1 2 3 4 5 6 7 8 9 10 11 Supplementary Information for Impacts of a warming marginal sea on torrential rainfall organized under the Asian summer monsoon 12 13 14 Atsuyoshi Manda 1, Hisashi Nakamura 2,4, Naruhiko Asano 2, Satoshi Iizuka 3, Toru Miyama 4, Qoosaku Moteki 5, Mayumi K. Yoshioka 6, Kazuaki Nishii 2, and Takafumi Miyasaka 2 15 16 17 18 19 20 21 22 23 24 25 26 1: Graduate School of Fisheries Science and Environmental Studies, Nagasaki University, Nagasaki 852-8521, Japan 2: Research Center for Advanced Science and Technology, The University of Tokyo, 153-8904, Tokyo, Japan 3: Monitoring and Forecast Research Department, National Research Institute for Earth Science and Disaster Prevention, Tsukuba, 305-0006, Japan 4: Application Laboratory, Japan Agency for Marine-Earth Science and Technology, Yokohama, 236-0001, Japan 5: Department of Coupled Ocean-Atmosphere-Land Processes Research, Japan Agency for Marine-Earth Science and Technology, Yokosuka, 237-0061, Japan 6: Center for Atmospheric and Oceanic Studies, Tohoku University, Sendai, 980-8578, Japan e-mail: manda@nagasaki-u.ac.jp 1
27 28 29 30 31 32 33 34 35 36 Supplementary Information S1: Precipitation, moisture flux and SST data The precipitation climatology in Fig. 1a is based on the Tropical Rainfall Measurement Mission (TRMM) 3B43 dataset (Huffman et al., 2007) for 2002 2010 at 0.25 grid intervals in latitude and longitude, where TRMM satellite radar measurements have been blended with other satellite data. The climatological moisture flux in Fig. 1a has been computed from the 6-houly ERA-Interim Reanalysis data (Dee et al., 2011) at 0.75 grid intervals in latitude and longitude for the period 2002 2010, in which high-resolution SST data are incorporated into atmospheric data assimilation. The precipitation statistics in Figs. 1c-d are based on rain-gauge measurements at observatories and weather stations of the Japan Meteorological Agency (JMA) for the period 1950 2012 and on the JMA automated (AMeDAS) measurements for the period 1974 2012. These data were available at JMA Web site (http://www.jma.go.jp/jma/). 37 38 39 40 41 References: Dee, D. P., et al. The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Q. J. Roy. Meteorol. Soc., 137, 553-597 (2011). Huffman, G. J., et al. The TRMM multisatellite precipitation analysis (TMPA): Quasi-global, multiyear, combined-sensor precipitation estimates at fine scales. J. Hydrometeor., 8, 38-55 (2007). 42 43 Supplementary Information S2: Simulations for a rain event in mid-june 2001 44 45 46 47 48 49 50 In addition to the control simulation (CTRL01) that is designed to reproduce a particular heavy rain event that occurred over Kyushu on June 19th, 2001, the corresponding seasonal march simulations (SMCH01) are conducted. Each of these simulations was initialized with the atmospheric condition observed at 0000 UTC on June 18th, 2001. As in our simulations of the rain event in mid-july 2012, high-resolution daily SST data for mid-june 2001 produced by the JCOPE reanalysis system 27 were used for the model lower-boundary condition of CTRL01, while the SMCH01 simulation utilizes the daily OISST 28 climatology for the period 1985 2004 with 1/4 resolution in latitude and longitude. The SMCH01 runs are designated as S01MonDD, 2
51 52 53 54 55 56 following the convention for the corresponding SMCH for the rainfall event in July 2012. The SMCH01 simulation yields qualitatively the same sensitivity of precipitation over Kyushu to SST over ECS as the SMCH simulation for the rain event in mid-july, 2012 (Supplementary Fig. S5), although a substantial fraction of rainfall occurred over ECS and precipitation over Kyushu is thus rather modest. Nevertheless, the SMCH01 experiment suggests that rainfall averaged over Kyushu would have increased by 11 16% if the event had occurred in late July. 57 58 Supplementary Information S3: CMIP5 SST data 59 60 61 62 63 64 65 Since the horizontal resolution and detailed configuration of the coastal boundaries differ from one climate model to another, we did not use simple averaging among the CMIP5 models to obtain the MME field of SST. Instead, it was mimicked by first choosing MIROC5 (see Supplementary Table S1) as a representative model whose SST around ECS is found similar to MME. For each of the decades (i.e., the 2040s and 2090s), we then obtained SST at [30 N, 127.5 E] from its value at the neighbouring grid points of a particular model through linear interpolation, before evaluating the difference from the MIROC5 value. Finally, the differences were averaged over all the models before added uniformly to the SST field of MIROC5 within ECS. 66 3
67 Supplementary Table S1 Description of 32 CMIP5 models. Model name ACCESS1-0 ACCESS1-3 BCC-CSM1.1 BCC-CSM1.1(m) CCSM4 CESM1-BGC CESM1-CAM5 CMCC-CM CMCC-CMS CNRM-CM5 CSIRO-Mk3.6.0 FIO-ESM GFDL-CM3 GFDL-ESM2G GFDL-ESM2M GISS-E2-H GISS-E2-H-CC Institution(s) Commonwealth Scientific and Industrial Research Organization (CSIRO) and Bureau of Meteorology (BOM), Australia Beijing Climate Center, China Meteorological Administration National Center for Atmospheric Research Community Earth System Model Contributors Centro Euro-Mediterraneo per I Cambiamenti Climatici Centre National de Recherches Meteorologiques / Centre Europeen de Recherche et Formation Avancees en Calcul Scientifique Commonwealth Scientific and Industrial Research Organisation in collaboration with the Queensland Climate Change Centre of Excellence The First Institute of Oceanography, SOA, China NOAA Geophysical Fluid Dynamics Laboratory Geophysical Fluid Dynamics Laboratory NASA Goddard Institute for Space Studies GISS-E2-R-CC 68 HadGEM2-CC HadGEM2-ES INMCM4 IPSL-CM5A-LR IPSL-CM5A-MR IPSL-CM5B-LR MIROC5 MIROC-ESM MIROC-ESM-CHEM MPI-ESM-LR MPI-ESM-MR MRI-CGCM3 NorESM1-M NorESM1-ME Met Office Hadley Centre Institute for Numerical Mathematics Institut Pierre-Simon Laplace Atmosphere and Ocean Research Institute (The University of Tokyo), National Institute for Environmental Studies, and Japan Agency for Marine-Earth Science and Technology Japan Agency for Marine-Earth Science and Technology, Atmosphere and Ocean Research Institute (The University of Tokyo), and National Institute for Environmental Studies Max Planck Institute for Meteorology (MPI-M) Meteorological Research Institute Norwegian Climate Centre 4
69 70 71 72 73 Supplementary Figure S1 Maps of climatological bi-pentad SST within ECS. Based on daily OISST data 28 for the period 1984 2012. The Generic Mapping Tools were used for creating the maps in this figure. 5
a b 74 75 76 77 78 79 80 81 82 Supplementary Figure S2 Thermodynamic conditions during the heavy rainfall event in mid-july 2012. a, Vertical profiles of equivalent potential temperature (K; color shade) and water vapor mixing ratio (g kg -1 ; white contour) along the backward trajectory of an air parcel released at 1200 UTC on July 10, 2012 at [33.2 N, 131 E], a location within the area of heavy rainfall. Dots indicate the altitude of the parcel at 30-minute intervals. b, horizontal distribution of 950-hPa moisture flux (vector) and locations of the air parcel at 30-minute intervals, as marked with circles whose color indicates its pressure level. Each of the numerals along the trajectory indicates the elapse time (in hours) for the parcel from 0000 UTC July 10, 2012. SST distribution used in this simulation is indicated with black contours. The Generic Mapping Tools were used for creating the map in this figure. 6
83 84 85 86 87 88 Supplementary Figure S3 Influence of seasonal SST rising on convective instability and evaporation over ECS for the heavy rainfall event in mid-july 2012. A measure of convective instability (bar), evaluated as the difference in equivalent potential temperature ( e; left axis) from the 975 hpa to 900 hpa levels, and surface evaporation (E; red square; right axis), both averaged in time from 0000 UTC 11 July to 0000 UTC 15 July and spatially within [25 N 35 N, 123 E 132 E]. 89 7
90 d01 91 92 93 94 95 96 Supplementary Figure S4 Model domain. The map labeled by d01 indicates the outer domain, and the two inset black squares labeled by d02 and d03 indicate the intermediate and inner subdomains, respectively. The yellow square indicates the Kyushu area for averaging precipitation in Figs. 3 and 4, in addition to Supplementary Fig. S5. The NCAR Command Languague was used for creating the map in this figure. 8
a S01Jun01 S01Jun11 S01Jun21 S01Jul01 S01Jul10 S01Jun21 S01Jul31 S01Aug10 mm/4-d b 97 98 99 100 101 102 103 Supplementary Figure S5 Impact of seasonal ECS warming on the rainfall event in mid-june 2001. a, Maps of 4-day precipitation (mm) obtained by the seasonal march simulations (SMCH01), in which the climatological-mean SST fields for bi-pentad periods, as indicated, from June 1st to August 10 are assigned while atmospheric conditions are kept the same as in CTRL01. b, Area-averaged 4-day precipitation (mm) over Kyushu [31.0 N 34.1 N, 129.5 E 131.8 E]. The Grid Analysis and Display System was used for creating the maps in this figure. 9
104 105 106 107 108 109 110 Supplementary Figure S6 Influence of SST rise in ECS on convectively unstable stratification. Vertical profiles of equivalent potential temperature ( e, red), potential temperature, (green), and water vapor mixing ratio (w, blue), averaged over Kyushu [32 N-33 N, 129 E-130 E], at (a) 0000 and (b) 1200 UTC on July 11, 2012. The vertical profiles of e below the 900 hpa level show intensified convectively unstable stratification in JulMME90(S), in which only SST is raised, than those in JulMME90(A), in which only the air temperature is raised. Note that both e and w between the 950 and 400 111 hpa levels in JulMME90(S) are less than those in JulMME90(A). 10
112 113 114 115 116 117 Supplementary Figure S7 Influence of SST rise in ECS on surface evaporation, lower-tropospheric stratification and winds. Latitude-time plots of equivalent potential temperature ( e; contour), winds (vector) both at 975 hpa and surface evaporation (mm/h, color) simulated in (a) CTRL and (b) JulMME90(S). Thick green contours correspond to e = 365 K. All the variables are averaged longitudinally from 126.0 E to129.5 E. 118 11
119 120 121 122 Supplementary Figure S8 Influence of SST rise in ECS on mid-troposheric winds. Latitude-time plots of 500-hPa wind speed (m s 1 ; color) and vectors simulated in (a) CTRL and (b) JulMME90(S). All the variables are averaged longitudinally from 126.0 E to129.5 E. 123 12
124 125 126 127 128 129 130 Supplementary Figure S9 Moisure flux into Kyushu during the July 2012 evnet. a, Time series of moisture flux (kg m 1 s 1 ) simulated in Jul90MME(S) at 129.5 E, vertically integrated from the surface to 100 hpa and meridionally averaged from 30 N to 34 N. b, Same as in a but for the difference between Jul90MME(S) and CTRL. 131 13
132 133 134 135 Supplementary Figure S10 Daily precipitation simulated in the two other members of the ensemble experiment in the CTRL. Initial time of the simulation is shifted six hours (a) earlier or (b) later from 0000 UTC on July 10, 2012. 136 14