Inter-model Spread in Summer Precipitation: Connections and Causes Across Monsoon Regions in AMIP6 Simulations

Inter-model Spread in Summer Precipitation: Connections and Causes Across Monsoon Regions in AMIP6 Simulations | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Inter-model Spread in Summer Precipitation: Connections and Causes Across Monsoon Regions in AMIP6 Simulations Na Wen, Kai Huan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6711488/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study investigates the interconnections and underlying causes of the inter-model variability in summer precipitation across global monsoon regions, based on historical AMIP6 simulations. The results reveal significant connections among the leading modes of precipitation inter-model deviations in different monsoon domains, with the western North Pacific (WNP) emerging as a key driver. Deviations over the WNP modulate adjacent monsoon systems -- including the South Asian, North African, and Australian monsoons -- primarily through the Walker and Hadley circulations. In addition, WNP precipitation anomalies, via jet stream disturbances originating from the tropical Indian Ocean, generate zonal wave trains in the mid-to-high latitudes of both hemispheres, influencing the Somali and South American monsoon. The WNP also affects East Asian monsoon precipitation mainly through meridional wave trains along the East Asian coast, triggered by perturbations in the monsoon trough. These findings underscore the central role of WNP precipitation deviations -- via atmospheric dynamic process -- in shaping the global pattern of monsoon precipitation inter-model variability in AMIP6, and highlight the need for improved model performance over this region to enhance global monsoon simulation fidelity. AMIP6 Inter-model variability Monsoon precipitation Global monsoon Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction Monsoon activity is a critical driver of regional water resources, affecting nearly two-thirds of the global population. Accurate prediction of monsoon precipitation is therefore essential for effective water resource management, food security, disaster mitigation, and infrastructure planning. A widely used approach for such prediction is the multi-model ensemble mean (MME) method (Molteni et al., 1996 ; Krishnamurti, 1999 ; Chen et al., 2025), which reduces prediction uncertainty by averaging outputs from multiple climate models. The MME has been extensively adopted as an effective strategy for climate forecasting. However, substantial challenges remain in realistically simulating monsoon precipitation, particularly with regard to model performance and uncertainty (Lenartz et al., 2010 ; Zhi et al., 2012 ; Zhu et al., 2021 ). To enhance forecast skill, it requires a better understanding of the sources of uncertainty in monsoon predictions. Beyond the inherent complexity of the monsoon system, the uncertainty in monsoon predictions is further compounded by internal variability, model uncertainty, and scenario uncertainty that is associated with future climate projections (Lehner et al. 2020 ). Uncertainty due to internal variability originates from the chaotic character of the climate system (Lorenz 1963 ), which imposes a fundamental limit on climate predictability. Internal variability in a climate model can be effectively estimated using long-time mean or large initial-condition ensembles (Maher et al. 2018 ). Model uncertainty stems from structural differences across models, variations in parameterization schemes, and the differing responses of models to external forcing (Murphy et al., 2004 ; Bowler et al., 2008 ; Sanderson et al., 2008 ; Xu et al., 2022 ). These uncertainties reflect the diverse assumptions and design choices made during model development and parameter tuning. In practice, model uncertainty can be reduced through continuous advancements in model development. Under global warming, scenario uncertainty arises from the unknown trajectory of future greenhouse gas emissions. Zhou et al. ( 2020 ) utilized both Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations and single-model initial-condition large ensembles (SMILEs) to systematically assess the relative contributions of these uncertainties to the ensemble spread of projected global land monsoon precipitation. They found that for mean precipitation, model uncertainty is the dominant contributor—accounting for approximately 90% of total uncertainty—while the influence of internal variability decreases over time and scenario uncertainty becomes more significant. Their results emphasize the critical role of model uncertainty in determining the predictability of monsoon climates. To further investigate the monsoon precipitation uncertainty, several studies have focused on specific regional monsoon systems. For instance, Huang et al. ( 2022 ) found pronounced inter-model variability in East Asian summer precipitation within CMIP6 simulations, with roughly one-third of the models overestimating and two-thirds underestimating precipitation. These deviations were closely associated with discrepancies in the simulated western North Pacific anticyclone and the positioning of the subtropical westerly jet. Similarly, Hagos et al. ( 2019 ) reported that CMIP5 models exhibit a dry bias and substantial inter-model spread in simulating South Asian monsoon precipitation. This was largely attributed to differences in the representation of convection over the equatorial Indian Ocean during the pre-monsoon season. Their analysis suggested that model-to-model variations in precipitable water in this region perturb the large-scale circulation, thereby amplifying both the dry bias and the spread over the Indian subcontinent. These findings highlight the critical role of improving convective parameterizations over the equatorial Indian Ocean to enhance the fidelity of South Asian monsoon simulations. In contrast, Monerie et al. ( 2020 ) examined uncertainties in projected precipitation changes over the Sahel using outputs from 29 CMIP5 and 11 CMIP6 models. They found that future Sahel rainfall projections are subject to large inter-model uncertainties, primarily stemming from differences in the simulated atmospheric circulation response to global warming (i.e., dynamical changes), whereas thermodynamic contributions to uncertainty were relatively modest. Previous studies examining inter-model spread in monsoon regions have predominantly focused on regional scales, emphasizing precipitation deviations within individual monsoon systems. Although regional monsoons exhibit distinct characteristics shaped by geography and topography, growing evidence suggests that their variability is not entirely independent. Trenberth et al. ( 2000 ) introduced the concept of global monsoon, which arises from and reflects the coupled atmosphere–land–ocean system’s response to annual cycle of solar forcing. Wang and Ding ( 2008 ) further refined this concept, identifying the global monsoon as the dominant mode of annual variability in the tropics. Building on this foundation, Wang and Wang (2014) synthesized earlier studies and emphasized the coherent variability among regional monsoons across a range of timescales from interannual to tectonic, thus substantiating the global monsoon framework at multiple temporal scales. Expanding on this work, Wang et al. ( 2017 ) concluded that the co-variability of global monsoon on long time scales is primarily driven by external forcing, whereas on shorter timescales it is predominantly influenced by internal feedbacks within the climate system. For example, with regard to internal mechanisms underlying monsoon variability, the Indo-Pacific or Western Pacific Warm Pool is recognized as the heat engine of the global climate system. Its extensive exchange of heat and moisture exerts a significant influence on climate patterns worldwide. Even subtle variations in sea surface temperature (SST) within this region can alter the location and intensity of convection in the ascending branches of the Hadley and Walker circulations, disrupt planetary-scale atmospheric circulation patterns, and modify tropical hydrology (Qu et al., 2005 ). Within this context, a critical question arises: Does a similar coherence exist in the uncertainties of simulated monsoon precipitation across regions? If so, what are the underlying mechanisms responsible for such coordinated uncertainty? This study, therefore, focuses on examining the connections of the inter-model variability in precipitation across different monsoon regions, as well as the underlying causes driving the co-variability, using historical simulations from AMIP6. The paper is organized as follows: Section 2 describes the data and methodology. Section 3 evaluates the performance of simulated precipitation in individual monsoon regions, investigates the inter-regional linkages in inter-model variability, and explores the causes underlying these connections. Section 4 offers a summary and discussion. 2. Data and Methods This study utilizes historical simulations from 23 AMIP6 models spanning 1979–2014 to evaluate summer precipitation inter-model variability across different monsoon regions. The models, sourced from various institutions (listed in Appendix 1), were selected to ensure relative independence. To reduce the influence of internal atmospheric variability, ensemble means were calculated for all realization members of each model (Deser et al., 2012 ). For consistency, all model outputs were interpolated to a horizontal resolution of 2.5° × 2.5°. Summer climatological precipitation refers to the multi-year average for June, July, and August (JJA). Inter-model spread is defined as the deviation of each model’s summer climatological precipitation from the multi-model ensemble mean (MME). The MME was calculated as the unweighted average of all models. Following the global monsoon region classification by Wang et al. ( 2017 ), where local summer-minus-winter precipitation exceeds 300 mm and local summer precipitation exceeds 55% of the annual total, eight monsoon regions were selected for analysis: Western North Pacific (WNP): 10°N–20°N, 110°E–150°E; East Asia (EA): 20°N–45°N, 110°E–135°E; South Asia (SA): 10°N–33°N, 70°E–105°E; Australia (AUS): 5°S–20°S, 110°E–150°E; North Africa (NAF): 5°N–15°N, 30°W–30°E; Somalia (SAF): 7°S–25°S, 25°E–70°E; South America (SAM): 5°S–25°S, 70°W–40°W; and North America (NAM): 5°N–23°N, 110°W–80°W. To facilitate comparison of precipitation deviations across different monsoon regions, deviations for each model were normalized by the MME climatological precipitation. Empirical Orthogonal Function (EOF) analysis (Hannachi et al., 2007 ) was used to extract the leading modes of summer precipitation deviations in the monsoon regions. Principal component indices derived from the leading EOFs (based on ± 0.75 standard deviations) were used to perform composite analyses of circulation and precipitation patterns. The statistical significance of the composite results was assessed using the Student’s t -test. Wave activity flux, as formulated by Takaya and Nakamura ( 2001 ), was also calculated to investigate wave train dynamics. The horizontal components of the wave activity flux in spherical coordinates are expressed as: $$\:W=\frac{\text{pcos}\varphi\:}{2\left|\overline{U}\right|}\left(\begin{array}{c}\frac{\overline{u}}{{a}^{2}{\text{cos}}^{2}\varphi\:}\left[{\left(\frac{\partial\:{\phi\:}^{{\prime\:}}}{\partial\:\lambda\:}\right)}^{2}-{\phi\:}^{{\prime\:}}\frac{{\partial\:}^{2}{\phi\:}^{{\prime\:}}}{\partial\:{\lambda\:}^{2}}\right]+\frac{\overline{v}}{{a}^{2}\text{cos}\varphi\:}\left[\frac{\partial\:{\phi\:}^{{\prime\:}}}{\partial\:\lambda\:}\frac{\partial\:{\phi\:}^{{\prime\:}}}{\partial\:\varphi\:}-{\phi\:}^{{\prime\:}}\frac{{\partial\:}^{2}{\phi\:}^{{\prime\:}}}{\partial\:\lambda\:\partial\:\varphi\:}\right]\\\:\frac{\overline{u}}{{a}^{2}\text{cos}\varphi\:}\left[\frac{\partial\:{\phi\:}^{{\prime\:}}}{\partial\:\lambda\:}\frac{\partial\:{\phi\:}^{{\prime\:}}}{\partial\:\varphi\:}-{\phi\:}^{{\prime\:}}\frac{{\partial\:}^{2}{\phi\:}^{{\prime\:}}}{\partial\:\lambda\:\partial\:\varphi\:}\right]+\frac{\overline{v}}{{a}^{2}}\left[{\left(\frac{\partial\:{\phi\:}^{{\prime\:}}}{\partial\:\varphi\:}\right)}^{2}-{\phi\:}^{{\prime\:}}\frac{{\partial\:}^{2}{\phi\:}^{{\prime\:}}}{{\partial\:}^{2}\varphi\:}\right]\end{array}\right)$$ where 𝜙 is the stream function, (u, v) are horizontal wind components, a is the Earth’s radius, and (λ, Φ) are longitude and latitude, respectively. Overbars represent the climatological mean of the ensemble mean, and primes denote deviations from the MME. 3. Results a. Evaluation of Summer Precipitation Deviations in Monsoon Regions We first examine the inter-model deviations of summer climatological precipitation across various monsoon regions. As shown in Fig. 1 , substantial differences exist in monsoon precipitation simulated by different AMIP6 models. Relative to the multi-model ensemble mean precipitation, Fig. 1 b displays the standard deviation of inter-model differences. The most pronounced inter-model variability in precipitation is concentrated in the tropics, with maximum variability exceeding 4 mm/day. Centers of high variability are