Analysis Of Mid-summer Rainfall Patterns Over Northern China Monsoon Region And Projection Its Changes By CMIP6

The northern China monsoon region(NCMR)covering a vast area(including North China,Northeast China and eastern Inner Mongolia),is an important agricultural,industrial and energy base in China.Because of its location on the northeast of the eastern China,the weather and climate are affected not only by the tropical and subtropical monsoon circulation systems but also related to the mid-to high-latitude circulation,which result the large interannual variability of summer precipitation,and the short term climate prediction is difficult.In recent years,The real-time forecast accuracy of summer precipitation are significant low in the NCMR.Especially,when the prediction of the northern rainfall pattern(rainfall pattern I),it is still difficult to determine whether the main rain belt in North China or Northeast China,due to the classification of summer rainfall patterns that is too rough in the past research over the NCMR.In addition,under the background of global warming,previous studies have shown that NCMR is more sensitive to warming than southern China,and the frequency of extreme precipitation and heavy rainfall will show increase trend.Therefore,it is urgent to strengthen scientific research on the the location of the rain band in the NCMR.At present,many climate models have been participated in CMIP6,in order to scientifically improve and utilize climate models,the systematical evaluation of midsummer precipitation in the NCMR,which is critical by CMIP6 models.In this study,the mid-summer rainfall patterns are objectively classified in the NCMR during 1961-2019,and their associated with atmospheric circulation and SST evolution are analyzed,in order to investigate their formation mechanisms;the instability of interannual relationship between early summer tropical eastern Pacific(TEP)SST and North China midsummer precipitation(NCMSP)and its behind mechanism have been studied;A time-scale decomposition(TSD)approach of physical statistical prediction model is described for midsummer precipitation in the Northeast China,which is a new forecasting technology and improve the prediction skill.Besides,this paper also evaluates the performance in reproducing climatological spatial distribution and major modes of mid-summer precipitation by CMIP6 models over NCMR.Finally,the sources of projection uncertainty by CMIP6 models are investigated.The main conclusions are as follows:(1)Classification of typical midsummer rainfall patterns in the NCMR and their associated with the atmospheric circulation and SST evolutionThe midsummer rainfall patterns in the NCMR during 1961-2019 are objectively classified into four typical categories: pattern A,pattern B,pattern C,and pattern D.Pattern A(B)corresponds to the positive(negative)anomaly midsummer precipitation in the whole area,while pattern C(D)corresponds to positive(negative)anomaly precipitation in North China(Northeast China).The results confirm that the different rainfall patterns possess obvious differences.Pattern A: The East Asia subtropical westerly jet stream(EAJS)shows a northward trend relative to its normal position.The ridge line of the western Pacific subtropical high(WPSH)is located to the north of its normal position,accompanying stronger southerly winds,and the zonal circulation is dominant in the middle and high latitudes of Eurasia.There are significant convergences(divergences)at the low(high)level in the troposphere,cold and warm air convergence in the NCMR.Besides,the negative phase of the North Atlantic SST tripole and the eastern pattern of La Ni?a develop from the previous winter to summer.Meanwhile,the characteristics of atmospheric circulation and SST evolution of pattern B are basically opposite to those of pattern A.Pattern C(D):EAJS shows an anomalous northward(slight southward)trend relative to its normal position,and the longitudinal position of WPSH is located westward(eastward),their ridge lines both tend to be northward.East Asian-Pacific(EAP)teleconnection shows a typically positive phase.North China is controlled by stronger southwest(northwest)winds,while northeast(southeast)winds prevail over northeast China.The lower troposphere converges(diverges)and the upper troposphere diverges(converges)in North China,while the lower troposphere diverges(converges)and the upper troposphere converges(diverges)in northeast China,accompanying a weak(active)northeast cold vortex.Meanwhile,the warm SST transforms to cold SST(El Ni?o pattern gradually forms)in the middle eastern tropical Pacific.(2)Instability of the interannual relationship between NCMSP and early summer TEP SSTEnhanced interannual relationship between NCMSP and early summer TEP SST after the late 1970 s.For 1951-1972,the relationship between early summer TEP SST and NCMSP were statistically insignificant.In contrast,for 1977-2018,they were positively significant.Since the late 1970 s,the early-summer negative TEP SST anomaly was generally followed by a significant anomalous anticyclone over the Korean Peninsula and its environs in the following midsummer,which facilitated anomalous southerly winds over North China(NC),conveying more moisture from the Western North Pacific(WNP),leading more precipitation over NC.Further studies show that the enhanced relationship between early summer TEP SST and tropical Indian Ocean to WNP SST are suggested to be responsible for significantly anticyclonic anomalies over WNP after the late 1970 s.(3)A TSD approach of physical statistical prediction model is established for midsummer precipitation in the Northeast China.The downscaling model for the interannual rainfall variability is closely related to the WPSH and SST evolution from previous winter to following summer over the tropical Indian Ocean and North Atlantic,while the one for the interdecadal rainfall variability is related to the sea level pressure(SLP)gradient from east to west over southern Hemisphere,and the strength of Siberian(Masklin)high in the winter(spring).The results show that the time-scale decomposition(TSD)approach achieved a good skill to predict the interannual variability of precipitation observation in the independent validation period(2001-2009)by potential optimal interannual predictors in the winter and spring,and the annual mean of root-mean-square-error(RMS)is only 9.8% and 10.8% corresponding to the climatological precipitation,and the anomaly sign consistency rate(ASCR)is 84.3% and 73.7%, respectively;The interdecadal of spring predictors are better than those of winter predictors,which are good skill to predict the interdecadal midsummer precipitation in Northeast China.Taking the downscale interannual and interdecadal components together which are equivalent to total midsummer precipitation in Northeast China.Besides,The forecast skill of TSD approach of physical statistical prediction model for midsummer precipitation in the Northeast China is significantly enhanced compared with CSM1.1,CFSv2 and ECMWF4 operational models.(4)Evaluation of the performance on the simulation climate and major modes of mid-summer precipitation by CMIP6 models over NCMR.Using the 30 global climate models(GCMs)from Coupled Model Intercomparison Project Phase6(CMIP6),we evaluates the performance in reproducing climatological spatial distribution and major modes of mid-summer precipitation over NCMR.Based on the evaluation indexes such as spatial pattern coefficient,root mean square error,interannual variability score and relative deviation,the four models can well capture the spatial distributions of mid-summer precipitation over NCMR,which are MPI-ESM-1-2-HAM,MPI-ESM1-2-HR, MPI-ESM1-2-LR and E3SM-1-0,respectively;The models which are the good skill to comprehensively evaluate the climatological spatial distribution,EOF1(EOF2), PC1(PC2)and interannual variability of mid-summer precipitation over NCMR are NESM3,E3SM-1-0,GFDL-ESM4 and MMI-ESM1-2-LR(EC-Earth3-Veg,EC-Earth3,E3SM-1-0,NESM3 and MPI-ESM1-2-HR),respectively.(5)The future changes of mid-summer precipitation over NCMR and its sources of projection uncertainty under four scenarios are investigated by CMIP6 modelsThe projection of mid-summer precipitation trend is slight increasing in the near-term(2021-2040),maximum in the mid-term(2041-2060)and slowing down in the long-term(2080-2099)over NCMR,and its mean increasing trends are projected to be 3%(5%)under different scenarios in the near-term(middle-term),and slightly decreases to 2% in the long-term.Especially,it is sensitive to temperature change under SSP1-2.6 scenario,which may be related to the small amount of temperature change under this scenario.Under the background of global warming,the optimal model projection enhancement of NCMSP is caused by dynamic responses due to increased circulation,which is partly offset by thermodynamic responses.The projection uncertainty of mid-summer precipitation is the largest under SSP5-8.5 in the long-term over NCMR.