This study investigates the subseasonal intensity variation of South Asian high (SAH) and the dynamic linkage with pre- cipitation induced diabatic heating during the summer of 1979–2005, based on the observation and 18 models from the Coupled Model Intercomparison Project (CMIP5). The SAH’s intensity variation with a period of approximately 10–36 days is identified both in the observation and 18 models, by performing an empirical orthogonal function (EOF) analysis on the standardized subseasonal anomalies in geopotential height at 100 hPa over the SAH’s main body region. With the observed strengthening of the SAH, an enhanced rainfall belt between 70°–120°E migrates northward from the equatorial region. The northward propagating rainfall belt occupies almost the whole Indian subcontinent, Indochina Peninsula and subtropical east Asian regions between 20°–40°N when the SAH reaches the strongest. The relative vorticity diagnosis reveals the dynamical linkage between the subseasonal SAH’s intensity variation and the precipitation anomalies over the Asian monsoon region: (1) The observed northward propagating rainfall band forms a horizontal gradient of diabatic heating, favoring an enhance- ment of the SAH over its southern part; (2) When the rainfall band approaches the regions of South and East Asia between 20°–40°N, the increased vertical gradient of diabatic heating in the upper level contributes significantly to the intensification of the SAH; (3) The horizontal advection process favors the westward expansion of the SAH. The observation-to-model comparisons indicate that the model’s capacity to reproduce the SAH’s intensity variation is highly associated with simula- tions of the anomalous rainfall band and its northward propagation. A realistic reproduction of precipitation anomalies is fundamental to simulating the dynamical processes related to diabatic heating, and thus the simulation of the SAH’s sub- seasonal intensity variation. Keywords South Asian high · CMIP5 · Subseasonal intensity variation · Diabatic heating 1 Introduction (Mason and Anderson 1963; Tao and Zhu 1964; Krishna- murti 1973). As a member of the Asian summer monsoon The South Asian high (SAH) is a planetary-scale anticy- system, the SAH’s spatiotemporal variability and impacts clonic circulation in the upper troposphere, with its main on the weather and climate over Asia and beyond have been body zonally elongated over the Iranian Plateau (IP)–Tibetan investigated (Zhang et al. 2002, 2005, 2016; Liu et al. 2004, Plateau (TP) and the surrounding areas during summer 2007, 2013; Duan and Wu 2005; Randel and Park 2006; Zarrin et al. 2010; Huang et al. 2011; Jiang et al. 2011; Qu and Huang 2012; Wei et al. 2014; Yang et al. 2014; Ren * Xuejuan Ren et al. 2015; Wu et al. 2015; Chen and Zhai 2016; Shi and email@example.com Qian 2016; Yang and Li 2016). Especially, variations of the SAH on subseasonal time scale have attracted more atten- CMA-NJU Joint Laboratory for Climate Prediction Studies, School of Atmospheric Sciences, Nanjing University, tion in recent years. Because some high-impact events (e.g. Nanjing 210023, People’s Republic of China persistent heavy rainfall or long-lived heat wave) are often Industrial Ecology Programme, Department of Energy induced by anomalies in atmospheric circulation and water and Process Engineering, Norwegian University of Science vapor transport accompanied by the SAH’s subseasonal vari- and Technology, Trondheim, Norway ations (Ren et al. 2015; Yang et al. 2017; Chen and Zhai Institute of Meteorology, Freie Universität Berlin, Berlin, 2016; Ge et al. 2017). Germany Vol.:(0123456789) 1 3 W. Shang et al. Previous studies (Tao and Zhu 1964; Zhang et al. 2002; variation and the corresponding dynamical mechanisms. Jia and Yang 2013; Ren et al. 2015; Yang and Li 2016) The World Climate Research Programme (WCRP) Working have identified two dominant modes of the SAH’s varia- Group on Coupled Modelling (WGCM) has employed the tions on subseasonal time scale. One mode demonstrates phase 5 of Coupled Model Intercomparison Project (CMIP5) zonal movement of the SAH’s main body (denoted as “zonal since September 2008 (Taylor et al. 2012). Evaluating the oscillation”). Another one depicts the intensity variation ability of the CMIP5 models in simulating the Asian sum- of the SAH’s main body (hereafter “intensity variation”). mer monsoon has been carried out (Duan et al. 2013; Huang The SAH’s zonal oscillation around its climatology posi- et al. 2013; Cherchi et al. 2014; Lee and Wang 2014; Frey- tion for periods of weeks were firstly revealed by Mason chet et al. 2015; Gao et al. 2015; Sooraj et al. 2016; Li et al. and Anderson (1963), and Tao and Zhu (1964). Zhang et al. 2017). Some studies have assessed the capacity of climate (2002) suggested the bimodality of the SAH at 100 hPa and models in simulating the SAH’s climatology and its varia- identified them as the Iranian pattern and the Tibetan pat- tions. Xue et al. (2017) studied the climatology and inter- tern, respectively. Further investigations have studied the annual variations of the SAH in 38 CMIP5 models. They features of subseasonal zonal shift of the SAH and its con- pointed out that two-thirds of these models can simulate nection with precipitation and diabatic heating over east- the observed relationship between the El Niño–Southern ern Asia (Ren et al. 2015; Shi and Qian 2016; Yang and Li Oscillation (ENSO) and the SAH. Qu and Huang (2015) 2016). Other literature mentioned about the SAH’s subsea- evaluated the performance of CMIP5 models in reproducing sonal intensity variation. For example, Jia and Yang (2013) the decadal relationship between the SAH and sea surface demonstrated that the northwestward propagation of the temperature in the tropical Indian Ocean. The above investi- quasi-biweekly oscillation over the western North Pacific is gations assessed the SAH’s interannual to decadal variations accompanied by the strengthening of SAH, which contrib- simulated by the CMIP5 models. The present evaluation utes to the upper level divergence pattern over the subtropi- focuses on the subseasonal intensity variation of the SAH. cal east Asian region. Yang et al. (2017) suggested that the The paper is organized as follows. Section 2 introduces SAH is enhanced with eastward and westward extensions the datasets, CMIP5 models and analysis methods. Section3 when the quasi-biweekly oscillation in the eastern TP sum- presents the features of climatology and the subseasonal mer precipitation transits from its dry to wet phase. Com- intensity variation of the SAH, and investigates the rela- pared to the studies about the SAH’s zonal oscillation, our tionship between precipitation and the SAH in the observa- understanding of characteristics and mechanisms for SAH’s tion and CMIP5 models. Section 4 reveals the corresponding intensity variation on subseasonal time scale is incomplete. dynamical mechanism. A summary and discussion is shown This makes up the core subject of the present study. in Sect. 5. The SAH is considered as a thermal-driven circulation (Jin and Hoskins 1995; Liu et al. 2004; Lin 2009). Diabatic heating induced by precipitation is an important dynamical 2 