The Arctic is experiencing rapid warming and sea ice loss, with an ice-free Arctic projected by the end of this century. Arctic cyclones, synoptic weather events transporting heat and moisture into the Arctic, play a crucial role in sea ice dynamics and the climate system. However, the influence of climate change on these cyclones remains poorly understood. This study addresses this gap by employing high-resolution (4 km) regional modeling, combined with downscaled global climate reconstructions and projections from CMIP6, to analyze how recent and future climate changes modify cyclone behavior during spring (March and April). Spring is a critical period for sea ice-atmosphere interactions, influencing sea ice survivability and preconditioning it for the summer melt season. The study uses a Lagrangian cyclone finding and tracking scheme, with case studies validated against data from the 2019-2020 Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAIC) expedition. The research focuses on understanding how changes in large-scale dynamic and thermodynamic environments affect storm-scale changes in cyclone characteristics, addressing whether sufficient climate change has already impacted spring Arctic cyclones and what future changes are expected with continued warming and sea ice loss.

Recent decades have witnessed a rapid decline in Arctic sea ice extent, concentration, thickness, and snowpack depth, alongside a lengthening melt season and a shift from perennial to seasonal ice. This sea ice decline has significant feedback effects on atmospheric processes, potentially influencing Arctic cyclone intensification and behavior. Studies have shown an increase in Arctic Eady growth rates (rapid increases in baroclinic disturbance) in winter and spring due to decreased low-level static stability. While global climate models (GCMs) from CMIP6 project continued Arctic sea ice loss and Arctic amplification, they exhibit substantial biases and discrepancies in projecting Arctic change under different climate scenarios and particularly regarding winter cyclone activity. These discrepancies may stem from the coarse spatial resolution of global models, inadequate to accurately represent local-scale processes and cyclone characteristics. Previous research suggests that altered land-sea temperature contrasts and Arctic atmospheric thermal structures might increase baroclinicity, leading to enhanced cyclogenesis and cyclone activity. Increased surface turbulent fluxes with climate change may also foster overturning and deep convection, resulting in more intense cyclones. However, observational data are sparse, making it difficult to draw conclusive insights on long-term trends in Arctic cyclone activity from observational and reanalysis data alone. Existing literature reveals significant spatio-temporal variability in cyclone frequency and characteristic changes, with conflicting findings on future Arctic cyclone activity.

This study utilizes a regional configuration of the Weather Research and Forecasting (WRF) model (v3.9.1.1) at convection-permitting resolutions (≤4 km) to simulate Arctic cyclones during spring. High spatial and temporal resolutions are crucial for accurately capturing cyclone intensification, decay, and complex interactions with the local environment. The study employs a “Storylines” approach and a Pseudo Global Warming (PGW) technique to assess the contributions of climate change to extreme events. Large-scale climate perturbations are derived from CMIP6 and downscaled onto the regional modeling framework to simulate cyclones under historical (1885–1914), current (1985–2014), and future (2070–2099) climate conditions. The PGW technique involves perturbing the current climate boundary and initial conditions using climate change deltas calculated from CMIP6 multi-model ensembles. This isolates the response of cyclone characteristics to changes in the large-scale climate system. The CMIP6 30-year climatologies used for delta calculations are derived from 8 model realizations across 6 modeling centers under the SSP5-8.5 scenario, representing the highest emission scenario in CMIP6. Nine Arctic cyclones (six in March and three in April) identified using the Melbourne University cyclone finding and tracking scheme applied to ERA5 SLP data were selected as case studies. The selection criteria included timing (spring), geographic distribution, and intensity. Three of the cyclones (A-C) coincided with the MOSAIC expedition, enabling validation against in-situ observational data. The WRF model was configured using specific physical parameterizations (RRTMG radiation, Eta similarity surface layer, Unified Noah Land Surface Model, Mellor-Yamada-Janjic PBL, Morrison 2-moment microphysics), with daily updates of sea ice concentration (SIC) and sea surface temperatures (SSTs). Simulations were initialized shortly after cyclone genesis and run to capture their full lifecycle. The WRF model simulations were evaluated by comparing simulated atmospheric pressure and temperature to MOSAIC expedition data.

The analysis of climate change deltas reveals a decrease in sea ice concentration and an increase in surface and atmospheric temperatures since the pre-industrial era. These changes are particularly pronounced in the Barents Sea region. By the end of the 21st century, even more significant sea ice decline and warming are projected. The study finds that recent climate change has not yet significantly impacted spring Arctic cyclone characteristics. Cyclones in the historical climate show slightly higher sea ice concentrations and lower surface temperatures compared to the current climate, resulting in minimal changes in air-surface temperature differences and surface turbulent fluxes. Cyclone intensities and lifecycles remain largely unchanged. However, future climate conditions show a substantial response. With projected sea ice loss and warming, cyclones will traverse areas with drastically reduced sea ice and higher surface temperatures. This leads to significant increases in the near-surface temperature gradient and surface turbulent fluxes (sensible and latent heat). The increased vertical temperature gradient, moisture availability, and atmospheric water vapor capacity result in a dramatic enhancement of these fluxes. The future climate also supports the development of convective available potential energy (CAPE), indicating atmospheric instability and energy for convection and overturning. While cyclone size does not consistently change, the thermodynamic alterations result in increased and prolonged cyclone intensity, with enhanced wind speeds (up to 17 m s⁻¹ increases), decreased minimum sea level pressure, and extended periods at maximum intensity. Precipitation is also significantly affected. In March, snowfall increases considerably, while warmer near-surface air temperatures bring cyclones closer to the freezing/melting point. This facilitates more mixed-phase and liquid precipitation, with small amounts of rain even at high latitudes. In April, when air temperatures are more likely to be above freezing, the precipitation shifts decisively to rainfall. These changes have significant implications for sea ice mass balance: increased snowfall in March could temporarily delay melt, but warming temperatures and rain-on-snow events could accelerate it. Finally, changes in upper-level atmospheric steering flows alter cyclone trajectories, potentially expanding their geographical reach and exposing previously unaffected areas to intense cyclone activity.

The findings address the research questions by demonstrating that while recent climate change has yet to substantially alter spring Arctic cyclone characteristics, future climate change will significantly impact their intensity and trajectories. The lack of appreciable impact from recent climate change is linked to the fact that, despite warming, temperatures remain largely below freezing in spring, and substantial sea ice changes are limited to marginal seas. However, the projected future sea ice loss and warming will lead to substantial changes in the air-surface temperature gradients and resulting fluxes, driving intensified cyclones. The increased intensity and extended durations of these cyclones, coupled with their expanded geographical reach, will have profound implications. Increased snowfall might provide a temporary buffer to sea ice melt, but warmer temperatures and rainfall could outweigh this effect. Increased wind speeds and ocean mixing will enhance the positive feedback loop between cyclone intensity and climate change. The compounding of extreme weather variables (wind, precipitation, temperature) will intensify the stress on vulnerable ecosystems and coastal communities. The study's implications are significant for Arctic ecosystems, communities dependent on sea ice, and commercial activities.

This study reveals a complex interaction between Arctic cyclones and climate change. While recent climate change has not yet significantly altered spring cyclone characteristics, future climate change will dramatically increase cyclone intensity and alter their trajectories, expanding their influence to previously unaffected areas. The findings underscore the importance of mitigating climate change and sea ice loss to lessen the impacts of extreme polar weather events. Further research is needed to investigate the seasonal variations in cyclone responses and explore other climate scenarios.

The study uses a pseudo-global warming approach with inherent assumptions. It does not directly assess changes in cyclone genesis or frequency with climate change. The model's accuracy depends on the reliability of CMIP6 climate projections and the selected parameterizations. The resolution of the model might not capture all small-scale processes influencing cyclone behavior.