The IPWP as a capacitor for autumn sea ice loss in Northeastern Canada | npj Climate and Atmospheric Science

npj Climate and Atmospheric Science volume 7, Article number: 259 (2024) Cite this article

Metrics details

The Indo-Pacific Warm Pool (IPWP) has been warming due largely to increasing greenhouse gas emissions, but its impact on Arctic sea ice remains unclear. Our study finds a significant negative correlation between the IPWP index and sea ice concentration in northeastern Canada during boreal autumn (October-December). Our results suggest that IPWP warming statistically accounts for 45% of sea ice loss observed in this region. We introduce the “Arctic capacitor effect of the IPWP”, a novel concept that expounds upon the distant connection between greenhouse gas emissions and Arctic sea ice loss. Specifically, as greenhouse gases elevate temperatures in the IPWP, increasing temperature gradient and tropical convection, a planetary wavetrain is initiated. This wavetrain, along with transit eddy feedback, traverses towards the Arctic and thereby influences the strength of the Arctic vortex and its associated effects on Arctic sea ice. Our findings highlight the crucial role of tropical oceans in the broader context of global climate change, emphasizing the necessity of accounting for their impact on polar climate.

Over the past several decades, the Arctic sea ice has exhibited a significant decreasing trend, particularly since the late 1990s, as noted by Parkinson1 and Simmonds2. The reduction in sea ice is largely attributed to an increase of surface temperature resulting from an increase in greenhouse gas concentrations. On the basis of observations from 1952 to 2015, Notz and Stroeve3 detected a robust relationship: each metric ton of global CO2 emission has resulted in a September sea ice loss of 3 ± 0.3 m2. Considering the complexity of processes of atmosphere-snow-ice-ocean interactions4 and the numerous positive feedbacks5, the existence of a robust linear relationship is surprising, and calls for further research on the role of greenhouse gas forcing in the Arctic. Besides direct local positive feedbacks, greenhouse gas forcing indirectly affects the Arctic climate system through remote and indirect effects. These indirect effects are felt via remote drivers on atmospheric and oceanic large-scale circulation and air- and water-mass transports from the south to the Arctic6,7,8. Some large-scale processes through which remote greenhouse gas concentrations may influence Arctic sea ice have already been suggested by previous studies9,10. Further research is needed to better understand the relative importance of these and other processes in driving the observed sea ice loss in the Arctic.

The Indo-Pacific Warm Pool (IPWP) is a warm water mass with sea surface temperature (SST) above 28 °C in the eastern Indian and western and central Pacific Oceans11,12. The SST anomalies in the IPWP can exert strong effects on African and Asian monsoon climate13,14, the characteristics of El Niño–Southern Oscillation (ENSO)15,16 and the Arctic climate17,18,19,20. In recent decades, the IPWP has exhibited increased warming and expansion, which has been attributed mainly to increases in greenhouse gas concentrations, although the Pacific Decadal Oscillation (PDO) has played a smaller yet significant role21,22. Recent studies have suggested that the warming and expansion of the IPWP have altered the Madden-Julian Oscillation (MJO) life cycle23. The overarching influence of the warming and expansion of the entire IPWP on Arctic sea ice remains a subject of inquiry. While previous studies have explored the role of specific portions or isolated events within the IPWP on Arctic sea ice variability24,25,26,27, the comprehensive impact of the IPWP’s broader dynamics on Arctic sea ice remains to be further elucidated.

The purpose of this study is to investigate the potential remote greenhouse forcing on the sea ice loss in the broader Arctic region through the IPWP. We employ a multifaceted approach that involves using the IPWP index, defined as the area-averaged SST over the region between 60°E-170°E longitude and 15°S-15°N latitude, to quantify SST variations within the IPWP. We first estimate the trend in Arctic sea ice loss associated with warming IPWP conditions. This involves calculating the trend in normalized IPWP index and multiplying it by sea ice anomalies regressed onto the normalized detrended IPWP index. Our aim is to demonstrate a statistically significant correlation between the trend of sea ice concentration in Arctic regions and the IPWP index. Subsequently, we delve deeper into the mechanism linking IPWP warming and Arctic sea ice loss by conducting regression analyses of atmospheric variables onto the detrended IPWP index. By detrending the IPWP index, we aim to isolate its influence from the direct effects of CO2 increases on Arctic sea ice. We show that changes in the IPWP index statistically account for nearly half of the sea ice loss in the broader Arctic region encompassing Hudson Bay, Baffin Bay and Labrador Sea. We further support the statistical relationship through a controlled numerical experiment. Finally, we explain the underlying physical mechanism that establishes this link and introduce a new concept, referred to as the “Arctic capacitor effect of the IPWP”. This concept posits that the warming signal associated with escalating greenhouse gas emissions finds a repository within the IPWP. Subsequently, the intensified warming of the IPWP induced primarily by greenhouse gas increases, indirectly contribute to sea ice loss through anomalous planetary wavetrain and Arctic vortex. Our study provides insight into the complex mechanisms driving Arctic sea ice loss and underscores the critical role that tropical oceans, particularly the IPWP, may play in this process.

Over the past four decades, the IPWP region (60oE-170oE, 15oS-15oN) has experienced a warming trend of 0.0146 oC yr−1 (p < 0.01) in the area-averaged SST during boreal autumn (October–December), indicating an increase in the IPWP index (Fig. 1a). During the same period, Arctic sea ice concentration in boreal autumn has exhibited a declining trend (Fig. 1b, d). A regression analysis of the detrended IPWP index, quantified through area-averaged SST, and detrended Arctic sea ice concentration in boreal autumn reveals statistically significant (p < 0.05) negative sea ice anomalies across the broader Arctic region encompassing Hudson Bay, Baffin Bay and Labrador Sea. Although significant positive sea ice anomalies are observed in the Denmark Strait and a small area just west of Novaya Zemlya in the Barents Sea (Fig. 1c), their spatial extent is notably smaller when contrasted with areas affected by negative anomalies. Therefore, our emphasis rests on the Hudson Bay, Baffin Bay and Labrador Sea region in the eastern part of the Canadian Arctic and sub-Arctic, where a discernable and inverse relationship exists between the sea ice concentration and the IPWP index.