Datasets and methods factor responsible for the SAH’s variations (Liu et al. 2004, 2007, 2013; Ren et al. 2015; Chen and Zhai 2016; Wang and 2.1 Datasets Ge 2016; Ge et al. 2017). Ren et al. (2015) suggested that a north–south dipolar structure of the condensation heating/ The datasets used include (1) daily reanalysis data on a rainfall over eastern Asia performed a positive feedback in 1° × 1° horizontal resolution in 1979–2005 derived from the SAH’s subseasonal zonal oscillation. Similar relation- the European Center for Medium-Range Weather Forecasts ship between condensation heating over eastern Asia and (ECMWF) (ERA-Interim, Dee et al. 2011), (2) daily mean the zonal oscillation of the SAH were also advocated by outgoing longwave radiation (OLR) data on a 2.5° × 2.5° Yang and Li (2016), Chen and Zhai (2016) and Wang and Ge grid from the National Oceanic and Atmospheric Adminis- (2016). The strengthening/weakening of SAH is also accom- tration (NOAA) (Liebmann and Smith 1996) for 1979–2005, panied by precipitation anomalies over monsoon region and (3) gauge-based analysis of global daily precipitation (Jia and Yang 2013; Yang et al. 2017). The corresponding data from NOAA Climate Prediction Center (CPC) for the change in diabatic heating should be a no-negligible fac- period 1979–2005 on 0.5° × 0.5° horizontal resolution. Fur- tor responsible for the SAH’s subseasonal intensity varia- ther details for the precipitation dataset can be found in Xie tion. The present study focuses on this factor. We intend to et al. (2007) and Chen et al. (2008). The observational and reveal the role of precipitation induced diabatic heating on reanalysis datasets are called “observation” in the present the SAH’s intensity variation through performing a diagno- study. sis on relative vorticity. The historical experiments for daily mean fields of 18 Besides, we also evaluate the model performances in CMIP5 models from 1979 to 2005 are used in this study reproducing the features of the SAH’s subseasonal intensity (Taylor et al. 2012). Table 1 shows the detail information, 1 3 Subseasonal intensity variation of the South Asian high in relationship to diabatic heating:… Table 1 Information of CMIP5 climate models Models name Institute/country Resolution ACCESS1-0 Commonwealth Scientific and Industrial Research Organization (CSIRO) and Bureau of Meteorology 192 × 144 (BOM)/Australia ACCESS1-3 CSIRO and BOM/Australia 192 × 144 BNU-ESM Beijing Normal University/China 128 × 64 CanESM2 Canadian Center for Climate Modeling and Analysis/Canada 128 × 64 FGOALS-g2 State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics 128 × 60 (LASG)/China GFDL-CM3 National Oceanic and Atmospheric Administration (NOAA) GFDL/United States 144 × 90 GFDL-ESM2G NOAA/GFDL/United States 144 × 90 HadGEM2-CC Met Office Hadley Centre (MOHC)/United Kingdom 192 × 144 HadGEM2-ES MOHC/United Kingdom 192 × 144 IPSL-CM5A-LR Institute Pierre Simon Laplace/France 96 × 96 IPSL-CM5A-MR Institute Pierre Simon Laplace/France 144 × 143 MIROC5 University of Tokyo, National Institute for Environmental Studies and Japan Agency for Marine-Earth Sci- 256 × 128 ence and Technology/Japan MIROC-ESM-CHEM University of Tokyo, National Institute for Environmental Studies and Japan Agency for Marine-Earth Sci- 128 × 64 ence and Technology/Japan MPI-ESM-LR Max-Planck Institute for Meteorology (MPI-M)/Germany 192 × 96 MPI-ESM-MR MPI-M/Germany 192 × 96 MRI-CGCM3 Meteorological Research Institute/Japan 320 × 160 MRI-ESM1 Meteorological Research Institute/Japan 320 × 160 NorESM1-M Norwegian Climate Center (NCC)/Norway 144 × 96 including the institutes, horizontal resolutions of these mod- analysis and the EOF results are similar. The main body els. Given the large volume of data, only one run from each region (20°–40°N, 35°–110°E) covers the climatological model and experiment was used in this study. The model 16,750 gpm contour line of the SAH at 100 hPa (Fig. 1a). outputs are interpolated onto 1° × 1° resolution to match A power spectrum analysis (Gilman et al. 1963) is used to the ERA-Interim reanalysis data. More information about identify the dominant periodicity on the principal compo- the 18 CMIP5 models can be referred to http://cmip-pcmdi nents of EOF. .llnl.gov/cmip5 /. The observed first EOF mode (EOF1) demonstrates a monopole pattern centered over the IP–western TP. This 2.2 Extracting subseasonal daily anomalies mode indicates the SAH’s intensity variation on subsea- and the SAH’s subsesonal intensity variation sonal time scale. The observed second EOF mode displays a positive center of anomalous Z100 over East Asia and a We use the method adopted by Krishnamurthy and Shukla negative one over the IP. It represents the subseasonal zonal (2000) and Ren et al. (2015) to extract the subseasonal daily oscillation of the SAH, which has been intensively studied anomalies both for observation and for model data for all the (Zhang et al. 2002; Ren et al. 2015; Shi and Qian 2016; variables. The process is as follows. First, a 5-day running Yang and Li 2016; Chen and Zhai 2016). We focus on the mean is preformed to remove the high-frequency fluctua- EOF1 in this study. Time-lagged regressions of the observed tions in the daily data, and then remove the daily climatology subseasonal variables on the normalized first principal com- to get daily anomalies, finally, remove the interannual signal ponents of EOF (observed normalized PC1) are calculated. by subtracting seasonal anomaly. The same process is then done on each of the 18 models’ The amplitudes of 100 hPa geopotential height (Z100) simulation. The time-lagged regression is used. Day 0 is anomalies are increased rapidly from the tropical to extra- simultaneous in regression. The simultaneously regression tropical latitudes. To avoid the variance of the EOF being fields can show the anomalous pattern when the SAH gets it dominated by the signals in the middle latitudes, the EOF strongest stage (the positive peak of normalized PC1). The analysis is performed on standardized subseasonal anomalies lead-lag regression can demonstrate the evolution pattern of Z100 over the SAH’s main body region during summer of anomalies accompanied by the evolution of the SAH. A (June, July and August) of 1979–2005 (27 × 92 = 2484 days negative lag day (a positive lag day) is obtained by shifting in total). We also perform the latitudinal weight before EOF backward (forward) the number of the leading (lag) days. 1 3 W. Shang et al. (a) (k) (p) (f) (b) (g) (q) (l) (h) (c) (r) (m) (s) (d) (i) (n) (e) (j) (o) Fig. 1 Climatological summer (June, July and August) fields of Z100 CMIP5 models over the region of 15°–45°N,10°–140°E are shown (contour and shaded with interval of 50 gpm) during 1979–2005 for at each of the top-right corners. The red box in a denotes the region a ERA-Interim and b–s results of the 18 CMIP5 models. The spa- for EOF analysis in Fig. 2. The grey line in each is the outline of the tial correlation coefficients between the ERA-interim and individual Tibetan Plateau heating, respectively. According to Eq. (1), the changes of 2.3 Diagnosis on relative vorticity equation are contributed by the following processes: the horizontal advection of relative