1. Introduction
The South Asian summer monsoon (SASM) and the East Asian summer monsoon (EASM), which jointly influence the production and livelihoods of over half of the global population (Wang et al. 2021), are to a certain extent independent from each other, yet they also interact with one another (Ding and Chan 2005). When the SASM is stronger, rainfall increases over northern China but decreases in the Yangtze River Valley (YRV), southern Korea, and southern Japan (Guo and Wang 1988; Zhang 1999; Kripalani and Kulkarni 2001; Krishnan and Sugi 2001; Day et al. 2015). Understanding the physical connections between these two monsoon systems is essential for improving the simulation of climate models and climate prediction for these densely populated regions (Wu 2017; Ha et al. 2018; H.-J. Chen et al. 2023; W. Chen et al. 2023).
With the prolonged efforts by previous studies, the dynamical connections for the SASM–EASM relationship can generally be concluded as the midlatitude atmospheric wave pattern pathway (Krishnan and Sugi 2001; Kim et al. 2002; Lu et al. 2002; Wu 2002; Ding and Wang 2005; Greatbatch et al. 2013; Wu 2017; Zhou et al. 2020; Kosaka 2021) and the coherent water vapor transport pathway in the lower latitudes (Zhang 1999, 2001; Liu and Ding 2008; Wu 2017; X.-F. Li et al. 2019). In the midlatitudes, an upper-level wave pattern, triggered by increased latent heat release associated with enhanced SASM rainfall, propagates along the westerly jet stream with anticyclonic anomalies over central Asia and Northeast Asia and an anomalous cyclone over the northern Indo-China Peninsula (Kim et al. 2002; Lu et al. 2002; Wu 2002; Greatbatch et al. 2013). This wave pattern, also referred to as the Silk Road pattern (Kosaka et al. 2009; Kosaka 2021), can influence EASM rainfall significantly (L. Wang et al. 2017), representing the interaction between the SASM and the EASM (Wei et al. 2014, 2015). In the low latitudes, less moisture is transported along the west edge of the weaker western North Pacific subtropical high toward the YRV when stronger water vapor flow heads to the SASM region and vice versa (Zhang 1999, 2001; Liu and Ding 2008; Cao et al. 2012; X.-F. Li et al. 2019). This phenomenon results in a negative relationship between the SASM rainfall and the YRV rainfall.
The SASM–EASM relationship experienced an interdecadal change at the end of the 1970s (Guo 1992; Kripalani and Kulkarni 2001; Wang and Huang 2006; Wu 2017), and the relationship between monsoon rainfalls displayed a declining trend after the 1970s (Guo 1992; Wang and Huang 2006; Wu 2017). This phenomenon may be attributed to the weakened midlatitude wave teleconnection due to reduced interannual variability of the SASM rainfall (Wang et al. 2012; Wu 2017), the northward shift of westerly jet steam over East Asia (Lin et al. 2017), the mid- and high-latitude disturbances (Sun and Ming 2019), the change in El Niño–Southern Oscillation (ENSO)–related EASM circulation anomalies during the decaying phase of ENSO (Wu and Wang 2002; Ha et al. 2018), the stochastic processes (Wu et al. 2018), and so on. A further question is whether the SASM–EASM relationship will change in the future under global warming.
In the warming climate, the SASM–EASM relationship is undergoing new changes, particularly in the context of increasing extreme climate events (You and Ting 2021a,b). There is an urgent need to understand the new changes in the SASM–EASM relationship to better comprehend the connection between the extreme events in the two monsoon regions and to predict their future changes. In the Coupled Model Intercomparison Project (CMIP) models, weakening SASM circulation (Ueda et al. 2006; Li et al. 2022) and strengthening EASM circulation (Jiang and Tian 2013; Z. Li et al. 2019, Li et al. 2022; Tian et al. 2022) are projected, whereas both SASM and EASM rainfalls are expected to increase (Wang et al. 2014; Chen et al. 2020; Ha et al. 2020; Li et al. 2022; Wang et al. 2021). These features also imply potential future changes in the relationship between the SASM and the EASM (Ha et al. 2018). However, CMIP5 models cannot reach a consensus on the future changes in the SASM–EASM rainfall relationship, suggesting multidecadal variations with alternate epochs in different models (Preethi et al. 2017; Wu and Jiao 2017). A further study indicates that the SASM-related rainband will extend southward from northern China to the YRV region in the future based on CMIP6 models (H.-J. Chen et al. 2023). This change in the rainfall pattern is probably caused by the synergistic influence of the distinct future changes in the midlatitude and low-latitude pathways (H.-J. Chen et al. 2023). The result suggests the necessity of analyzing the underlying physical mechanisms for the different changes in these two pathways to gain a deeper insight into the future SASM–EASM relationship. Additionally, CMIP6 models show varying performances in both the midlatitudes and the low latitudes (Song et al. 2024), also pointing out the need for distinct discussions of these pathways.
Therefore, this study will separately identify the midlatitude pathway and the low-latitude pathway connecting the SASM and the EASM, as well as the associated rainfall patterns, and investigate their future changes using the results from CMIP6 models. Additionally, a water vapor transport diagnosis will be applied to clarify the roles of circulation changes and atmospheric moisture changes due to the modulation of global warming on the SASM–EASM relationship.
The remainder of this paper is organized as follows. Section 2 describes the data and analysis methods applied in this study. Section 3 presents the midlatitude and low-latitude pathways between the SASM and the EASM and evaluates the performance of CMIP6 models in reproducing these pathways. Section 4 outlines the future projections of the pathways, and section 5 summarizes and discusses the main findings.
2. Data and methodology
a. Data
In this study, 26 CMIP6 models (see Table 1) are used, including many popular coupled general circulation models such as the CESM2. In addition to the historical run for model simulation evaluation, the moderate-emission Shared Socioeconomic Pathway 2-4.5 run (SSP2–4.5) and high-emission Shared Socioeconomic Pathway 5-8.5 run (SSP5–8.5) are considered to investigate the responses of the SASM–EASM relationship under different warming conditions (O’Neill et al. 2016). The multimodel ensemble (MME) mean is obtained by averaging the outputs from 26 CMIP6 models, using a single member from each model. This method helps to reduce the uncertainties associated with climate model simulations and projections. The MME mean results are considered statistically significant if they align with the signs over 70% of the models (He et al. 2019).
Details of the CMIP6 models analyzed in this study. The six models (denoted in bold) and seven models (marked with asterisks) that better reproduce the midlatitude and low-latitude pathways, respectively, are selected to conduct the future projection analysis. Models that are selected for both pathways are denoted in bold and with asterisks.
The observation data and reanalysis used for the period of 1979–2014 are as follows: 1) the fifth major global reanalysis produced by European Centre for Medium-Range Weather Forecasts (ERA5) with a 1° × 1° resolution (Hersbach et al. 2020) for geopotential height and winds and 2) the Global Precipitation Climatology Project (GPCP) data, version 2.2, with a 2.5° × 2.5° resolution for precipitation (Adler et al. 2003).
A 36-yr period is applied for the analysis of both the present climate from 1979 to 2014 and the future climate from 2064 to 2099. We calculate the summer mean values using the monthly data from June to August (JJA). To remove the impact of the long-term trend on the SASM–EASM relationship, the linear trends in the period of 1979–2014 are removed from observations and the model results from historical runs. The same is done for both SSP2–4.5 and SSP5–8.5 results during the period of 2015–99. To validate the performances of CMIP6 models in a historical run with respect to observations, both observation data and model outputs are bilinearly interpolated to the resolution of 2.5° × 2.5°. The differences between the SSP scenarios and the historical run represent the future changes projected by the models.
b. Partial regression analysis
According to previous studies, the midlatitude wave teleconnection is strongly connected to the diabatic heating over the northern Indian Peninsula (Greatbatch et al. 2013; Wei et al. 2014; Kosaka 2021). And the low-latitude coherent water vapor transport pathway is observed between the southern Indian Peninsula and the YRV region (Zhang 1999, 2001; Kosaka 2021). In this study, we apply a partial regression analysis with northern SASM rainfall and southern SASM rainfall to separately identify the midlatitude and low-latitude connections between the SASM and the EASM.
c. Wave strength index
d. Idealized sensitivity experiment design
Two idealized experiments are designed to emphasize the impact of projected basic-state changes on the midlatitude atmospheric wave pattern pathway. The idealized experiments are based on the linear baroclinic model (LBM), which is developed by Watanabe and Kimoto (2000).
The LBM model has a horizontal resolution of T42 with 20 vertical levels. Following Greatbatch et al. (2013), this study adopts a vertical profile of idealized heating anomaly (Fig. 1), with the largest anomaly (2.6 K day−1) placed over the northern Indian Peninsula at the level σ = 0.5 and a horizontal center at 25°N, 80°E. Its horizontal distribution is shown in Fig. 1b. The two idealized experiments are conducted with the same diabatic heating forcing. One uses the JJA climatology for the period of 1979–2014 in the historical run as the basic states (referred to as the CTRL experiment), and the other uses the JJA climatology for the period of 2064–99 in SSP5–8.5 as the basic state (labeled the BS585 experiment). The JJA climatology in the two experiments is obtained from the ensemble mean (marked as mbMME mean) of six models (denoted in bold in Table 1) that better reproduce the midlatitude pathway.
(a) Vertical profile of the ideal heating anomaly (K day−1) prescribed in the northern SASM region. (b) The mbMME mean of 200-hPa JJA climatological u wind (m s−1) for the period of 1979–2014 in historical. The black contours in (b) indicate the horizontal distribution of the diabatic heating (K day−1). (c) As in (b), but for the period of 2064–99 in SSP5–8.5. The green contours denote the mbMME mean of the 200-hPa climatological u wind in SSP5–8.5 minus that in historical, with intervals of 1.5 (−1.5, 0, and 1.5 m s−1). The solid, thick solid, and dashed green lines denote the positive, zero, and negative values, respectively.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
The experiments use biharmonic horizontal diffusion, with an e-folding time of 2 h for the smallest wave. The Rayleigh friction and Newtonian damping are set as in Kosaka et al. (2009), with 30-day damping time in the free troposphere, and 1-day damping for the three lowest and two uppermost levels. For the fourth and fifth lowest levels, damping times are set to 5 and 15 days, respectively. To obtain the steady-state atmospheric response to heating, the model is integrated over 60 days, with the averages from the last 20 days used for analysis.
e. Moisture transport analysis
Under global warming, circulation changes occur alongside the increase in atmospheric moisture due to the Clausius–Clapeyron relationship (Moon and Ha 2020). To reveal their respective roles in future variations of the SASM–EASM relationship, a moisture transport diagnosis is employed in this study.
In this study, the precipitation, evaporation, and dynamic and thermodynamic moisture transport terms are first calculated based on nondetrended CMIP6 data for 1979–2014 in the historical run and 2064–99 in the SSP2–4.5 and SSP5–8.5 runs. These terms are then applied with partial regression mentioned above.
According to the moisture budget equation, precipitation anomalies may be attributed to evaporation anomalies and dynamic and thermodynamic moisture transport anomalies. Thus, the regressed rainfall anomalies associated with northern or southern SASM rainfall may be contributed by these terms related to the SASM rainfall.
3. SASM–EASM connections in observations and CMIP6 models
a. The SASM–EASM pathways in observations
The two pathways connecting the SASM and the EASM include the midlatitude wave train pattern along the westerly jet stream (Lu et al. 2002; Wu 2002) and the low-latitude coherent moisture transport relationship (Zhang 1999, 2001; Cao et al. 2012). The former is closely associated with the northern SASM rainfall (Greatbatch et al. 2013; Wei et al. 2014; Kosaka 2021) and the latter mainly connects the southern SASM rainfall and the YRV rainfall (Zhang 1999, 2001; Kosaka 2021). A partial regression analysis with northern and southern SASM rainfalls is conducted to identify the two pathways separately in both observations and models. The results from the partial regression with northern SASM rainfall clearly show the midlatitude pathway at 200 hPa (Fig. 2c), which is characterized by an anomalous anticyclone over southern central Asia (hereafter SCAAC), an anomalous cyclone to the north of the Indo-China Peninsula, and an anomalous anticyclone over Northeast Asia (hereafter NEAAC). Besides, the low-latitude pathway, which involves one flow heading westward to the southern SASM region and another flow along the west edge of western North Pacific subtropical high to the YRV at 850 hPa, can be clearly observed in the partial regression with southern SASM rainfall (Fig. 2d). However, the midlatitude pathway cannot be observed in the partial regression with southern SASM rainfall (figure not shown). Similarly, the low-latitude pathway cannot be observed in the partial regression with northern SASM rainfall (Fig. 2a). These features indicate that the partial regression analysis with northern and southern SASM rainfalls can clearly separate the midlatitude and low-latitude connections between the SASM and the EASM.
Partial regression of the GPCP summer rainfall (mm day−1) on (a) northern SASM rainfall and (b) southern SASM rainfall. Vectors in (a) depict the partially regressed ERA5 850-hPa horizontal winds on northern SASM rainfall. Partial regressions of ERA5 geopotential height (shading; gpm) and horizontal winds (vectors; m s−1) (c) at 200 hPa on northern SASM rainfall and (d) at 850 hPa on southern SASM rainfall. Green contours in (c) show the climatological westerly jet at 200 hPa during 1979–2014 in 20, 25, and 30 m s−1. The dots denote the significant values at the 90% confidence level in (a) and (b) and 95% in (c) and (d). Only the wind anomalies significantly surpassing the 95% confidence level are displayed. Dashed boxes outline the range of northern China (32.5°–40°N, 105°–120°E) in (a), the range of YRV (27.5°–32.5°N, 105°–120°E; upper box) and MC (from −5°S to 5°N, 105°–120°E) in (b), and the range for calculating PCCs in midlatitude (15°–47.5°N, 50°–137.5°E) and low-latitude pathways (15°–35°N, 110°–137.5°E) in (c) and (d), respectively.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
Based on the partial regression results with northern SASM rainfall, rainfall anomalies over East Asia exhibit a tripole distribution (Fig. 2a). This pattern shows excessive rainfall over northern China, deficient rainfall over the mei-yu–baiu region, and excessive rainfall over southern China and the South China Sea associated with the positive northern SASM rainfall anomalies (Fig. 2a). The rainfall anomalies over northern China reach 0.13 mm day−1. In response to the diabatic heating in the northern SASM region, a wave train propagates along the westerly jet stream at 200 hPa (Fig. 2c). Due to the influence of NEAAC, a barotropic anomalous anticyclone forms over the Yellow Sea at 850 hPa, resulting in moisture transport to northern China and contributing to positive rainfall anomalies there (Fig. 2a). This wave pattern acts as the midlatitude connection bridging the SASM and the EASM.
When increased rainfall occurs over the southern SASM region, rainfall anomalies over East Asia are characterized by a dipole distribution with positive anomalies over the YRV and negative anomalies over southern China (Fig. 2b). The rainfall anomalies over the YRV reach 0.40 mm day−1. Associated with these rainfall relationships, anticyclonic anomalies over the western North Pacific (hereafter WNPAC) and extending easterlies from the southern Indo-China Peninsula to the Arabian Sea are observed (Fig. 2d). The coherent water vapor transport pathway involves one flow heading westward to the SASM region and another flow along the west edge of WNPAC to the EASM region. It contributes to the positive rainfall relationship between the YRV and the SASM and acts as the low-latitude connection linking the SASM and the EASM.
This positive rainfall relationship does not contradict the negative rainfall relationship between the SASM and mei-yu–Baiu regions reported by previous studies (Guo and Wang 1988; Zhang 1999, 2001; Krishnan and Sugi 2001; Day et al. 2015). Due to the influence of NEAAC on the midlatitude wave pattern, a barotropic anomalous anticyclone forms over the Yellow Sea at 850 hPa, extending the northern edge of the WNPAC (Fig. 2a). This suppresses the rainfall over the YRV (Fig. 2a) and thus establishes a negative relationship between the rainfalls over the SASM and YRV regions. Since this study aims to analyze the underlying physical mechanisms for the differing changes in the midlatitude and low-latitude pathways, a partial regression analysis is conducted to isolate the influence of NEAAC from the low-latitude pathway. It reveals an increased water vapor transport along the west edge of the WNPAC, directed toward the YRV when southern SASM rainfall is stronger. This feature indicates a positive rainfall relationship between the SASM and YRV regions.
b. Evaluation and selection of CMIP6 models
To reach convincing conclusions in future projections on the SASM–EASM relationship, it is necessary to evaluate the performance of CMIP6 models in reproducing this relationship. Given that the midlatitude and low-latitude pathways are key physical processes linking the SASM and the EASM, this study prioritizes evaluating the performance of models in capturing these circulation features. As seen from the MME mean, models well reproduce the midlatitude wave pattern along the westerly jet stream (Fig. 3a). For a quantitative assessment, the pattern correlation coefficient (PCC) of the MME mean is calculated between the observed and model-simulated geopotential height anomalies related to the northern SASM rainfall over 15°–47.5°N, 50°–137.5°E (the dashed line box in Fig. 2c) at 200 hPa (Fig. 4a). The PCC of the MME mean reaches 0.94. However, the normalized standard deviation of the MME mean, calculated by dividing the spatial standard deviation of the MME mean by that in observations, is only 0.77, indicating an underestimation of the wave pattern. The PCCs of individual models range from 0.42 to 0.96. The underestimation in the MME mean and the wide model spread highlight the importance of selecting those reliable models for a more reliable representation and projection of the SASM–EASM relationship. Thus, six models with their PCCs higher than 0.85, including CESM2-WACCM, CNRM-ESM2-1, EC-Earth3, HadGEM3-GC31-LL, NESM3, and UKESM1-0-LL, are selected. Their MME mean is recorded as mbMME mean hereafter. Compared with MME mean, the mbMME mean better reproduces the midlatitude wave pattern in both spatial pattern and strength (Fig. 3c), with a PCC of 0.96 and normalized standard deviations of 0.92 (Fig. 4a).