The time coefficients of the IPWP index, represented by the area averaged SST in the region (60°E-170°E, 15°S-15°N) (°C) (a) and sea ice concentration in the region (95°W-50°W, 50°N-80°N) (b), IPWP-related sea ice concentration anomalies (c), trend in sea ice concentration (d), ratio of the IPWP-explained sea ice concentration trend to the total trend (e). The green region in panel (c) indicates the region of sea ice concentration in panel (b). The dotted regions in panel (c, d) indicate above 90% confidence level. All plots refer to boreal autumn.

To simplify the computation, we placed a latitude-longitude box (95°W-50°W, 50°N-80°N) (Fig. 1c) over the region and calculated the averaged correlation over different historical periods before and after detrending is applied to eliminate the impact of trends on the correlation. Over the period from 1979 to 2020, the correlations were −0.71 (p < 0.05) before detrending and −0.40 (p < 0.05) after detrending (Table S1). Extending the period from four decades to over a century (1900–2015) yielded slightly smaller but still significant correlations, −0.49 (p < 0.05) and −0.33 (p < 0.05) before and after detrending (Table S1). These results establish a statistically significant inverse relationship between the sea ice variations in the region and the IPWP index.

To confirm this link, we also calculated correlations using historical model simulations from 49 CMIP6 models (Table S2). The majority of the models produced statistically significant negative correlations between averaged sea ice concentration over the box and IPWP index during boreal autumn, with 46, 45, and 37 models (or 94%, 92%, and 75%) showing such correlations for the 1850–2014, 1900–2014, and 1970–2014 periods, respectively. Averaging across all 49 model simulations, the detrended correlation coefficients were −0.44 (p < 0.05) for the 1979–2014 period, −0.40 (p < 0.05) for the 1900–2014 period, and −0.38 (p < 0.05) for the 1850–2014 period. These values were found to be similar to the calculations using reanalysis data. Together, the observational and model simulation results clearly demonstrate the relationship between the variability in Arctic sea ice and the IPWP.

We have also investigated the interrelation between the boreal autumn IPWP index and Arctic sea ice concentration in boreal winter (January-February-March, JFM) period on the interannual time scale. The result (Fig. S1) reveals a notable inverse correlation in sea ice concentration across the northern Atlantic Ocean, both east and west of 30°W, as well as a similar dipole pattern in the northern Pacific Ocean. These findings suggest a lead-lag relationship, wherein the boreal autumn IPWP index influences the subsequent boreal winter sea ice concentration in these regions.

The quantitative impact of the warming IPWP on the sea ice reduction across the broader Arctic region, encompassing the Hudson Bay, Baffin Bay and Labrador Sea, can be gauged through the ratio of sea ice trend related to the IPWP to the overall trend (Fig. 1e). The IPWP-related sea ice trend is estimated by the product of the sea ice anomalies regressed onto the normalized detrended IPWP index, and the trend in normalized IPWP index (Eq. 2). Averaging across the designated box, the ratio is 45%, suggesting that nearly half of the sea ice loss in this region is statistically linked to the warming of the IPWP. Other regions in the Arctic Ocean exhibit smaller ratios, and there is even a negative contribution from IPWP warming in the Greenland and Barents Seas.

The relationship between the warming of the IPWP and the loss of Arctic sea ice across the Hudson Bay, Baffin Bay and Labrador Sea region can be explained by examining changes in SST and atmospheric circulations related to the IPWP (as shown in Figs. 2–4). The sea level pressure pattern regressed on the IPWP index (Fig. 2a) reveals the coexistence between the warm phase of the IPWP and the negative phase of the Arctic Oscillation (AO). This corresponds with a significantly negative correlation between the IPWP index and the AO index (−0.36, p < 0.05). Additionally, a significant positive correlation (0.39, p < 0.05) is found between the AO index and the sea ice concentration anomalies in this Arctic region. The negative AO results in anomalous high pressure over Greenland, producing an anomalous anticyclonic surface circulation (Fig. 2b). The circulation transports warm and moisture-laden air over the northern Atlantic Ocean into this region, contributing to surface warming (Fig. 2c). This warming is further amplified by stronger downward longwave radiation (Fig. 2d), which is caused by increased moisture in the atmosphere (Fig. 3a). Adiabatic warming associated with sinking motion plays a less crucial role in sea ice loss in the region relative to advection (Fig. 3b). The on-shore wind stress also decreases the sea ice concentration. Ultimately, these factors lead to a reduction in sea ice in the region (Fig. 1c). A previous study24 also suggested a connection between anomalous high pressure over Greenland and Arctic sea ice loss during boreal summer.

Regression map of autumnal mean sea level pressure (MSLP) (Pascal) (a), 10-m wind field (vector, m s−1) (b), 2-m air temperature (oC) (c), and accumulated downward longwave radiation (×105W m−2) (d) onto the normalized detrended IPWP index. The dotted regions in panels (a, c, and d) indicate above 90% confidence level. The green vectors in panel (b) denote above 90% confidence level.

Regression map of autumnal low level (750–1000 hPa) specific humidity (g kg−1) (a), 500-hPa vertical velocity (Pa s−1) (b) onto the normalized detrended IPWP index. The dotted regions indicate above 90% confidence level.

The same as Fig. 2, but for anomalous sea surface temperature (SST) (oC) (a), outgoing longwave radiation (OLR) (W m−2) (b), Rossby wave source (shading) (10−10 s −2) and 200-hPa divergent wind (vector, m s−1) (c) and wave activity flux (vector, m2 s−2) and 200-hPa geopotential height (contour) (gpm) (d). Black dotted regions in panels (a, b, and d) and green dotted regions in panel cindicate above 90% confidence level.