located in North Africa, South Asia, the Northwest Pacific, and North America, corresponding to regions of high climatological precipitation (Fig. 1 a). Another notable feature of the inter-model variability is its gradual decline from the South Asia and Northwest Pacific regions toward adjacent monsoon areas. Regardless of whether the variability in a monsoon region is large or small, the standard deviation exceeds 30% of the local climatological precipitation (Fig. 1 c). Additionally, the regions with the most pronounced inter-model variability relative to climatology are the Middle East and Southern Africa. These regions act as critical nodes where source deviations trigger inter-model variability in other regions through atmospheric dynamical processes, as discussed later. Overall, the issue of inter-model precipitation deviations is significant and cannot be overlooked. To investigate the main characteristics of inter-model deviations in monsoon regions, we employ principal component analysis (PCA) to identify the leading modes of precipitation deviations and their corresponding time series for each monsoon region. As shown in Fig. 3 , the spatial patterns of precipitation deviations in most monsoon regions exhibit consistent overestimations or underestimations across the entire region. For example, the leading mode of precipitation deviations in the WNP monsoon region is characterized by consistent overestimations, with the largest deviation center located east of the South China Sea and over the Philippine Sea. This mode explains 68% of the variance, capturing the majority of the WNP summer precipitation deviation information. Surrounding monsoon regions, such as East Asia, Australia, and North Africa, show underestimations in their leading modes, with explained variances ranging from 47–54%. In contrast, the leading mode of precipitation deviations in the South Asian monsoon region exhibits a dipole pattern with opposite deviations between the eastern and western parts. However, this mode only explains 28% of the variance, indicating relatively weak representativeness for the region’s precipitation deviation patterns. The South American monsoon region's leading mode displays consistent deviations across the entire region, with an explained variance of 56%. For other regions, such as the Somali and North American monsoon regions, the key modes of precipitation deviation are the third and second modes, respectively, explaining only ~ 15% of the variance. These modes were selected because they show stronger connections with other monsoon regions. Notably, except for these two regions, the first leading EOF mode of the precipitation deviations in all other monsoon regions passes the 95% significance level of the “North” test. b. Connections Across Monsoon Regions To examine the relationships between precipitation deviations across monsoon regions, we calculated the correlation coefficients between the principal components (PCs) of the leading modes in each region. As summarized in Table 1 , significant correlations were found across most monsoon regions, with the exception of the North American monsoon region. The WNP monsoon region demonstrates the strongest correlations with other regions, including East Asia (0.44), South Asia (0.43), Australia (0.53), and North Africa (0.55), all of which are statistically significant at the 95% confidence level. The WNP also shows notable correlations with Southern Hemisphere monsoon regions, exhibiting a correlation of 0.62 with the third mode of the Somali monsoon region and 0.4 with the first mode of the South American monsoon region, both exceeding the 90% confidence level. In contrast, the WNP’s linkage with the North American monsoon is weak, with a maximum correlation of only ~ 0.2 for the second mode. Interestingly, despite their geographic proximity, the South American and North American monsoon regions share a low correlation of only 0.13. However, the South American monsoon region is significantly correlated with more distant monsoon regions, such as South Asia, Australia, North Africa, and the Somali monsoon region, with correlations ranging from 0.35 to 0.6. These results, as shown in Table 1 , indicate that precipitation deviations across different monsoon regions are not independent. Except for the North American monsoon, all other systems exhibit substantial interconnections. Table 1 Correlation coefficients of the principal components (PCs) of precipitation deviations for each monsoon region (as shown in Fig. 2 i). Bold values with double asterisks in the upper-right corner indicate significance at the 95% confidence level, while those with a single asterisk denote significance at the 90% confidence level. Corr. WNP (PC1) EA (PC1) SA (PC1) AUS (PC1) NAF (PC1) SAF (PC3) SAM (PC1) NAM (PC2) WNP _PC1 1 0.44 ** 0.43 ** 0.53 ** 0.55 ** 0.62 ** 0.4 * 0.18 EA _PC1 1 0.09 0.44 ** 0.49 ** 0.31 -0.09 0.25 SA _PC1 1 0.56 ** 0.19 0.3 0.44 ** -0.36 * AUS _PC1 1 0.56 ** 0.56 ** 0.58 ** 0.13 NAF _PC1 1 0.33 0.51 ** 0.18 SAF _PC3 1 0.35 * 0.27 SAM _PC1 1 0.13 NAM _PC2 1 To further explore these interrelationships, we examined the spatial patterns of global precipitation and circulation deviations. Based on indices derived from the PCs of precipitation deviations (Fig. 2 i), we selected events exceeding ± 0.75 standard deviations and composited global precipitation fields accordingly. As shown in Fig. 3 , although the largest anomalies appear within the respective monsoon domains, the broader spatial patterns associated with different monsoon regions display striking similarity. For example, anomalously high precipitation over the WNP is accompanied by dry anomalies across the tropical Indian Ocean, the Maritime Continent, and East Asia, as well as over North Africa and Central America. This out-of-phase pattern in both zonal and meridional directions suggests a possible bridging influence of the Hadley and Walker circulations. Furthermore, the spatial configuration of precipitation deviations (Fig. 3 ) closely resembles the leading EOF mode of global boreal summer precipitation deviations reported by Wen et al. (2025b), highlighting the robustness of this structure. These consistent large-scale patterns reinforce the notion of coordinated variability among regional monsoon systems, in line with the global monsoon concept proposed in earlier studies (Wang et al., 2014). Corresponding to the precipitation anomalies in Fig. 3 , Fig. 4 presents the composite 200-hPa geopotential height deviations associated with each monsoon region. These upper-level circulation anomalies are primarily confined to the mid- and high-latitudes and exhibit distinct zonal wave train structures. In some cases, the wave trains are limited to one hemisphere, while in others they span both. Notably, the wave patterns in the Northern Hemisphere closely resemble those induced by the WNP monsoon’s northern branch, whereas those in the Southern Hemisphere broadly mirror its southern counterpart. As shown in Fig. 4 a, the circulation anomalies associated with WNP precipitation deviations exhibit characteristics of Rossby wave propagation along the subtropical jets. In the Northern Hemisphere, this manifests as 3–4 large-scale wave fluctuations, with low-pressure centers over the Mediterranean, northern East Asia, and central North America, interspersed with weaker high-pressure anomalies. This structure closely resembles the first leading EOF mode of upper-tropospheric circulation deviations across the Northern Hemisphere (figure not shown). In the Southern Hemisphere, the wave train comprises alternating high- and low-pressure systems extending from western South Africa, across the southern Indian Ocean and northern Australia, through the Ross Sea and the South Pacific, to Argentina. This pattern closely aligns with the second EOF mode of 200hPa geopotential height in the Southern hemisphere. Similar wave train patterns are induced by the South Asian, Australian, and Somali monsoons (Figs. 4 c, 4 d, and 4 f). In contrast, circulation anomalies linked to the East Asian and North African monsoons are predominantly confined to the Northern Hemisphere (Figs. 4 b and 4 e), aligning with the WNP’s northern wave train. While the circulation responses associated with the South and North American monsoons are relatively weaker, they still exhibit Southern Hemisphere features that resemble those of the WNP’s southern branch. Collectively, the upper-level circulation anomalies induced by precipitation deviations in different monsoon regions—whether in the Northern or Southern Hemisphere—show strong resemblance to those associated with the subtropical WNP monsoon region, suggesting a common dynamical pathway mediated by large-scale teleconnections. Overall, these results underscore the strong interconnections among precipitation deviations across global monsoon regions, with the WNP monsoon emerging as a particularly influential center. The large-scale precipitation anomaly patterns associated with different monsoon systems bear remarkable resemblance to those linked to the WNP, and the corresponding upper-tropospheric circulation anomalies similarly echo the wave train structures induced by WNP precipitation deviations. This consistent alignment in both precipitation and circulation fields highlights the WNP monsoon's role as a key dynamical hub, facilitating hemispheric teleconnections and coupling among global monsoon systems. c. Potential causes of these connections The above analysis reveals significant linkages in precipitation deviations across monsoon regions, particularly between the WNP and other monsoon systems. To further assess the relationship between precipitation deviations in the WNP and those in other monsoon regions, we examine the correlation between the first principal component (PC1) of WNP precipitation deviations and the global precipitation deviations (Fig. 5 ). Results show not only strong local and nearby autocorrelations, but also pronounced negative correlations with surrounding monsoon regions—including East Asia, North Africa, and Indonesia—where correlation coefficients range from − 0.6 to − 0.8. Additionally, WNP precipitation deviations exhibit distinct positive and negative correlations with areas in the Southern Hemisphere, notably the southern Indian Ocean and the South Pacific. A significant positive correlation is also observed with the South American monsoon region, reinforcing the strong connections between the WNP and other monsoon systems. Comparison of Fig. 5 with Fig. 2 indicates that the spatial correlation patterns associated with the WNP PC1 align closely with the leading modes of precipitation deviations in each respective region. This consistency supports the interpretation that these leading modes, derived via empirical orthogonal function (EOF) analysis, reflect physically meaningful patterns rather than statistical artifacts. The observed connections imply the underlying dynamical linkages between the WNP and global monsoon systems. To elucidate the relationship between precipitation deviations in the WNP monsoon region and adjacent monsoon systems, we conducted a composite analysis of the Walker and western Pacific Hadley circulations, based on the first principal component (PC1) of WNP precipitation deviations (± 0.75 standard deviations). As shown in Fig. 6 a, strong ascending motion over the tropical western Pacific is accompanied by broad subsidence over the tropical Indian Ocean and a secondary subsidence branch over equatorial Central America. This configuration of the Walker circulation suggests that WNP precipitation deviations serve as a primary driver of the observed circulation perturbations. Through modulation of the Walker circulation, WNP precipitation anomalies exert influences on the South Asian, North African, and North American monsoon regions, thereby contributing to precipitation deviations in those areas. In the meridional direction, WNP precipitation deviations also influence surrounding monsoon regions via the Hadley circulation. As illustrated in Fig. 6 b, anomalous ascent over the western Pacific induces descending motion south of the equator, forming a clockwise meridional circulation cell. In contrast, a broader but weaker counterclockwise cell develops to the north of the WNP. This asymmetry, characterized by a stronger southern branch and a weaker northern branch, reflects the typical northward displacement of the thermal equator during boreal summer. Through this meridional circulation, WNP precipitation deviations are transmitted to the Indonesian and Australian monsoon regions. Additionally, to a lesser extent, the East Asian monsoon is also modulated via Hadley-type circulation associated with WNP precipitation deviations. Beyond adjacent monsoon regions, a critical question concerns the remote teleconnections between precipitation deviations over the WNP and distant monsoon regions, such as those in Somalia and South America. To investigate this, we performed a composite analysis of convergence–divergence wind field and wave activity fluxes, using the WNP precipitation deviation index. As illustrated in Figs. 7 a and 7 b, anomalous precipitation over the WNP generate strong local low-level convergence and upper-level