vorticity, effect, divergence of atmos- Following the Ertel potential vorticity equation (Ertel 1942; pheric circulation, and vorticity sources induced by vertical Hoskins et al. 1985), the relatively vorticity equation is and horizontal gradients of Q . expressed as (Wu and Liu 1998; Wu et al. 1999; Liu et al. Diabatic heating Q (also known as apparent heating 2001, 2004, 2013). source) and apparent moisture sink Q and their vertically f + Q integrated values ⟨Q ⟩ and ⟨Q ⟩ are calculated based on the =−V ⋅ ∇ − v −(f + )∇ ⋅ V + + S, h h h h 1 2 t z following formulas (Yanai et al. 1973; Luo and Yanai 1984) (1) T T RT Q Q 1 v 1 u 1 1 Q = + V ⋅ ∇ T + − , 1 h h (3) S =− + , (2) t p c P z x z y z z where is relative vorticity, V =(u, v) the horizontal wind, V the horizontal gradient operator, f the planetary q q Q =−L + V ⋅ ∇ q + , (4) 2 h h vorticity, and and Q are the static stability and diabatic t P t z 1 1 3 Subseasonal intensity variation of the South Asian high in relationship to diabatic heating:… p 3 Characteristics of subseasonal intensity variation of SAH ⟨Q ⟩ = Q dp, (5) 1 1 3.1 Subseasonal intensity variation of the SAH Ps Before showing the EOF results, we depict the observed and simulated climatological summer mean fields of ⟨Q ⟩ = Q dp, (6) 2 2 Z100 for 1979–2005. The observed SAH is located over the IP–TP and its surrounding areas, with its center value slightly higher than 16,800 gpm (Fig. 1a). The simula- where T is the air temperature; q is the water vapor, is the tions by 18 CMIP5 models reproduce the zonally elon- vertical velocity in pressure ( P ) coordinates; k = , R and c gated structure of the SAH in general (Fig. 1b–s). Among the 18 models, CanESM2 slightly overestimates the inten- c are the gas constant and specific heat at constant pressure sity of the SAH with the center higher than 16,850 gpm, of dry air, respectively; L is the latent heat of condensation; while the SAH’s centers of FGOALS-g2, GFDL-ESM2G, P = 1000 hPa. g is the gravitational acceleration; P and P 0 t s IPSL-CM5A-LR, MRI-CGCM3 are less than 16,700 gpm. are tropopause pressure and surface pressure, respectively. Based on the pattern correlation between the observation The components of diabatic heating Q include: radiative and the 18 models, MPI-ESM-LR, BNU-ESM, CanESM2, heating, turbulent sensible heating at the earth surface, and MPI-ESM-MR and NorESM1-M show better skills on condensation latent heating induced by precipitation. The simulation of the climatological SAH. Their pattern cor- apparent moisture sink Q is mainly related to the condensa- relation coefficients are above 0.95. HadGEM2-CC and tion latent heating. A similar pattern of ⟨Q ⟩ and ⟨Q ⟩ over 1 2 HadGEM2-ES produce a slightly northward center of SAH summer monsoon region indicates that condensational heat- compared to the observation. Their pattern correlations ing induced by precipitation is the major component of dia- are around 0.6. batic heating (Yanai and Tomita 1998; Jin et al. 2013). The spatial pattern of the observed EOF1 is plotted in A variable is decomposed into its time mean (denoted Fig. 2a. It demonstrates a monopole pattern with a center with an overbar) and subseasonal (denoted with a prime), over the IP–western TP region. As mentioned earlier, this and non-subseasonal components (Zhang and Ling 2012; pattern exhibits the spatial structure of subseasonal inten- Ren et al. 2015). The horizontal advection of relative vor- sity variation of SAH. The variance of the observed EOF1 ticity and the two vorticity generation terms in Eq. (1) is 43%. Figure 2b–s depict the EOF1 patterns of each which contribute to the subseasonal anomaly of local vor- CMIP5 models. The 18 models simulate the monopole ticity change , can be expressed as follows: pattern to a certain extent with variances ranging from 33 to 44%. We calculate the subseasonal pattern correla- � � � ̄ ̄ 𝜕𝜁 𝜕𝜁 𝜕𝜁 𝜕 𝜁 𝜕 𝜁 � � adv =− u ̄ + v̄ − u + v + residue, tion coefficients (r ) between the observed EOF1 and EOF 𝜕 t 𝜕 x 𝜕 y 𝜕 x 𝜕 y each model’s. ACCESS1-0, ACCESS1-3, BNU-ESM, (7) HadGEM2-CC, HadGEM2-ES, MIROC-ESM-CHEM, MRI-CGCM3, and NorESM1-M reasonably reproduce a 𝜕 Q 𝜕 Q 𝜕𝜁 1 1 1 1 similar EOF1 pattern to the observation’s with r ≥ 0.8. Q _z = f + 𝜁 + 𝜁 + residue, (8) EOF ̄ ̄ 𝜕 t 𝜕 z 𝜕 z 𝜃 𝜃 z z Some other models show differences in the location of the monopole center, compared to the observation. For � � � example, the anomalous centers simulated by CanESM2, 𝜕 Q 𝜕 Q 𝜕𝜁 1 𝜕̄u 𝜕̄v 1 1 S = − GFDL-CM3, IPSL-CM5A-LR and IPSL-CM5A-MR are 𝜕 t 𝜕 z 𝜕 y 𝜕 z 𝜕 x located slightly poleward. Their subseasonal pattern cor- (9) � � 𝜕 Q 𝜕 Q 1 𝜕 u 𝜕 v 1 1 relation coefficients r are under 0.5. EOF + − + residue 𝜕 z 𝜕 y 𝜕 z 𝜕 x We divide the 18 models into three groups, accord- ing to the value of r . (1) Better-simulated group (BG) EOF with r ≥ 0.8, includes ACCESS1-3 (0.98), HadGEM2- EOF Where residue terms represent the contributions from CC (0.96), ACCESS1-0 (0.93), HadGEM2-ES (0.93), nonlinear interaction between subseasonal and non-sub- MIROC-ESM-CHEM (0.90), BNU-ESM (0.89), MRI- seasonal components. The residue terms in (7)–(9), CGCM3 (0.84), NorESM1-M (0.8). (2) General-simulated effect, and divergence term in (1 ) are much smaller than group (GG) with r between 0.8 and 0.5, includes MRI- EOF the others and thus ignored hereinafter. ESM1 (0.72), MIROC5 (0.68), MPI-ESM-MR (0.68), 1 3 W. Shang et al. (a) (k) (p) (f) (b) (g) (q) (l) (h) (c) (r) (m) (s) (d) (i) (n) (e) (j) (o) Fig. 2 Spatial pattern of the first EOF mode of normalized subsea- Interim and those of the CMIP5 models (r ) are displayed at each EOF sonal anomalies of Z100 over the region of 20°–40°N, 35°–110°E. of the top-right corners. The purple contour in a is the climatological a ERA-Interim and b–s the results of the 18 CMIP5 models. The 16,750 gpm contour line of the SAH at 100 hPa spatial correlation coefficients between the first EOF of the ERA- MPI-ESM-LR (0.67), GFDL-ESM2G (0.67), FGOALS-g2 LG models, CanESM2, GFDL-CM3, IPSL-CM5A-LR and (0.58). (3) The rest of the models with r less than 0.5 IPSL-CM5A-MR, all simulate no significant peak period. EOF are lower-simulated group (LG), which includes GFDL- Figure 4a–e plot the regression of subseasonal Z100 CM3 (0.46), CanESM2 (0.44), IPSL-CM5A-LR (0.43) and 100 hPa wind field anomalies on the normalized PC1 and IPSL-CM5A-MR (0.41). We select one model from for the observation, from days − 9 to + 3 with interval of each group as their individual representatives. They are 3 days. In the observation, it is seen that an anomalous ACCESS1-3 from the BG, MRI-ESM1 from the GG, and high is in control of the SAH area at 100 hPa centered IPSL-CM5A-MRall the three models from the LG. over the IP–western TP and East Asia at day 0. Meanwhile, Figure 3 displays the power spectrum of PC1 for the a cyclonic one exists over the Barents Sea and Novaya observation and 18 models. A 10-36-day periodicity is Zemlya poleward of the anomalous high. Such anoma- seen in the observation (Fig. 3a) with the peak around lous pattern is seen on day − 9 with a weak amplitude. 