New climate models reveal faster and larger increases in Arctic precipitation than previously projected

The Arctic is experiencing dramatic environmental changes due to its disproportionately rapid warming compared to the global average. A key aspect of this change is the intensification of the hydrological cycle, characterized by increased evaporation from expanding open water areas (resulting from sea ice loss) and a projected increase in precipitation. Previous studies, primarily based on CMIP5 (Coupled Model Intercomparison Project Phase 5) climate models, estimated Arctic precipitation increases ranging from 30% to 60% by 2100. These increases were attributed to several factors: increased evaporation from open water, higher air temperatures enhancing the atmosphere's moisture-carrying capacity, and increased poleward moisture transport. Furthermore, a transition from a snow-dominated to a rain-dominated precipitation regime was anticipated. However, uncertainties remained regarding the regional extent and seasonality of these changes. The availability of data from CMIP6, which offers improved simulations of sea-ice, snow cover, and global precipitation intensities compared to CMIP5, provides an opportunity to refine projections of Arctic precipitation change and assess the associated uncertainties. This study focuses on comparing CMIP6 and CMIP5 projections of Arctic precipitation change through 2100, with a particular emphasis on the transition to a rain-dominated regime and the underlying drivers of these changes. The implications of these projected changes on various aspects of the Arctic environment, including the Greenland ice sheet, sea ice extent, permafrost, and ecosystems, are substantial and necessitate a thorough investigation.

Existing literature consistently points towards an increase in Arctic precipitation throughout the 21st century. Estimates vary, but a significant rise is expected, ranging from 30% to 60% by 2100. Several studies have linked this anticipated increase to several factors: amplified evaporation due to the reduction of sea ice, leading to more open water; rising air temperatures, increasing the atmosphere's capacity to hold moisture; and intensified poleward moisture transport. The shift towards a rain-dominated Arctic, particularly during the warmer months, has also been highlighted, with observations already supporting this transition in the Atlantic sector. The literature explores the implications of a wetter Arctic, anticipating significant impacts on Greenland ice sheet mass balance and global sea level, river discharge, Arctic sea-ice extent and thickness, permafrost thaw, and the broader Arctic flora and fauna. However, regional variations and seasonal patterns remain sources of uncertainty. While the consensus points towards increased rainfall in autumn and winter, discrepancies persist in spring precipitation projections, indicating a need for improved modeling and data analysis.

This study examines projections of Arctic precipitation change through 2100 using data from CMIP6 and CMIP5 climate models. The analysis focuses on total precipitation, snowfall, and rainfall (derived as the difference between total precipitation and snowfall) for four seasons: December–February (DJF), March–May (MAM), June–August (JJA), and September–November (SON). The Arctic region is defined as the area north of 70°N. The highest emissions scenarios (RCP8.5 for CMIP5 and SSP5-8.5 for CMIP6) were used for future projections. The researchers compared the CMIP6 and CMIP5 model outputs with observational data from ERA5 reanalysis and the Global Precipitation Climatology Project (GPCP). Historical changes (1979–2005) were assessed, examining both annual-mean and seasonal precipitation patterns and the snowfall-to-precipitation ratio. End-of-century (2091–2100) projections relative to the start of the century (2005–2014) were computed for precipitation types, open water area, surface air temperature, and vertically integrated moisture flux (VIMF). Statistical analyses, including correlation analyses and Student's t-tests, were employed to assess the significance of the findings. The VIMF was calculated using equation (1) from the paper, which involves the acceleration due to gravity, surface pressure, pressure at 400 hPa, meridional wind component, and specific humidity. The transition decade to a rain-dominated regime was defined as the decade when the annual or seasonal snowfall-to-total precipitation ratio fell below 50%, using 10-year windows. The impact of global warming on the snowfall-to-precipitation ratio was investigated using global temperature anomalies. The paper notes the importance of staying within 1.5°C or 2°C global warming to mitigate climate change, and compares snowfall ratios at different warming levels (1.5°C, 2°C, and 3°C).

The key findings of the study demonstrate that CMIP6 models project significantly larger and faster increases in Arctic precipitation compared to CMIP5 models. This difference is particularly pronounced in autumn, where CMIP6 projects a 422% increase in rainfall by 2100 compared to 260% in CMIP5. The transition to a rain-dominated Arctic in summer and autumn is projected to occur much earlier in CMIP6, approximately one to two decades earlier than in CMIP5. This earlier shift is evident across much of the Arctic Ocean, Siberia, and the Canadian Archipelago. CMIP6 also projects a more rapid reduction in the snowfall-to-precipitation ratio. The increased precipitation in CMIP6 is attributed to a combination of factors: stronger projected increases in global and Arctic air temperature; more rapid sea-ice decline resulting in larger areas of open water; and a greater sensitivity of precipitation to warming, especially in autumn. The correlation analysis showed a significant relationship between the magnitude of rainfall increases (and snowfall decreases) and factors like Arctic warming, open water area, and VIMF in both CMIP ensembles, but more prominent in CMIP6. When analyzing precipitation changes relative to specific global warming levels (1.5°C, 2°C, and 3°C), CMIP6 consistently projects more rainfall at all warming levels. The analysis of the transition to a rain-dominated regime reveals that under a 1.5°C or 2°C warming scenario, some regions like the Beaufort, Chukchi, Bering, Laptev, and East Siberian Seas may remain snow-dominated, while others, such as the Greenland and Norwegian Seas, are likely to transition regardless of the warming level. At 3°C warming, most Arctic regions would transition to rainfall-dominated precipitation, except possibly those on the Pacific side, but with winter and spring largely remaining snow-dominated.

The findings highlight a substantial discrepancy between CMIP5 and CMIP6 projections of Arctic precipitation changes, with CMIP6 indicating a more rapid and intense shift towards a rain-dominated regime. This discrepancy is not solely attributable to the larger global warming projected in CMIP6, but also to increased Arctic amplification, faster sea-ice loss, and greater precipitation sensitivity to warming. The earlier transition to a rain-dominated Arctic has significant implications for Arctic ecosystems and human societies. The reduced snow cover duration will alter seasonality, affect tundra greening, impact wildlife populations, and influence human livelihoods, leading to increased risks such as flooding due to changing soil moisture and groundwater conditions. The increased frequency of rain-on-snow events poses a substantial threat to caribou, reindeer, and muskoxen populations. While some potential benefits, such as increased migratory bird populations, exist, the negative impacts far outweigh the positive ones. The changes in precipitation and snow cover will impact sea-ice extent and thickness, affecting the Arctic Ocean's physical properties and primary productivity, leading to cascading changes through marine food webs. The projected increased winter snowfall over Greenland could potentially mitigate mass loss from increased melting in central parts of the ice sheet, but greater rainfall at the southern and coastal edges may accelerate ice loss. The study reinforces the urgency of stringent mitigation policies to limit global warming, as projected changes previously anticipated at 2°C global warming now appear plausible even at 1.5°C.

This study demonstrates that the latest generation of climate models (CMIP6) project significantly larger and faster increases in Arctic precipitation and an earlier shift to a rain-dominated regime compared to previous models (CMIP5). This accelerated transition, primarily driven by increased warming, sea-ice loss, and greater precipitation sensitivity to warming, underscores the need for more aggressive climate change mitigation efforts. Further research should focus on refining regional projections, improving understanding of the impacts on specific ecosystems and human communities, and developing adaptation strategies to mitigate the negative consequences of these profound environmental changes.

The study relies on climate model outputs, which inherently have limitations. While CMIP6 represents an improvement over CMIP5, uncertainties remain in the models' representation of Arctic processes, particularly concerning cloud cover, precipitation processes, and sea-ice dynamics. The use of a single ensemble member per model might not fully capture the range of model uncertainty. Additionally, the observational data used for comparison (ERA5 and GPCP) also have limitations, including potential biases in high-latitude precipitation measurements. The interpretation of the results should consider these inherent uncertainties.