Xinyu ZHANG,Zhicong YIN,2,3,Huijun WANG,2,3,and Mingkeng DUAN

1Key Laboratory of Meteorological Disaster, Ministry of Education/Joint International Research Laboratory of Climate and Environment Change/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters,Nanjing University of Information Science &Technology, Nanjing 210044, China

2Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai 519080, China

3Nansen-Zhu International Research Centre, Institute of Atmospheric Physics,Chinese Academy of Sciences, Beijing 100029, China

Table S1.Definitions of the East Asian early winter (December and January) monsoon indices.Note:ALI,MW850I,and EADTI have already been multiplied by ?1,in order to ensure the consistency of high indices and high strength.

Fig.S1.(a) The emissions inventory (total black carbon,ammonia,organic carbon,sulfur dioxide,PM10,and PM2.5;unit:g;left) and HDs (unit:d;right) difference between January during 1980–2014 and December during 1979–2013 over the YRD.(b) As in (a) but for NC.(c) The December emissions inventory (total black carbon,ammonia,organic carbon,sulfur dioxide,PM10,and PM2.5;unit:g;left) and HDs (unit:d;right) difference between the YRD and NC during 1979–2013.(d) As in (c) but for January variables during 1980–2014.Numbers in the upper right are the CCs between two lines.The linear trends of all data were removed.

Fig.S2.The CCs between December HDs in NC and December (a) UV speed,(b) potential temperature and (c) RH from 1981 to 2017.White slanted lines indicate that the CCs represented by the shading exceed the 95% confidence level.The gray box indicates the location of NC.The linear trends of all data were removed.

Fig.S3.The CCs between January HDs in NC and the January (a) PBLH,(b) potential temperature and (c) RH from 1982 to 2018.White slanted lines indicate that the CCs represented by the shading exceed the 95% confidence level.The gray box indicates the location of NC.The linear trends of all data were removed.