27-day. Figure 3b to s are sorted according to r from the From days − 6 to − 3, the anomalous high and low centers EOF greatest to least. Most of the BG models and GG models become more significant with intensified amplitudes. The (Fig. 3b-o) simulate the peaks within 15-30-day, while above pattern matures on day 0 and fades away locally the GG models show slightly lower variance. The four after day 0. 1 3 Subseasonal intensity variation of the South Asian high in relationship to diabatic heating:… The regression of subseasonal anomalies in Z100 the eastern Asian continent is too weak. IPSL-CM5A- and 100 hPa wind field for the three selected models, MR’s simulations of rainfall belt and its northward propa- ACCESS1-3 (BG), MRI-ESM1 (GG) and IPSL-CM5A- gation are much weaker, compared to the observation and MR (LG) on each model’s normalized PC1, are shown in the other two models’ (Fig. 5m to p). Fig. 4f–t. In general, all the three models can simulate the We plot latitude–time section of the precipitation anomalous anticyclonic band over the SAH’s main body anomalies for the observation (Fig. 6a) and 18 models region. However, the simulation results show differences (Fig. 6b–s), averaged over 70°–120°E, to further dem- in the intensity of the anomalous anticyclone and its center onstrate the simulated northward propagating feature of location. ACCESS1-3 (BG, Fig. 4f–j) simulates most of the precipitation anomaly. Figure 6b–s are sorted according observed features of the Z100 anomalous circulation, espe-to r from the greatest to least. The BG models (Fig. 6b EOF cially the anomalous anticyclonic center anchored over the to i) can overall simulate well the northward propagation IP–western TP region with an amplitude close to the obser- of the precipitation anomalies from tropical areas to sub- vation. The evolution process simulated by ACCESS1-3 tropical Asia. The GG models (Fig. 6j–o) also can display from days − 9 to + 3 also bears resemblance with that in the the northward propagation of precipitation anomalies to observation. MRI-ESM1 (GG, Fig. 4k–o) reproduces the a certain extent. However, the propagation patterns are anomalous anticyclonic band over the SAH’s main body slightly different with that of the observation. The prop - region with slightly weaker amplitude and with the center agation signals in the LG models (Fig. 6p–s) are rather shifting to the western part of TP compared to ACCESS1-3 weak, compared to the BG and GG models. Meanwhile, and the observation. IPSL-CM5A-MR (LG, Fig. 4p–t) the propagation signals in the LG models are not alike simulates a much weaker anomalous anticyclone over the those in the observation. IP–TP region from days − 9 to + 3, compared to the obser- To show the relationship between intensity anomaly vation, ACCESS1-3 (BG) and MRI-ESM1 (GG). of the SAH and precipitation anomaly over the south and east Asian monsoon regions, we average the simultaneous 3.2 T he precipitation anomaly associated with SAH regression of subseasonal Z100 anomalies on the normalized PC1 over the region of (20°–40°N, 35°–110°E) (denoted as Figure 5a–d plot the time-lagged regression of precipita- Z100’ ). The same process is done for precipitation anom- SAH tion and OLR anomalies on PC1 for the observation. The alies over the region of (20°–40°N, 70°–120°E) (denoted northward propagation of the anomalous rainfall belt is as R’ ). Figure 7 displays the scatter diagrams between ASIA seen in the observation. During the early stage of the SAH’s Z100’ and R’ for the observation and 18 models. It SAH ASIA strengthening (prior to day − 6), positive precipitation is seen that an enhanced SAH is always accompanied by anomalies are located between the equator and 20°N, con- an above-normal precipitation over south and east Asian sistent with the local negative OLR anomalies. Increased regions. Furthermore, a quasi-linear relationship is shown precipitation also scatters over the regions to the east of TP, between Z100’ and R’ . The correlation coefficient SAH ASIA southern Korea and southern Japan. During days − 6 to − 4, between Z100’ and R’ in BG, GG, and LG group SAH ASIA the area with increased rainfall expands northward to almost are 0.69, 0.61 and 0.62, respectively. This result means all over the Indian subcontinent, Indo-China Peninsula and the stronger (weaker) precipitation is accompanied by the subtropical east Asian region. When the SAH’s intensity enhanced (damped) SAH. For example, the BG (models reaches its peak stage (day 0), the positive rainfall belt prop- showing in Fig. 7 as ‘*’) overall reproduces reasonably agates northward furthermore covering almost the whole the enhanced SAH and accompanied positive precipitation of south Asian and east Asian summer monsoon regions anomaly. The GG (showing as ‘O’) simulate smaller val- between 20° and 40°N. After day 0, the above increased ues of positive Z100′ and positive R’ compared to the ASIA precipitation region becomes shrink, especially over the observation and the BG models. For the LG (showing as area to the east of TP. ‘+’), their positive values of Z100’ and R’ are both SAH ASIA The simulated time-lagged regression of precipita- the smallest. tion anomalies on PC1 for the three models are shown in Fig. 5e to p. ACCESS1-3 reproduces the northward propa- gation of the precipitation anomalies with the strengthen- ing of the SAH, though with a slightly stronger amplitude 4 Mechanism for the subseasonal intensity and weaker signals over the Indian subcontinent compared variation of SAH to the observation (Fig. 5e–h). MRI-ESM1 simulates the northward propagation of the precipitation anomalies in In this section, we diagnose the three dynamical processes general (Fig. 5i–l). However, its zonal rainfall belt is not contributing to the subseasonal relatively vorticity anomaly well organized. The simulated precipitation anomaly over at 100 hPa over the SAH’s region: horizontal advection of 1 3 W. Shang et al. 1 3 Subseasonal intensity variation of the South Asian high in relationship to diabatic heating:… ◂Fig. 3 The 27-year averaged mean power spectra (blue solid line) of 4.1 The dynamic processes in observation the principal component of the first EOF (PC1) for a ERA-Interim and b–s results of the 18 CMIP5 models. The pink dashed line is the Figure 8a, b show the latitude-time sections of regressed red noise spectrum and red dashed line is the spectrum of 95% confi- dence level. The values at each of the top-right corners is each mod- anomalies in ⟨Q ⟩, ⟨Q ⟩, Q _z at 100 hPa and precipi- 1 2 1 els’ r . b–s are sorted according to r from the greatest to least EOF EOF tation against PC1 averaged over 70°–120°E for the obser- vation. As mentioned before, a northward propagation of the positive rainfall band can be seen with the strengthening relativity vorticity adv, vorticity sources induced by of the SAH. The positive diabatic heating as well as positive vertical gradients Q _z and by horizontal gradients 1 moisture sink also propagate from the tropical to the south and east Asian regions. The similar northward propagation S of diabatic heating, respectively. of ⟨Q ⟩ and ⟨Q ⟩ anomalies indicates that the anomalous 1 2 condensational heating induced by the rainfall anomalies dominates the anomalous diabatic heating (Yanai and (a) (f) (k) (p) (g) (b) (l) (q) (c) (h) (m) (r) (d) (i) (n) (s) (e) (j) (o) (t) Fig. 