Supporting data for “Post-Laki Hydroclimate in East Asia Modulated by Westerlies under Different Climate States”

We employ the fully-coupled CESM1.2 consisting of the latest version of the Community Atmosphere Model (version 5, CAM5), the Community Land Model (Version 4, CLM4), the Parallel Ocean Program (Version 2, POP2) and the Community Ice Code (Version 4, CICE4). CAM5 has adopted the new parameterizations that show great amelioration of the physical process itself and the mutual feedback within the climate system (Gettelman et al., 2012; Hurrell et al., 2013). The horizontal resolution of our simulations is 1.9° × 2.5° at finite volume grid for atmosphere and land components, approximately 1° for ocean and sea-ice components. We design 30 sets of prescribed Laki-size volcanic forcing experiments under the pre-industrial (PI) and Representative Concentration Pathway 8.5 (RCP8.5) scenarios, respectively (Table 2). A 100-year PI spin-up run (Hurrell et al., 2013) with all climatic parameters (e.g. solar variability, ozone and aerosol forcing) fixed at the 1850 CE values is performed for initializing. Another 30-year run continues which then serve as the starting conditions for 30 ensemble runs. The experiments without (Ctrl-PI) or with added volcanic forcing (Laki-PI) are run for 5 years. For the RCP8.5 simulations, we begin performing the by-default 20^th century run (Hurrell et al., 2013) without climate change policy intervention from 2005 to 2079 as the baseline to establish anthropogenic warming condition. Then, we run 10 sets of experiments up to 2080 from 10 initial states from 2070 to 2079, assuming that they reached the same warming level but had different internal phases. For example, the inter-ensemble standard deviation of annual-averaged global-mean surface temperature is 0.14 K for these ten 2080 startup conditions, while that of the Niño3.4 index can reach 1.36 ℃. Three sets of El Niño, La Niña and Neural states are selected from the ten 2080 startup conditions as the initial conditions for subsequent experiments. A total of 10 different temperature perturbations (1×10^-14 K, 2×10^-14 K, 3×10^-14 K,…....,1×10^-13 K) are set up, respectively, under three different initial conditions, and 30 sets of simulations can be obtained by the corresponding runs without (Ctrl-RCP8.5) or with (Laki-RCP8.5) added volcanic forcing for 5 years.