Interannual Variability Of The Tropics And Its Response To The Global Warming

Tropical interannual variation is the strongest signal of the global climate change,and plays an important role in modulating the global climate.The interannual variation in the tropics can not only affect the neighboring area of the tropical oceans,but also influence the global climate through the atmosphere bridge.Thus,investigating the physical mechanism and predicting the tropical interannual variation are always the hot spots in the recent decades.In this study,by using different complexity coupled models,we investigate the model simulation and prediction skill for the tropical interannual variation and the response of the tropical Pacific Ocean to the global warming.First,an intermediate ocean-atmosphere coupled model is developed to simulate and predict the tropical interannual variability.Originating from the basic physical framework of the Zebiak-Cane(ZC)model,this tropical intermediate couple model extends to the entire global tropics,with a surface heat flux parameterization and a surface wind bias correction added to improve model performance and inter-basin connections.Several numerical tests were conducted to exam the model performance for the tropical interannual variation.The results of the forced run and the coupled run suggest that the model well reproduces the variabilities in the tropical Pacific and Indian basins,such as ENSO,IOBM and IOD.The relationship between the Pacific Ocean and Indian Ocean is close to the observation but the relationship between the Pacific Ocean and Atlantic Ocean is overestimated in the model,which may cause the systematic bias in simulating the Atlantic Nino.The results of a reasonable simulation of tropical interannual variation show that this model can be a powerful tool for studying and predicting tropical interannual variation.Next,by using a 136-year hindcast experiment,we investigate the prediction skill for the tropical interannual variation and the decadal variation of the predictability for ENSO and IOD.The results show that model can predict ENSO events at one year ahead similar to the skill of the latest version of the LDE05 model.In the tropical Indian Ocean,the predictability seems much higher in the west than in the east.The correlation skill of IOD prediction reaches 0.5 at a 5-month lead,which is comparable to that of the state-of-the-art coupled general circulation models.The predictability of ENSO was higher before 1980-2000 and the 1900s,and lower after 1960 and 2000.Similar to previous conclusions,ENSO intensity is an important factor affecting the ENSO predictability.The IOD predictability of IOD was highest during the perIOD of 1981-2000 and 1921-1940,and relatively low during the perIOD of 1941-1960 and 1961-1980.Unlike ENSO.the main factor affecting the predictability of IOD is the relationship between ENSO-IOD,while the effects of the strength of IOD and ENSO are re ative weak.Finally,we studied the tropical Pacific response to global warming using models of different degrees of complexity.Our results suggest that,in both the ZC model and the simple conceptual model,ENSO's response to global warming is consistent with the "oceanic thermostat" mechanism,which is characterized by an increase in the temperature gradient between the east and the west Pacific.The cause for this response is the stronger upwelling in the eastern Pacific Ocean than in the western Pacific Ocean.When considering the non-linearity of the subsurface temperature response to the thermocline anomaly,the rising thermocline in the eastern Pacific Ocean will weaken the response of SST to the change of thermocline depth and further weaken the intensity of ENSO.The bifurcation analysis of the conceptual model shows that there is a critical external forced heat flux for a particular coupling strength.When the external forced heat flux exceeds this critical value,the coupled system will change from self-sustained oscillation to a stable La Nina equilibrium state.The transition of this state will result in a rapidly nonlinear change in the temperature gradient.The current tropical climate state is near this bifurcation point,which makes the coupling system more sensitive to changes in forced heat flux and coupling strength.