simulation of the interface between the indian summer ... · mpi-esm is more skillful than...

ADVANCES IN ATMOSPHERIC SCIENCES, VOL. 33, MARCH 2016, 294–308

Simulation of the Interface between the Indian Summer Monsoon and theEast Asian Summer Monsoon: Intercomparison between

MPI-ESM and ECHAM5/MPI-OM

Yiran GUO1, Jie CAO∗1,2, Hui LI1, Jian WANG1, and Yuchao DING1

1Department of Atmospheric Sciences, Yunnan University, Kunming 6500912Yunnan Key Laboratory of International Rivers and Transboundary Eco-security, Kunming 650091

(Received 15 March 2015; revised 13 September 2015; accepted 28 September 2015)

ABSTRACT

The time-mean and interannual variability of the interface between the Indian summer monsoon and East Asian summermonsoon (IIE) was assessed using both Max-Planck-Institute Earth System Model (MPI-ESM) and ECHAM5/MPI-OM andby calculating diagnostics and skill metrics around the IIE area. Progress has been made in modeling these aspects by movingfrom ECHAM5/MPI-OM to MPI-ESM. MPI-ESM is more skillful than ECHAM5/MPI-OM in modeling the time-mean stateand the extreme condition of the IIE. Though simulation of the interannual variability significantly deviates to some extentin both MPI-ESM and ECHAM5/MPI-OM, MPI-ESM-LR shows better skill in reflecting the relationship among sea surfacetemperature anomalies over the Pacific, circulation anomalies over East Asia, and IIE variability. The temperature becomeswarmer under the RCP2.6 and RCP8.5 scenarios in comparison with the historical experiments, but the position of the IIEand the key physical process in relation to the IIE variability almost remains the same, suggesting that the Indian summermonsoon tends to change in phase with the East Asian summer monsoon under each RCP scenario. The relatively realisticdescription of the physical processes modulated by terrain in MPI-ESM may be one of the most important reasons whyMPI-ESM performs better in simulating the IIE.

Key words: Asian summer monsoon, IIE, MPI-ESM, ECHAM5/MPI-OM, intercomparison

Citation: Guo, Y. R., J. Cao, H. Li, J. Wang, and Y. C. Ding, 2016: Simulation of the interface between the Indian summermonsoon and the East Asian summer monsoon: Intercomparison between MPI-ESM and ECHAM5/MPI-OM. Adv. Atmos.Sci., 33(3), 294–308, doi: 10.1007/s00376-015-5073-z.

1. IntroductionThe Asian monsoon influences more than 60% of the

world’s population, through controlling droughts and floods,and plays a key role in the Earth’s climate system (Flohn,1957; Ding, 1994; Webster et al., 1998; Huang et al., 2012;Wu et al., 2012). Since the Asian monsoon varies greatly onmultiple temporal and spatial scales, the performance of sim-ulating the activity of the Asian monsoon has become oneof the most important indicators to assess the quality of anumerical model. ECHAM5, released in 2003, has been ex-tensively used in studying the variation of the Asian summermonsoon because of its high simulation efficiency (Roeckneret al., 2003, 2006; Marsland et al., 2003). For example, Anna-malai et al. (2007) studied the relationship between the Indiansummer monsoon and ENSO with ECHAM5/MPI-OM. Liand Zhang (2009) found that ECHAM5/MPI-OM simulatesthe wind onset of the Asian summer monsoon reasonably

∗ Corresponding author: Jie CAOEmail: [email protected]

well. Lu and Fu (2010) compared 12 CMIP3 models in-cluding ECHAM5/MPI-OM. They found that these mod-els can successfully reproduce the essential features of thepresent-day interannual variability in rainfall-related circula-tions. Xie et al. (2009) introduced the Indian Ocean capacitoreffect on the Indo–western Pacific climate using ECHAM5.Zhou et al. (2009) tested the ability of ECHAM5 and 10other GCMs in capturing the leading modes of the interan-nual variability of the Asian–Australian monsoon. Zhao andFu (2010) found ECHAM5/MPI-OM can reproduce the gen-eral distributions of precipitation, specific humidity and windvectors for the years 1978–2000. Huang et al. (2014) pointedout that a new global climate model integrating ECHAM5with Nucleus for European Modelling of the Ocean (NEMO)can successfully capture the distribution of SST, precipita-tion, the seasonal cycle of equatorial Pacific SST and precip-itation, and the circulation of the East Asian summer mon-soon. The main interannual variations of the tropical PacificSST, the East Asian and western North Pacific climate, andthe East Asia–Pacific Pattern and El Nino, are reproducedwell. In 2012, MPI-ESM, whose atmospheric component is

© Institute of Atmospheric Physics/Chinese Academy of Sciences, and Science Press and Springer-Verlag Berlin Heidelberg 2016

MARCH 2016 GUO ET AL. 295

ECHAM6, was released by the Max Planck Institute for Me-teorology in Germany (Giorgetta et al., 2013). The land sur-face model (JSBACH) was extended by dynamic vegetation,for climate-consistent development of the geographic distri-bution of vegetation (Brovkin et al., 2009; Reick et al., 2013).The surface albedo scheme was improved to use a weightedaverage of the ice albedo and the water albedo, depending onsolar zenith angle (Roeckner et al., 2012). The representationof shortwave radiative transfer, the height of the model top,model tuning and convective triggering have also changedin ECHAM6 (Hagemann et al., 2013; Stevens et al., 2013).These changes lead ECHAM6 to provide a better represen-tation than ECHAM5 of the mean climate (Stevens et al.,2013). For the ocean model, some improvements have beenmade in MPI-ESM-MR, including increased vertical resolu-tion, while in MPI-ESM-LR it remains the same resolutionas ECHAM5/MPI-OM (Jungclaus et al., 2013).

