Marika M. Holland

We present Community Earth System Model (CESM2) simulations of cloud perturbations representing idealized marine cloud brightening (MCB) climate intervention in the Arctic. Against a greenhouse gas (GHG) warming scenario (“baseline”), we simulate three MCB strategies by increasing low cloud droplet number concentration (N _d ) in ocean grid cells inside the Arctic Circle: year-round (“Annual”), summer-only (“Summer”), and a summer-only “dynamic mask” which implements cloud seeding only in the open ocean and the marginal ice zone (“Dynamic”). For each of these, we implement low-, medium-, and high-intensity interventions which increase N _d of low clouds to 100, 200, and 400 cm ^-3 , respectively. The Summer interventions provide the greatest degree of local and global cooling, with Dynamic in the middle and Annual cooling the least. Within the first year, all nine perturbations result in increased Arctic sea ice extent and volume relative to baseline; this includes Annual-100, which does not meaningfully cool the Arctic. The other eight perturbations cool the Arctic by ~1.0-3.5 ^∘ C relative to baseline, and seven of the nine (all but Annual-100 and Dynamic-100) have a statistically significant difference in global mean temperature compared to the baseline ensemble. The MCB scenarios also have increased NH midlatitude permafrost extent and depth relative to baseline, and two of the 400-strength perturbations also have increased Greenland Ice Sheet mass balance. Some of the stronger perturbations drive shifts in tropical precipitation, and all nine of the interventions also forestall the simulated weakening of the Atlantic Meridional Overturning Circulation (AMOC).

We present Community Earth System Model (CESM2) simulations of cloud perturbations representing idealized marine cloud brightening (MCB) climate intervention in the Arctic. Against a greenhouse gas (GHG) warming scenario (“baseline”), we simulate three MCB strategies by increasing low cloud droplet number concentration (N _d ) in ocean grid cells inside the Arctic Circle: year-round (“Annual”), summer-only (“Summer”), and a summer-only “dynamic mask” which implements cloud seeding only in the open ocean and the marginal ice zone (“Dynamic”). For each of these, we implement low-, medium-, and high-intensity interventions which increase N _d of low clouds to 100, 200, and 400 cm ⁻³ , respectively. The Summer interventions provide the greatest degree of local and global cooling, with Dynamic in the middle and Annual cooling the least. Within the first year, all nine perturbations result in increased Arctic sea ice extent and volume relative to baseline; this includes Annual-100, which does not meaningfully cool the Arctic. The other eight perturbations cool the Arctic by ∼1.0-3.5°C relative to baseline, and seven of the nine (all but Annual-100 and Dynamic-100) have a statistically significant difference in global mean temperature compared to the baseline ensemble. The MCB scenarios also have increased NH midlatitude permafrost extent and depth relative to baseline, and two of the 400-strength perturbations also have increased Greenland Ice Sheet mass balance. Some of the stronger perturbations drive shifts in tropical precipitation, and all nine of the interventions also forestall the simulated weakening of the Atlantic Meridional Overturning Circulation (AMOC).

Marine cloud brightening (MCB) has been proposed to offset warming by increasing cloud reflectivity, but its efficacy in the Arctic is uncertain due to strong seasonal and surface-dependent cloud radiative effects. Rapid and harmful Arctic-amplified warming and sea ice loss have increased interest in targeted regional interventions. Furthermore, Arctic MCB may be amplified by the sea ice albedo feedback, providing additional cooling beyond the imposed cloud forcing. We assess the potential for sea ice recovery in the fully-coupled Community Earth System Model version 2 (CESM2), under three intervention strategies at three perturbation magnitudes. We use idealized cloud droplet number (N _d ) perturbations to approximate marine boundary layer cloud responses to aerosol seeding. The intervention is strongly seasonal and dominated by shortwave effects: increasing N _d produces substantial cooling in summer, but weak winter warming due to longwave cloud radiative effects. This makes summer-only intervention more effective than year-round seeding. The response scales approximately logarithmically, producing similar gains for each doubling of N _d . Increases in liquid water path enhance the radiative response beyond the direct brightening from droplet size reduction. Sea ice expansion reduces latent heat fluxes to the atmosphere and cools the surface, contributing to a shift from rain to snow in summer and reduced total precipitation in fall. In the strongest intervention, targeting only open ocean and marginal ice zone regions reduces the perturbed area while maintaining similar sea ice gains. These results highlight the importance of seasonal and spatial targeting for Arctic MCB.

The Arctic is undergoing rapid and disproportionate climate change, driven by tightly coupled interactions among the ocean, sea ice, and atmosphere. This review synthesizes current understanding of historical and projected changes in the Arctic ocean–ice system, emphasizing the role of variability across timescales—from seasonal to multidecadal—in shaping observed trends. We highlight the interconnected nature of Arctic system components, focusing on feedbacks—including emerging coupled ocean–ice processes—pathways, and mechanisms that link variability and long‐term change. Particular attention is given to state‐dependent and potentially nonlinear responses, as well as to the roles of remote forcing and increasing connectivity with sub‐Arctic regions. We assess the relative contributions of internal variability and anthropogenic forcing, highlighting key challenges in attribution, and identify priorities for improving future projections. This synthesis provides a process‐based framework for understanding Arctic change and its growing influence on the global climate system.

Historically, polar marine phytoplankton were thought to primarily grow after the seasonal breakup of sea ice, when there is plentiful light available in the surface ocean. However, observations of substantial productivity under sea ice has called this assumption into question. Using a global Earth system model, we quantify under‐ice phytoplankton productivity in the Arctic and Southern Oceans. We find that phytoplankton growing under sea ice, which are invisible to remote sensing‐derived estimates of productivity, generate ∼100 Tg C yr ⁻¹ in each polar region. Additionally, while the sea ice conditions that permit under‐ice growth differ between the polar regions, in both poles the under‐ice blooms shift polewards and decline in importance near the end of the 21st century as sea ice loss accelerates. These changes likely have implications for food availability for the benthic and pelagic consumers in polar ecosystems and for global carbon cycling.

The Southern Ocean around Antarctica is one of the fastest changing regions on the planet and an emerging resource frontier for fisheries. Here, we present the Antarctic Ecosystem Value Index created by merging ecosystem information across food web trophic levels, from phytoplankton to fish and penguins, to quantify the ecological value of marine areas around the Antarctic continent. We find that coastal polynyas - areas of reduced sea-ice - have Index values 31–72% higher than surrounding areas, suggesting that these areas are biologically valuable hot spots for a number of ice-dependent Antarctic Species. Using output from an Earth system model to generate future projections of the Index, we find that high-value locations, often within polynyas, are likely to continue to be valuable throughout the 21st century despite environmental changes. The Antarctic Ecosystem Value Index indicates that penguins lose importance as their habitat becomes increasingly unsuitable, so protecting high-value habitat areas may be critical for these species. This study also shows that while many high-value Index areas are within existing or proposed Marine Protected Areas, there are several opportunities for adopting additional protection, particularly in East Antarctica and the Amundsen Sea.

The primary sources of recent summer Arctic moistening trends in reanalysis are uncertain, hindering attribution of observed Arctic warming due to radiative effects from water vapor changes. Here, we use a combined online numerical water tracer and circulation nudging approach in the Community Earth System Model to track the sources of water vapor beyond its initial sources. Trends in boreal summer large-scale circulation have driven moistening of the Arctic over recent decades, having a large impact on the Arctic radiative budget, accounting for 94% of the strengthening water vapor radiative feedback. We identify two key regions supplying the Arctic water vapor feedback: Northeast North America and western/central Eurasia. In both regions, anticyclonic circulations over the southwest Atlantic and eastern Europe move moisture from the tropical oceans poleward to high latitude land through precipitation in winter and spring. During summer, evapotranspiration over land releases this water vapor, and it is transported by winds into the Arctic. We refer to this sequence of terrestrial moisture storage and release as the land capacitor effect. Thus, the impacts of circulation changes on poleward moisture transport and land-atmosphere interactions over high latitudes represent the underlying mechanisms of the recent moistening and warming in the Arctic.

Phytoplankton net primary production in the Arctic has historically been constrained to a short, intense summer bloom that sustains fish, seabird, and marine mammal populations. However, climate change is altering Arctic phytoplankton bloom phenology. We use an ensemble of Earth system model simulations to isolate the impact of climate change on the timing, duration, and importance (relative contribution to total net primary production) of the bloom. Earlier blooms emerge across 71% of the Arctic Ocean by 2100, when blooms begin 34 days earlier and last 15 days longer than in 1970. Productivity is less concentrated in a single bloom in sub-Arctic seas and on Arctic inflow shelves by 2100, indicating that the bloom declines in importance. In contrast, bloom phenology and productivity exhibit only small changes by 2020. Our study demonstrates that anthropogenic climate change will greatly alter the timing and importance of the Arctic Ocean phytoplankton bloom by 2100.

The Arctic is a prime example of global climate change, experiencing shifts on an unparalleled scale. This review summarizes the current state of knowledge on long-term changes and variations—both historical and projected—in the oceanic and sea ice components of the Arctic climate system. Using observations and modeling results, it examines the role of climate variability modes, from multidecadal to seasonal time scales, in shaping the Arctic climate. Emphasis is placed on the interconnectivity between Arctic system components, with a focus on feedbacks (including newly emerging ones), pathways, and processes that link long-term changes and variations in the oceanic and sea ice systems. The review also explores the relationships of Arctic climate change and variability with atmospheric forcing and sub-Arctic regions. It provides a critical assessment of the roles of natural variability and human-induced climate change, as well as projections for the future. Understanding these changes and their drivers is crucial for assessing the Arctic’s role in the global climate system.