4 Regression fields of subseasonal Z100 anomalies (contour MR. The dashed lines denote negative values. The negative (posi- with interval of 4 gpm) and 100 hPa wind field anomalies (vector; tive) lag days mean Z100 and wind fields anomalies leading (lagging) − 1 m s , only values above the 95% confidence level are shown) on PC1. Shaded areas denote regions of statistically significant at 95% PC1, from day − 9 to day + 3 with interval of 3 days, for a–e ERA- confidence level Interim, f–j ACCESS1-3, k–o MRI-ESM1, and p–t IPSL-CM5A- 1 3 W. Shang et al. (a) (i) (e) (m) (b) (f) (j) (n) (o) (k) (c) (g) (p) (d) (h) (l) Fig. 5 Regression fields of subseasonal precipitation anomalies regressed OLR anomalies are plotted by contours in a–d (with inter- − 1 − 2 (shaded; mm d ) on PC1, averaged between days − 9 and − 7, − 6 val of 1 W m ), with the dashed contours being negative. Shaded and − 4, − 3 and 0, + 1 and + 3, for a–d the observation (CPC), e– areas denote regions of statistically significant at 90% confidence h ACCESS1-3, i–l MRI-ESM1, and m–p IPSL-CM5A-MR. The level Tomita 1998; Jin et al. 2013). Thus a northward propagation dient dominates the forcing, whereas at the border of the diabatic heating the horizontal heating gradient plays a of negative Q _z band is produced in Fig. 8b due to more important role (Wu and Liu 1998; Liu et al. 2001). the increased vertical gradient of Q in the upper level. In Fig. 8c, on the north flank of the heating, a negative When the negative Q _z band reaches the SAH’s main vorticity source which is induced by the horizontal gradi- ent of diabatic heating is generated and propagates north- body region, it contributes directly to the strengthening of ward as well. When the positive rainfall belt propagates the SAH. We also used a Linear Baroclinic Model (LBM) northward and reaches around 15°N (about day − 9), the experiment to examine the locally responses of circulation generated weak negative tendency of vorticity anomalies to the condensation heating. The results (figure omitted) begins to favor the enhancement of the southern part of verify again the effect of condensation heating to the SAH, SAH. The observed regression of adv, is plotted in which is similar to the results showing in Zhang et al. (2016). Fig. 8d. During the summertime, the climatological east- The contour in Fig. 8c plots observed regression of erly flow is located over the southern part of the SAH. It S . Within the heating region the vertical heating gra- can be seen that the negative vorticity anomalies with the 1 3 Subseasonal intensity variation of the South Asian high in relationship to diabatic heating:… Fig. 6 Lag-regression of subseasonal precipitation anomalies the top-right corners are the same as those in Fig. 3. The regressed − 1 (shaded; mm d ) on PC1 averaged over 70°–120°E, for a the obser- observation OLR anomaly is plotted in a (contour; with dashed con- − 2 vation (CPC) and b–s the 18 CMIP5 models. The values at each of tour being negative and interval of 1 W m ) 1 3 W. Shang et al. is mainly located between 90°and 120°E (Fig. 10c). MRI- ESM1 reproduces the northward propagation of Q _z and S in general (Fig. 10d, e). However, the signal is weaker than the observation and ACCESS1-3, due to the weaker simulation of the rainfall band shown in Fig. 6j. The three dynamical processes simulated by IPSL-CM5A- MR are rather weak (Fig. 10g–i). Figure 11 shows the three dynamic processes of the three models averaged over the SAH’s main body region. The evolutions and amplitudes of the three area-averaged dynamic processes simulated by ACCESS1-3 resem- ble those of the observation (Fig. 9). The amplitudes Fig. 7 Scatter diagrams showing the relationship between subsea- simulated by MRI-ESM1 are weaker compared to the sonal Z100 anomalies (gpm; x-axis) averaged over 20°–40°N, 35°– − 1 observation and ACCESS1-3, though the evolutions are 110°E and precipitation anomalies (mm d ; y-axis) averaged over 20°–40°N, 70°–120°E both at day 0, for the ERA-Interim and 18 similar to the observation. The evolutions of the three CMIP5 models dynamic processes simulated by IPSL-CM5A-MR are too weak. Meanwhile, the amplitudes are underestimated dramatically. climatological easterly flow together induce a westward advection of the negative vorticity, which is conducive to the westward expansion of the SAH. 5 Summary and discussion To demonstrate further the evolutions of above dynami- cal processes, the three dynamic processes are averaged The present study investigates the features and dynamical over the SAH’s main body region and plotted in Fig. 9. In mechanism of the subseasonal intensity variation of SAH the observation, the northward propagating of rainfall at 100 hPa based on the observational data and 18 CMIP5 band during days − 9 to − 6 is on the south flank of SAH. models during 1979–2005. In the observation, the SAH at Thus, the process of S plays a negative vorticity 100 hPa is located over the IP–TP and its surrounding areas, with its center value slightly higher than 16,800 gpm. An source in favor of enhancement of the SAH’s southern EOF analysis is carried out on the standardized subseasonal part. The process of Q _z acts to contribute to the Z100 anomalies over the region of (20°–40°N, 35°–110°E). The observed EOF1 pattern demonstrates the strengthen- enhancement of SAH after day − 6. It has the most signifi- ing/weakening of the SAH’s main body with its center over cant role in the strengthening of SAH around day 0, when the IP–western TP region. The intensity variation of SAH the anomalous rainfall band occupies the south and east displays a periodicity of 10–36-day with the peak around Asian regions. The horizontal advection process of 27-day. adv is also favorable to the SAH’s enhancement, The regressions of the observed anomalies in Z100, especially during days − 6 to days 0. 100 hPa wind fields, and precipitation onto the observed normalized PC1 reveal the following features: The SAH’s strengthening begins with a weak anomalous high centered 4.2 T he models performances over the IP–western TP, and a low poleward of the high located over the Barents Sea and Novaya Zemlya. Above We next evaluate the selected three models’ performances pattern is invigorated gradually and matures on day 0. Before of above dynamical processes. Figure 10 displays the lat- day − 6, an enhanced rainfall belt migrates northward con- itude-time sections of regression of simulated subseasonal spicuously from the equatorial Indian Ocean, the Bay of anomalies of Q _z, S and adv onto each Bengal, the South China Sea and the Western North Pacific t t t and finally occupies almost the whole Indian subcontinent, models’ normalized PC1. ACCESS1-3 simulates the Indochina Peninsula and subtropical East Asian regions northward propagation of negative vorticity tendency gen- between 20°–40°N on day 0. erated by Q _z and S (Fig. 10a, b). The simu- 1 Three dynamical processes responsible for the subsea- t t sonal intensity variation of SAH are investigated, lated westward advection of negative vorticity anomalies 1 3 Subseasonal intensity variation of the South Asian high in relationship to diabatic heating:… respectively: vertical gradient of diabatic heating The process of Q _z contributes most significantly to Q _z , horizontal gradient of diabatic heating S, t t the enhancement of SAH during days − 3 to + 3 when the and horizontal advection of relative vorticity adv. anomalous rainfall band occupies 20°–40°N of the south and east Asian monsoon regions. The horizontal advection The dynamical process of S plays a role in the process of adv induces a westward transportation of enhancement of the southern part of SAH around days − 9 the negative vorticity, being favorable to the westward to -6 due to the horizontal gradient of diabatic heating. expansion of SAH. Fig. 8 latitude-time section (averaged over 70°–120°E) of regressed subseasonal anomalies on PC1 for ERA- Interim in a vertically integrated apparent heating source ⟨Q ⟩ − 2 (shaded; W m ) and vertically integrated apparent mositure sink ⟨Q ⟩ (purple contour with − 2 interval of 2 W m ); b − 11 − 2 Q _z (shaded; 10 s ) and precipitation (purple contour with interval of − 1 0.2 mm d ); c Q _z − 11 − 2 (shaded; 10 s ) and S (contour with interval of − 11 − 2 2 × 10 s ); d longitude-time section (averaged over 20°–35°N) of regressed subseasonal anomalies on PC1 of vorticity advection − 11 − 2 ( adv ) (shaded; 10 s ) 1 3 W. Shang et al. Fig. 10 Upper panels: latitude-time section of regressed subseasonal − 11 − 2 anomalies in Q _z (10 s ) on PC1 averaged over 70°– 120°E for a ACCESS1-3, d MRI-ESM1 and g IPSL-CM5A-MR. Middle panels: the same as upper panels, but for S. Bottom: the same as upper panels, but for longitude-time section (averaged over 20°–35°N) of regressed subseasonal anomalies in adv for the reproduction of the subseasonal variation of the SAH. Why some of the models cannot reproduce this northward propagation of precipitation anomalies? Com- parison of Figs. 1 and 2, it seems that the model’s ability to reproduce subseasonal intensity variation of the SAH has no robust connection with its capacity in simulating the Fig. 9 Regression of observed subseasonal anomalies in S SAH climatology. One of the possible reasons may result from the horizontal resolution of models. In CMIP5, the − 11 − 2 − 11 − 2 (green line; 10 s ), adv (purple line; 10 s ), Q _z t t simulation of precipitation has been further improved to a − 11 − 2 − 6 − 1 (blue line; 10 s ) and (red line; 10 s ) on PC1 averaged certain extent compared to the CMIP3. However, the bias over 20°–40°N, 40°–120°E of the precipitation simulation is still seen in current gen- eral couple models (Chen and Frauenfeld 2014; Yao et al. 2017; Fang et al. 2017). In this paper, it can be found that The 18 CMIP5 models capture the zonally elongated the top four models in BG models have the same horizontal structure of the climatological SAH in general. All the 18 resolution (192 × 144), which could be a suitable horizon- CMIP5 models simulate the SAH’s monopole pattern of tal resolution for the precipitation simulation related to the EOF1 to a certain extent. The models are divided into three SAH subseasonal variation. The lower or higher horizon- groups according to the subseasonal pattern correlation coef- tal resolution can both influence the models performance, ficient (r ) between the observed EOF1 and the models’. EOF which may due to the convective parameterization scheme One model is selected from each group as their individual or microphysis scheme (Giorgi and Marinucci 1996; Chan representatives: ACCESS1-3 from the BG with r above EOF et al. 2013; Huang et al. 2013; Kan et al. 2015; Yang et al. 0.8, MRI-ESM1 from the GG with r between 0.8 and 0.5, EOF 2015). and IPSL-CM5A-MR from the LG with r less than 0.5. EOF Moreover, the boreal summer intraseasonal oscillation ACCESS1-3 (BG) simulates a realistic subseasonal intensity (BSISO) is a dominant signal over the tropical and Asian anomaly of SAH. MRI-ESM1 (GG) reproduces a slightly monsoon region (Yasunari 1979; Krishnamurti and Subra- weaker one with the center of anticyclonic anomaly shifting manian 1982; Lau and Chan 1986; Jiang et al. 2004; Anna- to the western part of TP. IPSL-CM5A-MR (LG) simulates a malai and Sperber 2005; Wang et al. 2005; Lee et al. 2013; much weaker one compared to the observation, ACCESS1-3 Demott et al. 2013). The simulation of large-scale circula- and MRI-ESM1. tion such as mean wind, moisture fields, or western North Our results indicate that, the capacity of the 18 mod- Pacific subtropical high (WNPSH), and the ocean boundary els to simulate the SAH’s intensity variation has a close conditions such as sea surface temperature (SST) could also conjunction with simulations of the strength as well as the lead to the bias of models reproduction of northward propa- northward propagating features of anomalous rainfall band gation BSISO signal (Fu and Wang 2004; Park et al. 2009; over the south and east Asian monsoon regions. The real- Levine and Turner 2012; Demott et al. 2013, 2014; Fang istic simulation of precipitation anomalies in BG models et al. 2017), and thus may influence the ability of model to befits reproduction of the dynamical processes related to the simulation of SAH variation. The relationship between diabatic heating. When the simulated subseasonal precipi- the subseasonal variations of SAH and the BSISO is not tation anomalies (for example, LG models) are too weak analyzed thoroughly in this study and need a further inves- and show deficiency in propagating features, the simulated tigation both in observation and models. More exploration SAH’s subseasonal intensity anomalies are dramatically about the above issues should be taken into account in future weakened. studies. The present study shows that the strength and northward propagation of precipitation from the tropics is important 1 3 Subseasonal intensity variation of the South Asian high in relationship to diabatic heating:… 1 3 W. Shang et al. Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creat iveco mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. References Annamalai H, Sperber KR (2005) Regional heat sources and the active and break phases of boreal summer intraseasonal (30–50 day) variability. J Atmos Sci 62:2726–2748 Chan SC, Kendon EJ, Fowler HJ, Blenkinsop S, Ferro CAT, Stephen- son DB (2013) Dose increasing the spatial resolution of a regional climate model improve the simulatated daily precipitation in a regional climate model? Clim Dyn 41:1475–1495. https ://doi. org/10.1007/s0038 2-012-1568-9 Chen L, Frauenfeld OW (2014) A comprehensive evaluation of pre- cipitation simulations over China based on CMIP5 multimodel ensemble projections. J Geophys Res Atmos 119:5767–5786. https ://doi.org/10.1002/2013J D0211 90 Chen Y, Zhai PM (2016) Mechanisms for concurrent low-latitude cir- culation anomalies responsible for persistent extreme precipitation in the Yangtze River Valley. Clim Dyn 47:989–1006. https ://doi. org/10.1007/s0038 2-015-2885-6 Chen MY, Shi W, Xie PP, Silva VBS, Kousky VE, Higgins RW, Janow- iak JE (2008) Assessing objective techniques for gauge-based analyses of global daily precipitation. J Geophys Res 113:D04110. https ://doi.org/10.1029/2007J D0091 32 Cherchi A, Annamalai H, Masina S, Navarran A (2014) South Asian summer monsoon and the eastern Mediterranean climate: the monsoon-desert mechanism in CMIP5 simulations. J Clim 15:6877–6903. https ://doi.org/10.1175/JCLI-D-13-00530 .1 Dee DP, Uppala SM, Simmons AJ, Berrisford P, Poli P, Kobayashi S, Andrae U, Balmaseda MA, Balsamo G, Bauer P, Bechtold P (2011) The ERA-Interim reanalysis: configuration and per - formance of the data assimilation system. Q J R Meteorol Soc 137:553–597. https ://doi.org/10.1002/qj.828 Demott CA, Stan C, Randall DA (2013) Northward propagation mechanisms of the Boreal Summer Intraseasonal Oscillation in the ERA-Interim and SP-CCSM. J Clim 26:1973–1992. https :// doi.org/10.1175/JCLI-D-12-00191 .1 Demott CA, Stan C, Randall DA, Branson MD (2014) Instraseasonal variation in coupled GCMs: the roles of ocean feedbacks and model physics. J Clim 27:4970–4995. ht tp s : // do i .o rg / 10 . 11 75 / JCLI-D-13-00760 .1 Fig. 11 Regression of subseasonal anomalies in S (green line; Duan AM, Wu GX (2005) Role of the Tibetan Plateau thermal forcing in the summer climate patterns over subtropical Asia. Clim Dyn − 11 − 2 − 11 − 2 10 s ), adv (purple line; 10 s ), Q _z(blue line; t t 24:793–807. https ://doi.org/10.1007/s0038 2-004-0488-8 − 11 − 2 − 6 − 1 Duan AM, Hu J, Xiao ZX (2013) The Tibetan Plateau summer mon- 10 s ) and (red line; 10 s ) on PC1 averaged over 20°– soon in the CMIP5 simulations. J Clim 1:7747–7766. https ://doi. 40°N, 40°–120°E for a ACCESS1-3, b MRI-ESM1 and c IPSL- org/10.1175/JCLI-D-12-00685 .1 CM5A-MR, respectively Ertel H (1942) Ein neuer hydrodynamische wirbdsatz. Metror Z Braun- schweig 59:33–49 Fang YJ, Wu PL, Wu TW, Wang ZZ, Zhang L, Liu XW, Xin XG, Acknowledgements We thank anonymous reviewers for comments and Huang AN (2017) An evaluation of boreal summer intra- suggestions that have substantially improved the manuscript. This work seasonal oscillation simulated by BCC_AGCM2.2. Clim Dyn was jointly supported by the National Natural Science foundation of 48:3409–3423. https ://doi.org/10.1007/s0038 2-016-3275-4 China under Grants 41621005, 41675067 and 41330420. The ERA- Freychet N, Hsu H-H, Chou C, Wu C-H (2015) Asian summer mon- Interim daily data is downloaded from http://apps.ecmwf.int/dat ase ts/. soon in CMIP5 project: a link between the change in extreme The CMIP5 data is available at the Institute of Meteorology at Freie precipitation and monsoon dynamics. J Clim 15:1477–1493. Universität Berlin and German Climate Computing Center (DKRZ). https ://doi.org/10.1175/JCLI-D-14-00449 .1 The authors thank Dr. Patricia Margerison for editing the manuscript. 1 3 Subseasonal intensity variation of the South Asian high in relationship to diabatic heating:… Fu XH, Wang B (2004) Differences of boreal summer intraseasonal Lee J-Y, Wang B, Wheeler MC, Fu X, Waliser DE, Kang I-S (2013) oscillations simulated in an atmosphere-ocean coupled model Real-time multivariate indices for the boreal summer intraseasonal and an atmosphere-only model. J Clim 17:1263–1271 oscillation over the Asian summer monsoon region. Clim Dyn Gao Y, Wang HJ, Jiang DB (2015) An intercomparison of CMIP5 40:493–509. https ://doi.org/10.1007/s0038 2-012-1544-4 and CMIP3 models for interannuanl variability of summer Levine RC, Turner AG (2012) Dependence of Indian monsoon rain- precipitation in Pan-Asian monsoon region. Int J Climatol fall on moisture fluxed across the Arabian Sea and the impact of 35:3770–3780. https ://doi.org/10.1002/joc.4245 coupled mode sea surface temperature biases. Clim Dyn 38:2167– Ge J, You QL, Zhang YQ (2017) The influence of the Asian sum- 2190. https ://doi.org/10.1007/s0038 2-011-1096-z doi mer monsoon onset on the northward movement of the South Li RQ, Lv SH, Han B, Gao YH, Meng XH (2017) Projections of South Asian high towards the Tibetan Plateau and its thermodynamic Asian summer monsoon precipitation based on 12 CMIP5 models. mechanism. Int J Climatol 38:543–553. https://doi.or g/10.1002/ Int J Climatol 37:94–108. https ://doi.org/10.1002/joc.4689 joc.5192 Liebmann B, Smith CA (1996) Description of a complete (interpo- Gilman DL, Fuglister LF, Mitchell JM (1963) On the power spectrum lated) outgoing longwave radiation dataset. Bull Am Meteorol of red noise. J Atmos Sci 20:182–184 Soc 77:1275–1277 Giorgi F, Marinucci MR (1996) Improvements in the simulation of Lin H (2009) Global extratropical response to diabatic heating vari- surface climatology over the European region with a nested ability of the Asian summer monsoon. J Atmos Sci 66:2697–2713. modeling system. Geophys Res Lett 23:273–276. https ://doi.https ://doi.org/10.1175/2009J AS300 8.1 org/10.1029/96GL0 0050 Liu YM, Wu GX, Liu H, Liu P (2001) Condensation heating of the Hoskins BJ, Mclntyre M, Robertson AW (1985) On the use and Asian Summer monsoon and the subtropical anticyclones in the significance of isentropic potential vorticity maps. Quart J R eastern hemispher. Clim Dyn 17:327–338 Meteor Soc 111:877–946 Liu YM, Wu GX, Ren RC (2004) Relationship between the subtropi- Huang G, Qu X, Hu KM (2011) The impact of the tropical Indian cal anticyclone and diabatic heating. J Clim 17: 682–698. https:// Ocean on the South Asian high in boreal summer. Adv Atmos doi.org/10.1175/1520-0442(2004)017,0682:RBTSA A.2.0.CO;2 Sci 28:421–432. https ://doi.org/10.1007/s0037 6-010-9224-y Liu YM, Hoskins BJ, Blackburn M (2007) Impact of Tibetan orogra- Huang D-Q, Zhu J, Zhang Y-C, Huang A-N (2013) Uncertainties on phy and heating on the summer flow over Asia. J Meteor Soc Jpn the simulated summer precipitation over Eastern China from the 85B:1–19. https ://doi.org/10.2151/jmsj.85B.1 CMIP5 models. J Geophys Res Atmos 118:9035–9047. https:// Liu BQ, Wu GX, Mao JY, He JH (2013) Genesis of the South Asian doi.org/10.1002/jgrd.50695 high and its impact on the Asian summer monsoon onset. J Clim Jia XL, Yang S (2013) Impact of the quasi-biweekly oscillation over 26:2976–2991. https ://doi.org/10.1175/JCLI-D-12-00286 .1 the western North Pacific on East Asian subtropical monsoon Luo HB, Yanai M (1984) The large-scale circulation and heat sources during early summer. J Geophys Res Atmos 118:4421–4434. over the Tibetan Plateau and surrounding areas during the early https ://doi.org/10.1002/jgrd.50422 summer of 1979. Part II: heat and moisture budgets. Mon Weather Jiang XA, Li T, Wang B (2004) Structures and mechanisms of the Rev 112:966–989 northward propagating boreal summer intraseasonal oscillation. Mason RB, Anderson CE (1963) The development and decay of J Clim 17:1022–1039 the 100-mb summertime anticyclone over southern Asia. Jiang XW, Li YQ, Yang S, Wu RG (2011) Interannual and interdec- Mon Weather Rev 91:3–12. ht tp s : // d oi . or g/ 1 0. 1 17 5 /1 5 20 - adal variations of the South Asian and western Pacific subtropical 0493(1963)091,0003:TDADO T.2.3.CO;2 highs and their relationships with Asian-Pacific summer climate. Park S, Hong S-Y, Byun Y-H (2009) Precipitation in boreal sum- Meteor Atmos Phys 113:171–180. https ://doi.org/10.1007/s0070 mer simulated by a GCM with two convective parameteriza- 3-011-0146-8 tion schemes: implications of the Intraseasonal Ocsillation for Jin FF, Hoskins BJ (1995) The direct response to tropical heating in dynamic seasonal predicition. J Clim 23:2801–2816. https ://doi. a baroclinic atmosphere. J Atmos Sci 52:307–319. https ://doi.org/10.1175/2010J CLI32 83.1 org/10.1175/1520-0469(1995)052,0307:TDRTT H.2.0.CO;2 Qu X, Huang G (2012) An enhanced influenced of tropical Indian Jin Q, Yang X-Q, Sun X-G, Fang J-B (2013) East Asian summer mon- Ocean on the South Asian high after the late 1970s. J Clim soon circulation structure controlled by feedback of condensa- 25:6930–6941. https ://doi.org/10.1175/JCLI-D-11-00696 .1 tional heating. Clim Dyn 41:1885–1897. https://doi.or g/10.1007/ Qu X, Huang G (2015) The decadal variability of the tropical Indian s0038 2-012-1620-9 Ocean SST-the South Asian High relation: CMIP5 model Kan MY, Huang AN, Zhao Y, Zhou Y, Yang B, Wu HM (2015) Evalu- study. Clim Dyn 45:273–289. https ://doi.or g/10.1007/S0038 ation of the summer precipitation over the China simulated by 2-014-2285-3 BCC_CSM model with different horizontal resolutions during the Randel WJ, Park M (2006) Deep convective influence on the Asian recent half century. J Geophys Res Atmos 120:4657–4670. https summer monsoon anticyclone and associated tracer variability ://doi.org/10.1002/2015J D0231 31 observed with Atmospheric Infrared Sounder (AIRS). J Geophys Krishnamurthy V, Shukla J (2000) Intraseasonal and interannual vari- Res 111:271–273. https ://doi.or g/10.1029/2005J D0064 90 D12314 ability of rainfall over India. J Clim 13:4366–4377. https ://doi. Ren XJ, Yang DJ, Yang X-Q (2015) Characteristics and mechanisms org/10.1175/1520-0442(2000)013,0001:IAIVO R.2.0.CO;2 of the subseasonal eastward extension of the South Asian high. J Krishnamurti TN (1973) Tibetan High and upper tropospheric tropi- Clim 28:6799–6822. https://do i.org/10.1175/JCLI-D-14-00682.1 cal circulation during northern summer. Bull Am Meteorol Soc Shi J, Qian WH (2016) Connection between anomalous zonal activities 54:1234–1249 of the South Asian high and Eurasian summer climate anomalies. Krishnamurti TN, Subramanian D (1982) The 30–50 day mode at J Clim 15:8249–8267. https://doi.or g/10.1175/JCLI-D-15-0823.1 850mb during MONEX. J Atmos Sci 39:2088–2095 Sooraj KP, Terray P, Xavier P (2016) Sub-seasonal behavior of Asian Lau K-M, Chan PH (1986) Aspects of the 40–50 day oscillation during summer monsoon under a changing climate: assessments using the northern summer as inferred from outgoing long wave radia- CMIP5 models. Clim Dyn 46:4003–4025. https://doi.or g/10.1007/ tion. Mon Weather Rev 114:1354–1367s0038 2-015-2817-5 Lee J-Y, Wang B (2014) Future change of global monsoon in Tao SY, Zhu FK (1964) The 100-mb flow patterns in southern Asia in summer and its relation to the advance and retreat of the CMIP5. Clim Dyn 42:101–119. https ://doi.org/10.1007//s0038 2-012-1564-0 1 3 W. Shang et al. West-Pacific subtropical anticyclone over the far east. Acta Mete - Yang SY, Li T (2016) Zonal shift of the South Asian High on subsea- orol Sin 34:385–396 (in Chinese) sonal timescale and its relation to the summer rainfall anomaly Taylor K, Stouffer R, Meehl G (2012) An overview of CMIP5 and in China. Quart J R Meteor Soc 142:2324–2335. https ://doi. experiments design. Bull Am Meteor Soc 93:485–498. https :// org/10.1002/qj.2826 doi.org/10.1175/BAMS-D-11-00094 .1 Yang J, Qing B, Wang B, Gong D-Y, He H, Gao M-N (2014) Distinct Wang LJ, Ge J (2016) Relationship between low-frequency oscillations quasi-biweekly features of the subtropical East Asian monsoon of atmospheric heat source over the Tibetan Plateau and longi- during early and late summer. Clim Dyn 42:1469–1486. https :// tudinal oscillations of the South Asia high in the summer. Chin doi.org/10.1007/s0038 2-013-1728-6 J Atmos Sci 40(4):853–863. https ://doi.org/10.3878/j.issn.1006- Yang B, Zhang YC, Qian Y, Wu TW, Huang AN, Fang YJ (2015) 9895.1509.15164 (in Chinese) Parametric sensitivity analysis for the Asian summer monsoon Wang B, Webster PJ, Teng H (2005) Antecedents and self-induction precipitation simulation in the Beijing Climate Center AGCM, of the active-break Indian summer monsoon. Geophys Res Lett version 2.1. J Clim 28:5622–5644. https ://doi.org/10.1175/JCLI- 32:L04704. https ://doi.org/10.1029/2004G L0209 96D-14-00655 .1 Wei W, Zhang RH, Wen M, Rong XY, Li T (2014) Impact of Indian Yang J, Qing B, Wang B, He HZ, Gao MN, Gong DY (2017) Char- summer monsoon on the South Asian high and its influence on acterizing two types of transient intraseasonal oscillations in the summer rainfall over China. Clim Dyn 43:1257–1269. https://doi. Eastern Tibetan Plateau summer rainfall. Clim Dyn 48:1749– org/10.1007/s0038 2-013-1938-y 1768. https ://doi.org/10.1007/s0038 2-016-3170-z Wu GX, Liu HZ (1998) Vertical vorticity development owing to Yao JC, Zhou TJ, Guo Z, Chen XL, Zou LW, Sun Y (2017) Improved down-sliding at slantwise isentropic surface. Dyn Atmos Oceans performance of high-resolution tropospheric models in simulating 27:715–743 the East-Asian summer monsoon rainbelt. J Clim 30:8825–8840. Wu GX, Liu YM, Liu P (1999) The effect of spatially nonuniform heat-https ://doi.org/10.1175/JCLI-D-16-0372.1 ing on the formation and variation of subtropical high. Part I: scale Yasunari T (1979) Cloudiness fluctuations associated with the northern analysis. Acta Meteorol Sin 57:257–263 (in Chinese) hemisphere summer monsoon. J Meteor Soc Jpn 57:227–242 Wu GX, He B, Liu YM, Bao Q, Ren RC (2015) Location and variation Zarrin A, Ghaemi H, Azadi M, Farajzadeh M (2010) The spatial of the summertime upper-troposphere temperature maximum over pattern of summertime subtropical anticyclones over Asia and South Asia. Clim Dyn 45:2757–2774. https ://doi.org/10.1007/ Africa: a climatological review. Int J Climatol 30:159–173. https s0038 2-015-2506-4 ://doi.org/10.1002/joc.1879 Xie PP, Yatagai A, Chen MY, Hayasaka T, Fukushima Y, Liu CM, Zhang CD, Ling J (2012) Potential vorticity of the Madden–Julian Yang S (2007) A gauge-based analysis of daily precipitation over oscillation. J Amos Sci 69:65–78. https ://doi.or g/10.1175/ East Asia. J Hydrometeor 8:607–627. https ://doi.org/10.1175/ JAS-D-11-081.1 JHM58 3.1 Zhang Q, Wu GX, Qian YF (2002) The bimodality of the 100 hPa Xue X, Chen W, Chen SF (2017) The climatology and interannual vari- South Asia high and its relationship to the climate anomaly over ability of the South Asian high and its relationship with ENSO in East Asia in summer. J Meteor Soc Jpn 80:733–744 CMIP5 models. Clim Dyn 48:3507–3528. https://doi.or g/10.1007/ Zhang PQ, Yang S, Kousky VE (2005) South Asian high and Asian- s0038 2-016-3281-6 Pacific-American climate teleconnection. Adv Atmos Sci 22:915– Yanai M, Tomita T (1998) Seasonal and interannual variability of 923. https ://doi.org/10.1007/BF029 18690 atmospheric heat sources and moisture sinks as determined from Zhang PF, Liu YM, He B (2016) Impact of East Asian summer mon- NCEP-NCAR reanalysis. J Clim 11:463–482 soon heating on the interannual variation of the South Asian high. Yanai M, Esbensen S, Chu J-H (1973) Determination of bulk proper- J Clim 29:159–173. https ://doi.org/10.1175/JCLI-D-15-0118.1 ties of tropical cloud clusters from large-scale heat and moisture budgets. J Atmos Sci 30: 611–627. https://doi.or g/10.1175/1520- 0442(1998)011,0463:SAIVO A.2.0.CO;2 1 3
Climate Dynamics – Springer Journals
Published: May 30, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera