution and the SST forcing is uniquely bias corrected from the multi-model averaged forecasted SST from the two coupled ocean–atmosphere models. Our analysis reveals: (a) The two-tier forecast system has seasonal prediction skill for precipitation that is comparable (over the Southeast Asian monsoon) or even higher (over the South Asian monsoon) than the coupled ocean–atmosphere. For seasonal anomalies of the surface temperature the results are more comparable across models, with all of them showing higher skill than that for precipitation. (b) Despite the improvement from the uncoupled AGCM all models in this study display a deterministic skill for seasonal precipitation anomalies over the Asian summer monsoon region to be weak. But there is useful probabilistic skill for tercile anomalies of precipitation and surface temperature that could be harvested from both the coupled and the uncoupled climate models. (c) Seasonal predictability of the South Asian summer monsoon (rainfall and temperature) does seem to stem from the remote ENSO forcing especially over the Indian monsoon region and the relatively weaker seasonal predictability in the Southeast Asian summer monsoon could be related to the comparatively weaker teleconnection with ENSO. The uncoupled AGCM with the bias corrected SST is able to leverage this teleconnection for improved seasonal prediction skill of the South Asian monsoon relative to the coupled models which display large systematic errors of the tropical SST’s.

Appendix: Signal to noise ratio

The signal to noise ratio is an objective measure of an AGCM’s predictability (Straus and Shukla 2000). It is basically a measure of the variance displayed by the ensemble mean relative to the ensemble spread of the seasonal hindcast. So higher values of this ratio correspond to higher predictability of the phenomenon by the AGCM. This measure of predictability however does not reflect on the verification of the hindcast or forecast. The ensemble mean for a given climate variable (say Y), for a given climate model, and for a given year j is:

$$ \overline{{Y_{j} }} = \frac{1}{K}\sum\limits_{i = 1}^{K} {Y_{{\overline{N} }} } $$

where i is the index for number of ensemble members and K is the total number of ensemble members (in our study it is 6).

The variance of the ensemble spread for a given year j is given by:

$$ \mathop \sigma \nolimits_{j}^{2} = \frac{1}{K}\sum\limits_{i = 1}^{K} {(Y{}_{{\overline{N} }} - \overline{{Y_{j} }} )^{2} } $$

The variance of the ensemble spread is a measure of the noise in the forecast system, which is averaged over all years of the hindcast to obtain:

$$ \mathop \sigma \nolimits_{noise}^{2} = \frac{1}{L}\sum\limits_{j = 1}^{L} {\mathop \sigma \nolimits_{j}^{2} } $$

The variance of the signal component is given by:

$$ \mathop \sigma \nolimits_{signal}^{2} = \frac{1}{L}\sum\limits_{j = 1}^{L} {(\overline{{Y_{j} }} - \overline{Y} )^{2} } $$

where

$$ \overline{Y} = \frac{1}{LK}\sum\limits_{j = 1}^{L} {\sum\limits_{i = 1}^{K} {Y_{{\overline{N} }} } } $$

Then, predictability (Π) or signal to noise ratio is defined as:

$$ \prod = \frac{{\sigma_{signal}^{2} }}{{\sigma_{signal}^{2} + \sigma_{noise}^{2} }} $$