“…Comparison of proxy records of sea surface temperature (SST) gradients and sea surface salinity (SSS). (a) SST anomalies of the western equatorial Pacific (WEP) (MD2140, ODP806; de Garidel-Thoron et al, 2005; Medina-Elizalde & Lea, 2005), ODP871(Dyez & Ravelo, 2013 this study), mid-NH (ODP1208, ODP1020, ODP1021, ODP1014, ODP1012;Brierley et al, 2009;Dekens et al, 2007;LaRiviere et al, 2012;Liu et al, 2008) and mid-SH (DSDP593, ODP1125, ODP1123, ODP1237;Crundwell et al, 2008;Dekens et al, 2007;McClymont et al, 2016;Peterson et al, 2020); (b) meridional SST gradient anomaly(Fedorov et al, 2015) and SST gradient anomaly between the WEP and northeast Pacific (NEP, ODP1020, ODP1021, ODP1012, ODP1014), and SST gradient anomaly between the eastern equatorial Pacific (EEP, U1337 and ODP846;Herbert et al, 2010;Liu et al, 2019) and southeast Pacific (southeastern Pacific, ODP1237;Dekens et al, 2007); (c) SST gradient anomalies between northern and southern margin of WPWP; (d) SSS records in the WPWP; (e) East Asian summer monsoon rainfall index in the Chinese Loess Plateau(Meng et al, 2018); (f) pollen A/C in Qaidam Basin(Cai et al, 2012). The SST anomaly is calculated from the 200 kyr running average subtracting the maximum value over the whole-time window.…”