Arctic sea ice has been dramatically declining due to global warming, with feedback mechanisms accelerating the warming trend. The future climate trends associated with sea ice melting remain uncertain due to a lack of understanding of the retreat mechanisms and the possibility of a future with ice-free Arctic summers. Paleoclimate studies, by examining warmer periods in the past like the mid-Holocene (MH), can provide insights into potential future scenarios. Previous research has attributed Holocene Arctic sea ice changes to variations in solar insolation, poleward moisture transport and associated latent heat, and inflows of relatively warm water from the Atlantic and Pacific Oceans. More recent studies, based on limited instrumental records and numerical simulations, have also suggested a role for pan-Arctic rivers in delivering heat from continents into the Arctic Ocean through the discharge of warmer waters. In early summer (June-July), when solar energy input is at its peak, the thermal flux from pan-Arctic rivers can be substantial, representing approximately 10% of the total heat input from the Atlantic and Pacific Oceans into the Arctic Ocean. While seemingly small, this discharge is concentrated and the warm freshwater directly interacts with Arctic sea ice due to buoyancy. The Arctic climate during the MH summer was warmer, had less sea ice, and experienced stronger solar insolation compared to the present. A stronger thermal flux associated with pan-Arctic river discharge would have contributed to the MH summer sea ice loss, a mechanism largely overlooked until this study. The East Siberian Arctic Shelf (ESAS), with its shallow waters and significant Russian pan-Arctic river discharge, is an ideal region to investigate the thermal impact of pan-Arctic rivers on Arctic sea ice.

Existing literature on Arctic sea ice retreat primarily focuses on the influence of extrapolar thermal energy import. Studies have linked changes in Arctic sea ice extent during the Holocene to various factors, including solar insolation, poleward moisture transport with latent heat transfer, and warm water inflows from the Atlantic and Pacific Oceans. Some research has begun to explore the contribution of pan-Arctic river heat discharge to recent sea ice decline, using limited instrumental records and numerical modeling. However, a comprehensive understanding of the role of pan-Arctic river heat discharge in past and future Arctic sea ice loss has remained elusive. This study addresses this gap by investigating the relationship between pan-Arctic river heat discharge, early summer solar insolation, and Arctic sea ice extent during the Holocene, leveraging both modern observational data and paleoclimate proxies.

This study reconstructed the Holocene variations in Arctic sea ice and Russian pan-Arctic river heat discharge using multiple proxies. Ice-rafted debris (IRD) records, primarily composed of sand sediment (>63 µm), served as a proxy for sea ice extent. Sedimentation rates in the ESAS region, mainly controlled by river material supply, were used as a proxy for river thermal discharge. Two sediment cores (LV77-36-1 and LV77-41-1) were collected from the ESAS and analyzed. Age-depth models were established using radiocarbon (¹⁴C) dating of bivalve shells and optically stimulated luminescence (OSL) dating of quartz. Rare earth element data were used to trace the core sediment source. The study also synthesized published data on sedimentation rates, sea ice extent, and other climate variables to provide a comprehensive picture of Holocene Arctic climate variability. The OSL dating results were recalibrated using a high-confidence polynomial equation to align with the calibrated ¹⁴C ages, improving the accuracy of the age-depth models and enabling direct comparison between the MH and LH. The chosen cores were located in areas with relatively stable paleo-coastal lines since the MH, minimizing the impact of sea-level change on the results. Additional data on surface clay minerals, Holocene carbon isotopes, and terrigenous biomarkers were integrated to corroborate the findings. Modern instrumental records of pan-Arctic river water discharge and temperature were also utilized to establish correlations between freshwater warming trends and surface air temperature.

The IRD records indicate an increasing trend in sea ice from the MH to the LH in the ESAS region. This finding is consistent with IRD records from the Beaufort Sea and Fram Strait, suggesting that the ESAS sea ice export was a major factor influencing these records. Conversely, dinocyst-based reconstructions of sea ice in Alaskan marginal seas show longer sea ice cover during the MH compared to the LH. This disparity suggests that external western Arctic sea ice did not significantly influence the ESAS IRD records. Regional sea ice biomarker values also show a gradual increase from the MH to the LH, suggesting more intense sea ice melting in the MH spring and summer. Micropaleontological records indicate sea ice-free conditions in the ESAS during the MH summer. Sedimentation rates in the ESAS were higher in the MH than in the LH, suggesting a larger input of Russian pan-Arctic river discharge water during the MH. This increase is further supported by higher total organic carbon levels in the shelf sediments during the MH. The enhanced Russian pan-Arctic river heat discharge during the MH resulted from both increased river runoff and higher freshwater temperatures. Stronger summer solar insolation in the middle and high latitudes during the MH led to higher surface air temperatures, intensified thawing of land snow/ice and permafrost, and increased river basin precipitation. Paleo-pollen records suggest increased regional precipitation in high-latitude Russia and mid-latitude Asia during the MH. This increased precipitation, along with the enhanced thawing of permafrost, contributed to higher levels of river freshwater discharge and sediment transport into the ESAS. The reconstructed sea ice growth since the MH aligns with the decreasing trend of early summer (June-July) solar insolation at 75°N. This finding highlights the crucial role of early summer solar heat energy in determining Arctic sea ice growth. The significant loss of Arctic sea ice during the MH, along with sediment- and phytoplankton-rich freshwater, notably reduced the early summer sea ice albedo, resulting in higher absorption of solar heat energy. This created a positive feedback loop, further accelerating sea ice melt and expanding ice-free regions in early summer. The study notes that Atlantic and Pacific water inflows to the Arctic, while influencing sea ice melt, typically peak later in the season (winter and August-October) compared to the pan-Arctic river heat discharge, which peaks in early summer. The relatively stable Atlantic and Pacific inflows throughout the Holocene further emphasize the role of solar insolation and river heat discharge in the MH sea ice reduction.

The study's findings demonstrate that the enhanced Russian pan-Arctic river heat discharge, driven by increased early summer solar insolation during the MH, significantly contributed to the melting of Arctic sea ice. This mechanism operated through increased river runoff, higher freshwater temperatures, and reduced sea ice albedo, creating a positive feedback loop. The results highlight the importance of considering the indirect effects of increased solar radiation on Arctic sea ice through enhanced river discharge. While other factors such as Atlantic and Pacific water inflows also contribute to Arctic sea ice melt, their influence is shown to be less dominant compared to early-summer river heat discharge. The findings also underscore the need for incorporating the feedback mechanisms between permafrost thawing, increased river runoff, and sea ice decline into climate models, as this is currently not well represented.

This study provides compelling evidence that the increased heat discharge from Russian pan-Arctic rivers, amplified by higher early summer solar insolation, played a significant role in the enhanced Arctic sea ice melting during the mid-Holocene. The findings emphasize the importance of considering this mechanism in future climate projections and improving the representation of permafrost-related feedback processes in climate models. Future research could focus on spatially extending the findings to the entire Arctic Ocean and refining the modeling of this complex interaction between river discharge, solar radiation, and sea ice extent under different warming scenarios.

The study's focus on the ESAS region limits the generalizability of the findings to the entire Arctic Ocean. The proxies used may not capture all aspects of sea ice and river discharge variability. While the study successfully integrated various data sources, further improvements in the accuracy of age-depth models, particularly for OSL dating, could enhance the reliability of the results.