divergence. These circulation anomalies are transmitted via the Walker circulation to the tropical Indian Ocean and equatorial Central America. Notably, Fig. 7 a reveals a broad upper-level convergence center over the tropical Indian Ocean, extending meridionally toward the Mediterranean westerly jet and southeastward into the subtropical jet over the southern Indian Ocean. This large-scale perturbation of the subtropical jet acts as a source of zonally propagating Rossby waves in both hemispheres. As depicted in Fig. 4 a, the resultant wave trains propagate eastward along the Northern and Southern Hemisphere midlatitude jets. The wave activity fluxes in Fig. 7 c clearly demonstrate a northward propagation of energy from the North Africa–Middle East region, followed by eastward dispersion along the midlatitude jet in the Northern Hemisphere. In the Southern Hemisphere, wave energy emanates from the Southern Africa–tropical South Indian Ocean sector, travels southward into the southern Indian Ocean, reaches the Ross Sea, and subsequently curves northward to follow the path of the subtropical jet. The detailed physical mechanisms governing these processes are discussed in a companion paper (Wen et al. 2025b). Furthermore, the regions identified as sources of wave activity fluxes associated with jet stream disturbances in both hemispheres coincide with areas of maximum inter-model precipitation variability relative to climatology—specifically, the Middle East and Southern Africa (Fig. 1 c). This spatial correspondence reinforces the pivotal role of these regions as dynamic hubs in the transmission of the WNP precipitation deviations across the global monsoon system. In brief, precipitation anomalies over the WNP induce mid-to-high-latitude zonal wave trains via disturbances over the tropical Indian Ocean, thereby facilitating the remote transmission of precipitation deviations to other monsoon regions such as Somalia and South America. In addition to above two pathways, Wen et al. (2025b) also noted that monsoon trough disturbances triggered by WNP precipitation anomalies can excite meridional wave trains along the East Asian coast (Fig. 7 d). This meridional wave train may serve as a primary conduit for conveying WNP-related precipitation deviations to the East Asian monsoon region. From above analysis, we conclude that precipitation deviations in the WNP monsoon region serve as a primary driver of the inter-model spread relationships among monsoon regions within AMIP6 simulations. Through the Walker and Hadley circulations, these deviations exert significant influence on precipitation inter-model variability in adjacent monsoon regions, including South Asia, North Africa, and Australia. Moreover, midlatitude zonal wave trains in both hemispheres, excited by perturbations over the tropical Indian Ocean, facilitate the transmission of precipitation deviations of the WNP to remote monsoon regions such as those over Somalia and South America. In addition, disturbances originating from the WNP monsoon trough give rise to meridional wave trains along the East Asian coast, acting as conduits linking WNP precipitation deviations to the East Asian monsoon system. These multiple pathways highlight the central role of the WNP in organizing global monsoon inter-model variability and underscore its importance for accurately simulating precipitation in this region. 4. Conclusion and Discussion This study utilizes historical AMIP6 simulations to systematically investigate the connections and underlying drivers of climatological summer precipitation deviations across major monsoon regions, including the western North Pacific, East Asia, South Asia, Australia, North Africa, Somalia, South America, and North America. The results demonstrate that summer precipitation deviations across these regions are not isolated but exhibit pronounced interconnections, with particularly strong linkages to the western North Pacific (WNP) monsoon system. Further analysis identifies precipitation deviations over the WNP as a key driver of these inter-regional connections. Through atmospheric dynamical pathways—including the Walker and Hadley circulations, meridional wave trains along the East Asian coast, and zonal wave trains propagating through the mid-to-high latitudes of both hemispheres—the precipitation deviations in this region are effectively transmitted to other monsoon regions, thereby shaping global monsoon inter-model variability. Using principal component analysis (PCA), we investigated the dominant characteristics of summer precipitation deviations in various monsoon regions. While the leading modes of precipitation deviations were extracted independently for each region, their corresponding principal components exhibit significant correlations, particularly between the WNP and other monsoon regions, with correlation coefficients ranging from 0.4 to 0.6. To consolidate the connections among the monsoon regions, we conducted a comparative analysis of global precipitation and circulation deviations associated with the principal components of each monsoon region. The results reveal a high degree of similarity in the large-scale precipitation and circulation patterns across regions. In particular, precipitation deviations often display an out-of-phase relationship between the WNP and its surrounding monsoon domains. Upper-level circulation patterns in both hemispheres consistently align with the wave trains induced by precipitation deviations over the WNP. These findings suggest that the co-variability in precipitation deviations among monsoon regions is governed by common large-scale dynamical processes, with the WNP monsoon region serving as a key driver. Through the Walker and Hadley circulations, precipitation deviations in the WNP propagate to adjacent monsoon regions, including South Asia, North Africa, and Australia. Further analysis of convergence-divergence wind fields and wave activity fluxes demonstrates that the zonal wave trains in the subtropical jets of both hemispheres, generated by disturbances over the tropical Indian Ocean, establish teleconnections between WNP precipitation deviations and those over Somalia and South America. Additionally, as detailed in a companion study (Wen et al. 2025b), WNP precipitation anomalies excite disturbances within the monsoon trough that trigger meridional wave trains along the East Asian coast, thereby modulating precipitation deviations in the East Asian monsoon region. In addition, we note that while the WNP precipitation deviations exhibit only a weak linkage to NAM precipitation via the Walker circulation, the inter-model variability of precipitation over the NAM displays substantial independence. As shown in Supplementary Fig. 1a, the leading EOF mode of summer precipitation deviations over the NAM is characterized by a northwestward-oriented band of reduced rainfall across the monsoon domain. The associated global precipitation deviations (SFig. 1c) bear some resemblance to the large-scale pattern depicted in Fig. 3 a, but with notably amplified local anomalies over the NAM region. The suppressed diabatic heating over the NAM region gives rise to a circulation response that differs substantially from that shown in Fig. 4 a. As illustrated in SFig. 1d, the tropical response features a prominent low pressure belt and a pair of anomalous centers over the equatorial eastern Pacific, resembling the canonical Matsuno–Gill pattern induced by tropical heating (Matsuno, 1966 ; Gill, 1980 ). In the midlatitudes of both hemispheres, high-pressure anomalous centers emerge, nearly out of phase with those associated with the WNP precipitation anomalies (Fig. 4 a). This circulation response may be influenced by the local convergence–divergence wind over Central America and jet stream fluctuations triggered by upstream disturbances (SFig. 2). As shown in SFig. 2c, the primary sources of energy emission are located over the west coasts of North and South America respectively, corresponding to the secondary centers of inter-model precipitation variability relative to climatology in Fig. 1 c. Via atmospheric teleconnections, the precipitation deviations over the NAM region are also linked to inter-model variability in other monsoon systems. As indicated in Supplementary Table 2, the leading model of NAM inter-model variability is significantly correlated with the second EOF mode in most monsoon regions such as the WNP, EA, SA, NAF, SAM, with absolute correlation coefficients ranging from 0.35 to 0.51, all significant at the 90% confidence level. These results suggests that precipitation variability over the NAM play a secondary yet important role in the co-variability of precipitation deviations across global monsoon regions. Overall, the circulation responses to precipitation deviations over the WNP and NAM exhibit a competitive relationship, with the prevailing influence determined by their relative intensities. The WNP, characterized by a strong inter-model variability, shows robust connections with other monsoon systems, primarily through associations with the leading EOF mode of precipitation deviations in those regions. In contrast, the magnitude of inter-model variability over the NAM (Fig. 1 b), along with its corresponding atmospheric response (SFig. 1d), is relatively weak, resulting in secondary linkages with other monsoon regions. Beyond the WNP and NAM, additional interregional connections may arise as indicated by the notable linkage between PC1 of Somali monsoon precipitation anomalies and PC2 over the South American monsoon region. This study highlights the critical role of precipitation deviations over the WNP in driving inter-model variability within the global monsoon system. This finding aligns with previous research (Bony et al., 2015 ; Ren et al., 2023 ), which indicates that in the WNP region, simulation biases – stemming from differences in model parameterization schemes -- can be amplified due to the atmosphere's heightened sensitivity to underlying warm pool sea surface temperature (SST) anomalies. As such, the fidelity of simulated precipitation over the WNP emerges as a key determinant of simulation accuracy in other monsoon regions. This highlights the critical need to enhance model performance in this region to improve the overall reliability of climate simulations. It is also worth noting that the data used in this study are derived from AMIP6 historical simulations, which prescribe observed sea surface temperatures (SSTs). In these simulations, precipitation deviations over the WNP are primarily attributed to factors such as the parameterization of convective processes and model resolution. However, in coupled atmosphere–ocean systems, biases in simulated SSTs can also influence precipitation deviations in monsoon regions through air–sea interactions. Accordingly, further investigation is required to assess whether the relationships among precipitation deviations in monsoon regions are altered under coupled model configurations. Declarations Acknowledgement This work is supported by National Key R&D Program of China (2020YFA0608901). Data Availability Statement The model data used in this study is openly available at the website (https://aims2.llnl.gov/search/cmip6/). Conflict of Interest The authors declare that they have no conflict of interest. References Bony, S., Stevens, B., Frierson, D. M. W., et al. (2015). Clouds, circulation and climate sensitivity [J]. Nature Geoscience, 8(4): 261−268. https://doi:10.1038/ngeo2398 Bowler N E, Arribas A, Mylne K R, et al. (2008). The MOGREPS short-range ensemble predictio system. Quart J Roy Meteor Soc, 134, 703-722. https://doi.org/10.1002/qj.234 Chen Jing, Zhu Yuejian, Duan Wansuo, Zhi Xiefei, Min Jinzhong, Li Xiaoli, Deng Guo, Yuan Huiling, Feng Jie, Du Jun, Li Qiaoping, Gong Jiandong, Shen Xueshun, MuMu. (2025). A review on development, challenges and future perspectives of ensemble forecast. Acta Meteorologica Sinica, 83(3): 1-23. https://doi.org/10.11676/qxxb2025.20240151. Deser, C., Phillips, A., Bourdette, V., & Teng, H. (2012). Uncertainty in climate change projections: The role of internal variability. Climate Dynamics, 38(3), 527–546. https://doi.org/10.1007/s00382-010-0977-x Gill, A. E. (1980). Some simple solutions for heat-induced tropical circulation[J]. Quart J Roy Meteor Soc, 106(449): 447-462. Hagos, S., Leung, L. R., Ashfaq, M., & Balaguru, K. (2019). South Asian monsoon precipitation in CMIP5: A link between inter-model spread and the representations of tropical convection. Climate Dynamics, 52(2), 1049–1061. https://doi.org/10.1007/s00382-018-4177-4 Hannachi, A., Jolliffe, I. T., & Stephenson, D. B. (2007). Empirical orthogonal functions and related techniques in atmospheric science: A review. International Journal of Climatology, 27(9), 1119–1152. https://doi.org/10.1002/joc.1499 Huang, D., Liu, A., Zheng, Y., & Zhu, J. (2022). Inter-model spread of the simulated East Asian summer monsoon rainfall and the associated atmospheric circulations from the CMIP6 models. Journal of Geophysical Research: Atmospheres, 127, e2022JD037371. https://doi.org/10.1029/2022JD037371 Krishnamurti T N. (1999). Improved weather and seasonal climate forecasts from multimodel superensemble. Science, 285(5433): 