Effects And Prediction Of AMV And PDV On The Interdecadal Variation Of East Asian Summer Precipitation

Decadal prediction,which focuses on timescales between 10 and 30 years,is getting growing attention in recent years.Climatology of East Asia features remarkable interdecadal variability,Atlantic Multidecadal Variation(AMV)and Pacific Decadal Variation(PDV)are two major signals that influence the interdecadal variability of summer precipitation over East Asia.This study discusses the individual and combined effects of AMV and PDV on East Asian summer precipitation,evaluates the prediction ability of near-surface air temperature and precipitation over East Asia based on the decadal hindcasts from CMIP6 experiments and examines the influences of AMV and PDV on decadal variability of East Asian summer rainfall in the hindcasts of CMIP6.The main conclusions are as follows:(1)AMV mainly affects the interdecadal variability of summer precipitation over the Huang He-Huai River valley(HHRV),PDV influences both the Yangtze River valley(YRV)and the HHRV,and PDV plays a dominant role.AMV induces the anomalies of atmospheric circulations from Lake Baikal to the Sea of Okhotsk through the AMV-Northern Hemisphere(ANH)teleconnection,while the sea surface temperature anomalies(SSTAs)in the Northwest Pacific linked with PDV cause the anomalies of atmospheric circulations from southeast of Japan to the South China Sea(SCS).When the AMV and PDV are in the same phase,there are opposite anomalies of atmospheric circulations between Lake Baikal to the Sea of Okhotsk and southeast Japan to the SCS.In this situation,the Pacific-Japan(PJ)teleconnection enhances these two anomalies,which in turn amplifies precipitation anomalies over the YRV and the HHRV.When AMV and PDV are in opposite phases,the same anomalies of atmospheric circulations occurr in Lake Baikal to the Sea of Okhotsk and southeast Japan to the SCS,the precipitation anomalies are weaker than the in-phase combination of the AMV and PDV.(2)The multi-model ensemble(MME)mean of the decadal hindcasts in CMIP6 can capture the variations of annual mean and seasonal mean near-surface temperature that have not been detrended,the prediction ability of the seasonal mean near-surface temperature is much lower than the annual mean.MME's prediction ability of detrended near-surface temperature is significantly reduced,the prediction skill of the temperature during autumn mainly originates from the reproduction of the long-term trend.The MME decadal hindcasts show less skill over precipitation than temperature,the gap between prediction abilities of annual and seasonal mean precipitation is narrower than temperature.The difference between the prediction skills of un-detrended and detrended precipitation is also smaller than temperature,there is the poorest skill over summer precipitation among the four seasons.The2019-2023 prediction of near-surface temperature and precipitation indicates a 1.13? national average warming with respect to 1970–2010 average,and increasing precipitation with a national average 0.16 mm/day above climatology.(3)Decadal hindcasts can capture the phase transition time of AMV and PDV roughly,the SSTAs over the North Atlantic corresponding to AMV in decadal hindcasts are consistent with the observation,while SSTAs over the northwestern Pacific discord with observation.Under individual and combined AMV and PDV phases,aside from AMV-/PDV-phase,the summer precipitation anomalies in hindcasts consist with observation over HHRV,but discord with observation over YRV.As for atmospheric circulations,decadal hindcasts only reproduce the ANH teleconnection as well,which demonstrate AMV mainly affects summer precipitation over HHRV,and SSTAs over the northwestern Pacific corresponding to PDV play a key role in summer precipitation over YRV and HHRV.