It is widely accepted that the Asian summer monsoon hastwo subsystems: the Indian summer monsoon system (Krish-namurti and Bhalme, 1976) and the East Asian summer mon-soon system (Tao and Chen, 1987). There is an interface be-tween the Indian and East Asian summer monsoons, referredto as the IIE (Jin and Chen, 1982; Lau and Li, 1984; Zhu etal., 1986; Tao and Chen, 1987; Ding, 1994; Wang and LinHo,2002). Cao et al. (2012) found that the IIE is approximatelylocated along (18◦N, 102.98◦E) at 900 hPa, (20◦N, 101.68◦E)at 850 hPa, (22◦N, 100.88◦E) at 800 hPa, (24◦N, 100.96◦E) at750 hPa, (26◦N, 101.08◦E) at 650 hPa, and (28◦N, 99.32◦E)at 600 hPa, and its variability can be quantified by the IIEindex. As the IIE is a variable that describes the interac-tion between the East Asian summer monsoon and Indiansummer monsoon, simulation of IIE variation may be cru-cial in understanding the impact of the two monsoon subsys-tems on local weather and climate. If the temporal and spa-tial evolution of the IIE can be better reproduced in GCMs,the skill of numerical models in simulating the Asian sum-mer monsoon may also be improved. However, the perfor-mance of both ECHAM5/MPI-OM and MPI-ESM in simu-lating the IIE is not yet clear. The aim of this study, there-fore, was to establish whether MPI-ESM can provide a betterrepresentation in modeling the IIE, and, if this is the case,then also determine the kinds of improvements that can beachieved.

The remainder of this paper is organized as follows: Themodel and observation data are described in section 2. Insection 3, the abilities of ECHAM5/MPI-OM, MPI-ESM-LRand MPI-ESM-MR to reproduce the mean state of the atmo-spheric reanalysis data of the equivalent potential tempera-ture around the IIE are analyzed. Section 4 investigates thecapabilities of the three numerical models to represent themean state and extreme condition of the IIE. The relation-ship between SST, circulation and middle–upper troposphereanomalies with IIE interannual variation is also discussed.The future changes of the IIE under the RCP2.6 and 8.5 sce-narios are explored in section 5. A summary and discussionare provided in section 6.

2. Model and dataThree models—ECHAM5/MPI-OM, MPI-ESM-LR and

MPI-ESM-MR—were employed in this study. In MPI-ESM-LR, the GCM, ECHAM6, is run at T63 horizontal resolu-tion (approximately 1.875◦ on a Gaussian grid) with 47 hy-brid sigma-pressure levels, and the ocean circulation model,MPI-OM, is run at a 1.5◦ horizontal resolution (near theequator) and 40 z-levels in a bipolar grid. In MPI-ESM-MR, ECHAM6 is run with 95 hybrid sigma-pressure lev-els with the same horizontal resolution as MPI-ESM-LR,and MPI-OM is run at 0.4◦ horizontal resolution with thesame vertical levels as MPI-ESM-LR. The top of the GCMreaches 0.01 hPa in both MPI-ESM-LR and MPI-ESM-MR.ECHAM5/MPI-OM has the same horizontal resolution asMPI-ESM-LR, except the 31 vertical levels of the GCM ex-tend only to 10 hPa in ECHAM5. Further details of the threemodels can be found in Stevens et al. (2013), Jungclaus et al.(2013), Reick et al. (2013) and Ilyina et al. (2013). The modeldata were downloaded from http://www-pcmdi.llnl.gov/ andhttps://esg.llnl.gov:8443/.

The 20th century reanalysis data (version 2) were pro-vided by NOAA/OAR/ESRL Physical Sciences Division forthe summers (June–July–August; JJA) (Whitaker et al., 2004;Compo et al., 2006, 2011). The resolution of the reanalysisdata is 2◦ × 2◦ and there are 16 pressure levels from 1000to 200 hPa. The SST data were from ERSST.v3b (Smithet al., 2008). The resolution of the SST data is 2◦ × 2◦ ona latitude–longitude grid. The study area was selected as(16◦–30◦N, 90◦–110◦E). The historical run spanned a totalof 27 years from 1979 to 2005, and the future scenario runsspanned from 2006 to 2100. The IIE data were provided byCao et al. (2012).

3. Comparisons of the longitude–pressure sec-tion of equivalent potential temperaturebetween observational data and modelingdata around the IIE area

Figure 1a shows that the observational summer equiva-lent potential temperature along 21.45◦N has a typical saddle-shaped pattern. The equivalent potential temperature at thesame isobaric level has a low–high–low pattern with an in-crease in longitude. The equivalent potential temperature atthe same longitude features a high–low–high pattern as airpressure decreases. Two high centers with values above 351K occur at the lower and mid–high levels of the troposphere,respectively. The saddle-shaped pattern of equivalent poten-tial temperature is reproduced well in the three models butwith different levels of superiority. For the positions of thetwo low centers around the mid-troposphere, the modelingresults of both MPI-ESM-MR and MPI-ESM-LR are moreaccurate than those of ECHAM5/MPI-OM (Figs. 1b and 1c).However, for the intensity of equivalent potential tempera-ture, the modeling result of ECHAM5/MPI-OM is closer toobservations. The negative deviations appear at the mid–high

296 SIMULATION INTERCOMPARISON OF THE IIE VOLUME 33

Fig. 1. Section of JJA equivalent potential temperature along a Gaussian grid (21.45◦N) averaged between 1979 and2005 from (a) 20th century reanalysis data (version 2), (b) MPI-ESM-MR, (c) MPI-ESM-LR, and (d) ECHAM5/MPI-OM. Shaded areas denote the difference between modeling results and observations. Units: K.

levels of the troposphere in the three models, and are mostconspicuous in MPI-ESM-MR, with center values below −6K. The positive deviations mainly appear at the lower lev-els of the troposphere around 95◦E and 110◦E in the threemodels, and are most obvious in ECHAM5/MPI-OM. Thesections of JJA equivalent potential temperature along fiveother latitudes—17.72◦N, 19.59◦N, 23.32◦N, 25.18◦N and27.05◦N (figures omitted)—share a similar pattern as Fig. 1.

A Taylor diagram (Taylor, 2001), which quantitativelydemonstrates the modeling skill of the three models, is shownin Fig. 2. Comparing the correlation coefficients, standard de-viation and root-mean-square deviation of the three models,it can be seen that the modeling skill of MPI-ESM-MR, MPI-ESM-LR and ECHAM5/MPI-OM are similar at 19.59◦N,21.45◦N and 23.32◦N, but MPI-ESM-MR and MPI-ESM-LRpossess greater skill than ECHAM5/MPI-OM at 25.18◦N and27.05◦N.

4. Comparisons of the IIE state between ob-servations and simulations

4.1. Position of the IIEAccording to the definition of the IIE developed by Cao

et al. (2012) (the surface where the change of the equivalentpotential temperature, θE, with respect to the longitude, λ ,is zero, i.e., ∂θE/∂λ = 0), we obtained the modeled IIE po-sition along the Earth’s surface, averaged over 1979–2005,

in MPI-ESM-MR, MPI-ESM-LR and ECHAM5/MPI-OM(Fig. 3a). Figure 3a shows that at the Earth’s surface the IIEis distributed around 100◦E with a wavy pattern. The normal

0.5

1

0

1

0

1.50.1 0.2 0.3

0.40.5

0.6

0.7

0.8

0.9

0.95

0.99

Stan

dar

d d

evia

tion

RM

S D

Standard deviation

s1s1s1 s2s2s2 s3s3s3s4s4s4

s5s5

s5

s6s6

s6

REF 1.5

Fig. 2. Taylor diagram illustrating a statistical comparison withobservations of six equivalent geopotential temperature sectionsaround the IIE area. The labels s1–s6 denote the sections ofequivalent geopotential temperature along six Gaussian grids:17.72◦N, 19.58◦N, 21.45◦N, 23.35◦N, 25.18◦N and 27.05◦N.Blue coloring denotes MPI-ESM-MR, red denotes MPI-ESM-LR, and green denotes ECHAM5/MPI-OM. The black circledenotes the observations.


Fig. 3. Positions of the IIE: (a) normal; (b) more westward; (c)more eastward. Magenta denotes the observation, yellow de-notes MPI-ESM-MR, cyan denotes MPI-ESM-LR, and orangedenotes ECHAM5/MPI-OM. Shaded areas denote the terrain(units: m).

position of the IIE is reproduced well by the three models.In comparison with the observation, the simulated positionof the IIE is farther west in all three models. The positionof the IIE simulated by MPI-ESM-LR is the nearest to theobservation. The main errors appearing in the model relateto the wavy pattern of the IIE changing with latitude. Themaximum deviation is located in the northern part of the IIE.The anomalies of the IIE position are further shown in Figs.3b and 3c. Figure 3b shows that the negative anomalous po-sition of the IIE is reproduced by the three models, but theyare situated more westward than observed. The deviation inmodeling the wavy pattern of the IIE’s change with latitudeis largest in negative anomaly years, followed by normal andpositive anomaly years. The simulated pattern of the positiveanomalous position of the IIE mirrors the negative conditionto some extent. The three simulated positive anomalous posi-tions are situated more eastward than in the observation. Thewavy pattern of the IIE’s change with latitude in the positiveanomalous condition is reproduced well (Fig. 3c).

The corresponding Taylor diagram (Fig. 4) quantitativelyshows the capacity of the three models in simulating the IIEposition. The evaluation results presented in Fig. 4 coincidewith the results in Fig. 3. The models best simulate the pos-itive anomalous position of the IIE, followed by the normalposition of the IIE, and then the negative anomalous posi-tion of the IIE. However, even the lowest correlation coef-ficient (0.75) passes the significance test at the 99% confi-dence level. If we consider all three conditions of the IIEposition together, it is apparent (Fig. 4) that MPI-ESM-LRhas the highest capacity in simulating the IIE position, be-cause the distance between the evaluation and the observa-tional points is the shortest. MPI-ESM-MR takes secondplace. The capacities of the two new models are better thanthat of ECHAM5/MPI-OM.

4.2. SST and circulation anomalies associated with IIEvariability

Figure 5a shows observed JJA SST anomalies (SSTAs)regressed onto the IIE index. The SSTAs are similar to thedecaying phase of an El Nino event (or the developing phaseof La Nina). Previous studies have indicated that the decay-ing phase of El Nino tends to intensify the western Pacificsubtropical high (WPSH) (Huang et al., 1998; Chen, 2002;Sui et al., 2007), which can result in an enhanced WPSH witha farther south position and stronger Mei-yu front appearingalong the middle–lower reaches of the Yangtze River. Ananomalous cyclone in relation to the stronger Mei-yu frontfurther attracts the IIE to shift eastward (Cao et al., 2012). Allthree models capture the decaying El Nino pattern. However,ECHAM5/MPI-OM overestimates the strength of decayingEl Nino, while MPI-ESM-MR underestimates its strength.MPI-ESM-LR simulates the decaying El Nino pattern bestamong all three models. The fact that the results associatedwith ECHAM5/MPI-OM are quite different from those ofthe other two models may be caused by the large deviationin ECHAM5/MPI-OM in simulating the ENSO amplitude,which has been reduced in MPI-ESM (Jungclaus et al., 2013;


1

2

1

2

0

30.1 0.2 0.3

0.40.5

0.6

0.7

0.8

0.9

0.95

0.99

Stan

dar

d d

evia

tion

RM S D

Standard deviationREF 32

Fig. 4. Taylor diagram of normal (solid circles), more westward(triangles), and more eastward (squares) positions of the IIE.Diamonds denote the composition of the three positions of theIIE. Blue denotes MPI-ESM-MR, red denotes MPI-ESM-LR,and black denotes ECHAM5/MPI-OM. The black circle de-notes the observations.

Stevens et al., 2013). Actually, a general improvement fromCMIP3 to CMIP5 models in their representation of ENSO’samplitude, spectrum, and life cycle is apparent (Guilyardi etal., 2012; Bellenger et al., 2014).

Figure 6 shows the mean state of zonal wind and its

difference between the modeling results and observations at850 hPa. In summer, the interface between the easterly andwesterly flow appears east of 100◦E (Fig. 6a). Figures 6b–d show that all three models simulate the pattern of zonalwind well, but with deviation around 3.5 m s−1 in an ab-solute sense. The maximum negative deviation of the threemodels appears around the western coast of the Indo-ChinaPeninsula. The maximum positive deviation of MPI-ESM-MR and MPI-ESM-LR mainly appears in the eastern re-search domain, with a value above 1.5 m s−1. The positivedeviation of ECHAM5/MPI-OM is scattered in the north-western, north-central and southeastern domains. The skillof MPI-ESM-MR and MPI-ESM-LR in modeling summerzonal wind is similar, and is slightly improved comparedwith ECHAM5/MPI-OM. Because the IIE interface is de-fined along the surface isobaric level, we focus our analysishere at 24◦N, 850 hPa. The zero-line of zonal wind at 24◦Nlocated at 102◦E agrees with the IIE position at 24◦N (Fig.3a), suggesting the IIE position can be found in the obser-vational zonal wind field. The modeling results reproducethis distribution characteristic but with a certain level of de-viation. The position of the IIE varies significantly on theinterannual time scale (Cao et al., 2012). Figure 7a shows theobservational horizontal wind at 850 hPa regressed onto theobservational IIE time series for 1979–2005. The most pro-nounced feature in Fig. 6a is the enhanced anticyclone cen-tered around 20◦N, dominating the region from the westernPacific to East Asia. The easterly anomalies at the southernflank of the enhanced anticyclone appear around 10◦N, and

Fig. 5. SST regressed onto the IIE time series: (a) observation; (b) MPI-ESM-MR; (c)MPI-ESM-LR; (d) ECHAM5/MPI-OM. Shaded areas from light to dark denote statis-tical significance at the 90% and 95% confidence level.


Fig. 6. JJA zonal wind at 850 hPa: (a) 20th century reanalysis data (version 2); (b) MPI-ESM-MR; (c) MPI-ESM-LR; (d) ECHAM5/MPI-OM. Shaded areas denote the difference between the modeling results and observations. Units:m s−1. The green solid line represents the terrain.

the southwesterly anomalies at the northwestern and west-ern flank of the enhanced anticyclone appear between 20◦Nand 30◦N. In the remaining area, the anomalous signals arenot very significant. The enhanced anticyclonic pattern ap-pears in all three modeling results. These results indicate thatthe three models can reflect the key physical process impact-ing the interannual variability of the IIE. There are, however,some deviations in the three modeling results (Figs. 7b–d).The position of the enhanced anticyclone simulated by MPI-ESM-MR is located farther northward, with a center around25◦N, and southwesterly anomalies over the northwesternflank of the enhanced anticyclone and easterly anomaliesover the southern flank around the enhanced anticyclone arestronger than observed. The position of the enhanced anticy-clone simulated by MPI-ESM-LR is similar to the observa-tion. However, the southwesterly anomalies over the north-western flank of the enhanced anticyclone are weaker, andthe easterly anomalies over the southern flank around the en-hanced anticyclone are stronger than observed. The positionof the enhanced anticyclone simulated by ECHAM5/MPI-OM is more southerly, with its center around 16◦N, and boththe southwesterly anomalies over the northwestern flank ofthe enhanced anticyclone and the easterly anomalies overthe southern flank around the enhanced anticyclone, resem-

bling the wind anomalies simulated by MPI-ESM-MR, aretoo strong in comparison with the observation. This differ-ence is probably related to the overestimated ENSO SSTA inthe relationship between ENSO and IIE in ECHAM5/MPI-OM. If we calculate the spatial correlation coefficients be-tween the modeling results and observations around the IIEarea (0◦–40◦N, 90◦–140◦E) for both zonal and meridionalwinds, MPI-ESM-LR obtains the highest correlation coeffi-cients, with 0.65 for the zonal wind field and 0.67 for themeridional wind field. MPI-ESM-MR has the lowest cor-relation coefficients, with 0.52 for the zonal wind field and0.62 for the meridional wind field, and the correlation coeffi-cients associated with ECHAM5/MPI-OM are between thoseof MPI-ESM-LR and MPI-ESM-MR, at 0.64 for the zonalwind field and 0.55 for meridional wind field. Therefore,the circulation anomalies in relation to the IIE are best repre-sented by MPI-ESM-LR among the three models.

Figure 8 shows the geopotential height anomaly at 850hPa regressed onto the IIE time series. The anomalous highwith a center at around (20◦N, 130◦E) is consistent with theanomalous horizontal wind pattern around the same region(Figs. 7 and 8), suggesting that there is an enhanced WPSH.All three models capture the anomalous high over westernPacific. The maximum bias appears in ECHAM5/MPI-OM.


Fig. 7. Horizontal wind at 850 hPa regressed onto the IIE time series: (a) observation;(b) MPI-ESM-MR; (c) MPI-ESM-LR; (d) ECHAM5/MPI-OM. Shaded areas from lightto dark denote statistical significance at the 90% and 95% confidence level. The blacksolid line represents the terrain.

Fig. 8. Geopotential height at 850 hPa regressed onto the IIE time series: (a) obser-vation; (b) MPI-ESM-MR; (c) MPI-ESM-LR; (d) ECHAM5/MPI-OM. Shaded areasfrom light to dark denote statistical significance at the 90% and 95% confidence level.The black solid line represents the terrain.


However, while many areas pass the significance test at the95% confidence level in ECHAM5/MPI-OM, these areas—except those around the western Pacific—cannot be found inthe observation. In general, compared with ECHAM5/MPI-OM, MPI-ESM has made progress, especially in reducingfalse circulation signals associated with the IIE variabil-ity. For the vertical mean temperature at 200–500 hPa (Fig.9a), a positive center appears at 30◦N in East Asia, whichmight be related to the diabatic heating contributed by thesummer Mei-yu front (Liu et al., 1999; Wu et al., 1999).When the diabatic heating is stronger than normal along themiddle–lower reaches of the Yangtze River, the IIE is far-ther east than normal, and vice versa. Compared to MPI-ESM-LR and ECHAM5/MPI-OM, MPI-ESM-MR best cap-tures the observational pattern of the vertical mean temper-ature at 200–500 hPa. These results suggest that MPI-ESM-MR possesses a better simulation capability for the middle–upper troposphere, possibly because of the higher verticalresolution of MPI-ESM-MR compared with MPI-ESM-LRor ECHAM5/MPI-OM.

5. Future projections using MPI-ESM5.1. Position of the IIE

Figure 10 shows the simulated longitude–pressure sec-tions of equivalent potential temperature (along 21.45◦N) un-der the RCP2.6 and RCP8.5 scenarios during JJA. In all the

longitude–pressure sections (Figs. 10a–e), the equivalent po-tential temperatures exhibit a typical saddle-shaped patternwith the lower centers positioned around 100◦E, as in the his-torical experiment. Under the RCP2.6 scenario, the equiva-lent potential temperature tends to become warmer than thehistorical temperature at all isobaric surfaces, with more sig-nificant warming below 600 hPa. In comparison with theequivalent potential temperature under the RCP2.6 scenario,under the RCP8.5 scenario, the warming tendency of equiva-lent potential temperature is similar to that under the RCP2.6scenario but with a different warming pattern: the relativelyweak warming band (below 5 K) appears around the mid-dle troposphere in both MPI-ESM-MR and MPI-ESM-LR.Figures 10c and f show that the equivalent potential tempera-ture difference between RCP8.5 and RCP2.6 almost sharesthe same pattern: the equivalent potential temperature be-low 700 hPa becomes warmer than that above 700 hPa;namely, the positive difference has a stratified structure. Thelongitude–pressure sections of JJA equivalent potential tem-perature along five other latitudes (figures omitted) also sharethe similar pattern as Fig. 10.

Figure 11 shows the IIE’s position averaged from 2006to 2100 under both the RCP2.6 and RCP8.5 scenario. TheIIE’s position and its spatial structure under the RCP2.6 andRCP8.5 scenarios are generally similar to those in the his-torical experiments. These results, obtained by both MPI-ESM-MR and MPI-ESM-LR, suggest that global warmingmay not significantly change the IIE’s position; namely, the

Fig. 9. Vertical mean temperature at 200–500 hPa regressed onto the IIE time series:(a) observation; (b) MPI-ESM-MR; (c) MPI-ESM-LR; (d) ECHAM5/MPI-OM. Shadedareas from light to dark denote statistical significance at the 90% and 95% confidencelevel.


Fig. 10. Simulated JJA equivalent potential temperature along 21.45◦N averaged from 2006 to 2100: (a) MPI-ESM-MR under RCP2.6; (b)MPI-ESM-MR under RCP8.5; (c) difference between RCP8.5 and RCP2.6 obtained by MPI-ESM-MR; (d) MPI-ESM-LR under RCP2.6;(e) MPI-ESM-LR under RCP8.5; (f) difference between RCP8.5 and RCP2.6 obtained by MPI-ESM-LR. Units: K.

99°E 102°E 105°E 99°E 102°E 105°E

26°N 26°N

24°N 24°N

22°N 22°N

20°N20°N

18°N18°N

16°N 16°N

Fig. 11. Position of the IIE averaged from 2006 to 2100 under the RCP2.6 and RCP8.5 scenarios:(a) MPI-ESM-MR; (b) MPI-ESM-LR.

IIE’s mean distributions are insensitive to global warming.

5.2. Mean state of SST and circulationThe mean state of SST averaged from 2006 to 2100 un-

der the RCP2.6 and RCP8.5 scenarios is displayed in Fig. 12.

The SST mean states of MPI-ESM-LR and MPI-ESM-MRshow an overall warming trend, with generally more warm-ing in the RCP8.5 scenario (around 2 K) than the RCP2.6scenario (around 1 K), but the spatial distributions of SSTare similar. Figure 13 shows the mean state of horizontal


Fig. 12. Simulated JJA SST averaged from 2006 to 2100: (a) MPI-ESM-MR under RCP2.6; (b) MPI-ESM-LR under RCP2.6;(c) MPI-ESM-MR under RCP8.5; (d) MPI-ESM-LR under RCP8.5. Shading denotes the difference in SST between theRCP2.6/RCP8.5 scenario and the historical simulations. Units: K.

Fig. 13. Simulated JJA horizontal wind at 850 hPa averaged from 2006 to 2100: (a) MPI-ESM-MR under RCP2.6; (b) MPI-ESM-LR under RCP2.6; (c) MPI-ESM-MR under RCP8.5; (d) MPI-ESM-LR under RCP8.5. Shading denotes the differencein zonal wind between the RCP2.6/RCP8.5 scenario and the historical simulations. Units: m s−1.

wind at 850 hPa under the RCP2.6 and RCP8.5 scenarios.As with the spatial distributions of the SST mean state, thespatial distributions of the mean states of horizontal wind at850 hPa share a similar pattern. In Asia, the easterlies areintensified in the tropical Pacific (within 2 m s−1) and thewesterlies are intensified in the tropical Indian Ocean (within2 m s−1). This implies that the East Asian summer monsoonand Indian summer monsoon consistently enhance under theRCP2.6 and RCP8.5 scenarios. These results associated withthe horizontal wind at 850 hPa suggest that the overall SSTwarming may not significantly change the mean state of hor-izontal wind at 850 hPa.

5.3. SST and circulation anomalies associated with IIEvariability

Figure 14 shows the JJA SSTAs regressed onto the IIEtime series under the RCP2.6 and RCP8.5 scenarios. The re-sults show that a developing La Nina SSTA pattern appearsunder the RCP2.6 and RCP8.5 scenarios, suggesting that therelationship between SSTAs and the IIE remains unchanged

against a warmer climate background. As with the historicalexperiments, the negative SSTAs over the tropical Pacific inMPI-ESM-LR are stronger than those in MPI-ESM-MR.

Figure 15 shows the averaged 850 hPa zonal wind underthe RCP2.6 and RCP8.5 scenarios. Basically, the projectedwind field under both scenarios has a very similar distribu-tion to that of the historical experiments in all four experi-ments (Figs. 5 and 15). These patterns are consistent with theresults that the IIE position in future projections shares a sim-ilar pattern with the historical experiments, implying that thezonal thermal gradient in this domain would not be signifi-cantly changed under global warming. Figure 16 shows thefuture projection of 850 hPa horizontal wind anomalies underthe RCP2.6 and RCP8.5 scenarios. The results indicate thatthe enhanced anticyclone around 20◦N in East Asia will stillbe the major circulation feature associated with the IIE vari-ability in the future. The 850 hPa geopotential height anoma-lies associated with the IIE variability in the observation andhistorical experiments is not reflected in the future projec-tions (figures omitted), possibly because of the influence of


Fig. 14. Future projection of SST onto the IIE time series: (a) MPI-ESM-MR underRCP2.6; (b) MPI-ESM-LR under RCP2.6; (c) MPI-ESM-MR under RCP8.5; (d) MPI-ESM-LR under RCP8.5. Shaded areas from light to dark denote statistical significanceat the 90% and 95% confidence level.

global warming (Yang and Sun, 2003; Lu et al., 2008). Thefuture projected vertical mean temperature anomalies at 200–500 hPa are shown in Fig. 17. A significant positive center islocated around 30◦N in South China and a significant nega-tive center is located around Indonesia in MPI-ESM-MR un-der RCP2.6. The future projection of vertical mean temper-ature anomalies in MPI-ESM-MR is generally similar to theobservational results, but with larger negative anomalies thanthose seen in the historical experiments. In MPI-ESM-LR,the positive anomaly center does not appear, and the nega-tive anomaly center moves farther west than MPI-ESM-MR,which resembles the results of MPI-ESM-LR in the historicalexperiments. Under RCP8.5, the magnitudes of the 200–500hPa mean temperature anomalies are obviously larger thanthose under RCP2.6. However, these anomalous centers donot pass the significance test at the 90% confidence levelin both MPI-ESM-LR and MPI-ESM-MR, suggesting thatthe relationship between the IIE and middle- and high-leveltemperature may change significantly under stronger globalwarming.

6. Summary and discussionIn this study, the capacities of MPI-ESM-LR, MPI-ESM-

MR and ECHAM5/MPI-OM in modeling the variability ofthe IIE were investigated with their integrated data from 1979to 2005. Among the three models, MPI-ESM-LR outper-

forms the other two models for most of the diagnostics. Incomparison to the 20th century reanalysis data (version 2),the IIE’s position is located around 100◦E under normal con-ditions and the position under extreme conditions is repro-duced by the three models. The three models also reflect thekey physical process: when the decaying phase of El Ninoappears, an enhanced WPSH will occur with a farther southposition, and a significant anomalous cyclone controls themiddle–lower reaches of the Yangtze River, which further re-sults in a farther east position of the IIE. When the decay-ing phase of La Nina appears, the opposite conditions willhappen. The three models are sensitive to the different RCPscenarios. The equivalent potential temperature simulated bythe three models is warmer under the RCP8.5 scenario thanunder the RCP2.6 scenario over the research domain, con-sistent with global warming. However, the saddle structureof the equivalent potential temperature vertical distribution isnot sensitive to the RCP scenario used. This important fea-ture directly implies that the IIE’s position simulated underdifferent RCP scenarios by the three models is similar, and al-most remains the same as in their historical experiments. TheSSTA–anomalous circulation patterns associated with the IIEvariability under different future scenarios resemble their his-torical experiments, suggesting that the key physical processremains unchanged against a global warming background.Through studying global monsoon change in CMIP5, Leeand Wang (2012) found that the monsoon domain does not


Fig. 15. 850 hPa zonal wind averaged from 2006 to 2100: (a) MPI-ESM-MR under RCP2.6; (b) MPI-ESM-MR un-der RCP8.5; (c) MPI-ESM-LR under RCP2.6; (d) MPI-ESM-LR under RCP8.5. Shaded areas denote the differencebetween the modeling results and observations. Units: m s−1. The green solid line represents the terrain.

change appreciably, implying that the Indian summer mon-soon tends to change in phase with the East Asian summermonsoon under each RCP scenario. This may be the mainreason why the IIE only presents conspicuous variability onthe interannual, but not on the interdecadal, timescale, in boththe historical evolution and future changes of the IIE (Cao etal., 2012).

Even so, the three models also produce errors in mod-eling the IIE’s spatial distribution and the physical processassociated with the interannual variability of the IIE. Impor-tantly, the deviation between the IIE’s position simulated bythe three models and the observation generally become largerwith latitude, increasing under normal and extreme condi-tions. If we compare the deviation for each condition, themaximum deviation appears under the condition in whichthe IIE is situated more westward than normal. The devia-tion under the normal position of the IIE varies less, and theminimum deviation occurs under the condition in which theIIE is situated more eastward than normal. Because the ter-rain becomes more complex with increasing latitude and lon-gitude over the research domain, the error distribution sug-gests that terrain may be one of the most important factorsimpacting the formation of the IIE. The fact that the differ-

ence between MPI-ESM-MR and MPI-ESM-LR only man-ifests in the vertical resolution, and the physical processesmodulated by terrain are described more realistically in MPI-ESM than in ECHAM5/MPI-OM, may also be used to inter-pret why the MPI-ESM-LR almost performs as efficiently asMPI-ESM-MR in modeling the IIE, but MPS-ESM presentshigher efficiency than ECHAM5/MPI-OM. Meanwhile, thewavy pattern of the IIE is not sufficiently reproduced by thethree models. All of these results imply that an appropriateincrease in the horizontal resolution of a numerical model andan improvement in describing the physical processes associ-ated with terrain may easily achieve a better result in model-ing the spatiotemporal distribution of the IIE.

Through intercomparing the capacities of numerical mod-els in simulating the Asian summer monsoon betweenCMIP3 and CMIP5, Sperber et al. (2013) demonstrated thatthere is a large deviation in the simulation of the WPSH,which, unfortunately, is one of the most important compo-nents of the East Asian summer monsoon system (Tao andChen, 1987). It significantly modulates the IIE’s interannualvariability: when the WPSH’s position is more southwardand westward than normal, the IIE will move more westwardthan normal, and vice versa (Cao et al., 2012). Figure 7 shows


Fig. 16. Future projection of 850 hPa horizontal wind regressed onto the IIE time series:(a) MPI-ESM-MR under RCP2.6; (b) MPI-ESM-LR under RCP2.6; (c) MPI-ESM-MRunder RCP8.5; (d) MPI-ESM-LR under RCP8.5. Shaded areas from light to dark de-note statistical significance at the 90% and 95% confidence level. The black solid linerepresents the terrain.

Fig. 17. Future projection of vertical mean temperature at 200–500 hPa regressed ontothe IIE time series: (a) MPI-ESM-MR under RCP2.6; (b) MPI-ESM-LR under RCP2.6;(c) MPI-ESM-MR under RCP8.5; (d) MPI-ESM-LR under RCP8.5. Shaded areas fromlight to dark denote statistical significance at the 90% and 95% confidence level.


almost identical results to these previous studies, suggestingthat an improvement in the simulation of the WPSH may bean efficient way to further improve the performance of MPI-ESM in modeling the interannual variability of the IIE.

Acknowledgements. We thank the anonymous reviewers fortheir valuable comments that led to improvement of the manuscript.This work was supported by the National Natural Science Founda-tion of China (Grant No. 41375097) and the Jiangsu CollaborativeInnovation Center for Climate Change. The data of MPI-ESM andECHAM5/MPIOM were downloaded from http://www-pcmdi.llnl.gov/ and https://esg.llnl.gov:8443/. The twentieth century reanalysisdata (version 2) were downloaded from http://www.esrl.noaa.gov/psd/.

REFERENCES

Annamalai, H., K. Hamilton, and K. R. Sperber, 2007: The SouthAsian summer monsoon and its relationship with ENSO inthe IPCC AR4 simulations. J. Climate, 20, 1071–1092.

Bellenger, H., E. Guilyardi, J. Leloup, M. Lengaigne, and J.Vialard, 2014: ENSO representation in climate models: FromCMIP3 to CMIP5. Climate Dyn., 42(7–8), 1999–2018, doi:10.1007/s00382-013-1783-z.

Brovkin, V., T. Raddatz, C. H. Reick, M. Claussen, and V. Gayler,2009: Global biogeophysical interactions between forest andclimate. Geophys. Res. Lett., 36, L07405, doi: 10.1029/2009GL037543.

Cao, J., J. M. Hu, and Y. Tao, 2012: An index for the inter-face between the Indian summer monsoon and the East Asiansummer monsoon. J. Geophys. Res., 117(D18), D18108, doi:10.1029/2012JD017841.

Chen, W., 2002: Impacts of El Nino and La Nina on the cycle ofthe East Asian winter and summer monsoon. Chinese J. At-mos. Sci., 26, 595–610. (in Chinese)

Compo, G. P., J. S. Whitaker, and P. D. Sardeshmukh, 2006: Fea-sibility of a 100 year reanalysis using only surface pressuredata. Bull. Amer. Meteor. Soc., 87, 175–190, doi: 10.1175/BAMS-87-2-175.

Compo, G. P., and Coauthors, 2011: The twentieth century reanal-ysis project. Quart. J. Roy. Meteor. Soc., 137, 1–28.

Ding, Y. H., 1994: The summer monsoon in East Asia. Monsoonsover China, Kluwer Acad., 90 pp.

Flohn, H., 1957: Large-scale aspects of the “summer monsoon” inSouth and East Asia. J. Meteor. Soc. Japan, 75, 180–186.

Giorgetta, M., and Coauthors, 2013: The atmospheric general cir-culation model ECHAM6: Model description. [Available on-line at http://www.mpimet.mpg.de/fileadmin/publikationen/Reports/WEB BzE 135.pdf]

Guilyardi, E., H. Bellenger, M. Collins, S. Ferrett, W. J. Cai, and A.T. Wittenberg, 2012: A first look at ENSO in CMIP5. ClivarExchanges, 17(1), 29–32.

Hagemann, S., A. Loew, and A. Andersson, 2013: Combined eval-uation of MPI-ESM land surface water and energy fluxes. J.Adv. Model. Earth Syst., 5, 259–286.

Huang, P., P. F. Wang, K. M. Hu, G. Huang, Z. H. Zhang, Y. Liu,and B. L. Yan, 2014: An introduction to the integrated cli-mate model of the center for monsoon system research andits simulated influence of El Nino on East Asian-westernNorth Pacific climate. Adv. Atmos. Sci., 31(5), 1136–1146,

doi: 10.1007/s00376-014-3233-1.Huang, R. H., Y. H. Xu, P. F. Wang, and L. T. Zhou, 1998: The fea-

tures of the catastrophic flood over the Changjiang river basinduring the summer of 1998 and cause exploration. Climaticand Environmental Research, 3(4), 300–313. (in Chinese)

Huang, R. H., J. L. Chen, L. Wang, and Z. D. Lin, 2012: Char-acteristics, processes, and causes of the spatio-temporal vari-abilities of the East Asian monsoon system. Adv. Atmos. Sci.,29, 910–942.

Ilyina, T., K. D. Six, J. Segschneider, E. Maier-Reimer, H. M.Li, and I. Nunez-Riboni, 2013: Global ocean biogeochem-istry model HAMOCC: Model architecture and performanceas component of the MPI-Earth system model in differentCMIP5 experimental realizations. Journal of Advances inModeling Earth Systems, 5, 287–315.

Jin, Z. H., and L. X. Chen, 1982: On the medium-range oscillationof the East Asian monsoon circulation system and its relationwith the Indian monsoon system. The National SymposiumCollections on the Tropical Summer Monsoon, People’s PressYunnan Province, Kunming, China, 204–215. (in Chinese)

Jungclaus, J. H., and Coauthors, 2013: Characteristics of theocean simulations in the Max Planck Institute Ocean Model(MPIOM) the ocean component of the MPI-Earth systemmodel. J. Adv. Model. Earth Syst., 5, 422–446.

Krishnamurti, T. N., and H. N. Bhalme, 1976: Oscillations of amonsoon system. Part I. Observational aspects. J. Atmos. Sci.,33, 1937–1954.

Lau, K. M., and M. T. Li, 1984: The monsoon of East Asia and itsglobal associations—A survey. Bull. Amer. Meteor. Soc., 65,114–125.

Lee, J.-Y., and B. Wang, 2012: Future change of global mon-soon in the CMIP5. Clim. Dyn., 42(1–2), 101–119, doi:10.1007/s00382-012-1564-0.

Li, J. P., and L. Zhang, 2009: Wind onset and withdrawal of Asiansummer monsoon and their simulated performance in AMIPmodels. Climate Dyn., 32, 935–968.

Liu, Y. M., G. X. Wu, H. Liu, and P. Liu, 1999: The effect ofspatially nonuniform heating on the formation and variationof subtropical high. Part III: Condensation heating and SouthAsia high and western Pacific subtropical high. Acta Meteo-rologica Sinica, 57(5), 525–538. (in Chinese)

Lu, J., G. Chen, and D. M. Frierson, 2008: Response of the zonalmean atmospheric circulation to El Nino versus global warm-ing. J. Climate, 21(22), 5835–5851.

Lu, R. Y., and Y. H. Fu, 2010: Intensification of East Asian sum-mer rainfall interannual variability in the twenty-first centurysimulated by 12 CMIP3 coupled models. J. Climate, 23(12),3316–3331.

Marsland, S. J., H. Haak, J. H. Jungclaus, M. Latif, and F. Roske,2003: The Max-Planck-Institute global ocean/sea ice modelwith orthogonal curvilinear coordinates. Ocean Model., 5,91–127.

Reick, C. H., T. Raddatz, V. Brovkin, and V. Gayler, 2013: Repre-sentation of natural and anthropogenic land cover change inMPI-ESM. J. Adv. Model. Earth Syst., 5, 459–482.

Roeckner, E., and Coauthors, 2003: The atmospheric general cir-culation model ECHAM5 Part I: Model description. MaxPlanck Institute fur Meteorology Rep, No. 349, 127 pp.

Roeckner, E., and Coauthors, 2006: Sensitivity of simulated cli-mate to horizontal and vertical resolution in the ECHAM5atmosphere model. J. Climate, 19, 3771–3791.

Roeckner, E., T. Mauritsen, M. Esch, and R. Brokopf, 2012: Im-


pact of melt ponds on Arctic sea ice in past and future cli-mates as simulated by MPI-ESM. J. Adv. Model. Earth Syst.,4, M00A02, doi: 10.1029/2012MS000157.

Smith, T. M., R. W. Reynolds, T. C. Peterson, and J. Lawrimore,2008: Improvements to NOAA’s historical merged land–ocean temp analysis (1880–2006). J. Climate, 21, 2283–2296.

Sperber, K. R., H. Annamalai, I.-S. Kang, A. Kitoh, A. Moise,A. Turner, B. Wang, and T. Zhou, 2013: The Asian summermonsoon: An intercomparison of CMIP5 vs. CMIP3 simula-tions of the late 20th century. Climate Dyn., 41, 2711–2744.

Stevens, B., and Coauthors, 2013: Atmospheric component of theMPI-M earth system model: ECHAM6. J. Adv. Model. EarthSyst., 5, 146–172.

Sui, C.-H., P.-H. Chung, and T. Li, 2007: Interannual andinterdecadal variability of the summertime western NorthPacific subtropical high. Geophys. Res. Lett., 34(11), doi:10.1029/2006GL029204.

Tao, S. Y., and L. X. Chen, 1987: A review of recent research onthe East Asian summer monsoon in China. Monsoon Mete-orology, C. P. Chang and T. N. Krishnamurti, Eds., OxfordUniversity Press, Oxford, U. K., 60–92.

Taylor, K. E., 2001: Summarizing multiple aspects of model per-formance in a single diagram. J. Geophys. Res., 106(D7),7183–7192, doi: 10.1029/2000JD900719.

Wang, B., and LinHo, 2002: Rainy season of the Asian–Pacificsummer monsoon. J. Climate, 15, 386–398.

Webster, P. J., V. O. Magana, T. N. Palmer, J. Shukla, R. A. Tomas,M. Yanai, and T. Yasunari, 1998: Monsoons: Processes, pre-dictability, and the prospects for prediction. J. Geophy. Res.,103, 14451–14510.

Whitaker, J. S., G. P. Compo, X. Wei, and T. M. Hamill, 2004:Reanalysis without radiosondes using ensemble data assimi-lation. Mon. Wea. Rev., 132, 1190–1200.

Wu, G. X., Y. M. Liu, and P. Liu, 1999: The effect of spatially non-uniform heating on the formation and variation of subtropicalhigh. Part I: Scale analysis. Acta Meteorologica Sinica, 57(3),257–263. (in Chinese)

Wu, G. X., Y. M. Liu, B. He, Q. Bao, A. M. Duan, and F. F. Jin,2012: Thermal controls on the Asian summer monsoon. Sci-entific Reports, 2, 404, doi: 10.1038/srep00404.

Xie, S. P., K. M. Hu, J. Hafner, H. Tokinaga, Y. Du, G. Huang,and T. Sampe, 2009: Indian ocean capacitor effect on indo–Western Pacific climate during the summer following ElNino. J. Climate, 22, 730–747.

Yang, H., and S. Q. Sun, 2003: Longitudinal displacement ofthe subtropical high in the western Pacific in summer andits influence. Adv. Atmos. Sci., 20(6), 921–933, doi: 10.1007/BF02915515.

Zhao, D. M., and C. B. Fu, 2010: Comparisons of low-level cir-culation characteristics between ECHAM5/MPI-OM resultsand NCEP/NCAR re-analysis data in East Asia. Atmos. Oce.Sci. Lett., 3, 189–194.

Zhou, T. J., B. Wu, and B. Wang, 2009: How well do atmosphericgeneral circulation models capture the leading modes of theinterannual variability of the Asian–Australian monsoon? J.Climate, 22, 1159–1173.

Zhu, Q. G., J. H. He, and P. X. Wang, 1986: A study of circula-tion differences between east-Asian and Indian summer mon-soons with their interaction. Adv. Atmos. Sci., 3, 466–477, doi:10.1007/BF02657936.

simulation of the interface between the indian summer ... · mpi-esm is more skillful than...

Documents