Asymmetric Impacts Of ENSO Cycles On The Rainfall Over Southern China In Winter Half Year

In this paper, the asymmetric impacts of ENSO cycles on the rainfall over southern Chinain winter half year (November–April) are investigated based on observed and reanalysisdatasets by using various analysis and diagnostic methods. And the main results in this paperare examined with numerical simulation. The main conclusions are derived as follows:(1) The analysis results of observed rainfall data show that the impacts of El Ni(?) o andNi(?)a on the rainfall over southern China are asymmetric in winter half year. It’s found thatpositive rainfall anomalies with statistical significance appear over southern China in the ElNi o episodes. But in the La Ni(?)a episodes, no reverse rainfall anomalies with statisticalsignificance appear over southern China. The statistical significance of positive correlationsbetween ENSO and rainfall over Southern China are mainly contributed by their correlationsin El Ni(?) o episodes, and the relationships are weak in La Ni(?)a episodes.(2) The asymmetric responses of atmospheric circulations in lower troposphere to ElNi o and La Ni(?)a are the main mechanisms of asymmetric impacts of rainfall over southernChina, possibly. The anticyclone and cyclone anomalies over western North Pacific (WNP)and South China Sea (SCS) give rise to the anomalies of anticyclonic and cyclonic watervapor transport during El Ni(?) o and La Ni(?)a episodes, respectively. In the El Ni(?) o episodes,the important feathers are the obvious southwesterly anomalies along the northwestern marginof anticyclone anomaly over SCS and southern China. The enchanced southwesterliestransport more water vapor to southern China, and the convergence of water vapor oversouthern China increase the precipitable water and specific humidity in the atmosphere. Atsame time, these positive anomalies are statistical significance. It’s one of the main reasons ofpositive rainfall over southern China during El Ni(?) o episodes. But in the La Ni(?)a episodes, the important feathers are the weaker norhteasterlies anomalies along the northwestern marginof cyclonic anomaly over SCS and southern China. The northeasterlies weaken the watervapor transport to southern China, and the divergence of water vapor over southern Chinadecrease the precipitable water and specific humidity in the atmosphere. No reverse rainfallanomalies with statistical significance appear over southern China. Although these physicalelements are changed in opposite direction, its negative anomalies aren’t as obvious as thosein El Ni(?) o period. Above mentioned results are verified by using different methods withvarious datasets.(3) The possible physical mechanism of the asymmetric impacts of ENSO cycles on therainfall over southern China are investigated by analyzing the differences of IntraseasonalOscillation (ISO) between El Ni(?) o and La Ni(?)a in winter half year. In the El Ni(?) o episodes,the colder SSTs suppress the growth of convective activities over WNP and SCS. Theperturbation kinetic energy and ISOs over WNP and SCS are weak because of the inactiveconvective activities during the El Ni(?) o episodes. But in the La Ni(?)a episodes, the warmerSSTs benefit the development of convective activities over WNP and SCS. So the activeconvective activiites motivate the strong oscillation kinetic energy and ISOs. These results aretested and verified by the results of computing the OLR10~20d and20~50d oscillationvariances and perturbation kinetic energy in850hPa. The correlation relationships amongvarious indices also have similar conclusions.(4) A physical mechanism is proposed to explain the asymmetric responses to ENSOsignals of the rainfall over southern China in winter half year. In the El Ni(?) o episodes, theISOs activities over WNP and SCS are inactive, which exert little effect on the stronganticyclone anomaly over WNP and SCS in the lower troposphere in the winter half year.Therefore the transportation and convergence of water vapor could be maintained steadilybecause of the existing of anticyclone anomaly, inducing the positive rainfall anomalies withstatistical significance over southern China in the El Ni(?) o episodes. However in the La Ni(?)aepisodes, the active ISOs activities weaken cyclone anomalies over WNP and SCS atintraseasonal time scale. The interannual variation signals of ENSO are disturbed byintraseasonal variation signals over WNP and southern China. The interannual signals related to ENSO are weaker than the signals of intraseasonal variation. Consequently, the interannualsignals over WNP and SCS cann’t be responsed by the atmosphere, the statistical significancerelated to interannual signals of ENSO cann’t be tested. The results of phase composites showthat the active and strong ISOs lead to the emergences of anticyclone and cyclone anomaliesby turns in the La Ni(?)a episode. So the cyclone anomalies cann’t been maintained steadilybecause the interannual variation signals of ENSO are weakened by the signals of ISOs greatly.At the same time, the southerly and northerly anomalies periodically appear over SCS andsouthern China during the period of oscillation. Finally, the asymmetries impacts of rainfall onENSO cycles over southern China are possibly caused by the differences of ISOs over WNPand SCS between El Ni(?) o and La Ni(?)a in winter half year.(5) Using the retrospective forecasts dataset of CFSv2model, the climatic modes andseasonal cycles of rainfall, OLR and wind in850hPa are analyzed. And their characteristics ofISOs are diagnosed in this study. The simulation output of CFSv2show that the simulatedrainfall, OLR and wind fields match the main feathers with observed dataset. Theirdistribution patterns of simulated and observed coincide in climatic modes and seasonal cycles.The distribution and intensity of ISOs could be simulated in CFSv2. The modeling capabilitiesweaken gradually with longer lead times. The simulated rainfall, circulation anomalies andISOs in CFSv2coincide with observed dataset during El Ni(?) o and La Ni(?)a episodes. Somedeficiencies existing in model need be improved in the future update. The analyzed results inthis paper are verified by the model simulated output further.