Greenland's warming and subsequent ice loss significantly contribute to global sea-level rise. Until 2012, warming was accelerated by an intensified anticyclone linked to the tropical Pacific. This anticyclone, associated with the negative phase of the North Atlantic Oscillation (NAO), promoted warmer conditions and enhanced ice melt by reducing cloud cover and surface albedo. However, Greenland's warming and ice loss have slowed in the last decade, while Arctic warming continued elsewhere. This uneven warming is likely due to natural variability, which most climate models struggle to simulate accurately. Atmospheric teleconnections from the tropics are crucial to Arctic natural variability, but most studies focus on winter teleconnections. While summer tropical SSTs can influence the Arctic via atmospheric teleconnections, the mechanism remains unclear due to weak subtropical jets and the northward displacement of westerly winds. This study investigates the mechanism behind the recent slowdown of Greenland warming by focusing on summertime teleconnections from the tropics.

Previous research highlighted the role of intensified anticyclonic circulation and a negative NAO in accelerating Greenland warming until 2012, primarily focusing on the influence of wintertime teleconnections from the tropics. While some studies suggested that summer tropical SSTs affect Arctic sea ice through atmospheric teleconnections, most climate models failed to replicate this. The impact of summer tropical variability on high-latitude atmospheric circulation remains poorly understood due to the prevailing easterly winds near the equator that typically inhibit remote forcing from the tropics. This study builds upon this existing literature by focusing specifically on summer teleconnections and their role in the recent slowdown of Greenland warming.

The study utilized observational data from JRA55 reanalysis, HadISST, NSIDC, GPCP, DMI synoptic stations, GHCN CAMS, and ERA5 reanalysis. The Greenland Blocking Index (GBI) was used to quantify the relationship between Greenland temperature and atmospheric circulation. Empirical orthogonal function (EOF) analysis was employed to identify major modes of upper tropospheric circulation variations. ECHAM5 atmospheric model simulations, with observed and 1880s radiative forcing, were used for comparison. The El Niño Modoki Index (EMI) represented central Pacific El Niño events. A linear baroclinic model (LBM) was used to diagnose the atmospheric circulation response to tropical heating, isolating the impact of tropical heating from extratropical factors. The LBM experiments involved applying idealized heating at different locations and analyzing the resulting atmospheric responses, including changes in geopotential height and temperature.

The study found a slowdown in Greenland summer warming since 2012, despite ongoing anthropogenic greenhouse gas emissions. ECHAM5 simulations accurately reproduced the warming trend but not the recent slowdown, suggesting a significant role for non-anthropogenic forcing. The Greenland Blocking Index (GBI), which peaked in 2012, decreased afterward, coinciding with the temperature anomaly. EOF analysis revealed a Pacific-North America (PNA)-like teleconnection, closely linked to decadal variability and inversely correlated with Greenland temperature and GBI. This teleconnection showed an upward trend over the last decade, corresponding to cooling over northeastern Canada and Greenland, North Atlantic SST cooling, and tropical Pacific SST warming characteristic of Central Pacific (CP) El Niño events, which have become more frequent than Eastern Pacific (EP) events since the late 1990s. Comparison of periods before and after 2000 revealed a distinct wave train propagation from the tropical Pacific towards Greenland during the later period, with the negative anomaly extending into the Arctic Ocean. The LBM successfully simulated the teleconnection, highlighting the role of increased rainfall north of the ITCZ in driving Rossby waves towards Greenland. Removal of the EMI-related component eliminated the intensified cyclonic circulation and Greenland cooling, indicating the critical role of CP El Niño. The intensified cyclonic circulation related to EMI extended into the Arctic Ocean in observations, possibly impacting Arctic sea ice through downwelling longwave radiation and ice transport. The timing of the abrupt slowdown of September sea ice loss corresponded to the upward EMI trend since the late 2000s. While the LBM reproduced the Greenland teleconnection, the amplification of the cyclonic wave to the Arctic Ocean was not simulated, implying an indirect forcing mechanism might be involved.

The findings address the research question by identifying the frequent occurrence of Central Pacific El Niño events and the associated shift in tropical rainfall as the primary driver of the slowdown in Greenland summer warming. The successful simulation of this teleconnection using a simplified model strengthens the causal link. The results highlight the importance of accurately representing tropical rainfall patterns in climate models for better simulation of Arctic climate. The observed discrepancy between model simulations and observations regarding Arctic cyclonic wave amplification suggests the involvement of indirect forcing mechanisms or internal atmospheric variability.

This study demonstrates a summertime teleconnection from the tropics to Greenland driven by frequent Central Pacific El Niño events, leading to the slowdown in Greenland warming. The model successfully captures this teleconnection, highlighting the crucial role of tropical rainfall patterns. Improving the representation of tropical rainfall in climate models is essential for more accurate Arctic climate simulations. Future research should focus on further investigating the indirect forcing mechanisms contributing to Arctic cyclonic wave amplification and on understanding the role of both natural variability and anthropogenic forcing in the changing frequency of CP El Niño events.

The study primarily focuses on summer conditions, neglecting potential interactions with other seasons. The simplified atmospheric circulation model used may not fully capture the complexities of the climate system. While the study suggests a link between CP El Niño and Arctic sea ice loss, the exact mechanisms and the relative contributions of direct and indirect forcing require further investigation. The nonstationary relationship between the subtropical Pacific and Arctic sea ice needs further exploration.