1548-1550. https://doi.org/10.1126/science.285.5433.1548 Lehner F., Deser, C., Maher, N., et al. (2020). Partitioning climate projection uncertainty with multiple large ensembles and CMIP5/6, Earth Syst. Dynam., 11, 491–508, https://doi.org/10.5194/esd-11-491-2020 Lenartz F. Mourre B. Barth A. et al. (2010). Enhanced ocean temperature forecast skills through 3-D super-ensemble multi-model fusion. Geophys Res Lett, 37, L19606. https://doi.org/10.1029/2010GL044591 Lorenz, E. N. (1963). Deterministic Nonperiodic Flow, J. At-mos. Sci., 20, 130–141. https://doi.org/10.1175/1520-0469(1963)0202.0.co;2. Maher, N., Matei, D., Milinski, S., and Marotzke, J. (2018). ENSO Change in Climate Projections: Forced Response or Internal Variability?, Geophys. Res. Lett., 45, 11390-11398. https://doi.org/10.1029/2018GL079764 Matsuno, T. (1966). Quasi-geostrophic motions in the equatorial area. Journal of the Meteorological Society of Japan, 44(1), 25-43. https://www.jstage.jst.go.jp/article/jmsj1965/44/1/44_1 25 Molteni F. Buizza R, Palmer T N, et al. (1996). The ECMWF Ensemble Prediction System: Methodology and validation. Quart J Roy Meteor Soc, 122: 73-119. https://doi.org/10.1002/qj.49712252905 Monerie, P. A., Wainwright, C. M., Sidibe, M., et al. (2020). Model uncertainties in climate change impacts on Sahel precipitation in ensembles of CMIP5 and CMIP6 simulations. Climate Dynamics, 55(5), 1385–1401. https://doi.org/10.1007/s00382-020-05332-0 Murphy, J. M., Sexton, D. M. H., Barnett, D. H., Jones,G. S., Webb, M. J., Collins, M., and Stainforth, D. A. (2004). Quantification of modelling uncertainties in a large ensemble of climate change simulations, Nature, 430, 768–772. https://doi.org/10.1038/nature02771 Qu, T., Du, Y., Strachan, J., Meyers, G., Slingo, J. (2005). Sea surface temperature and its variability in the Indonesian region. Oceanography 18, 50–61. https://doi.org/10.1029/2008JC005150 Sanderson, B. M., Piani, C., Ingram, W. J., Stone, D. A., and Allen, M. R. (2008). Towards constraining climate sensitivity by linear analysis of feedback patterns in thousands of perturbed-physics GCM simulations, Clim. Dynam., 30, 175–190. https://doi.org/10.1007/s00382-007-0280-7. Ren Z, Zhou T, Guo Z, et al. (2023). Relationship between Vertical Convection Structure and Precipitation Simulation Bias in the Tropical Atmosphere: An Analysis Based on GAMIL3 Model [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 47(2): 239−258. doi:10.3878/j.issn.1006-9895.2109.21098 Takaya, K., & Nakamura, H. (2001). A formulation of a phase-independent wave-activity flux for stationary and migratory quasi-geostrophic eddies on a zonally varying basic flow. Journal of the Atmospheric Sciences, 58(6), 608–627. https://doi.org/10.1175/1520-0469(2001)0582.0.CO;2 Trenberth, K.E., Stepaniak, D.P., Caron, J.M. (2000). The global monsoon as seen through the divergent atmospheric circulation. J. Clim. 13, 3969–3993. https://doi.org/10.1175/1520-0442(2000)0132.0.CO;2 Wang, B., Ding, Q. (2008). Global monsoon: dominant mode of annual variation in the tropics. Dyn. Atmos. Oceans 44, 165–183. Wang, P. X., Wang, B., Cheng, H., et al. (2017). The global monsoon across time scales: Mechanisms and outstanding issues. Earth-Science Reviews, 174, 84–121. https://doi.org/10.1016/j.earscirev.2017.07.006 Wang, P.X., Wang, B., Cheng, H., Fasullo, J., Guo, Z., Kiefer, T., Liu, Z. (2014b). The global monsoon across time scales: coherent variability of regional monsoons. Clim. Past 10, 1–46. http://dx.doi.org/10.5194/cp-10-1-2014. Wen, N. and Huan, K., (2025b), AMIP6 Inter-model Spread analysis: Effects of precipitation deviation on circulation and its underlying mechanisms, preparing Xu Z Z, Chen J, Mu M, et al. (2022). A Nonlinear Representation of Model Uncertainty in a Convective-Scale Ensemble Prediction System. Adv Atmos Sci 39, 1432-1450. https://doi.org/10.1007/s00376-022-1341-x Zhi X F. Qi H X. Bai Y Q. et al. (2012). A comparison of three kinds of multimodel ensemble forecast techniques based on the TIGGE data. Acta Meteorol Sin, 26(1) : 41-51. https://doi.org/10.1007/s13351-012-0104-5 Zhou, T., Lu, J., Zhang, W., & Chen, Z. (2020). The sources of uncertainty in the projection of global land monsoon precipitation. Geophysical Research Letters, 47, e2020GL088415. https://doi.org/10.1029/2020GL088415 Zhu S P, Zhi X F, Ge F, et al. (2021). Subseasonal forecast of surface air temperature using superensemble approaches: Experiments over Northeast Asia for 2018. Wea Forecasting, 36(1): 39-51. https://doi.org/10.1175/WAF-D-20-0096.1 Supplementary Files Precintermaolspreadpart1Sup.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6711488","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":470136614,"identity":"1f2ea73a-2d5c-47a0-a7d8-24c9d8fafc3e","order_by":0,"name":"Na Wen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAqUlEQVRIiWNgGAWjYLACxgYbHn7+BtK0pMlIzjhAmpbDNgYNCUSqNjjee+zBzx3neQwYDjB++JhDjJYz59INe8/c5jFnbmCWnLmNCC1mN3LMJHjbbvNYNhxgY+YlSsv9N2aSf9vO8RgcSCBWyw0eM2netgMkaLE/k2MmLduWzCM542AzcX6RbD9jJvm2zc6en7/54IePxGhBAowNpKkfBaNgFIyCUYAbAADBEDW5CEgP1wAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-4262-1326","institution":"Nanjing University of information science \u0026 technology","correspondingAuthor":true,"prefix":"","firstName":"Na","middleName":"","lastName":"Wen","suffix":""},{"id":470136615,"identity":"2307d75f-1c72-48df-b6a4-6396d8ca71e0","order_by":1,"name":"Kai Huan","email":"","orcid":"","institution":"Nanjing University of information science \u0026 technology","correspondingAuthor":false,"prefix":"","firstName":"Kai","middleName":"","lastName":"Huan","suffix":""}],"badges":[],"createdAt":"2025-05-21 01:31:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6711488/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6711488/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":84563813,"identity":"bfe7ed48-87d2-4dcb-ad1b-b922bd766f03","added_by":"auto","created_at":"2025-06-13 13:50:52","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1339245,"visible":true,"origin":"","legend":"\u003cp\u003e(a) AMIP6 ensemble mean climatological summer precipitation. The contour interval (CI) is 2, with units in mm/day. (b) Standard deviation of precipitation inter-model variability in AMIP6 simulations (CI = 0.6, units: mm/day), relative to the MME results. (c) Ratio of the standard deviation of AMIP6 precipitation inter-model variability to the ensemble mean of climatological precipitation (CI = 0.3). In Fig. b, the red solid box denotes the western North Pacific monsoon region (WNP), while the blue dashed boxes represent the East Asian (EA), South Asian (SA), Australian (AUS), North African (NAF), Somali (SAF), South American (SAM), and North American (NAM) monsoon regions.\u003c/p\u003e","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/4eb8c20646e570baf7285a83.png"},{"id":84564804,"identity":"3d526633-fac4-40c8-b4eb-2014ab6ba620","added_by":"auto","created_at":"2025-06-13 13:58:52","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1160448,"visible":true,"origin":"","legend":"\u003cp\u003eLeading modes of summer climatological precipitation deviations and their corresponding principal components (PCs) for each monsoon region. Panels (a)–(h) depict the leading modes for the western North Pacific, East Asian, South Asian, Australian, North African, Somali, South American, and North American monsoon regions, respectively. Except for the Somali and North American monsoon regions, which show the third and second modes, respectively, all other regions present the first mode. Panel (i) shows the PCs corresponding to the leading modes for each monsoon region, with black dashed lines indicating ±0.75 standard deviations. The title of each panel includes the monsoon region name, the mode number, and the explained variance.\u003c/p\u003e","description":"","filename":"fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/06bf430a01a57d6856844be9.png"},{"id":84563815,"identity":"447ce64f-b8ee-4962-bbf0-e92e312d8494","added_by":"auto","created_at":"2025-06-13 13:50:52","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2908259,"visible":true,"origin":"","legend":"\u003cp\u003eComposite maps of global precipitation deviations (CI=0.5, units: mm/day) corresponding to the leading modes of precipitation inter-model variability for each monsoon region (as indicated in the figure captions). Stippled areas denote regions significant at the 90% confidence level.\u003c/p\u003e","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/8373e69a128e5f9795f159dd.png"},{"id":84565216,"identity":"1da0896c-6f06-4a77-a82c-dcc259cca36d","added_by":"auto","created_at":"2025-06-13 14:06:52","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":5069106,"visible":true,"origin":"","legend":"\u003cp\u003eComposite maps of global 200-hPa geopotential height deviations (CI = 10, units: gpm) corresponding to the leading modes of precipitation inter-model variability for each monsoon region (as indicated in the figure captions). Black solid contours represent regions significant at the 90% confidence level.\u003c/p\u003e","description":"","filename":"fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/d6ce8eaf12c75ed88533c7a8.png"},{"id":84563818,"identity":"ddffaae6-7f73-4ac0-9338-9582c1c8420a","added_by":"auto","created_at":"2025-06-13 13:50:52","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":847709,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation between the first principal component (PC1) of precipitation inter-model variability in the WNP monsoon region and global precipitation deviations. The red solid box indicates the WNP monsoon region, while the blue dashed boxes denote the East Asian, South Asian, Australian, North African, Somali, South American, and North American monsoon regions. Black solid contours represent correlations significant at the 90% confidence level.\u003c/p\u003e","description":"","filename":"fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/f7d770e63ba7eb9b9ee220be.png"},{"id":84563819,"identity":"c34bacde-cbb8-4952-8dbb-73e8e704f41b","added_by":"auto","created_at":"2025-06-13 13:50:52","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":673505,"visible":true,"origin":"","legend":"\u003cp\u003eComposite maps of (a) the meridionally averaged Walker circulation (10°N–20°N) and (b) the zonally averaged western Pacific Hadley circulation (110°E–150°E), based on the first principal component (PC1) index of precipitation deviations in the WNP monsoon region. Vertical velocity (-ω) is multiplied by 100 for clarity with units in m/s. Red and blue shading indicates ascending and descending motions significant at the 90% confidence level, respectively.\u003c/p\u003e","description":"","filename":"fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/bc16938b95d4a5b2130363cb.png"},{"id":84563821,"identity":"76e19132-36df-42a2-96ff-c061bf746c60","added_by":"auto","created_at":"2025-06-13 13:50:52","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":4280994,"visible":true,"origin":"","legend":"\u003cp\u003eComposite maps based on the first principal component (PC1) index of precipitation deviations in the WNP monsoon region. Panels (a) and (b) show the velocity potential (shaded, units: 10\u003csup\u003e6\u003c/sup\u003e m²/s) and divergent wind (vectors, units: m/s) at 200 hPa and 850 hPa, respectively. Panels (c) and (d) display wave activity fluxes (vectors, units: m²/s²) and vorticity (shaded, units: 10\u003csup\u003e−6\u003c/sup\u003e/s) at 200 hPa and 850 hPa, respectively. Brown contours represent the climatological zonal wind, with contour intervals of 10 m/s at 200 hPa and 5 m/s at 850 hPa.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e","description":"","filename":"fig7.png","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/e70332dd022029ed8905f843.png"},{"id":88202494,"identity":"aa327e87-a975-48d3-b388-2500a6c93107","added_by":"auto","created_at":"2025-08-04 01:07:13","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":12790142,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/1163543f-60de-41e4-b8a0-3d60f3c9a940.pdf"},{"id":84563831,"identity":"71d34479-93b8-4ee6-80ed-3cde709bdb1c","added_by":"auto","created_at":"2025-06-13 13:50:52","extension":"docx","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":732589,"visible":true,"origin":"","legend":"","description":"","filename":"Precintermaolspreadpart1Sup.docx","url":"https://assets-eu.researchsquare.com/files/rs-6711488/v1/2682e3c2e94bf1e506ce96bf.docx"}],"financialInterests":"","formattedTitle":"Inter-model Spread in Summer Precipitation: Connections and Causes Across Monsoon Regions in AMIP6 Simulations","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eMonsoon activity is a critical driver of regional water resources, affecting nearly two-thirds of the global population. Accurate prediction of monsoon precipitation is therefore essential for effective water resource management, food security, disaster mitigation, and infrastructure planning. A widely used approach for such prediction is the multi-model ensemble mean (MME) method (Molteni et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; Krishnamurti, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Chen et al., 2025), which reduces prediction uncertainty by averaging outputs from multiple climate models. The MME has been extensively adopted as an effective strategy for climate forecasting. However, substantial challenges remain in realistically simulating monsoon precipitation, particularly with regard to model performance and uncertainty (Lenartz et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Zhi et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Zhu et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). To enhance forecast skill, it requires a better understanding of the sources of uncertainty in monsoon predictions.