Further examination of the relationship between warmer IPWP and negative AO is useful. An analysis of IPWP-regressed SST patterns reveals significant warm SST anomalies in the tropical western Pacific Ocean, eastern Indian Ocean, and the South China Sea (Fig. 4a). SST anomalies related to the IPWP index (Fig. 4a), combined with the trend in tropical Pacific SST (Fig. S2) suggests strengthening SST gradient with decreasing SST values eastwards in the tropical Pacific Ocean. This stronger Walker circulation promotes increased convective activity, which is indicated by negative anomalies of outgoing longwave radiation at the top of the atmosphere (Fig. 4b), generating negative Rossby wave sources and 200-hPa divergent wind anomalies (Fig. 4c). These wave sources excite a planetary wavetrain that propagates northeastwards into the northwestern Pacific Ocean (Fig. 4d). While a small branch goes towards the Arctic Ocean, the main wavetrain continues to propagate towards the northeastern Pacific while being strengthened through interactions with local colder SSTs related to the aforementioned wavetrain. Some of the wavetrain over the northeastern Pacific enters directly north into the Arctic, and another portion goes into lower latitudes, while most of the wavetrain continues to propagate east and northeastwards to Greenland. The warmer SSTs in the North Atlantic also facilitate the wavetrain’s propagation to Greenland and lower latitudes. Finally, a part of the wavetrain reflects to Europe. The wavetrain excited by warmer SSTs results in negative 200-hPa height anomalies over northern mid-latitudes but in positive height anomalies over the Arctic (Fig. 4d), which corresponds to the negative AO. It is noteworthy that warming IPWP also corresponds to positive autumnal sea ice anomalies in some small areas east of Greenland.

In addition to the tropospheric pathway, there is also a stratospheric pathway through the anomalous Arctic vortex. Previous study noted that anomalous EP flux related to the IPWP can alter the intensity of stratospheric Arctic vortex through wave-flow interaction25. The warmer IPWP corresponds to a weakened stratospheric Arctic vortex (Fig. 5a). The stronger subtropical zonal wind and weaker mid-latitude zonal wind also indicate a weaker vortex (Fig. 5b). These findings suggest that the wavetrain enters the Arctic at three locations (the Siberia, Alaska, and Canadian Arctic) over the North Pacific and Atlantic. The EP flux from mid troposphere to stratosphere shows a stronger divergence at subtropics and convergence at mid-latitude (Fig. 5c), which is consistent with anomalous zonal wind speed and the weaker stratospheric Arctic vortex. The weaker vortex produces a negative AO in the troposphere according to the mechanisms outlined previously26.

The same as Fig. 2, but for anomalous 100-hPa geopotential height (shading) (gpm) (a), zonal wind speed (b) (m s−1), and Eliassen-Palm (EP) flux (c). Units for the horizontal and vertical vectors are 1 and 10−2 kg s−1, respectively. The red contour of wind speed in panel b denotes climatological zonal wind speed. Dotted regions in panels a and b indicate above 90% confidence level.

Prior research has established a linkage between autumnal Arctic sea ice loss and the negative phase of the North Atlantic Oscillation/Arctic Oscillation (NAO/AO)27,28,29, coupled with a weakening of the polar vortex30,31,32,33,34,35,36. To delve deeper, we conduct a regression analysis involving sea level pressure and 100-hPa geopotential height relative to the extent of Arctic sea ice. The sea level pressure (Fig. S3a) and 100-hPa geopotential height (Fig. S3b) corresponding to Arctic sea ice reduction equally manifest a negative NAO pattern and a debilitated polar vortex. Interestingly, their core positions shift towards the eastern Arctic Ocean, a deviation from those connected with the IPWP as shown in Figs. 2a and 5a. It is noteworthy that the correlation between the time series of Arctic sea ice loss and the IPWP remains insignificant (r = −0.14, p > 0.1). Evidently, our findings elucidate that the atmospheric circulation anomalies linked to the IPWP diverge from those intertwined with Arctic sea ice conditions, albeit they may potentially be influenced by overarching global warming trends.

In addition to Rossby wave dynamics, we also assessed the role of transit eddy feedback by calculating the contribution of the seasonal eddy momentum flux convergence, following the method of Nakamura and Zhu (2010), to the zonal mean 200-hPa zonal wind during boreal autumn related to the IPWP index. The results (Fig. S4) show that transient eddies play an important role in the decrease of 200-hPa zonal wind (50–80°N) and the weakening of the polar vortex, which complements the Rossby wave effects (Figs. 4d, 5).

To validate the aforementioned statistical findings, we conducted numerical experiments using the Community Atmosphere Model Version 5 (CAM5) atmospheric model. As shown earlier, the wavetrain’s origin is linked to SST anomalies in the northern Indian Ocean and the South China Sea (Fig. 4a). In the idealized numerical experiment, we introduced a 2 °C temperature increase within the designated region (80–160°E, 0–20°N) to scrutinize the IPWP’s influence on Arctic sea ice dynamics. Figure 6 portrays the anomalous 200-hPa geopotential height derived from the idealized experiment, juxtaposed with the control experiment without the temperature increase—an outcome congruent with the regression results (Fig. 4d).

The difference of 200-hPa geopotential height (gpm) between the idealized numerical experiment prescribed by the IPWP SST and the control experiment.

The spatial correlation coefficient between the patterns in Fig. 4d and Fig. 6 registers at 0.32 (p < 0.01) (across 6912 grid points). Notably, a high-pressure system emerges north of the SST anomaly zone, a response attributed to anomalous heating as delineated by the Gill-Mode. Anomalous low-pressure patterns emerge over the eastern Asian region and the central Pacific Ocean. Concurrently, the Arctic and northern Atlantic Ocean exhibit positive height anomalies. The numerical experiments also disclose a weakened Arctic polar vortex. Furthermore, the 100-hPa level north of 60°N reveals positive geopotential height anomalies and elevated air temperatures (Fig. 7), with a corresponding warming trend of the 850-hPa air temperature over Canada, Alaska, Greenland, and the Arctic Ocean (Fig. 8).

The difference of air temperature (oC) (a) and 100-hPa geopotential height (gpm) (b) between the idealized numerical experiment prescribed by the IPWP SST and the control experiment.

The difference of 850-hPa air temperature (oC) between the idealized numerical experiment prescribed by the IPWP SST and the control experiment.

Besides the IPWP region, the IPWP index is also associated with SST anomalies in other oceans of the Northern Hemisphere, particularly the tropical Atlantic Ocean and the North Pacific Ocean (Fig. 4a). To account for the effect of SST anomalies over these oceans, we conducted additional numerical experiments prescribed with SST anomalies from Fig. 4a. The simulated anomalous 200-hPa geopotential height patterns (Fig. 9) reveal positive height anomalies over the Arctic and south of 30°N and negative anomalies over eastern Asia, the central North Pacific, and the Atlantic. The spatial correlation between Fig. 4d and Fig. 9 is 0.61 (p < 0.01) across 6912 grid points. The spatial pattern of the 200-hPa height anomalies is more similar to Fig. 4d than to Fig. 6, where only IPWP SST anomalies were prescribed. Additionally, the anomalous 100-hPa heights (not shown) illustrate a weakened polar vortex. The anomalous positive 850-hPa temperatures (Fig. 10) also contribute to the Arctic sea ice loss in the study region.

The difference of 200-hPa geopotential height (gpm) between the SST idealized numerical experiment prescribed by the SST anomalies in Fig. 4a and the control experiment.

The difference of 850-hPa air temperature (oC) between the idealized numerical experiment prescribed by the SST anomalies in Fig. 4a and the control experiment.

While certain disparities in the location and intensity of geopotential height and air temperature anomalies exist between the reanalysis and model outcomes, the model results substantiate the discernible influence of SST anomalies related to the IPWP index on Arctic sea ice through anomalous atmospheric circulations.

The above analyses provide compelling evidence linking Arctic sea ice loss to the warming and expansion of the Indo-Pacific Warm Pool, which in turn has been linked to an increase in atmospheric greenhouse gas concentrations21,22. These findings have led us to introduce a novel concept called the “Arctic capacitor effect of the IPWP” which indicates that greenhouse warming in the tropics can be seen as an indirect greenhouse forcing on autumnal sea ice loss across extended Arctic expanse, spanning from Hudson Bay to Baffin Bay and to the Labrador Sea.

Specifically, the primarily human-induced warming of the IPWP and the associated increase in temperature gradient have triggered stronger convective activity over the tropical eastern Indian Ocean, tropical western Pacific Ocean, and South China Sea. As a result, a wavetrain travels into the Arctic, weakening the stratospheric Arctic vortex and causing a negative Arctic Oscillation (AO), which ultimately leads to sea ice loss across the broader Arctic region including Hudson Bay, Baffin Bay and the Labrador Sea.

The term “capacitor effect” traditionally denotes the cyclic storage and release of energy, a concept widely recognized in physics and climate science. For instance, it has been applied to describe the storage of El Niño-Southern Oscillation (ENSO) impacts in the tropical Indian Ocean28. Here, we extend this concept to encompass the role of the IPWP as a reservoir for signals of warming associated with greenhouse gas increases. This stored energy is released to enhance atmospheric convection, which in turn initiates teleconnection patterns and subsequently influences atmospheric circulation dynamics in the Arctic region. While our usage of the term “capacitor effect” deviates from its conventional application, it serves as a metaphorical representation of the process of energy storage and release within the IPWP in our study’s context.

In conclusion, our study provides new insights into the complex mechanisms driving Arctic sea ice loss. By analyzing long-term (1836–2020) global oceanic and atmospheric data augmented by historical simulations from 49 CMIP6 general circulation models, we found a significant negative correlation between autumnal (October–December) sea ice concentration in the extended Arctic expanse, including Hudson Bay, Baffin Bay and the Labrador Sea, and the IPWP index. We found correlation coefficients ranging from −0.33 to −0.71 (p < 0.01), depending on the time periods and datasets analyzed, and demonstrated that approximately 45% of the sea ice loss in this region can be attributed to the warming of the IPWP.

Our study proposes the “Arctic capacitor effect of the IPWP” as a mechanism indirectly linking sea ice loss in this broader Arctic region to rising greenhouse gas emissions. The warming of IPWP and the associated increase in SST gradient between the IPWP region and the east Pacific, which has been attributed primarily to increasing greenhouse gas concentrations, enhances convective activity over the tropical eastern Indian Ocean, western Pacific Ocean, and South China Sea. This stronger convection excites a wavetrain propagating into the Arctic and influences the stratospheric Arctic vortex through anomalous EP flux, leading to a negative AO. The anomalous high over the Arctic and Greenland related to the negative AO and its associated warm and moist advection and positive longwave radiation feedback ultimately lead to sea ice retreat in the extended Arctic expanse including Hudson Bay, Baffin Bay and Labrador Sea.

We acknowledge that the concept of the IPWP acting as a capacitor for autumn Arctic sea ice loss, which is largely based on statistical analyses, requires further validation. A valuable next step would be to conduct pacemaker experiments using a fully coupled atmosphere-ocean-sea ice-land model. These experiments could help verify the robustness of our findings and provide deeper insights into the underlying mechanisms. Future research should prioritize such tests to strengthen the connections proposed in this study.

It is important to note that our study does not rule out the well-known direct contribution of greenhouse gas increases to sea ice loss in the circumpolar Arctic. Further, IPWP does not represent the only mechanism of indirect greenhouse gas forcing on Arctic sea ice loss. Internal climate modes, such as the Pacific Decadal Oscillation (PDO)37 and the Atlantic Multidecadal Oscillation (AMO)38, also contribute to sea ice loss in this region and may play a more significant role in other Arctic regions39,40,41,42. The Arctic capacitor effect of the IPWP also leads to sea ice loss in Bering, Beaufort, and Kara Seas and the Sea of Okhotsk, but the contribution of this effect to sea ice change in those regions is smaller than in the Hudson Bay, Baffin Bay and Labrador Sea region.

It is pertinent to note that IPWP-related sea ice anomalies are positive in areas northeast of the Denmark Strait and west of Novaya Zemlya in the Barents Sea. Our analysis is primarily dedicated to elucidating the mechanisms underlying the inverse correlation between the IPWP index and sea ice concentration across the broader Arctic region, including the Hudson Bay, Baffin Bay and Labrador Sea. It is important to clarify that we do not posit a generalized extrapolation of the observed sea ice-IPWP association or potential physical mechanisms to other Arctic zones.

Past research has extensively explored the influence of tropical Pacific SST on Arctic atmospheric patterns and sea ice conditions43,44,45,46,47,48,49,50. These investigations predominantly centered on SST anomalies in the central and eastern Pacific Ocean linked to internal forcings, notably the Interdecadal Pacific Oscillation (IPO), affecting Arctic sea ice. In contrast, our present study brings to the forefront the impact of the IPWP, encompassing the Indian and western Pacific Oceans, which is primarily associated with CO2-induced effects on Arctic sea ice reduction. Notably, a significant proportion of the increase in the SST in the IPWP region over the past four decades remains independent of changes in the IPO and Atlantic Multidecadal Oscillation (AMO), arising primarily from the elevated levels of atmospheric CO2 concentrations21,22,49,50. In addition, previous studies indicated the effect of the Arctic sea ice concentration changes on the following AO variations or atmosphere-ocean systems over the tropics51,52,53,54,55,56,57,58,59,60,61. Further studies will be carried out to scrutinize the interaction of the IPWP and the Arctic sea ice loss. The regression of SST anomalies onto the sea ice loss in the target region indicates that the Arctic sea ice loss in the target region is related to a La Niña state in the tropical Pacific Ocean, leading to stronger IPWP (Not shown).

Our study significantly contributes to the current understanding of the drivers of Arctic sea ice change and emphasizes the crucial role of the IPWP. This study underscores the need for continued research to explore the complex processes that drive Arctic sea ice loss. Understanding these processes is crucial for accurately predicting the future of Arctic sea ice and its impacts on the global climate system, and for developing effective mitigation and adaptation strategies. The capacitor effect of the IPWP has the potential to influence not only the Arctic region but also other regions affected by IPWP variability, such as monsoon regions. This approach to studying the effect of greenhouse gas increase on global climate highlights the importance of tropical oceans in the global effect of increasing greenhouse emission. We recommend that future studies investigate whether other ocean regions in the tropics, subtropics or mid-latitudes exhibit similar capacitor effects.

Our study aimed to investigate the connection between the Indo-Pacific warm pool (IPWP) and Arctic sea ice concentration. To achieve this, we first employed a linear correlation analysis with historical sea surface temperature (SST) and sea ice data spanning from 1900 to the present. However, we focused on the recent four decades for subsequent process analyses, as the reliability of observational data, especially sea ice and SST, prior to the satellite era is considered to be lower. In addition, we narrowed our focus to the period of October through December, which is the time of year when sea ice grows most rapidly from its annual minimum in September. For clarity, we will refer to this period as boreal autumn.

To conduct our study, we utilized several publicly available gridded global reanalysis datasets with different time periods and spatial resolutions. For detailed information on each of the datasets used in our analyses, please refer to Table 1, which provides relevant information and references. Furthermore, we employed historical model simulations from 49 global climate models of the Coupled Model Intercomparison Project phase 6 (CMIP6) as described in Eyring et al.62. The IPWP index, which refers to the area-averaged SST in the region spanning from 60°E to 170°E and from 15°S to 15°N, was downloaded from the website https://psl.noaa.gov/gcos_wgsp/Timeseries/Data/pacwarmpool.ersst.data.

We also utilized two derived atmospheric variables to depict the generation and propagation of planetary waves: the Rossby wave source (RWS), which is defined by Sardeshmukh and Hoskins63, and the wave activity flux, described by Takaya and Nakamura64. To define the Eliassen-Palm (EP) flux, we followed the method proposed by Edmon et al.65. The statistical significance of correlation or regression was tested using the two-tailed Student’s t-test.

Effective degrees of freedom for the correlation coefficient is calculated based on the below formula66.

\({N}^{\ast }\) is the effective number of degrees of freedom; N is the sample size; and r1 and \({r}_{2}\) are the autocorrelations of two time coefficients at time lag one year.

We estimate the IPWP-related sea ice trend by multiplying the trend in normalized IPWP index with the sea ice anomalies regressed to the normalized detrended IPWP index:

Where

SIA represents sea ice anomalies

Trend IPWP represents IPWP-related trend

IPWPnorm represents normalized IPWP index

IPWPnorm∙detrend represents normalized detrended IPWP index

To further substantiate our statistical findings, numerical experiments were undertaken utilizing Version 5 of the Community Atmosphere Model (CAM5), an integral component of the Community Earth System Model (CESM). CAM5 features a vertical resolution of 30 levels and a horizontal resolution of 1.9 degrees latitude by 2.5 degrees longitude. Comprehensive details regarding the CAM5 model can be found in the report by Neale et al.67.

Two numerical experiments were conducted: a control run and an idealized experiment. The control run spanned 50 years and was forced by the Hadley Center’s climatological annual cycle of SST and sea ice concentration. The last 20 years served as the reference period for initializing the idealized experiment. For the idealized experiment, a 2 °C SST anomaly was introduced within the IPWP region (80–160°E, 0–20°N) from October 1 to December 31 annually, while the SST and sea ice conditions in the remaining timeframe mirrored climatological norms. As such, we present the difference between the idealized experiment and the control experiment to elucidate the impact of warmer IPWP.

To assess the impact of SST anomalies north of 20°S on Arctic atmospheric circulation, we conducted an idealized experiment using the SST anomalies shown in Fig. 4a. This experiment is similar to the one with a 2 °C SST anomaly, differing only in the specific SST anomalies prescribed. The differences between the idealized and control experiments are also presented.

The monthly Antarctic sea ice concentration data set is available from the (http://nsidc.org/data/NSIDC-0051). Monthly mean atmospheric variables are provided by ERA5 reanalysis (https://doi.org/10.24381/cds.6860a573). The monthly SST data in this study are derived from the NOAA Extended Reconstructed SST V5 (https://www1.ncdc.noaa.gov/pub/data/cmb/ersst/v5/netcdf/). The Niño3.4 index is available from the CPC, NOAA (https://www.cpc.ncep.noaa.gov/data/indices/ersst5.nino.mth.91-20.ascii). The IPWP index can be download from the website (https://psl.noaa.gov/gcos_wgsp/Timeseries/Data/pacwarmpool.ersst.data/). The OLR data are derived from the NOAA Interpolated OLR (https://psl.noaa.gov/data/gridded/data.uninterp_OLR.html). The CMIP6 model data are openly available at the following the website (https://esgf-node.llnl.gov/projects/esgf-llnl/). The sea ice concentration data from 20CR Version 3 are from the below website (https://rda.ucar.edu/datasets/ds131.3/).

Code is available upon request to corresponding author.

Parkinson, C. L. Global sea ice coverage from satellite data: annual cycle and 35-Yr trends. J. Clim. 27, 9377–9382 (2014).

Article Google Scholar

Simmonds, I. & Li, M. Trends and variability in polar sea ice, global atmospheric circulations and baroclinicity. Ann. N. Y. Acad. Sci. 1504, 167–186 (2021).

Article Google Scholar

Notz, D. & Stroeve, J. Observed Arctic sea-ice loss directly follows anthropogenic CO2 emission. Science 354, 747–750 (2016).

Article CAS Google Scholar

Vihma, T. et al. Advances in understanding and parameterization of small-scale physical processes in the marine Arctic climate system: a review. Atmos. Chem. Phys. 14, 9403–9450 (2014).

Article CAS Google Scholar

Wendisch, M. et al. Atmospheric and surface processes, and feedback mechanisms determining Arctic amplification: a review of first results and prospects of the (AC)3 project. Bull. Amer. Meteorol. Soc. 104, 208–242 (2022).

Article Google Scholar

Årthun, M., Eldevik, T., Smedsrud, L. H., Skagseth, Ø. & Ingvaldsen, R. B. Quantifying the influence of Atlantic heat on Barents Sea ice variability and retreat. J. Clim. 25, 4736–4743 (2012).

Article Google Scholar

Kapsch, M. L., Graversen, R. G. & Tjernström, M. Springtime atmospheric energy transport and the control of Arctic summer sea-ice extent. Nat. Clim. Change 3, 744–748 (2013).

Article Google Scholar

Polyakov, I. V. et al. Greater role for Atlantic inflows on sea-ice loss in the Eurasian basin of the Atlantic Ocean. Science 356, 285–291 (2017).

Article CAS Google Scholar

Yu, L. et al. Possible connections of the opposite trends in Arctic and Antarctic sea-ice cover. Sci. Rep. 7, 1–10 (2017).

Google Scholar

Yu, L., Zhong, S. & Sun, B. Synchronous variation patterns of monthly sea ice anomalies at the Arctic and Antarctic. J. Clim. 35, 2823–2847 (2022).

Article Google Scholar

Clement, A. C., Seager, R. & Murtugudde, R. Why are there tropical warm pools? J. Clim. 18, 5294–5311 (2005).

Article Google Scholar

de Deckker, P. The Indo-Pacific warm pool: critical to world oceanography and world climate. Geosci. Lett. 3, 1–12 (2016).

Article Google Scholar

Williams, A. P. & Funk, C. A. A westward extension of the warm pool leads to a westward extension of the Walker circulation, drying eastern Africa. Clim. Dyn. 37, 2417–2435 (2011).

Article Google Scholar

Ratna, S. B. et al. Moisture variability over the Indo-Pacific region and its influence on the Indian summer monsoon rainfall. Clim. Dyn. 46, 949–965 (2016).

Article Google Scholar

Lee, S.-K. et al. Pacific origin of theabrupt increase in Indian Ocean heat content during the warming hiatus. Nat. Geosci. 8, 445–449 (2015).

Article CAS Google Scholar

Marathe, S., Terray, P. & Karumuri, A. Tropical Indian Ocean and ENSO relationships in a changed climate. Clim. Dyn. 56, 3255–3276 (2021).

Article Google Scholar

Vecchi, G. A. & Bond, N. A. The Madden–Julian Oscillation (MJO) and northern high latitude wintertime surface air temperatures. Geophys. Res. Lett. 31, L04104 (2004).

Article Google Scholar

Lee, S., Gong, T., Johnson, N., Feldstein, S. & Pollard, D. On the possible link between tropical convectionand the Northern Hemisphere Arctic surface air temperature change between 1958 and 2001. J Clim. 24, 4350–4367 (2011).

Article Google Scholar

Yoo, C., Feldstein, S. & Lee, S. The impact of the Madden-Julian Oscillation trend on the Arcticamplification of surface air temperature during the 1979–2008 boreal winter. Geophys. Res. Lett. 38, L24804 (2011).

Article Google Scholar

Henderson, G. R., Barrett, B. S. & Lafleur, D. M. Arctic sea ice and the Madden–Julian Oscillation (MJO). Clim. Dyna 43, 2185–2196 (2014).

Article Google Scholar

Weller, E. et al. Human-caused Indo-Pacific warm pool expansion. Sci. Adv 2, e1501719 (2016).

Article Google Scholar

Bai, W., Liu, H., Lin, P., Hu, S. & Wang, F. Indo-Pacfiic warm pool present warming attribution and future projection constraint. Environ. Res. Lett. 17, 054026 (2022).

Article Google Scholar

Roxy, M. K. et al. Twofold expansion of the Indo-Pacific warm pool warps the MJO life cycle. Nature 575, 647–651 (2019).

Article CAS Google Scholar

Jeong, Y.-C., Yeh, S.-W., Lim, Y.-K., Santoso, A. & Wang, G. Indian Ocean warming as key driver of long-term positive trend of Arctic Oscillation. npj Clim. Atmos. Sci. 5, 56 (2022).

Article Google Scholar

Yang, Y.-M. et al. Increased Indian Ocean-North Atlantic Ocean warming chain under greenhouse warming. Nat. Commun. 13, 3978 (2022).

Article CAS Google Scholar

Jeong, H., Park, H.-S., Stuecker, M. & Yeh, S.-W. Record low Arctic sea ice extent in 2012 linked to two‐year la Niña‐Driven sea surface temperature pattern. Geophys. Res. Lett. 49, e2022GL098385 (2022).

Article Google Scholar

Jeong, H., Park, H.-S., Stuecker, M. & Yeh, S.-W. Distinct impacts of major El Niño events on Arctic temperatures due to differences in eastern tropical Pacific sea surface temperatures. Sci. Adv. 8, eabl8278 (2022).

Article Google Scholar

Xie, S.-P. et al. Indo-western Pacific Ocean capacitor and coherent climate anomalies in post-ENSO summer: a review. Adv. Atmos. Sci. 33, 411–432 (2016).

Article Google Scholar

Ding, Q. et al. Influence of high-latitude atmospheric circulation changes on summertime Arctic Sea ice. Nat. Clim. Change 7, 289–295 (2017).

Article Google Scholar

Zhou, X. et al. The effects of the Indo-Pacific warm pool on the stratosphere. Clim. Dyn. 51, 4043–4064 (2017).

Article Google Scholar

Baldwin, M. P. & Dunkerton, T. J. Propagation of the Arctic Oscillation from the stratosphere to the troposphere. J. Geophys. Res. 104, 30937–30946 (1999).

Article Google Scholar

Jaiser, R., Dethloff, K., Handorf, D., Rinke, A. & Cohen, J. Impact of sea ice cover changes on the Northern Hemisphere atmospheric winter circulation. Tellus 64A, 11595 (2012).

Article Google Scholar

Francis, J. A., Chan, W., Leathers, D. J., Miller, J. R. & Veron, D. E. Winter Northern Hemisphere weather patterns remember summer Arctic sea-ice extent. Geophys. Res. Lett. 36, L07503 (2009).

Article Google Scholar

Rasmus, A. P., Cvijanovic, I., Peter, L. L. & Vinther, B. M. The impact of regional Arctic sea ice loss on atmospheric circulation and the NAO. J. Clim. 29, 889–902 (2016).

Article Google Scholar

Kim, B.-M. et al. Weakening of the stratospheric polar vortex by Arctic sea-ice loss. Nat. Commun. 5, 4646 (2014).

Article CAS Google Scholar

Zhang, J., Tian, W., Chipperfield, M. P., Xie, F. & Huang, J. Persistent shift of the Arctic polar vortex towards the Eurasian continent in recent decades. Nat. Clim. Chang. 6, 1094–1099 (2016).

Article Google Scholar

Mantua, N. J., Hare, S., Zhang, Y., Wallace, J. M. & Francis, R. C. A. Pacific interdecadal climate oscillation with impacts on salmon production. B. Am. Meteorol. Soc. 78, 1069–1079 (1997).

2.0.CO;2" data-track-item_id="10.1175/1520-0477(1997)0782.0.CO;2" data-track-value="article reference" data-track-action="article reference" href="https://doi.org/10.1175%2F1520-0477%281997%29078%3C1069%3AAPICOW%3E2.0.CO%3B2" aria-label="Article reference 37" data-doi="10.1175/1520-0477(1997)0782.0.CO;2">Article Google Scholar

Enfield, D. B., Mestas-Nuñez, A. M. & Trimble, P. J. The Atlantic multidecadal oscillation and its relationship to rainfall and river flows in the continential U.S. Geophys. Res. Lett. 28, 2077–2080 (2001).

Article Google Scholar

Ding, Q. et al. Tropical forcing of the recent rapid Arctic warming in northeastern Canada and Greenland. Nature 509, 209–212 (2014).

Article CAS Google Scholar

Yu, L. & Zhong, S. Changes in sea-surface temperature and atmospheric circulation patterns associated with reductions in Arctic sea ice cover in recent decades. Atmos. Chem. Phys. 18, 14149–14159 (2018).

Article Google Scholar

Yu, L., Zhong, S., Zhou, M., Lenschow, D. H. & Sun, B. Revisiting the linkages between the variability of atmospheric circulations and Arctic melt-season sea ice cover at multiple time scales. J. Clim. 32, 1461–1482 (2019).

Article Google Scholar

Meehl, G. A., Chung, C. T. Y., Arblaster, J. M., Holland, M. M. & Bitz, C. M. Tropical decadal variability and the rate of Arctic sea ice decrease. Geophys. Res. Lett. 45, 11326–11333 (2018).

Article Google Scholar

McCrystall, M. R., Hosking, J. S., White, I. P. & Maycock, A. C. The impact of changes in tropical sea surface temperatures over 1979–2012 on Northern Hemisphere high-latitude climate. J. Clim. 33, 5103–5121 (2020).

Article Google Scholar

Topál, D. et al. An internal atmospheric process determining summertime Arctic sea ice melting in the next three decades: lessons learned from five large ensembles and multiple CMIP5 climate simulations. J. Clim. 33, 7431–7454 (2020).

Article Google Scholar

Baxter, I. et al. How tropical pacific surface cooling contributed to accelerated sea ice melt from 2007 to 2012 as ice is thinned by anthropogenic forcing. J. Clim. 32, 8583–8602 (2019).

Article Google Scholar

Matsumura, S. & Kosaka, Y. Arctic-Eurasin climate linkage induced by tropical ocean variability. Nat. Commun. 10, 3441 (2019).

Article Google Scholar

Wettstein, J. & Deser, C. Internal variability in projections of twenty-first-century Arctic sea ice loss: role of the large-scale atmospheric circulation. J. Clim. 27, 527–550 (2014).

Article Google Scholar

Stuecker, M. F. et al. Polar amplification dominated by local forcing and feedbacks. Nat. Clim. Change 8, 1076–1081 (2018).

Article CAS Google Scholar

Lu, J. H. & Cai, M. Quantifying contributions to polar warming amplification in an idealized coupled general circulation model. Clim. Dyn. 34, 669–687 (2010).

Article Google Scholar

Heede, U. K. & Fedorov, A. V. Colder eastern equatorial Pacific and stronger Walker circulation in the early 21st century: separating the forced response to global warming from natural variability. Geophys. Res. Lett. 50, e2022GL101020 (2023).

Article Google Scholar

Yang, H. et al. Satellite-oberved strong subtropical ocean warming as an early signature of global warming. Commun. Earth Environ. 4, 178 (2023).

Article Google Scholar

Chen, S., Wu, R. & Chen, W. A strengthened impact of November Arctic oscillation on subsequent tropical Pacific sea surface temperature variation since the late-1970s. Clim Dyn 51, 511–529 (2018).

Article Google Scholar

Chen, S., Yu, B. & Chen, W. An interdecadal change in the influence of the spring Arctic oscillation on the subsequent ENSO around the early 1970s. Clim Dyn 44, 1109–1126 (2015).

Article Google Scholar

Nakamura, T. et al. A negative phase shift of the winter AO/NAO due to the recent Arctic sea-ice reduction in late autumn. J. Geophys. Res. 120, 3209–3227 (2015).

Article Google Scholar

Hu, M., Gong, D. Y. & Mao, R. Possible influence of February–April Arctic oscillation on the ITCZ activity of western-central Pacific. J. Trop. Meteorol. 29, 55–65 (2013). (Chinese).

Google Scholar

Liu, J. P., Curry, J. A., Wang, H. J., Song, M. R. & Horton, R. M. Impact of declining Arctic sea ice on winter snowfall. Proc. Natl. Acad. Sci. USA 109, 4074–4079 (2012).

Article CAS Google Scholar

Jaiser, R., Dethloff, K., Handorf, D., Rinke, A. & Cohen, J. Impact of sea ice cover changes on the Northern Hemisphere atmospheric winter circulation. Tellus A 64, 595 (2012).

Article Google Scholar

Li, J. P. & Wu, Z. W. Importance of autumn Arctic sea ice to northern winter snowfall. Proc. Natl. Acad. Sci. USA 109, E1898 (2012).

CAS Google Scholar

Choi, K. S., Wu, C. C. & Byun, H. R. Possible connection between summer tropical cyclone frequency and spring Arctic oscillation over East Asia. Clim. Dyn. 38, 2613–2629 (2012).

Article Google Scholar

Nakamura, T., Tachibana, Y., Honda, M. & Yamane, S. Influence of the Northern Hemisphere annular mode on ENSO by modulating westerly wind bursts. Geophys. Res. Lett. 33, L07709 (2006).

Article Google Scholar

Zhou, S. & Miller, A. J. The interaction of the Madden-Julian oscillation and the Arctic Oscillation. J. Clim. 18, 143–159 (2005).

Article Google Scholar

Eyring, V. et al. Overview of the coupled model intercomparison project phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev. 9, 1937–1958 (2016).

Article Google Scholar

Sardeshmukh, P. D. & Hoskins, B. J. The generation of global rotational flow by steady idealizedtropical divergence. J. Atmos. Sci. 45, 1228–1251 (1988).

2.0.CO;2" data-track-item_id="10.1175/1520-0469(1988)0452.0.CO;2" data-track-value="article reference" data-track-action="article reference" href="https://doi.org/10.1175%2F1520-0469%281988%29045%3C1228%3ATGOGRF%3E2.0.CO%3B2" aria-label="Article reference 63" data-doi="10.1175/1520-0469(1988)0452.0.CO;2">Article Google Scholar

Takaya, K. & Nakamura, H. A formulation of a phase in dependent wave-activity flux for stationary and migratory quasi geostrophic eddies on a zonally varying basic flow. J. Atmos. Sci. 58, 608–627 (2001).

2.0.CO;2" data-track-item_id="10.1175/1520-0469(2001)0582.0.CO;2" data-track-value="article reference" data-track-action="article reference" href="https://doi.org/10.1175%2F1520-0469%282001%29058%3C0608%3AAFOAPI%3E2.0.CO%3B2" aria-label="Article reference 64" data-doi="10.1175/1520-0469(2001)0582.0.CO;2">Article Google Scholar

Edmon, H. J. Jr., Hoskins, B. J. & Mcintyre, M. Eliassen-Palm cross sections for the troposphere. J. Atmos. Sci. 37, 2600–2616 (1980).

2.0.CO;2" data-track-item_id="10.1175/1520-0469(1980)0372.0.CO;2" data-track-value="article reference" data-track-action="article reference" href="https://doi.org/10.1175%2F1520-0469%281980%29037%3C2600%3AEPCSFT%3E2.0.CO%3B2" aria-label="Article reference 65" data-doi="10.1175/1520-0469(1980)0372.0.CO;2">Article Google Scholar

Afyouni, S., Smith, S. M. & Nichols, T. E. Effective degrees of freedom of the Pearson’s correlation coefficient under autocorrelation. Neurolmage 199, 609–625 (2019).

Article Google Scholar

Neale, R. B. et al. Description of the NCAR community atmosphere Model (CAM5). Technical Report NCAR/TN-486+STR. Boulder, Colorado: National Center for Atmospheric Research, 268 (2011).

Huang, B. et al. Extended reconstructed sea surface temperature version 5 (ERSSTv5), upgrades, validations, and intercomparisons. J. Clim. 30, 8179–8205 (2017).

Article Google Scholar

Cavalieri, D. J., Parkinson, C. L. & Gloersen, P. Sea ice concentrations from Nimbus-7 SMMR andDMSP SSM/I-SSMIS passive microwave data, Version 1 (NASA). Boulder, CO: National Snow and Ice Data Center Distributed Active Archive Center (last access: 3 June 2021) (1996).

Slivinski, L. C. et al. An evaluation of the performance of the twentieth century reanalysis version 3. J. Clim. 34, 1417–1438 (2021).

Article Google Scholar

Hersbach, H. et al. The ERA5 global reanalysis. Q. J. Roy. Meteor. Soc. 146, 1999–2049 (2020).

Article Google Scholar

Liebmann, B. & Simth, C. A. Description of a complete (interpolated) outgoing longwave radiation dataset. B. Am. Meteor. Soc. 77, 1275–1277 (1996).

Google Scholar

Download references

We thank the European Centre for Medium-Range Weather Forecasts (ECMWF) for the ERA5 data. This study is financially supported by, the National Key R&D Program of China (2022YFE0106300), and the European Commission H2020 project Polar Regions in the Earth System (PolarRES; Grant101003590).

MNR Key Laboratory for Polar Science, Polar Research Institute of China, Shanghai, China

Lejiang Yu & Bo Sun

Department of Geography, Environment and Spatial Sciences, Michigan State University, East Lansing, MI, USA

Shiyuan Zhong

Finnish Meteorological Institute, Helsinki, Finland

Timo Vihma

Department of Atmospheric and Oceanic Sciences, and Institute of Atmospheric Sciences, Fudan University, Shanghai, China

Shuoyi Ding

National Marine Environmental Forecasting Center, Beijing, China

Cuijuan Sui

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

Lejiang Yu designed the research, analyzed the data, and wrote the first draft of the paper. Shiyuan Zhong and Timo Vihma revised the first draft and provided useful insights during various stages of the work. Shouyi Ding and Cuijuan Sui helped with the sea ice data analysis and CAM5 model simulation. Bo Sun provided some comments and helped with editing the paper. All authors reviewed the manuscript.

Correspondence to Lejiang Yu.

The authors declare no competing interests.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

Yu, L., Zhong, S., Vihma, T. et al. The IPWP as a capacitor for autumn sea ice loss in Northeastern Canada. npj Clim Atmos Sci 7, 259 (2024). https://doi.org/10.1038/s41612-024-00798-9

Download citation

Received: 21 November 2023

Accepted: 07 October 2024

Published: 25 October 2024

DOI: https://doi.org/10.1038/s41612-024-00798-9

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative