13A.2
Characteristics of the CMIP3 models simulating realistic response of tropical western Pacific precipitation to Nino3 SST variability

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Thursday, 27 January 2011: 1:45 PM
Characteristics of the CMIP3 models simulating realistic response of tropical western Pacific precipitation to Nino3 SST variability
608 (Washington State Convention Center)
Tomoaki Ose, MRI/Japan Meteorological Agency, Tsukuba, Ibaraki, Japan; and O. Arakawa

The 20th century simulations from the World Climate Research Programme's Coupled Model Intercomparison Project phase 3 (CMIP3) multi-model dataset are examined statistically using numerical skills to capture the characteristics of the models which realistically simulate responses of the tropical western Pacific (TWP) precipitation to sea surface temperature (SST) variability over the Niņo3 region [150°-90°W, 5°S-5°N].

The simultaneous correlation of precipitation anomaly over the TWP region of [90°-170°E, 20°S-20°N] with Niņo3 SST variability is successfully reproduced with relatively high skills for June to August (JJA) as compared with those for December to February (DJF) and March to May. The high skill models have common characteristics of realistically simulating the observed largest precipitation response to Niņo3 SST variability over the equatorial central Pacific east of the dateline. Furthermore, realistic simulation of the climatological mean equatorial precipitation west of the dateline seems to be responsible for the realistic response of the TWP precipitation to Niņo3 SST variability.

A few of the models successfully simulate the delayed response of the JJA precipitation over the TWP region of [90°-170°E, 10°S-30°N] to the preceding DJF Niņo3 SST variability with high skills. Those models reproduce statistically observed features of subtropical northwestern Pacific SST anomaly, precipitation and SST anomalies over the Indian Ocean in JJA following DJF Niņo3 SST variability. Another distinctive characteristic of those models is to reproduce almost null correlation of the equatorial central Pacific SST anomaly in JJA with the preceding DJF season Niņo3 SST variability.