\u003c/p\u003e \u003cp\u003eBeyond the inherent complexity of the monsoon system, the uncertainty in monsoon predictions is further compounded by internal variability, model uncertainty, and scenario uncertainty that is associated with future climate projections (Lehner et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Uncertainty due to internal variability originates from the chaotic character of the climate system (Lorenz \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e1963\u003c/span\u003e), which imposes a fundamental limit on climate predictability. Internal variability in a climate model can be effectively estimated using long-time mean or large initial-condition ensembles (Maher et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Model uncertainty stems from structural differences across models, variations in parameterization schemes, and the differing responses of models to external forcing (Murphy et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Bowler et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Sanderson et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Xu et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These uncertainties reflect the diverse assumptions and design choices made during model development and parameter tuning. In practice, model uncertainty can be reduced through continuous advancements in model development. Under global warming, scenario uncertainty arises from the unknown trajectory of future greenhouse gas emissions. Zhou et al. (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) utilized both Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations and single-model initial-condition large ensembles (SMILEs) to systematically assess the relative contributions of these uncertainties to the ensemble spread of projected global land monsoon precipitation. They found that for mean precipitation, model uncertainty is the dominant contributor\u0026mdash;accounting for approximately 90% of total uncertainty\u0026mdash;while the influence of internal variability decreases over time and scenario uncertainty becomes more significant. Their results emphasize the critical role of model uncertainty in determining the predictability of monsoon climates.\u003c/p\u003e \u003cp\u003eTo further investigate the monsoon precipitation uncertainty, several studies have focused on specific regional monsoon systems. For instance, Huang et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) found pronounced inter-model variability in East Asian summer precipitation within CMIP6 simulations, with roughly one-third of the models overestimating and two-thirds underestimating precipitation. These deviations were closely associated with discrepancies in the simulated western North Pacific anticyclone and the positioning of the subtropical westerly jet. Similarly, Hagos et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) reported that CMIP5 models exhibit a dry bias and substantial inter-model spread in simulating South Asian monsoon precipitation. This was largely attributed to differences in the representation of convection over the equatorial Indian Ocean during the pre-monsoon season. Their analysis suggested that model-to-model variations in precipitable water in this region perturb the large-scale circulation, thereby amplifying both the dry bias and the spread over the Indian subcontinent. These findings highlight the critical role of improving convective parameterizations over the equatorial Indian Ocean to enhance the fidelity of South Asian monsoon simulations. In contrast, Monerie et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) examined uncertainties in projected precipitation changes over the Sahel using outputs from 29 CMIP5 and 11 CMIP6 models. They found that future Sahel rainfall projections are subject to large inter-model uncertainties, primarily stemming from differences in the simulated atmospheric circulation response to global warming (i.e., dynamical changes), whereas thermodynamic contributions to uncertainty were relatively modest.\u003c/p\u003e \u003cp\u003ePrevious studies examining inter-model spread in monsoon regions have predominantly focused on regional scales, emphasizing precipitation deviations within individual monsoon systems. Although regional monsoons exhibit distinct characteristics shaped by geography and topography, growing evidence suggests that their variability is not entirely independent. Trenberth et al. (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2000\u003c/span\u003e) introduced the concept of global monsoon, which arises from and reflects the coupled atmosphere\u0026ndash;land\u0026ndash;ocean system\u0026rsquo;s response to annual cycle of solar forcing. Wang and Ding (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) further refined this concept, identifying the global monsoon as the dominant mode of annual variability in the tropics. Building on this foundation, Wang and Wang (2014) synthesized earlier studies and emphasized the coherent variability among regional monsoons across a range of timescales from interannual to tectonic, thus substantiating the global monsoon framework at multiple temporal scales. Expanding on this work, Wang et al. (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) concluded that the co-variability of global monsoon on long time scales is primarily driven by external forcing, whereas on shorter timescales it is predominantly influenced by internal feedbacks within the climate system. For example, with regard to internal mechanisms underlying monsoon variability, the Indo-Pacific or Western Pacific Warm Pool is recognized as the heat engine of the global climate system. Its extensive exchange of heat and moisture exerts a significant influence on climate patterns worldwide. Even subtle variations in sea surface temperature (SST) within this region can alter the location and intensity of convection in the ascending branches of the Hadley and Walker circulations, disrupt planetary-scale atmospheric circulation patterns, and modify tropical hydrology (Qu et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). Within this context, a critical question arises: Does a similar coherence exist in the uncertainties of simulated monsoon precipitation across regions? If so, what are the underlying mechanisms responsible for such coordinated uncertainty?\u003c/p\u003e \u003cp\u003eThis study, therefore, focuses on examining the connections of the inter-model variability in precipitation across different monsoon regions, as well as the underlying causes driving the co-variability, using historical simulations from AMIP6. The paper is organized as follows: Section \u003cspan refid=\"Sec2\" class=\"InternalRef\"\u003e2\u003c/span\u003e describes the data and methodology. Section \u003cspan refid=\"Sec3\" class=\"InternalRef\"\u003e3\u003c/span\u003e evaluates the performance of simulated precipitation in individual monsoon regions, investigates the inter-regional linkages in inter-model variability, and explores the causes underlying these connections. Section \u003cspan refid=\"Sec4\" class=\"InternalRef\"\u003e4\u003c/span\u003e offers a summary and discussion.\u003c/p\u003e"},{"header":"2. Data and Methods","content":"\u003cp\u003eThis study utilizes historical simulations from 23 AMIP6 models spanning 1979\u0026ndash;2014 to evaluate summer precipitation inter-model variability across different monsoon regions. The models, sourced from various institutions (listed in Appendix 1), were selected to ensure relative independence. To reduce the influence of internal atmospheric variability, ensemble means were calculated for all realization members of each model (Deser et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). For consistency, all model outputs were interpolated to a horizontal resolution of 2.5\u0026deg; \u0026times; 2.5\u0026deg;. Summer climatological precipitation refers to the multi-year average for June, July, and August (JJA). Inter-model spread is defined as the deviation of each model\u0026rsquo;s summer climatological precipitation from the multi-model ensemble mean (MME). The MME was calculated as the unweighted average of all models.\u003c/p\u003e \u003cp\u003eFollowing the global monsoon region classification by Wang et al. (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), where local summer-minus-winter precipitation exceeds 300 mm and local summer precipitation exceeds 55% of the annual total, eight monsoon regions were selected for analysis: Western North Pacific (WNP): 10\u0026deg;N\u0026ndash;20\u0026deg;N, 110\u0026deg;E\u0026ndash;150\u0026deg;E; East Asia (EA): 20\u0026deg;N\u0026ndash;45\u0026deg;N, 110\u0026deg;E\u0026ndash;135\u0026deg;E; South Asia (SA): 10\u0026deg;N\u0026ndash;33\u0026deg;N, 70\u0026deg;E\u0026ndash;105\u0026deg;E; Australia (AUS): 5\u0026deg;S\u0026ndash;20\u0026deg;S, 110\u0026deg;E\u0026ndash;150\u0026deg;E; North Africa (NAF): 5\u0026deg;N\u0026ndash;15\u0026deg;N, 30\u0026deg;W\u0026ndash;30\u0026deg;E; Somalia (SAF): 7\u0026deg;S\u0026ndash;25\u0026deg;S, 25\u0026deg;E\u0026ndash;70\u0026deg;E; South America (SAM): 5\u0026deg;S\u0026ndash;25\u0026deg;S, 70\u0026deg;W\u0026ndash;40\u0026deg;W; and North America (NAM): 5\u0026deg;N\u0026ndash;23\u0026deg;N, 110\u0026deg;W\u0026ndash;80\u0026deg;W. To facilitate comparison of precipitation deviations across different monsoon regions, deviations for each model were normalized by the MME climatological precipitation.\u003c/p\u003e \u003cp\u003eEmpirical Orthogonal Function (EOF) analysis (Hannachi et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) was used to extract the leading modes of summer precipitation deviations in the monsoon regions. Principal component indices derived from the leading EOFs (based on \u0026plusmn;\u0026thinsp;0.75 standard deviations) were used to perform composite analyses of circulation and precipitation patterns. The statistical significance of the composite results was assessed using the Student\u0026rsquo;s \u003cem\u003et\u003c/em\u003e-test. Wave activity flux, as formulated by Takaya and Nakamura (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), was also calculated to investigate wave train dynamics. The horizontal components of the wave activity flux in spherical coordinates are expressed as:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:W=\\frac{\\text{pcos}\\varphi\\:}{2\\left|\\overline{U}\\right|}\\left(\\begin{array}{c}\\frac{\\overline{u}}{{a}^{2}{\\text{cos}}^{2}\\varphi\\:}\\left[{\\left(\\frac{\\partial\\:{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:\\lambda\\:}\\right)}^{2}-{\\phi\\:}^{{\\prime\\:}}\\frac{{\\partial\\:}^{2}{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:{\\lambda\\:}^{2}}\\right]+\\frac{\\overline{v}}{{a}^{2}\\text{cos}\\varphi\\:}\\left[\\frac{\\partial\\:{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:\\lambda\\:}\\frac{\\partial\\:{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:\\varphi\\:}-{\\phi\\:}^{{\\prime\\:}}\\frac{{\\partial\\:}^{2}{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:\\lambda\\:\\partial\\:\\varphi\\:}\\right]\\\\\\:\\frac{\\overline{u}}{{a}^{2}\\text{cos}\\varphi\\:}\\left[\\frac{\\partial\\:{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:\\lambda\\:}\\frac{\\partial\\:{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:\\varphi\\:}-{\\phi\\:}^{{\\prime\\:}}\\frac{{\\partial\\:}^{2}{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:\\lambda\\:\\partial\\:\\varphi\\:}\\right]+\\frac{\\overline{v}}{{a}^{2}}\\left[{\\left(\\frac{\\partial\\:{\\phi\\:}^{{\\prime\\:}}}{\\partial\\:\\varphi\\:}\\right)}^{2}-{\\phi\\:}^{{\\prime\\:}}\\frac{{\\partial\\:}^{2}{\\phi\\:}^{{\\prime\\:}}}{{\\partial\\:}^{2}\\varphi\\:}\\right]\\end{array}\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u0026#120601; is the stream function, (u, v) are horizontal wind components, a is the Earth\u0026rsquo;s radius, and (λ, Φ) are longitude and latitude, respectively. Overbars represent the climatological mean of the ensemble mean, and primes denote deviations from the MME.\u003c/p\u003e"},{"header":"3. Results","content":"\u003cp\u003e \u003cb\u003ea. Evaluation of Summer Precipitation Deviations in Monsoon Regions\u003c/b\u003e \u003c/p\u003e \u003cp\u003eWe first examine the inter-model deviations of summer climatological precipitation across various monsoon regions. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, substantial differences exist in monsoon precipitation simulated by different AMIP6 models. Relative to the multi-model ensemble mean precipitation, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb displays the standard deviation of inter-model differences. The most pronounced inter-model variability in precipitation is concentrated in the tropics, with maximum variability exceeding 4 mm/day. Centers of high variability are located in North Africa, South Asia, the Northwest Pacific, and North America, corresponding to regions of high climatological precipitation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). Another notable feature of the inter-model variability is its gradual decline from the South Asia and Northwest Pacific regions toward adjacent monsoon areas. Regardless of whether the variability in a monsoon region is large or small, the standard deviation exceeds 30% of the local climatological precipitation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec). Additionally, the regions with the most pronounced inter-model variability relative to climatology are the Middle East and Southern Africa. These regions act as critical nodes where source deviations trigger inter-model variability in other regions through atmospheric dynamical processes, as discussed later. Overall, the issue of inter-model precipitation deviations is significant and cannot be overlooked.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo investigate the main characteristics of inter-model deviations in monsoon regions, we employ principal component analysis (PCA) to identify the leading modes of precipitation deviations and their corresponding time series for each monsoon region. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, the spatial patterns of precipitation deviations in most monsoon regions exhibit consistent overestimations or underestimations across the entire region. For example, the leading mode of precipitation deviations in the WNP monsoon region is characterized by consistent overestimations, with the largest deviation center located east of the South China Sea and over the Philippine Sea. This mode explains 68% of the variance, capturing the majority of the WNP summer precipitation deviation information. Surrounding monsoon regions, such as East Asia, Australia, and North Africa, show underestimations in their leading modes, with explained variances ranging from 47\u0026ndash;54%. In contrast, the leading mode of precipitation deviations in the South Asian monsoon region exhibits a dipole pattern with opposite deviations between the eastern and western parts. However, this mode only explains 28% of the variance, indicating relatively weak representativeness for the region\u0026rsquo;s precipitation deviation patterns. The South American monsoon region's leading mode displays consistent deviations across the entire region, with an explained variance of 56%. For other regions, such as the Somali and North American monsoon regions, the key modes of precipitation deviation are the third and second modes, respectively, explaining only\u0026thinsp;~\u0026thinsp;15% of the variance. These modes were selected because they show stronger connections with other monsoon regions. Notably, except for these two regions, the first leading EOF mode of the precipitation deviations in all other monsoon regions passes the 95% significance level of the \u0026ldquo;North\u0026rdquo; test.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eb. Connections Across Monsoon Regions\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo examine the relationships between precipitation deviations across monsoon regions, we calculated the correlation coefficients between the principal components (PCs) of the leading modes in each region. As summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, significant correlations were found across most monsoon regions, with the exception of the North American monsoon region. The WNP monsoon region demonstrates the strongest correlations with other regions, including East Asia (0.44), South Asia (0.43), Australia (0.53), and North Africa (0.55), all of which are statistically significant at the 95% confidence level. The WNP also shows notable correlations with Southern Hemisphere monsoon regions, exhibiting a correlation of 0.62 with the third mode of the Somali monsoon region and 0.4 with the first mode of the South American monsoon region, both exceeding the 90% confidence level. In contrast, the WNP\u0026rsquo;s linkage with the North American monsoon is weak, with a maximum correlation of only\u0026thinsp;~\u0026thinsp;0.2 for the second mode. Interestingly, despite their geographic proximity, the South American and North American monsoon regions share a low correlation of only 0.13. However, the South American monsoon region is significantly correlated with more distant monsoon regions, such as South Asia, Australia, North Africa, and the Somali monsoon region, with correlations ranging from 0.35 to 0.6. These results, as shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, indicate that precipitation deviations across different monsoon regions are not independent. Except for the North American monsoon, all other systems exhibit substantial interconnections.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCorrelation coefficients of the principal components (PCs) of precipitation deviations for each monsoon region (as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ei). Bold values with double asterisks in the upper-right corner indicate significance at the 95% confidence level, while those with a single asterisk denote significance at the 90% confidence level.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCorr.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWNP\u003c/p\u003e \u003cp\u003e(PC1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEA\u003c/p\u003e \u003cp\u003e(PC1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSA\u003c/p\u003e \u003cp\u003e(PC1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAUS\u003c/p\u003e \u003cp\u003e(PC1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNAF\u003c/p\u003e \u003cp\u003e(PC1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSAF\u003c/p\u003e \u003cp\u003e(PC3)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eSAM\u003c/p\u003e \u003cp\u003e(PC1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNAM\u003c/p\u003e \u003cp\u003e(PC2)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eWNP\u003c/b\u003e_PC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.44\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.43\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.53\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e0.55\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003e0.62\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003e0.4\u003c/b\u003e\u003csup\u003e\u003cb\u003e*\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.18\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eEA\u003c/b\u003e_PC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.44\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e0.49\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSA\u003c/b\u003e_PC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.56\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003e0.44\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e\u003cb\u003e-0.36\u003c/b\u003e\u003csup\u003e\u003cb\u003e*\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAUS\u003c/b\u003e_PC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e0.56\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003e0.56\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003e0.58\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNAF\u003c/b\u003e_PC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003e0.51\u003c/b\u003e\u003csup\u003e\u003cb\u003e**\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.18\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSAF\u003c/b\u003e_PC3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003e0.35\u003c/b\u003e\u003csup\u003e\u003cb\u003e*\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.27\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSAM\u003c/b\u003e_PC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNAM\u003c/b\u003e_PC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTo further explore these interrelationships, we examined the spatial patterns of global precipitation and circulation deviations. Based on indices derived from the PCs of precipitation deviations (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ei), we selected events exceeding\u0026thinsp;\u0026plusmn;\u0026thinsp;0.75 standard deviations and composited global precipitation fields accordingly. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, although the largest anomalies appear within the respective monsoon domains, the broader spatial patterns associated with different monsoon regions display striking similarity. For example, anomalously high precipitation over the WNP is accompanied by dry anomalies across the tropical Indian Ocean, the Maritime Continent, and East Asia, as well as over North Africa and Central America. This out-of-phase pattern in both zonal and meridional directions suggests a possible bridging influence of the Hadley and Walker circulations. Furthermore, the spatial configuration of precipitation deviations (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) closely resembles the leading EOF mode of global boreal summer precipitation deviations reported by Wen et al. (2025b), highlighting the robustness of this structure. These consistent large-scale patterns reinforce the notion of coordinated variability among regional monsoon systems, in line with the global monsoon concept proposed in earlier studies (Wang et al., 2014).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eCorresponding to the precipitation anomalies in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e presents the composite 200-hPa geopotential height deviations associated with each monsoon region. These upper-level circulation anomalies are primarily confined to the mid- and high-latitudes and exhibit distinct zonal wave train structures. In some cases, the wave trains are limited to one hemisphere, while in others they span both. Notably, the wave patterns in the Northern Hemisphere closely resemble those induced by the WNP monsoon\u0026rsquo;s northern branch, whereas those in the Southern Hemisphere broadly mirror its southern counterpart. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, the circulation anomalies associated with WNP precipitation deviations exhibit characteristics of Rossby wave propagation along the subtropical jets. In the Northern Hemisphere, this manifests as 3\u0026ndash;4 large-scale wave fluctuations, with low-pressure centers over the Mediterranean, northern East Asia, and central North America, interspersed with weaker high-pressure anomalies. This structure closely resembles the first leading EOF mode of upper-tropospheric circulation deviations across the Northern Hemisphere (figure not shown). In the Southern Hemisphere, the wave train comprises alternating high- and low-pressure systems extending from western South Africa, across the southern Indian Ocean and northern Australia, through the Ross Sea and the South Pacific, to Argentina. This pattern closely aligns with the second EOF mode of 200hPa geopotential height in the Southern hemisphere. Similar wave train patterns are induced by the South Asian, Australian, and Somali monsoons (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed, and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ef). In contrast, circulation anomalies linked to the East Asian and North African monsoons are predominantly confined to the Northern Hemisphere (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee), aligning with the WNP\u0026rsquo;s northern wave train. While the circulation responses associated with the South and North American monsoons are relatively weaker, they still exhibit Southern Hemisphere features that resemble those of the WNP\u0026rsquo;s southern branch. Collectively, the upper-level circulation anomalies induced by precipitation deviations in different monsoon regions\u0026mdash;whether in the Northern or Southern Hemisphere\u0026mdash;show strong resemblance to those associated with the subtropical WNP monsoon region, suggesting a common dynamical pathway mediated by large-scale teleconnections.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOverall, these results underscore the strong interconnections among precipitation deviations across global monsoon regions, with the WNP monsoon emerging as a particularly influential center. The large-scale precipitation anomaly patterns associated with different monsoon systems bear remarkable resemblance to those linked to the WNP, and the corresponding upper-tropospheric circulation anomalies similarly echo the wave train structures induced by WNP precipitation deviations. This consistent alignment in both precipitation and circulation fields highlights the WNP monsoon's role as a key dynamical hub, facilitating hemispheric teleconnections and coupling among global monsoon systems.\u003c/p\u003e \u003cp\u003e \u003cb\u003ec. Potential causes of these connections\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe above analysis reveals significant linkages in precipitation deviations across monsoon regions, particularly between the WNP and other monsoon systems. To further assess the relationship between precipitation deviations in the WNP and those in other monsoon regions, we examine the correlation between the first principal component (PC1) of WNP precipitation deviations and the global precipitation deviations (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Results show not only strong local and nearby autocorrelations, but also pronounced negative correlations with surrounding monsoon regions\u0026mdash;including East Asia, North Africa, and Indonesia\u0026mdash;where correlation coefficients range from \u0026minus;\u0026thinsp;0.6 to \u0026minus;\u0026thinsp;0.8. Additionally, WNP precipitation deviations exhibit distinct positive and negative correlations with areas in the Southern Hemisphere, notably the southern Indian Ocean and the South Pacific. A significant positive correlation is also observed with the South American monsoon region, reinforcing the strong connections between the WNP and other monsoon systems. Comparison of Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e with Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e indicates that the spatial correlation patterns associated with the WNP PC1 align closely with the leading modes of precipitation deviations in each respective region. This consistency supports the interpretation that these leading modes, derived via empirical orthogonal function (EOF) analysis, reflect physically meaningful patterns rather than statistical artifacts. The observed connections imply the underlying dynamical linkages between the WNP and global monsoon systems.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo elucidate the relationship between precipitation deviations in the WNP monsoon region and adjacent monsoon systems, we conducted a composite analysis of the Walker and western Pacific Hadley circulations, based on the first principal component (PC1) of WNP precipitation deviations (\u0026plusmn;\u0026thinsp;0.75 standard deviations). As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea, strong ascending motion over the tropical western Pacific is accompanied by broad subsidence over the tropical Indian Ocean and a secondary subsidence branch over equatorial Central America. This configuration of the Walker circulation suggests that WNP precipitation deviations serve as a primary driver of the observed circulation perturbations. Through modulation of the Walker circulation, WNP precipitation anomalies exert influences on the South Asian, North African, and North American monsoon regions, thereby contributing to precipitation deviations in those areas. In the meridional direction, WNP precipitation deviations also influence surrounding monsoon regions via the Hadley circulation. As illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb, anomalous ascent over the western Pacific induces descending motion south of the equator, forming a clockwise meridional circulation cell. In contrast, a broader but weaker counterclockwise cell develops to the north of the WNP. This asymmetry, characterized by a stronger southern branch and a weaker northern branch, reflects the typical northward displacement of the thermal equator during boreal summer. Through this meridional circulation, WNP precipitation deviations are transmitted to the Indonesian and Australian monsoon regions. Additionally, to a lesser extent, the East Asian monsoon is also modulated via Hadley-type circulation associated with WNP precipitation deviations.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBeyond adjacent monsoon regions, a critical question concerns the remote teleconnections between precipitation deviations over the WNP and distant monsoon regions, such as those in Somalia and South America. To investigate this, we performed a composite analysis of convergence\u0026ndash;divergence wind field and wave activity fluxes, using the WNP precipitation deviation index. As illustrated in Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb, anomalous precipitation over the WNP generate strong local low-level convergence and upper-level divergence. These circulation anomalies are transmitted via the Walker circulation to the tropical Indian Ocean and equatorial Central America. Notably, Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea reveals a broad upper-level convergence center over the tropical Indian Ocean, extending meridionally toward the Mediterranean westerly jet and southeastward into the subtropical jet over the southern Indian Ocean. This large-scale perturbation of the subtropical jet acts as a source of zonally propagating Rossby waves in both hemispheres. As depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, the resultant wave trains propagate eastward along the Northern and Southern Hemisphere midlatitude jets. The wave activity fluxes in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec clearly demonstrate a northward propagation of energy from the North Africa\u0026ndash;Middle East region, followed by eastward dispersion along the midlatitude jet in the Northern Hemisphere. In the Southern Hemisphere, wave energy emanates from the Southern Africa\u0026ndash;tropical South Indian Ocean sector, travels southward into the southern Indian Ocean, reaches the Ross Sea, and subsequently curves northward to follow the path of the subtropical jet. The detailed physical mechanisms governing these processes are discussed in a companion paper (Wen et al. 2025b). Furthermore, the regions identified as sources of wave activity fluxes associated with jet stream disturbances in both hemispheres coincide with areas of maximum inter-model precipitation variability relative to climatology\u0026mdash;specifically, the Middle East and Southern Africa (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec). This spatial correspondence reinforces the pivotal role of these regions as dynamic hubs in the transmission of the WNP precipitation deviations across the global monsoon system. In brief, precipitation anomalies over the WNP induce mid-to-high-latitude zonal wave trains via disturbances over the tropical Indian Ocean, thereby facilitating the remote transmission of precipitation deviations to other monsoon regions such as Somalia and South America.\u003c/p\u003e \u003cp\u003eIn addition to above two pathways, Wen et al. (2025b) also noted that monsoon trough disturbances triggered by WNP precipitation anomalies can excite meridional wave trains along the East Asian coast (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed). This meridional wave train may serve as a primary conduit for conveying WNP-related precipitation deviations to the East Asian monsoon region.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFrom above analysis, we conclude that precipitation deviations in the WNP monsoon region serve as a primary driver of the inter-model spread relationships among monsoon regions within AMIP6 simulations. Through the Walker and Hadley circulations, these deviations exert significant influence on precipitation inter-model variability in adjacent monsoon regions, including South Asia, North Africa, and Australia. Moreover, midlatitude zonal wave trains in both hemispheres, excited by perturbations over the tropical Indian Ocean, facilitate the transmission of precipitation deviations of the WNP to remote monsoon regions such as those over Somalia and South America. In addition, disturbances originating from the WNP monsoon trough give rise to meridional wave trains along the East Asian coast, acting as conduits linking WNP precipitation deviations to the East Asian monsoon system. These multiple pathways highlight the central role of the WNP in organizing global monsoon inter-model variability and underscore its importance for accurately simulating precipitation in this region.\u003c/p\u003e"},{"header":"4. Conclusion and Discussion","content":"\u003cp\u003eThis study utilizes historical AMIP6 simulations to systematically investigate the connections and underlying drivers of climatological summer precipitation deviations across major monsoon regions, including the western North Pacific, East Asia, South Asia, Australia, North Africa, Somalia, South America, and North America. The results demonstrate that summer precipitation deviations across these regions are not isolated but exhibit pronounced interconnections, with particularly strong linkages to the western North Pacific (WNP) monsoon system. Further analysis identifies precipitation deviations over the WNP as a key driver of these inter-regional connections. Through atmospheric dynamical pathways\u0026mdash;including the Walker and Hadley circulations, meridional wave trains along the East Asian coast, and zonal wave trains propagating through the mid-to-high latitudes of both hemispheres\u0026mdash;the precipitation deviations in this region are effectively transmitted to other monsoon regions, thereby shaping global monsoon inter-model variability.\u003c/p\u003e \u003cp\u003eUsing principal component analysis (PCA), we investigated the dominant characteristics of summer precipitation deviations in various monsoon regions. While the leading modes of precipitation deviations were extracted independently for each region, their corresponding principal components exhibit significant correlations, particularly between the WNP and other monsoon regions, with correlation coefficients ranging from 0.4 to 0.6. To consolidate the connections among the monsoon regions, we conducted a comparative analysis of global precipitation and circulation deviations associated with the principal components of each monsoon region. The results reveal a high degree of similarity in the large-scale precipitation and circulation patterns across regions. In particular, precipitation deviations often display an out-of-phase relationship between the WNP and its surrounding monsoon domains. Upper-level circulation patterns in both hemispheres consistently align with the wave trains induced by precipitation deviations over the WNP. These findings suggest that the co-variability in precipitation deviations among monsoon regions is governed by common large-scale dynamical processes, with the WNP monsoon region serving as a key driver. Through the Walker and Hadley circulations, precipitation deviations in the WNP propagate to adjacent monsoon regions, including South Asia, North Africa, and Australia. Further analysis of convergence-divergence wind fields and wave activity fluxes demonstrates that the zonal wave trains in the subtropical jets of both hemispheres, generated by disturbances over the tropical Indian Ocean, establish teleconnections between WNP precipitation deviations and those over Somalia and South America. Additionally, as detailed in a companion study (Wen et al. 2025b), WNP precipitation anomalies excite disturbances within the monsoon trough that trigger meridional wave trains along the East Asian coast, thereby modulating precipitation deviations in the East Asian monsoon region.\u003c/p\u003e \u003cp\u003eIn addition, we note that while the WNP precipitation deviations exhibit only a weak linkage to NAM precipitation via the Walker circulation, the inter-model variability of precipitation over the NAM displays substantial independence. As shown in Supplementary Fig.\u0026nbsp;1a, the leading EOF mode of summer precipitation deviations over the NAM is characterized by a northwestward-oriented band of reduced rainfall across the monsoon domain. The associated global precipitation deviations (SFig. 1c) bear some resemblance to the large-scale pattern depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea, but with notably amplified local anomalies over the NAM region. The suppressed diabatic heating over the NAM region gives rise to a circulation response that differs substantially from that shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea. As illustrated in SFig. 1d, the tropical response features a prominent low pressure belt and a pair of anomalous centers over the equatorial eastern Pacific, resembling the canonical Matsuno\u0026ndash;Gill pattern induced by tropical heating (Matsuno, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e1966\u003c/span\u003e; Gill, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1980\u003c/span\u003e). In the midlatitudes of both hemispheres, high-pressure anomalous centers emerge, nearly out of phase with those associated with the WNP precipitation anomalies (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). This circulation response may be influenced by the local convergence\u0026ndash;divergence wind over Central America and jet stream fluctuations triggered by upstream disturbances (SFig. 2). As shown in SFig. 2c, the primary sources of energy emission are located over the west coasts of North and South America respectively, corresponding to the secondary centers of inter-model precipitation variability relative to climatology in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec. Via atmospheric teleconnections, the precipitation deviations over the NAM region are also linked to inter-model variability in other monsoon systems. As indicated in Supplementary Table\u0026nbsp;2, the leading model of NAM inter-model variability is significantly correlated with the second EOF mode in most monsoon regions such as the WNP, EA, SA, NAF, SAM, with absolute correlation coefficients ranging from 0.35 to 0.51, all significant at the 90% confidence level. These results suggests that precipitation variability over the NAM play a secondary yet important role in the co-variability of precipitation deviations across global monsoon regions. Overall, the circulation responses to precipitation deviations over the WNP and NAM exhibit a competitive relationship, with the prevailing influence determined by their relative intensities. The WNP, characterized by a strong inter-model variability, shows robust connections with other monsoon systems, primarily through associations with the leading EOF mode of precipitation deviations in those regions. In contrast, the magnitude of inter-model variability over the NAM (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb), along with its corresponding atmospheric response (SFig. 1d), is relatively weak, resulting in secondary linkages with other monsoon regions. Beyond the WNP and NAM, additional interregional connections may arise as indicated by the notable linkage between PC1 of Somali monsoon precipitation anomalies and PC2 over the South American monsoon region.\u003c/p\u003e \u003cp\u003eThis study highlights the critical role of precipitation deviations over the WNP in driving inter-model variability within the global monsoon system. This finding aligns with previous research (Bony et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Ren et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), which indicates that in the WNP region, simulation biases \u0026ndash; stemming from differences in model parameterization schemes -- can be amplified due to the atmosphere's heightened sensitivity to underlying warm pool sea surface temperature (SST) anomalies. As such, the fidelity of simulated precipitation over the WNP emerges as a key determinant of simulation accuracy in other monsoon regions. This highlights the critical need to enhance model performance in this region to improve the overall reliability of climate simulations. It is also worth noting that the data used in this study are derived from AMIP6 historical simulations, which prescribe observed sea surface temperatures (SSTs). In these simulations, precipitation deviations over the WNP are primarily attributed to factors such as the parameterization of convective processes and model resolution. However, in coupled atmosphere\u0026ndash;ocean systems, biases in simulated SSTs can also influence precipitation deviations in monsoon regions through air\u0026ndash;sea interactions. Accordingly, further investigation is required to assess whether the relationships among precipitation deviations in monsoon regions are altered under coupled model configurations.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work is supported by National Key R\u0026amp;D Program of China (2020YFA0608901).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe model data used in this study is openly available at the website (https://aims2.llnl.gov/search/cmip6/).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u0026nbsp;\u003c/strong\u003eThe authors declare that they have no conflict of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBony, S., Stevens, B., Frierson, D. M. W., et al. (2015). Clouds, circulation and climate sensitivity [J]. Nature Geoscience, 8(4): 261\u0026minus;268. \u003cu\u003ehttps://doi:10.1038/ngeo2398\u003c/u\u003e\u003c/li\u003e\n\u003cli\u003eBowler N E, Arribas A, Mylne K R, et al. (2008). The MOGREPS short-range ensemble predictio system. Quart J Roy Meteor Soc, 134, 703-722. https://doi.org/10.1002/qj.234\u003c/li\u003e\n\u003cli\u003eChen Jing, Zhu Yuejian, Duan Wansuo, Zhi Xiefei, Min Jinzhong, Li Xiaoli, Deng Guo, Yuan Huiling, Feng Jie, Du Jun, Li Qiaoping, Gong Jiandong, Shen Xueshun, MuMu. (2025). A review on development, challenges and future perspectives of ensemble forecast. Acta Meteorologica Sinica, 83(3): 1-23. https://doi.org/10.11676/qxxb2025.20240151.\u003c/li\u003e\n\u003cli\u003eDeser, C., Phillips, A., Bourdette, V., \u0026amp; Teng, H. (2012). Uncertainty in climate change projections: The role of internal variability. Climate Dynamics, 38(3), 527\u0026ndash;546. https://doi.org/10.1007/s00382-010-0977-x\u003c/li\u003e\n\u003cli\u003eGill, A. E. (1980). Some simple solutions for heat-induced tropical circulation[J]. Quart J Roy Meteor Soc, 106(449): 447-462.\u003c/li\u003e\n\u003cli\u003eHagos, S., Leung, L. R., Ashfaq, M., \u0026amp; Balaguru, K. (2019). South Asian monsoon precipitation in CMIP5: A link between inter-model spread and the representations of tropical convection. Climate Dynamics, 52(2), 1049\u0026ndash;1061. https://doi.org/10.1007/s00382-018-4177-4\u003c/li\u003e\n\u003cli\u003eHannachi, A., Jolliffe, I. T., \u0026amp; Stephenson, D. B. (2007). Empirical orthogonal functions and related techniques in atmospheric science: A review. International Journal of Climatology, 27(9), 1119\u0026ndash;1152. https://doi.org/10.1002/joc.1499\u003c/li\u003e\n\u003cli\u003eHuang, D., Liu, A., Zheng, Y., \u0026amp; Zhu, J. (2022). Inter-model spread of the simulated East Asian summer monsoon rainfall and the associated atmospheric circulations from the CMIP6 models. Journal of Geophysical Research: Atmospheres, 127, e2022JD037371. https://doi.org/10.1029/2022JD037371\u003c/li\u003e\n\u003cli\u003eKrishnamurti T N. (1999). Improved weather and seasonal climate forecasts from multimodel superensemble. Science, 285(5433): 1548-1550. https://doi.org/10.1126/science.285.5433.1548\u003c/li\u003e\n\u003cli\u003eLehner F., Deser, C., Maher, N., et al. (2020). Partitioning climate projection uncertainty with multiple large ensembles and CMIP5/6, Earth Syst. Dynam., 11, 491\u0026ndash;508, https://doi.org/10.5194/esd-11-491-2020\u003c/li\u003e\n\u003cli\u003eLenartz F. Mourre B. Barth A. et al. (2010). Enhanced ocean temperature forecast skills through 3-D super-ensemble multi-model fusion. Geophys Res Lett, 37, L19606. https://doi.org/10.1029/2010GL044591\u003c/li\u003e\n\u003cli\u003eLorenz, E. N. (1963). Deterministic Nonperiodic Flow, J. At-mos. Sci., 20, 130\u0026ndash;141. https://doi.org/10.1175/1520-0469(1963)020\u0026lt;0130\u003c/li\u003e\n\u003cli\u003e:dnf\u0026gt;2.0.co;2.\u003c/li\u003e\n\u003cli\u003eMaher, N., Matei, D., Milinski, S., and Marotzke, J. (2018). ENSO Change in Climate Projections: Forced Response or Internal Variability?, Geophys. Res. Lett., 45, 11390-11398. https://doi.org/10.1029/2018GL079764\u003c/li\u003e\n\u003cli\u003eMatsuno, T. (1966). Quasi-geostrophic motions in the equatorial area. Journal of the Meteorological Society of Japan, 44(1), 25-43. https://www.jstage.jst.go.jp/article/jmsj1965/44/1/44_1 25\u003c/li\u003e\n\u003cli\u003eMolteni F. Buizza R, Palmer T N, et al. (1996). The ECMWF Ensemble Prediction System: Methodology and validation. Quart J Roy Meteor Soc, 122: 73-119. https://doi.org/10.1002/qj.49712252905\u003c/li\u003e\n\u003cli\u003eMonerie, P. A., Wainwright, C. M., Sidibe, M., et al. (2020). Model uncertainties in climate change impacts on Sahel precipitation in ensembles of CMIP5 and CMIP6 simulations. Climate Dynamics, 55(5), 1385\u0026ndash;1401. https://doi.org/10.1007/s00382-020-05332-0\u003c/li\u003e\n\u003cli\u003eMurphy, J. M., Sexton, D. M. H., Barnett, D. H., Jones,G. S., Webb, M. J., Collins, M., and Stainforth, D. A. (2004). Quantification of modelling uncertainties in a large ensemble of climate change simulations, Nature, 430, 768\u0026ndash;772. https://doi.org/10.1038/nature02771\u003c/li\u003e\n\u003cli\u003eQu, T., Du, Y., Strachan, J., Meyers, G., Slingo, J. (2005). Sea surface temperature and its variability in the Indonesian region. Oceanography 18, 50\u0026ndash;61. https://doi.org/10.1029/2008JC005150\u003c/li\u003e\n\u003cli\u003eSanderson, B. M., Piani, C., Ingram, W. J., Stone, D. A., and Allen, M. R. (2008). Towards constraining climate sensitivity by linear analysis of feedback patterns in thousands of perturbed-physics GCM simulations, Clim. Dynam., 30, 175\u0026ndash;190. https://doi.org/10.1007/s00382-007-0280-7.\u003c/li\u003e\n\u003cli\u003eRen Z, Zhou T, Guo Z, et al. (2023). Relationship between Vertical Convection Structure and Precipitation Simulation Bias in the Tropical Atmosphere: An Analysis Based on GAMIL3 Model [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 47(2): 239\u0026minus;258. doi:10.3878/j.issn.1006-9895.2109.21098\u003c/li\u003e\n\u003cli\u003eTakaya, K., \u0026amp; Nakamura, H. (2001). A formulation of a phase-independent wave-activity flux for stationary and migratory quasi-geostrophic eddies on a zonally varying basic flow. Journal of the Atmospheric Sciences, 58(6), 608\u0026ndash;627. https://doi.org/10.1175/1520-0469(2001)058\u0026lt;0608:AFOPIW\u0026gt;2.0.CO;2\u003c/li\u003e\n\u003cli\u003eTrenberth, K.E., Stepaniak, D.P., Caron, J.M. (2000). The global monsoon as seen through the divergent atmospheric circulation. J. Clim. 13, 3969\u0026ndash;3993. https://doi.org/10.1175/1520-0442(2000)013\u0026lt;3969:TGMAST\u0026gt;2.0.CO;2\u003c/li\u003e\n\u003cli\u003eWang, B., Ding, Q. (2008). Global monsoon: dominant mode of annual variation in the tropics. Dyn. Atmos. Oceans 44, 165\u0026ndash;183.\u003c/li\u003e\n\u003cli\u003eWang, P. X., Wang, B., Cheng, H., et al. (2017). The global monsoon across time scales: Mechanisms and outstanding issues. Earth-Science Reviews, 174, 84\u0026ndash;121. https://doi.org/10.1016/j.earscirev.2017.07.006\u003c/li\u003e\n\u003cli\u003eWang, P.X., Wang, B., Cheng, H., Fasullo, J., Guo, Z., Kiefer, T., Liu, Z. (2014b). The global monsoon across time scales: coherent variability of regional monsoons. Clim. Past 10, 1\u0026ndash;46. http://dx.doi.org/10.5194/cp-10-1-2014.\u003c/li\u003e\n\u003cli\u003eWen, N. and Huan, K., (2025b), AMIP6 Inter-model Spread analysis: Effects of precipitation deviation on circulation and its underlying mechanisms, preparing\u003c/li\u003e\n\u003cli\u003eXu Z Z, Chen J, Mu M, et al. (2022). A Nonlinear Representation of Model Uncertainty in a Convective-Scale Ensemble Prediction System. Adv Atmos Sci 39, 1432-1450. https://doi.org/10.1007/s00376-022-1341-x\u003c/li\u003e\n\u003cli\u003eZhi X F. Qi H X. Bai Y Q. et al. (2012). A comparison of three kinds of multimodel ensemble forecast techniques based on the TIGGE data. Acta Meteorol Sin, 26(1) : 41-51. https://doi.org/10.1007/s13351-012-0104-5\u003c/li\u003e\n\u003cli\u003eZhou, T., Lu, J., Zhang, W., \u0026amp; Chen, Z. (2020). The sources of uncertainty in the projection of global land monsoon precipitation. Geophysical Research Letters, 47, e2020GL088415. https://doi.org/10.1029/2020GL088415\u003c/li\u003e\n\u003cli\u003eZhu S P, Zhi X F, Ge F, et al. (2021). Subseasonal forecast of surface air temperature using superensemble approaches: Experiments over Northeast Asia for 2018. Wea Forecasting, 36(1): 39-51. https://doi.org/10.1175/WAF-D-20-0096.1\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"AMIP6, Inter-model variability, Monsoon precipitation, Global monsoon","lastPublishedDoi":"10.21203/rs.3.rs-6711488/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6711488/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study investigates the interconnections and underlying causes of the inter-model variability in summer precipitation across global monsoon regions, based on historical AMIP6 simulations. The results reveal significant connections among the leading modes of precipitation inter-model deviations in different monsoon domains, with the western North Pacific (WNP) emerging as a key driver. Deviations over the WNP modulate adjacent monsoon systems -- including the South Asian, North African, and Australian monsoons -- primarily through the Walker and Hadley circulations. In addition, WNP precipitation anomalies, via jet stream disturbances originating from the tropical Indian Ocean, generate zonal wave trains in the mid-to-high latitudes of both hemispheres, influencing the Somali and South American monsoon. The WNP also affects East Asian monsoon precipitation mainly through meridional wave trains along the East Asian coast, triggered by perturbations in the monsoon trough. These findings underscore the central role of WNP precipitation deviations -- via atmospheric dynamic process -- in shaping the global pattern of monsoon precipitation inter-model variability in AMIP6, and highlight the need for improved model performance over this region to enhance global monsoon simulation fidelity.\u003c/p\u003e","manuscriptTitle":"Inter-model Spread in Summer Precipitation: Connections and Causes Across Monsoon Regions in AMIP6 Simulations","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-13 13:50:47","doi":"10.21203/rs.3.rs-6711488/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0efe79eb-0e25-4344-bce8-9f7c681779b1","owner":[],"postedDate":"June 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-04T00:58:54+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-13 13:50:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6711488","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6711488","identity":"rs-6711488","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}