image: Schematic diagram illustrating the combined pathways by which cold sea surface temperature (SST) anomalies in the tropical central-east Pacific (TCEP; blue rectangle) and warm SST anomalies in the South Indian Ocean (SIO; red rectangle) drive sea ice reduction in the Okhotsk and Bering Seas (yellow rectangle). The TCEP cold SST anomaly enhances the Walker Circulation (blue dashed arrows), causing anomalous downward motion over the tropical Indian Ocean, which subsequently triggers a Rossby wave train (gray arrows) that induces an anomalous anticyclone over the Sea of Okhotsk (light purple semicircle). In parallel, the SIO warm SST anomaly generates local ascent, driving an anomalous Hadley Cell (red dashed arrows); the resultant downward motion anomalies over the tropical Indian Ocean further reinforce the similar Rossby wave train (gray arrows), thereby strengthening the Okhotsk anticyclone (light purple semicircle). These synergistic anticyclonic anomalies over the Okhotsk and Bering Seas create atmospheric conditions conducive to sea ice melt. Anticyclones and cyclones are denoted by purple and green circles, marked “AC” and “C”, respectively.
Credit: ©Science China Press
In February and March 2025, Arctic sea ice extent fell to record lows for those months, reaching 13.75 and 14.14 million square kilometers, respectively. The event occurred against a backdrop of a decadal slowdown in Arctic sea ice loss since the 2010s, making the 2025 winter minimum particularly notable.
Previous research on Arctic sea ice extremes has focused mainly on summer minima, especially the September record lows. Winter extremes have received less attention, even though winter sea ice can precondition the following summer minimum and influence mid-latitude climate through atmospheric teleconnections.
The winter of 2024/2025 featured only a weak La Niña, with modest cold sea surface temperature anomalies in the tropical Pacific. La Niña can affect the Aleutian Low and Arctic sea ice through teleconnections, but its influence alone appeared insufficient to explain the magnitude of the 2025 sea ice loss. Researchers therefore looked beyond the tropical Pacific and identified a significant warm sea surface temperature anomaly in the Southern Indian Ocean.
The study, led by researchers at The Hong Kong University of Science and Technology, examined the combined role of La Niña and the Southern Indian Ocean warm anomaly (SIOW). Using satellite observations, reanalysis data and dynamical diagnostics, together with simulations from a linear baroclinic model, the team showed that the two signals are largely independent, with a correlation of only 0.09. Yet each can induce an anticyclonic anomaly over the Sea of Okhotsk–Bering Strait region and reduce sea ice. When they occur together, the anticyclone is stronger and sea ice loss is greater.
Mechanistically, La Niña perturbs the Walker circulation, producing anomalous descending motion and upper-level convergence over the tropical Indian Ocean. SIOW alters the Hadley circulation and also generates descending motion and convergence over the tropical Indian Ocean. The two signals converge over the tropical Indian Ocean and jointly excite a northeast-propagating Rossby wave train that extends through southern China and Japan to the Sea of Okhotsk, where it produces a significant anticyclonic anomaly. This anticyclone enhances poleward heat and moisture transport and downward longwave radiation, accelerating local sea ice melt.
To test the independent role of SIOW, the researchers used a linear baroclinic model with prescribed anomalous diabatic heating over the Southern Indian Ocean warm anomaly region. The simulations reproduced the weakened Hadley circulation, tropical Indian Ocean descent and northeast-propagating Rossby wave train seen in observations and generated an anticyclonic anomaly over the Sea of Okhotsk, supporting the observationally derived mechanism.
Statistically, La Niña alone explains about 20% of the interannual variability in sea ice extent, while SIOW explains about 13%. A combined Niño3–SIO index correlates with sea ice extent at 0.60, corresponding to roughly 36% of the variance, indicating that SIOW adds explanatory power beyond ENSO alone. Among the four years on record when La Niña and SIOW co-occurred, three (1981, 2011 and 2025) rank among the lowest sea ice extent years on interannual timescales.
The record low in 2025 reflects the superposition of the long-term decline driven by anthropogenic forcing and the interannual extreme low. The study also notes that ENSO and SIOW explain only part of extreme events; not every La Niña or SIOW leads to extreme sea ice loss. Decadal variability and sea ice changes outside the Pacific sector, such as in the Barents–Kara Seas, may also contribute and warrant further investigation.
The findings reveal a cross-basin atmospheric bridge initiated by La Niña's zonal perturbation of the Walker circulation and SIOW's meridional alteration of the Hadley circulation. Their forcing converges over the tropical Indian Ocean, exciting a northeast-propagating Rossby wave that establishes an equivalent barotropic anticyclonic anomaly over the Sea of Okhotsk. The study highlights the essential role of synergistic Indo-Pacific forcing in driving extreme Arctic sea ice loss superimposed on a long-term decline and offers new clues for improving seasonal prediction of such extreme cryosphere events.
The study was led by first author Cen Wang and corresponding author Prof. Hui Su at The Hong Kong University of Science and Technology, with co-authors including Chengxing Zhai, Shiwei Yu, Huisi Mo, Luancheng Xu and Shuangchen Du from The Hong Kong University of Science and Technology, and Jianqiu Zheng and Jianan Huo from the University of Science and Technology of China.
The study was supported by the Hong Kong Jockey Club Charities Trust (FA123), the Innovation and Technology Commission (P0413), the Innovation and Technology Fund (P0456 and P0491), the Research Grants Council (16300823, 16308924 and 16308825), the NSFC/RGC Collaborative Research Scheme (N_HKUST680/25) and the State Key Laboratory of Climate Resilience for Coastal Cities (ITC-SKLCRCC26EG01).
Journal
National Science Review
Method of Research
Data/statistical analysis