The CMIP6 MME mean of the partially regressed (a) 200- and (b) 850-hPa geopotential height anomalies (shading; gpm) and horizontal wind anomalies (vectors; m s−1) onto (a) northern SASM rainfall and (b) southern SASM rainfall. (c) As in (a), but for the mbMME mean. (d) As in (b), but for the lbMME mean. The green contours illustrate the (a) MME mean and (c) mbMME mean of the 200-hPa climatological westerly jet stream (20, 25, and 30 m s−1). Dots and black vectors delineate the areas where over 70% of the selected models exhibit consistent signs with the MME mean. The dashed line boxes in (c) denote the range for calculating the vorticities of SCAAC (25°–50°N, 50°–80°E), anomalous cyclone (15°–37.5°N, 85°–115°E), and NEAAC (27.5°–50°N, 105°–137.5°E). The dashed line box in (d) denotes the range for calculating the vorticities of WNPAC (15°–35°N, 110°–137.5°E).
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
Taylor diagram of the comparison between ERA5 reanalysis and 26 CMIP6 models and their MME means (a) in the midlatitude wave pathway (as indicated by the dashed line box in Figs. 1c,b) and in the low-latitude pathway (as indicated by the dashed line box in Fig. 1d) for the period of 1979–2014. The REF points represent the ERA5 reanalysis data. The angular axis represents the PCC, while the radial axis represents the normalized standard deviation, which is calculated by dividing the standard deviation of models by that in observations. The red (blue) color in the legend denotes the models that better reproduce the midlatitude (low-latitude) pathway. The models selected for mbMME mean and lbMME mean simultaneously are shown in purple.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
It is found that the reproduction skills of CMIP6 models for the wave pattern are well correlated with their simulations in westerly jet stream. The correlation coefficient between the WSI (see section 2) and the climatological maximum 200-hPa zonal wind speed over 30°–50°N, 40°–140°E, which denotes the strength of climatological westerly jet stream, significantly reaches −0.53 (Fig. 5a). In the mbMME mean, both WSI and climatological maximum 200-hPa zonal wind speed are close to observations. These features imply that a better simulation on the strength of westerly jet stream may be helpful for improving the reproduction of the midlatitude wave pattern, because the westerly jet stream supplies potential energy to the wave pattern (Kosaka et al. 2009).
(a) Scatterplot of the climatological maximum 200-hPa zonal wind (CMU; x axis; m s−1) and WSI (y axis; 10−6 s−1) in observations and CMIP6 models for the period of 1979–2014. (b) Scatterplot of the partially regressed MC summer rainfall related to the southern SASM rainfall (x axis; mm day−1) and the vorticities of WNPAC (y axis; 10−6 s−1) in observations and CMIP6 models for the period of 1979–2014. The linear regression of the 26 CMIP6 models’ data on the x axis and y axis is shown by the red dashed line. The correlation coefficients between the data on the x axis and the y axis are shown in the upper-right corner, with those significantly surpassing the 95% confidence level marked with asterisks.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
However, CMIP6 models exhibit deficiencies in simulating the low-latitude moisture transport pathway with a large model spread. The MME mean notably underestimates the WNPAC and erroneously simulates a cross-equatorial flow over the Indian Ocean (Fig. 3b). The PCC for the low-latitude pathway is calculated between the observed and model-simulated geopotential height anomalies related to the southern SASM rainfall over 15°–35°N, 110°–137.5°E (the dashed line box in Fig. 2d) at 850 hPa. The PCC of the MME mean reaches 0.86 (Fig. 4b), but the PCCs of individual models range from −0.93 to 0.97 (Fig. 4b), indicating a large intermodel spread. In addition, 21 (80.7%) models’ normalized standard deviations are less than 1, implying that the models underestimate the WNPAC. These features also illustrate the necessity of choosing reliable models for the following analysis. Seven models whose PCCs are higher than 0.85 are selected, including BCC-CSM2-MR, CMCC-ESM2, CNRM-CM6-1, EC-Earth3, EC-Earth3-Veg, MIROC-ES2L, and UKESM1-0-LL. The MME mean of these models is recorded as the lbMME mean. It exhibits better simulations of the WNPAC and associated water vapor transport in the low-latitude pathway (Fig. 3d), with a PCC of 0.98 (Fig. 4b). It should be noted that the PCC of the mbMME mean for the low-latitude pathway is only 0.53 (Fig. 4b), while the PCC of the lbMME mean for simulating the midlatitude wave pattern reaches 0.88 (Fig. 4a). This result indicates that CMIP6 models have different simulation skills for midlatitudes and low latitudes, highlighting the necessity for separately discussing the midlatitude and low-latitude pathways.
The simulation of CMIP6 models for the WNPAC is related to the simulation of the relationship between rainfalls over the Maritime Continent (MC; from −5°S to 5°N, 105°–120°E) and the southern SASM region (Fig. 5b). The correlation coefficient between the rainfalls over MC and the vorticity of WNPAC (15°–35°N, 110°–137.5°E) reaches −0.55, indicating that the models with stronger WNPAC also have increased rainfall over the MC, except CESM2-WACCM. These features demonstrate that better simulation on the strength of the relationship between the rainfalls over the MC and the southern SASM region may be helpful for improving the reproduction of the WNPAC and water vapor transport along its western edge because the positive rainfall anomalies over the MC can drive an anticyclonic circulation over the western North Pacific through a Gill-type response (Gill 1980). Hu et al. (2024) also highlighted the role of the MC rainfall in the linkage between the WNPAC and southern SASM rainfall.
Considering that the circulation pattern should be physically consistent with precipitation distribution, it is also important to examine whether CMIP6 models can properly simulate the rainfall relationship between the SASM and the EASM. The area-mean partially regressed rainfall anomalies onto the northern and southern SASM rainfall over northern China and the YRV of CMIP6 models minus those in observations are shown in Fig. 6. The mbMME mean of anomalous rainfall over northern China (32.5°–40°N, 105°–120°E) partially regressed on northern SASM rainfall is close to that in observations (Fig. 6a). The lbMME mean of anomalous rainfall over the YRV (27.5°–32.5°N, 105°–120°E), partially regressed on southern SASM rainfall, is 0.29 mm day−1, lower than the observed value due to the underestimation of WNPAC. However, it shows better performance compared to the MME mean (Fig. 6b). Notably, there are large spreads among the selected models, which is mainly reflected in the values of the regressed precipitation anomalies (Figs. 6a,b) and the spatial distribution of precipitation over northern China (Fig. S1 in the online supplemental material) and the YRV (Fig. S2). This is probably due to the uncertainties from different physical parameterizations (Tett et al. 2022). The mbMME-selected models show positive area-mean rainfall regressed anomalies over northern China (Fig. S1), and similarly, the lbMME-selected models show positive anomalies over the YRV (Fig. S2). These results suggest that the selected models are capable of reproducing the precipitation relationship between the SASM and EASM. In summary, the mbMME mean and the lbMME mean, calculated by models with better skills in reproducing the midlatitude and low-latitude circulation pathways, also show better performance in simulating the rainfall relationship between the SASM and the EASM, suggesting more convincing projections for future climate.
(a) Area-mean partially regressed JJA rainfall anomalies (mm day−1) onto the northern SASM rainfall over northern China of CMIP6 models during 1979–2014 minus those in observations. (b) As in (a), but for area-mean partially regressed JJA rainfall anomalies (mm day−1) onto the southern SASM rainfall over the YRV. The red (blue) color of the x-axis label denotes the models that better reproduce the midlatitude (low-latitude) pathways. The models selected for mbMME mean and lbMME mean simultaneously are denoted in purple.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
4. Future changes in the midlatitude and low-latitude pathways
Given that CMIP6 models can reasonably reproduce the midlatitude and low-latitude pathways between the SASM and the EASM, their future changes, related factors, and possible impacts are examined in this section.
a. The midlatitude wave pattern pathway
In both SSP2–4.5 and SSP5–8.5 scenarios, the increase in the WSIs of the mbMME mean indicates a weakening midlatitude wave pattern (Fig. 7a). The 200-hPa geopotential height anomalies and horizontal wind anomalies related to the northern SASM rainfall in SSP2–4.5 (Fig. 8c) or SSP5–8.5 (Fig. 8d) minus those in historical run suggest the weakening of SCAAC and NEAAC. The mbMME mean of 200-hPa wave activity flux (Takaya and Nakamura 2001), partially regressed onto the northern SASM rainfall for both the historical and SSP5–8.5 periods, also suggests a weakening of the midlatitude wave pattern in future projections (Fig. S3). To quantitatively assess the future changes in NEAAC, which can influence northern China rainfall through the low-level barotropic anticyclone, the vorticity field is examined. The mbMME mean of the vorticity of NEAAC is expected to decrease in SSP2–4.5 and SSP5–8.5 (Fig. 7b). Five (83.3%) out of the six reliable models show a decreased vorticity of NEAAC (Figs. 7c,d). These features collectively demonstrate a weakening NEAAC in the future. It is noted that the WSIs and the vorticity of NEAAC in SSP2–4.5 are weaker than those in SSP5–8.5, likely due to some models such as CNRM-CM6-1 and INM-CM5-0 (Fig. 7). These models project that the wave teleconnection would turn into different circulation patterns in the future, indicating that their future-time results of WSIs and vorticity of NEAAC are unreliable (figure not shown). As emphasized in previous studies, the interdecadal variation in the midlatitude wave pattern is closely related to the interannual variability of SASM rainfall (e.g., Wang et al. 2012). In the historical period, the mbMME mean of the standard deviation of northern SASM rainfall is 0.40 mm day−1 and it becomes 0.43 mm day−1 under SSP5–8.5. This difference in the standard deviation between the two periods is statistically insignificant at the 95% confidence level, indicating no robust change in the interannual variability of the SASM. This may explain why the projected increase in SASM rainfall does not conflict with the weakening midlatitude wave teleconnection.
Boxplots of (a) the WSI (10−6 s−1) and (b) the vorticity of NEAAC (10−6 s−1). The vorticity of NEAAC is calculated using the 200-hPa horizontal wind anomalies partially regressed on the northern SASM rainfall over the region 27.5°–50°N, 105°–137.5°E. The WSI is calculated by the vorticity of SCAAC, NEAAC, and a cyclone anomaly (see section 2c). The horizontal solid lines and dashed lines inside the boxes indicate the median and mean respectively. (c) The scatterplot of results of the vorticity of NEAAC for the period of 2064–99 in SSP2–4.5 minus that for the period of 1979–2014 in the historical scenario (x axis; 10−6 s−1), and the result of JJA rainfall over northern China for the period of 2064–99 in SSP2–4.5 minus that for the period of 1979–2014 in historical (y axis; mm day−1). The NEAAC and rainfall anomalies over northern China are also shown as boxplots in vertical and horizontal orientations, respectively. The blue stars denote the results of the mbMME mean. The linear regression of the 26 CMIP6 models’ data on the x axis and y axis is shown by the red dashed line. The numbers at the upper-right corner are the correlation coefficients, with those significantly surpassing the 95% confidence level marked with asterisks. The numbers in parentheses in the upper-right corner correspond to the correlation coefficients across the mbMME models. (d) As in (c), but for the results in SSP5–8.5 minus that of the historical scenario. The red color in the legend denotes the models that better reproduce the midlatitude pathway.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
(a) The MME mean of partially regressed 200-hPa geopotential height (shading; gpm), horizontal wind anomalies (vectors; m s−1), and climatological 200-hPa u-wind (green contours; −1.5, 0, and 1.5 m s−1) on the northern SASM rainfall for the period of 2064–99 in SSP2–4.5 minus those for the period of 1979–2014 in historical. (b) As in (a), but for the difference between the SSP5–8.5 and historical runs. (c),(d) As in (a) and (b), but for the mbMME mean. The solid, thick solid, and dashed green contours denote the positive, zero, and negative values respectively. Dots and black vectors delineate the areas where over 70% of the selected models exhibit consistent signs with the MME mean.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
Considering that the changes in westerly jet stream can influence the wave pattern, the future climatological zonal wind is investigated. Figure 8 shows that the 200-hPa climatological zonal wind is projected to be weakened over Northeast Asia. This feature may be attributed to the increase in mid- and high-latitude temperatures, which reduces the meridional temperature gradient in the midlatitudes, consequently weakening the westerly jet stream (Fig. S4). It is noteworthy that the region of weakening climatological zonal wind highly overlaps with the decreasing wave pattern (Fig. 8). Eighteen (69.2%) models in SSP2–4.5 and 14 (53.8%) models in SSP5–8.5 show weakening NEAAC and 200-hPa climatological zonal wind over Northeast Asia (40°–47.5°N, 115°–135°E; Fig. 9a). The correlation coefficients between these two indices among models reach −0.24 in SSP2–4.5 and −0.50 in SSP5–8.5 (Fig. 9a), with the latter surpassing the 95% confidence level according to the Student’s t test. The linear relationship between these two indices among models suggests that the weakening of the wave pattern in the future can be attributed to the weakening climatological westerly jet stream, because the jet stream usually supplies potential energy to the wave pattern (Kosaka et al. 2009). This result is similar to that of Lin et al. (2017), who underscored the role of basic-state changes in the interdecadal weakening of the SASM–EASM relationship at the end of the 1970s.
The scatterplot of (a) the climatological westerly jet steam index (40°–47.5°N, 115°–135°E; x axis; m s−1) and the vorticity of NEAAC (y axis; 10−6 s−1) and (b) the partially regressed MC rainfall on the southern SASM rainfall (x axis; mm day−1) and the vorticity of WNPAC (y axis; 10−6 s−1). Blue and red colors denote the results for the period of 2064–99 in SSP2–4.5 minus that for the period of 1979–2014 in historical, and the results for the period of 2064–99 in SSP5–8.5 minus those for the period of 1979–2014 in historical, respectively. The triangles are the results of (a) mbMME mean and (b) lbMME mean, while the star represents the MME mean. The dashed lines are the linear regression of the 26 CMIP6 models’ data on the x axis and y axis. The correlation coefficients between the data on x axis and y axis are shown in the upper-right corner, with the former denoting the results of SSP2–4.5 and the latter representing the results of SSP5–8.5. Those correlation coefficients significantly surpassing the 95% confidence level are marked with asterisks.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
To emphasize the important role of projected basic-state changes on the weakening midlatitude wave teleconnection, two idealized experiments based on the LBM model are conducted. The two idealized experiments apply the same diabatic heating forcing. One uses the mbMME mean of JJA climatology for the period of 1979–2014 in historical as the basic states (CTRL experiment), and the other uses the JJA climatology for the period of 2064–99 in SSP5–8.5 as the basic state (BS585 experiment). The 200-hPa climatological u wind for these periods, as well as their differences, is shown in Figs. 1b and 1c. It is seen that the summer westerly jet stream is projected to be weakened over Northeast Asia (Fig. 1c). The responses of 200-hPa geopotential height and wind in CTRL experiment (Fig. 10a) well reproduce the midlatitude wave teleconnection linking the SASM and the EASM. This result indicates the LBM model can be used to verify the influence of future basic state on the midlatitude wave pattern. In the BS585 experiment, the weakening of the NEAAC and the SCAAC is observed (Figs. 10b,c), suggesting that the future changes in the basic state contribute to the weakening of these features.
Responses of 200-hPa geopotential height (shading; gpm) and 200-hPa winds (vectors; m s−1) in (a) CTRL experiment and (b) BS585 experiment, as well as (c) their differences (BS585 minus CTRL). In (c), the stippling denotes the values of geopotential height significantly exceeding the 95% confidence level. Only the wind anomalies surpassing the 95% confidence level are displayed in (c).
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
Given the expected weakening of the midlatitude teleconnection, the future changes in the relationship between the rainfalls over northern China and the northern SASM region are also investigated. The rainfall relationship is projected to weaken in the future in both SSP2–4.5 and SSP5–8.5 (Figs. 7c,d), with five (83.3%) out of the six well-simulating models suggesting the same conclusion. In SSP5–8.5, the correlation coefficient between the changes in NEAAC and the rainfall anomalies over northern China related to the northern SASM rainfall across the models reaches −0.52 (P < 0.05). This result indicates that the uncertainty in the rainfall relationship changes is linked to the spread in NEAAC anomaly changes. As discussed in section 3a, the NEAAC influences northern China rainfall anomalies through moisture transport, where a weakening upper-level NEAAC may decrease the low-level moisture transport to northern China through the barotropic anticyclone, consequently reducing the local rainfall anomalies associated with NEAAC variability.
According to the Clausius–Clapeyron relationship, an increase in the global temperature will lead to more moisture carried by the atmosphere. This may result in changes in the moisture relationship between northern China and the SASM, modifying the rainfall relationship. Therefore, to reveal the relative roles of circulation relationship changes and atmospheric moisture changes in regulating the future rainfall relationship between northern China and the SASM, a moisture transport diagnosis is applied over northern China (Fig. 11). Based on the moisture budget equation, regressed rainfall anomalies associated with northern SASM rainfall can be contributed by local evaporation anomalies and the dynamic and thermodynamic moisture transport related to the northern SASM rainfall. It is found that the future changes in rainfall anomalies over northern China associated with northern SASM rainfall are consistent with the changes in the dynamic terms among models in both SSP2–4.5 (Fig. 11a) and SSP5–8.5 (Fig. 11b). The correlation coefficient between the dynamic components in vertical moisture transport and the rainfall anomalies among models reaches 0.72 (P < 0.01; Table 2) in SSP2–4.5 and 0.81 (P < 0.01) in SSP5–8.5. This result shows that the decrease in rainfall relationship between northern China and the SASM is dominated by the dynamic term of vertical moisture transport, whose key role in future global and regional monsoon rainfall changes has also been highlighted by other studies (Chen and Zhou 2015; Chen et al. 2020). The dynamic term of horizontal moisture advection also contributes to the change in rainfall relationship, with a correlation coefficient of 0.37 (P < 0.10) in SSP5–8.5. Figure 11 shows the changes in NEAAC, whose correlations with the dynamic components of horizontal and vertical transports are, respectively, −0.62 (P < 0.01) and −0.50 (P < 0.01) in SSP5–8.5 (Table 2). These results suggest that the weakening rainfall relationship between northern China and the SASM in the future can be attributed to the decreased dynamic component in moisture transport and the suppressed vertical motions by the weakening NEAAC. Additionally, the projected increase in vertical stability over the EASM region may also suppress vertical motions and contribute to the projected decrease in northern China rainfall (Dai et al. 2022).
Six terms of q budget diagnosis over northern China partially regressed on the northern SASM rainfall (mm day−1) (a) for the period of 2064–99 in SSP2–4.5 minus that for the period of 1979–2014 in historical (refer to the left y axis) and (b) for the period of 2064–99 in SSP5–8.5 minus that for the period of 1979–2014 in historical (refer to the left y axis) in 26 CMIP6 models, MME mean, and mbMME mean. The blue cross represents the vorticity of the NEAAC (10−6 s−1) in the corresponding period minus that in the historical period (refer to the right y axis).
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
The correlation coefficients between rainfall anomalies and dynamic moisture transport terms over northern China among models, as well as the correlation coefficients between dynamic moisture transport terms and the future changes in the vorticity of NEAAC among 26 models. Similar correlation coefficients are calculated over YRV and the future changes in the vorticity of WNPAC. The upper (lower) number in each cell represents the results from SSP2–4.5 (SSP5–8.5). Correlation coefficient numbers marked with * indicate the significant values surpassing the 99% confidence level according to the Student’s t test.
b. The low-latitude moisture transport relationship
To understand the future changes in the low-latitude pathway connecting the SASM and the EASM, characterized by the coherent moisture transport with one flow heading westward to the SASM region and another flow along the west edge of the western North Pacific subtropical high to the YRV, it is important to examine the future changes in WNPAC. In both SSP2–4.5 and SSP5–8.5, the WNPAC is projected to be strengthened as shown by the boxplot in Fig. 12a, as supported by the median, MME mean, and lbMME mean values. The analysis of the 850-hPa geopotential height and horizontal wind anomalies related to the southern SASM rainfall in SSP2–4.5 (Fig. 13c) and SSP5–8.5 (Fig. 13d), minus those in historical runs, also supports the above result. The lbMME mean shows a strengthening WNPAC in the future, implying stronger moisture transport along its western edge. In SSP5–8.5, the lbMME mean also projects a northward expansion of the WNPAC.
(a) Boxplots of the vorticity of WNPAC (10−6 s−1). The vorticity of WNPAC is calculated using the 200-hPa horizontal wind anomalies partially regressed on the southern SASM rainfall over the region 15°–35°N, 110°–137.5°E. The horizontal solid and dash lines inside the boxes indicate the median and mean, respectively. (b) The scatterplot of the results of vorticity of WNPAC for the period of 2064–99 in SSP2–4.5 minus that for the period of 1979–2014 in the historical scenario (x axis; 10−6 s−1), and the result of JJA rainfall over the YRV for the period of 2064–99 in SSP2–4.5 minus that for the period of 1979–2014 in the historical scenario (y axis; mm day−1). The WNPAC and rainfall anomalies over the YRV are also shown as boxplots in vertical and horizontal orientations, respectively. The blue stars denote the results of the lbMME mean, and the hollow circles denote the individual data points beyond the whiskers of boxplots. (c) As in (b), but for the result of SSP5–8.5 minus that of the historical scenario. The blue color in the legend represents the models that better reproduce the low-latitude pathway.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
(a) The MME mean of partially regressed 850-hPa geopotential height (shading; gpm) and horizontal wind anomalies (vectors; m s−1) on the southern SASM rainfall for the period of 2064–99 in SSP2–4.5 minus that for the period of 1979–2014 in historical. (b) As in (a), but for the difference between the SSP5–8.5 and the historical experiment. (c),(d) As in (a) and (b), but for the lbMME mean. Dots and black vectors delineate the areas where over 70% of the selected models exhibit consistent signs with the MME mean.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
Considering that the MC rainfall anomalies related to the southern SASM rainfall can influence the WNPAC through the Gill response, the future changes in rainfall anomalies over the MC are also taken into account. It is found that the changes in MC rainfall anomalies and the changes in WNPAC are significantly and negatively correlated among models, with coefficients of −0.82 (P < 0.01) in SSP2–4.5 and −0.87 (P < 0.01) in SSP5–8.5, respectively (Fig. 9b). The linear relationship between these two indices among models suggests that the strengthening rainfall over the MC corresponds to the strengthening WNPAC in the future. This is likely because the anomalous rainfall over MC can drive an anticyclone Rossby wave over the western North Pacific, which is also suggested by Li et al. (2022).
We also assess the future changes in the rainfall relationship between the YRV and the SASM. In SSP2–4.5, the rainfall anomalies over YRV related to the southern SASM rainfall are expected to decrease (Fig. 12b). In SSP5–8.5, the lbMME mean projects diminished rainfall anomalies over the YRV (Fig. 12c). Twelve models in SSP2–4.5 (Fig. 12b) and six models in SSP5–8.5 (Fig. 12c) project a strengthening WNPAC and decreasing rainfall anomalies over the YRV. In SSP5–8.5, 10 models expect a strengthening WNPAC and an increasing rainfall relationship between the YRV and the SASM. The different results between SSP2–4.5 and SSP5–8.5 might be attributed to the location shift of the WNPAC, which will be discussed below.
A moisture transport diagnosis is carried out to further investigate the roles of future changing WNPAC and atmospheric moisture in contributing to the future rainfall relationship between the YRV and the SASM (Fig. 14). The future changes in rainfall anomalies over the YRV are highly correlated with the dynamic components in vertical moisture transport in both SSP2–4.5 and SSP5–8.5 (Table 2), with correlation coefficients of 0.85 (P < 0.01) and 0.95 (P < 0.01), respectively. These results indicate that the dynamic vertical moisture transport is the dominant term, whose importance for the future changes in global and regional monsoon rainfall has also been noted by Chen and Zhou (2015) and Chen et al. (2020). Additionally, in SSP5–8.5, the dynamic components in horizontal moisture advection also contribute to the change in rainfall relationship, with a correlation coefficient of 0.60 (P < 0.01). These results illustrate that the rainfall relationship changes between the YRV and the SASM in the future may be attributed to the circulation relationship changes, instead of atmospheric moisture changes.
As in Fig. 9, but for the result over the YRV partially regressed onto the southern SASM rainfall in 26 CMIP6 models, MME mean, and lbMME mean. The blue cross denotes the vorticity of WNPAC (10−6 s−1) in the corresponding period minus that in the historical period (refer to the right y axis).
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
As presented in Fig. 14, the vorticity projections in WNPAC are negatively correlated with the dynamic moisture advection in both SSP2–4.5 and SSP5–8.5 (Table 2), demonstrating that the strengthening WNPAC circulation corresponds to the increasing moisture advection to the YRV. It should be noted that the location and expansion of WNPAC can also influence the rainfall relationship between the YRV and the SASM through vertical motions. Figure 15 presents the longitudinal-mean (105°–120°E) zonal wind anomalies related to the southern SASM rainfall in historical, SSP2–4.5, and SSP5–8.5 runs. The MME mean shows a northward shift of the maximum westerlies and easterlies in SSP2–4.5, implying a northward shift of WNPAC. This result is consistent with the decrease in vertical moisture transport and thus reduces rainfall anomalies over the YRV (Fig. 15a). In SSP5–8.5, the WNPAC moves southward compared to that in SSP2–4.5, which corresponds to the relatively increasing YRV rainfall. The WNPAC in the lbMME mean is characterized by the continuous northward expansion in SSP2–4.5 and SSP5–8.5, consistent with the decreasing vertical moisture transport and reducing rainfall anomalies over the YRV (Fig. 15b). In summary, the future strengthening and northward shift of the WNPAC is projected, with a decreasing rainfall relationship between the YRV and the SASM.
The longitudinal-mean (105°–120°E) zonal wind anomalies (m s−1) related to the southern SASM rainfall in historical, SSP2–4.5, and SSP5–8.5 runs of (a) MME mean and (b) lbMME mean. The lower boxes show the rainfall anomalies (mm day−1), dynamic horizontal moisture advection (mm day−1), and dynamic vertical moisture transport (mm day−1) over the YRV in corresponding periods.
Citation: Journal of Climate 38, 10; 10.1175/JCLI-D-24-0357.1
5. Conclusions and discussion
In this study, the distinct future changes in midlatitude wave teleconnection and low-latitude coherent moisture transport, the two key pathways bridging the SASM and the EASM, are examined. The midlatitude wave pattern is driven by anomalous diabatic heating over the northern SASM region and propagates along the climatological westerly jet stream. It can influence the rainfall relationship between northern China and the SASM region by modulating moisture transport to northern China through the NEAAC. The low-latitude pathway is characterized by the coherent moisture transport with one flow toward the SASM region and the other along the western edge of the WNPAC to the YRV. This pathway is closely related to the relationship between the rainfall anomalies over the YRV and the southern SASM regions. CMIP6 models can well reproduce the midlatitude wave pattern while showing large model spread and underestimation in simulating the low-latitude moisture transport pathway. Improved simulations of the strength of climatological westerly jet stream and the Gill response related to the rainfall anomalies over the Maritime Continent may serve as key factors for better reproduction of the midlatitude and low-latitude pathways in models, respectively.
The midlatitude wave pattern and the NEAAC are projected to become weaker in the future in the mbMME mean. The weakening NEAAC is considered to be related to the weakening climatological westerly jet stream, which is suggested by the results of two idealized experiments based on the LBM. The rainfall relationship between northern China and the SASM region is projected to decrease in the future, which may be attributed to the diminishing dynamic components of moisture transport and decreasing vertical motions raised up by the weakening NEAAC.
The WNPAC, an important circulation system in the low-latitude moisture transport pathway, is expected to become stronger in the future. This result may be attributed to the increase in the rainfall anomalies over the MC due to the Gill response. The relationship between the rainfalls over the YRV and the southern SASM region is projected to be weakened in SSP2–4.5 and exhibits uncertain changes in SSP5–8.5. This rainfall relationship change is dominated by the dynamic vertical moisture transport term. When taking into account the circulation changes, the strengthening WNPAC corresponds to the increasing dynamic horizontal moisture transport, while the northward shift of WNPAC decreases the dynamic vertical moisture transport. In the lbMME mean, the northward shift of WNPAC may lead to a decrease in rainfall relationship between the YRV and the SASM region.
This study provides a valuable insight for better clarifying the future changes in the SASM–EASM relationship under global warming through separately analyzing the rainfall patterns associated with the two pathways and the underlying physical mechanisms. Grasping these future changes can help understanding extreme events in the two monsoon regions. For example, the unprecedented drought and heatwave over the YRV (Yin et al. 2023) and the record-shattering rainfall in Pakistan in 2022 (You et al. 2024) might be attributed to the intensification of the low-latitude pathway, which directs more moisture toward the SASM region while suppressing vertical motions over the YRV through the enhanced WNPAC (Tang et al. 2023).
Additionally, temperature changes exhibit nonuniform spatial distribution in the future warmer climate (Xia et al. 2014; Yao et al. 2016), which implies possible distinct future changes in different climate systems such as the weakening westerly jet stream (Dong et al. 2022), weakening South Asian high (Zhang et al. 2022), and strengthening western North Pacific subtropical high (Yang et al. 2022). Through investigating the future changes in the SASM–EASM relationship with separately analyzing the midlatitude and low-latitude pathways, this study provides a unique perspective on studying the inhomogeneous responses of important Asian climate systems under global warming.
However, it should be noted that the midlatitude and low-latitude pathways are not necessarily independent from each other. Previous studies have indicated that the western North Pacific subtropical high strengthens and extends westward when the South Asian high intensifies and extends eastward (Tao and Zhu 1964; Jiang et al. 2011; Wei et al. 2019; Lu et al. 2023). Under global warming, the northward shift of the WNPAC may also influence the midlatitude wave pattern (figure not shown), which may explain that the WSIs are weaker in SSP2–4.5 than in SSP5–8.5 (Figs. 6a,b). Additionally, other factors such as the future changes in interhemispheric warming patterns and land–sea thermal contrasts may also influence the relationship between the SASM and the EASM (Chen and Zhou 2015; Cao et al. 2020; Wang et al. 2020; Chen et al. 2022). The projected interhemispheric temperature gradients and associated low-level cross-equatorial flow, a key source of uncertainty in the future changes in the Afro-Asian summer monsoon, can affect moisture transport to the SASM and EASM regions (Cao et al. 2020; Wang et al. 2020; Chen et al. 2022). Furthermore, the enhanced land–ocean thermal contrast between the vast Eurasian–African landmass and adjacent oceans may also modulate the circulation and moisture transport in the SASM and EASM regions (Chen and Zhou 2015; Wang et al. 2020). These factors warrant further investigations in future studies.
Acknowledgments.
The authors are grateful to the three anonymous reviewers for providing thorough and insightful comments, which are helpful for improving the overall quality of the paper. This study was supported by the National Natural Science Foundation of China (42088101, 42375029), the Guangdong Basic and Applied Basic Research Foundation (2023A1515010908), the Science and Technology Planning Project of Guangdong Province (2023B1212060019), and the Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai) (SML2023SP209).
Data availability statement.
The CMIP6 model data are obtained at https://aims2.llnl.gov/search/cmip6/, the GPCP data at https://rda.ucar.edu/datasets/ds728.2/, and other reanalysis data of the ERA5 at https://cds.climate.copernicus.eu/#!/search?text=ERA5&type=dataset.
REFERENCES
Adler, R. F., and Coauthors, 2003: The Version-2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). J. Hydrometeor., 4, 1147–1167, https://doi.org/10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.
Cao, J., J. Hu, and Y. Tao, 2012: An index for the interface between the Indian Summer Monsoon and the East Asian Summer Monsoon. J. Geophys. Res., 117, D18108, https://doi.org/10.1029/2012JD017841.
Cao, J., B. Wang, B. Wang, H. Zhao, C. Wang, and Y. Han, 2020: Sources of the intermodel spread in projected global monsoon hydrological sensitivity. Geophys. Res. Lett., 47, e2020GL089560, https://doi.org/10.1029/2020GL089560.
Chen, H.-j., S. Yang, and W. Wei, 2023: Future changes in the relationship between the South and East Asian summer monsoons in CMIP6 models. J. Trop. Meteor., 29, 191–203, https://doi.org/10.46267/j.1006-8775.2023.015.
Chen, W., and Coauthors, 2023: Recent advances in understanding multi-scale climate variability of the Asian monsoon. Adv. Atmos. Sci., 40, 1429–1456, https://doi.org/10.1007/s00376-023-2266-8.
Chen, X., and T. Zhou, 2015: Distinct effects of global mean warming and regional sea surface warming pattern on projected uncertainty in the South Asian summer monsoon. Geophys. Res. Lett., 42, 9433–9439, https://doi.org/10.1002/2015GL066384.
Chen, Z., T. Zhou, L. Zhang, X. Chen, W. Zhang, and J. Jiang, 2020: Global land monsoon precipitation changes in CMIP6 projections. Geophys. Res. Lett., 47, e2019GL086902, https://doi.org/10.1029/2019GL086902.
Chen, Z., T. Zhou, X. Chen, W. Zhang, L. Zhang, M. Wu, and L. Zou, 2022: Observationally constrained projection of Afro-Asian monsoon precipitation. Nat. Commun., 13, 2552, https://doi.org/10.1038/s41467-022-30106-z.
Chou, C., J. D. Neelin, C.-A. Chen, and J.-Y. Tu, 2009: Evaluating the “rich-get-richer” mechanism in tropical precipitation change under global warming. J. Climate, 22, 1982–2005, https://doi.org/10.1175/2008JCLI2471.1.
Dai, L., T. F. Cheng, and M. Lu, 2022: Anthropogenic warming disrupts intraseasonal monsoon stages and brings dry-get-wetter climate in future East Asia. npj Climate Atmos. Sci., 5, 11, https://doi.org/10.1038/s41612-022-00235-9.
Day, J. A., I. Fung, and C. Risi, 2015: Coupling of South and East Asian monsoon precipitation in July–August. J. Climate, 28, 4330–4356, https://doi.org/10.1175/JCLI-D-14-00393.1.
Ding, Q., and B. Wang, 2005: Circumglobal teleconnection in the Northern Hemisphere summer. J. Climate, 18, 3483–3505, https://doi.org/10.1175/JCLI3473.1.
Ding, Y., and J. C. L. Chan, 2005: The East Asian summer monsoon: An overview. Meteor. Atmos. Phys., 89, 117–142, https://doi.org/10.1007/s00703-005-0125-z.
Dong, B., R. T. Sutton, L. Shaffrey, and B. Harvey, 2022: Recent decadal weakening of the summer Eurasian westerly jet attributable to anthropogenic aerosol emissions. Nat. Commun., 13, 1148, https://doi.org/10.1038/s41467-022-28816-5.
Draper, N. R., and H. Smith, 1998: Applied Regression Analysis. 3rd ed. Wiley, 736 pp.
Gill, A. E., 1980: Some simple solutions for heat-induced tropical circulation. Quart. J. Roy. Meteor. Soc., 106, 447–462, https://doi.org/10.1002/qj.49710644905.
Greatbatch, R. J., X. Sun, and X.-Q. Yang, 2013: Impact of variability in the Indian summer monsoon on the East Asian summer monsoon. Atmos. Sci. Lett., 14, 14–19, https://doi.org/10.1002/asl2.408.
Guo, Q., 1992: Teleconection between the floods/droughts in North China and Indian summer monsoon rainfall. Acta Geogr. Sin., 47, 394–402, https://doi.org/10.11821/xb199205002.
Guo, Q., and J. Wang, 1988: Comparative study of summer monsoon precipitation in China and India. J. Trop. Meteor., 4, 53–60, https://doi.org/10.16032/j.issn.1004-4965.1988.01.007.
Ha, K.-J., Y.-W. Seo, J.-Y. Lee, R. H. Kripalani, and K.-S. Yun, 2018: Linkages between the South and East Asian summer monsoons: A review and revisit. Climate Dyn., 51, 4207–4227, https://doi.org/10.1007/s00382-017-3773-z.
Ha, K.-J., S. Moon, A. Timmermann, and D. Kim, 2020: Future changes of summer monsoon characteristics and evaporative demand over Asia in CMIP6 simulations. Geophys. Res. Lett., 47, e2020GL087492, https://doi.org/10.1029/2020GL087492.
He, C., Z. Wang, T. Zhou, and T. Li, 2019: Enhanced latent heating over the Tibetan Plateau as a key to the enhanced East Asian summer monsoon circulation under a warming climate. J. Climate, 32, 3373–3388, https://doi.org/10.1175/JCLI-D-18-0427.1.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hu, P., W. Chen, S. Chen, R. Yang, L. Wang, and Y. Liu, 2024: Revisiting the linkage between the Pacific–Japan pattern and Indian summer monsoon rainfall: The crucial role of the Maritime Continent. Geophys. Res. Lett., 51, e2023GL106982, https://doi.org/10.1029/2023GL106982.
Jiang, D., and Z. Tian, 2013: East Asian monsoon change for the 21st century: Results of CMIP3 and CMIP5 models. Chin. Sci. Bull., 58, 1427–1435, https://doi.org/10.1007/s11434-012-5533-0.
Jiang, X., Y. Li, S. Yang, and R. Wu, 2011: Interannual and interdecadal variations of the South Asian and western Pacific subtropical highs and their relationships with Asian-Pacific summer climate. Meteor. Atmos. Phys., 113, 171–180, https://doi.org/10.1007/s00703-011-0146-8.
Kim, B.-J., S.-E. Moon, L. Riyu, and R. H. Kripalani, 2002: Teleconnections: Summer monsoon over Korea and India. Adv. Atmos. Sci., 19, 665–676, https://doi.org/10.1007/s00376-002-0006-z.
Kosaka, Y., 2021: Coupling of the Indian, western North Pacific, and East Asian summer monsoons. Indian Summer Monsoon Variability, J. Chowdary, A. Parekh, and C. Gnanaseelan, Eds., Elsevier, 263–286.
Kosaka, Y., H. Nakamura, M. Watanabe, and M. Kimoto, 2009: Analysis on the dynamics of a wave-like teleconnection pattern along the summertime Asian jet based on a reanalysis dataset and climate model simulations. J. Meteor. Soc. Japan, 87, 561–580, https://doi.org/10.2151/jmsj.87.561.
Kripalani, R. H., and A. Kulkarni, 2001: Monsoon rainfall variations and teleconnections over South and East Asia. Int. J. Climatol., 21, 603–616, https://doi.org/10.1002/joc.625.
Krishnan, R., and M. Sugi, 2001: Baiu rainfall variability and associated monsoon teleconnections. J. Meteor. Soc. Japan, 79, 851–860, https://doi.org/10.2151/jmsj.79.851.
Li, T., Y. Wang, B. Wang, M. Ting, Y. Ding, Y. Sun, C. He, and G. Yang, 2022: Distinctive South and East Asian monsoon circulation responses to global warming. Sci. Bull., 67, 762–770, https://doi.org/10.1016/j.scib.2021.12.001.
Li, X.-F., Z.-H. Jiang, and Y. Shi, 2019: Moisture transport paths and sources of South China annually first rainy season and their relationship with the onset of the South China Sea summer monsoon. J. Trop. Meteor., 25, 171–179, https://doi.org/10.16555/j.1006-8775.2019.02.004.
Li, Z., Y. Sun, T. Li, Y. Ding, and T. Hu, 2019: Future changes in East Asian summer monsoon circulation and precipitation under 1.5 to 5 °C of warming. Earth’s Future, 7, 1391–1406, https://doi.org/10.1029/2019EF001276.
Lin, Z., R. Lu, and R. Wu, 2017: Weakened impact of the Indian early summer monsoon on North China rainfall around the late 1970s: Role of basic-state change. J. Climate, 30, 7991–8005, https://doi.org/10.1175/JCLI-D-17-0036.1.
Liu, Y., and Y. Ding, 2008: Analysis and numerical simulation of the teleconnection between Indian summer monsoon and precipitation in North China. Acta Meteor. Sin., 66, 789–799, https://doi.org/10.11676/qxxb2008.072.
Lu, M., S. Yang, C. Zhu, J. Wang, S. Lin, W. Wei, and H. Fan, 2023: Thermal impact of the southern Tibetan Plateau on the Southeast Asian summer monsoon and modulation by the tropical Atlantic SST. J. Climate, 36, 1319–1330, https://doi.org/10.1175/JCLI-D-22-0493.1.
Lu, R.-Y., J.-H. Oh, and B.-J. Kim, 2002: A teleconnection pattern in upper-level meridional wind over the North African and Eurasian continent in summer. Tellus, 54A, 44–55, https://doi.org/10.3402/tellusa.v54i1.12122.
Moon, S., and K.-J. Ha, 2020: Future changes in monsoon duration and precipitation using CMIP6. npj Climate Atmos. Sci., 3, 45, https://doi.org/10.1038/s41612-020-00151-w.
O’Neill, B. C., and Coauthors, 2016: The Scenario Model Intercomparison Project (ScenarioMIP) for CMIP6. Geosci. Model Dev., 9, 3461–3482, https://doi.org/10.5194/gmd-9-3461-2016.
Preethi, B., M. Mujumdar, A. Prabhu, and R. Kripalani, 2017: Variability and teleconnections of South and East Asian summer monsoons in present and future projections of CMIP5 climate models. Asia-Pac. J. Atmos. Sci., 53, 305–325, https://doi.org/10.1007/s13143-017-0034-3.
Seager, R., N. Naik, and G. A. Vecchi, 2010: Thermodynamic and dynamic mechanisms for large-scale changes in the hydrological cycle in response to global warming. J. Climate, 23, 4651–4668, https://doi.org/10.1175/2010JCLI3655.1.
Song, Y. H., E.-S. Chung, and S. Shahid, 2024: Global future climate signal by latitudes using CMIP6 GCMs. Earth’s Future, 12, e2022EF003183, https://doi.org/10.1029/2022EF003183.
Sun, J., and J. Ming, 2019: Possible mechanism for the weakening relationship between Indian and central East Asian summer rainfall after the late 1970s: Role of the mid-to-high-latitude atmospheric circulation. Meteor. Atmos. Phys., 131, 517–524, https://doi.org/10.1007/s00703-018-0586-5.
Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58, 608–627, https://doi.org/10.1175/1520-0469(2001)058<0608:AFOAPI>2.0.CO;2.
Tang, S., and Coauthors, 2023: Linkages of unprecedented 2022 Yangtze River Valley heatwaves to Pakistan flood and triple-dip La Niña. npj Climate Atmos. Sci., 6, 44, https://doi.org/10.1038/s41612-023-00386-3.
Tao, S. Y., and F. K. Zhu, 1964: The 100-mb flow patterns in southern Asia in summer and its relation to the advance and retreat of the West-Pacific subtropical anticyclone over the Far East. Acta Meteor. Sin., 34, 385–396.
Tett, S. F. B., J. M. Gregory, N. Freychet, C. Cartis, M. J. Mineter, and L. Roberts, 2022: Does model calibration reduce uncertainty in climate projections? J. Climate, 35, 2585–2602, https://doi.org/10.1175/JCLI-D-21-0434.1.
Tian, Z., D. Hu, X. Lang, and D. Jiang, 2022: Index‐ and model‐dependent projections of East Asian summer monsoon in Coupled Model Intercomparison Project Phase 6 simulations. Int. J. Climatol., 42, 2208–2224, https://doi.org/10.1002/joc.7361.
Ueda, H., A. Iwai, K. Kuwako, and M. E. Hori, 2006: Impact of anthropogenic forcing on the Asian summer monsoon as simulated by eight GCMs. Geophys. Res. Lett., 33, L06703, https://doi.org/10.1029/2005GL025336.
Wang, B., S.-Y. Yim, J.-Y. Lee, J. Liu, and K.-J. Ha, 2014: Future change of Asian-Australian monsoon under RCP 4.5 anthropogenic warming scenario. Climate Dyn., 42, 83–100, https://doi.org/10.1007/s00382-013-1769-x.
Wang, B., C. Jin, and J. Liu, 2020: Understanding future change of global monsoons projected by CMIP6 models. J. Climate, 33, 6471–6489, https://doi.org/10.1175/JCLI-D-19-0993.1.
Wang, B., and Coauthors, 2021: Monsoons climate change assessment. Bull. Amer. Meteor. Soc., 102, E1–E19, https://doi.org/10.1175/BAMS-D-19-0335.1.
Wang, H., B. Wang, F. Huang, Q. Ding, and J.-Y. Lee, 2012: Interdecadal change of the boreal summer circumglobal teleconnection (1958–2010). Geophys. Res. Lett., 39, L12704, https://doi.org/10.1029/2012GL052371.
Wang, L., P. Xu, W. Chen, and Y. Liu, 2017: Interdecadal variations of the Silk Road pattern. J. Climate, 30, 9915–9932, https://doi.org/10.1175/JCLI-D-17-0340.1.
Wang, S. W., and J. B. Huang, 2006: Instability of precipitation teleconnection between North China and India. Prog. Nat. Sci., 16, 980–985.
Wang, Z., A. Duan, S. Yang, and K. Ullah, 2017: Atmospheric moisture budget and its regulation on the variability of summer precipitation over the Tibetan Plateau. J. Geophys. Res. Atmos., 122, 614–630, https://doi.org/10.1002/2016JD025515.
Watanabe, M., and M. Kimoto, 2000: Atmosphere-ocean thermal coupling in the North Atlantic: A positive feedback. Quart. J. Roy. Meteor. Soc., 126, 3343–3369, https://doi.org/10.1002/qj.49712657017.
Wei, W., R. Zhang, M. Wen, X. Rong, and T. Li, 2014: Impact of Indian summer monsoon on the South Asian High and its influence on summer rainfall over China. Climate Dyn., 43, 1257–1269, https://doi.org/10.1007/s00382-013-1938-y.
Wei, W., R. Zhang, M. Wen, B.-J. Kim, and J.-C. Nam, 2015: Interannual variation of the South Asian high and its relation with Indian and East Asian summer monsoon rainfall. J. Climate, 28, 2623–2634, https://doi.org/10.1175/JCLI-D-14-00454.1.
Wei, W., R. Zhang, M. Wen, S. Yang, and W. Li, 2019: Dynamic effect of the South Asian high on the interannual zonal extension of the western North Pacific subtropical high. Int. J. Climatol., 39, 5367–5379, https://doi.org/10.1002/joc.6160.
Wu, R., 2002: A mid-latitude Asian circulation anomaly pattern in boreal summer and its connection with the Indian and East Asian summer monsoons. Int. J. Climatol., 22, 1879–1895, https://doi.org/10.1002/joc.845.
Wu, R., 2017: Relationship between Indian and East Asian summer rainfall variations. Adv. Atmos. Sci., 34, 4–15, https://doi.org/10.1007/s00376-016-6216-6.
Wu, R., and B. Wang, 2002: A contrast of the East Asian summer monsoon–ENSO relationship between 1962–77 and 1978–93. J. Climate, 15, 3266–3279, https://doi.org/10.1175/1520-0442(2002)015<3266:ACOTEA>2.0.CO;2.
Wu, R., and Y. Jiao, 2017: The impacts of the Indian summer rainfall on North China summer rainfall. Asia-Pac. J. Atmos. Sci., 53, 195–206, https://doi.org/10.1007/s13143-017-0013-8.
Wu, R.-G., K.-M. Hu, and Z.-D. Lin, 2018: Perspectives on the non-stationarity of the relationship between Indian and East Asian summer rainfall variations. Atmos. Oceanic Sci. Lett., 11, 104–111, https://doi.org/10.1080/16742834.2018.1387758.
Xia, J., J. Chen, S. Piao, P. Ciais, Y. Luo, and S. Wan, 2014: Terrestrial carbon cycle affected by non-uniform climate warming. Nat. Geosci., 7, 173–180, https://doi.org/10.1038/ngeo2093.
Yang, K., W. Cai, G. Huang, K. Hu, B. Ng, and G. Wang, 2022: Increased variability of the western Pacific subtropical high under greenhouse warming. Proc. Natl. Acad. Sci. USA, 119, e2120335119, https://doi.org/10.1073/pnas.2120335119.
Yao, S.-L., G. Huang, R.-G. Wu, X. Qu, and D. Chen, 2016: Inhomogeneous warming of the Tropical Indian Ocean in the CMIP5 model simulations during 1900–2005 and associated mechanisms. Climate Dyn., 46, 619–636, https://doi.org/10.1007/s00382-015-2602-5.
Yin, Z., S. Yang, and W. Wei, 2023: Prevalent atmospheric and oceanic signals of the unprecedented heatwaves over the Yangtze River valley in July–August 2022. Atmos. Res., 295, 107018, https://doi.org/10.1016/j.atmosres.2023.107018.
You, Y., and M. Ting, 2021a: Observed trends in the South Asian monsoon low‐pressure systems and rainfall extremes since the late 1970s. Geophys. Res. Lett., 48, e2021GL092378, https://doi.org/10.1029/2021GL092378.
You, Y., and M. Ting, 2021b: Low pressure systems and extreme precipitation in Southeast and East Asian monsoon regions. J. Climate, 34, 1147–1162, https://doi.org/10.1175/JCLI-D-20-0206.1.
You, Y., M. Ting, and M. Biasutti, 2024: Climate warming contributes to the record-shattering 2022 Pakistan rainfall. npj Climate Atmos. Sci., 7, 89, https://doi.org/10.1038/s41612-024-00630-4.
Zhang, D., Y. Huang, B. Zhou, and H. Wang, 2022: Interdecadal changes of the South Asian high in CMIP5/6 and projection of its future changes. J. Climate, 35, 5661–5675, https://doi.org/10.1175/JCLI-D-21-0921.1.
Zhang, R., 2001: Relations of water vapor transport from Indian monsoon with that over East Asia and the summer rainfall in China. Adv. Atmos. Sci., 18, 1005–1017, https://doi.org/10.1007/BF03403519.
Zhang, R. H., 1999: The role of Indian summer monsoon water vapor transportation on the summer rainfall anomalies in the northern part of China during the El Niño mature phase. Plateau Meteor., 18, 567–574.
Zhou, F., R. Zhang, and J. Han, 2020: Influences of the East Asian summer rainfall on circumglobal teleconnection. J. Climate, 33, 5213–5221, https://doi.org/10.1175/JCLI-D-19-0325.1.
