The accelerating rate of global mean sea level rise over the last decade poses a significant threat, with projections indicating continued rise for centuries. The Amundsen Sea Embayment (ASE) in West Antarctica is identified as a critical region driving current and future sea level rise due to its accelerating ice discharge into the ocean. This acceleration is primarily attributed to reduced resistive stresses caused by ocean-driven ice shelf melting, leading to a state of disequilibrium where ice discharge surpasses mass gain through accumulation. Most studies focus on worst-case scenarios, neglecting the potential for mitigation through reduced emissions, geoengineering, or climate variability. This study investigates plausible scenarios where Antarctic ice loss could be limited by focusing on the ASE, a climate-sensitive tipping point where warm circumpolar deep water (CDW) infiltrates the continental shelf through bathymetric troughs, causing high melt rates. While the future climatology of the ASE is uncertain, changes in CDW volume within ASE ice shelf cavities will be a key determinant of future mass loss. The research explores the impact of multi-decadal periods of idealized ocean forcing on mass loss in the ASE and determines the conditions necessary to reverse mass loss trends and re-advance grounding lines, considering both increased and decreased melt rates.

Existing literature highlights the significant contribution of the Antarctic ice sheet to global sea level rise, with the rate of ice loss increasing. The Amundsen Sea Embayment (ASE) is identified as a primary contributor, with accelerating ice discharge driven by ocean-driven ice shelf melting. Studies project continued increases in ASE sub-ice shelf melt rates, potentially initiating unstable retreat and collapse. While worst-case scenarios are frequently modeled, the potential for mitigation through reduced emissions, geoengineering, or climate variability remains an area of exploration. Previous research demonstrates the sensitivity of ASE ice streams to ocean warming and the influence of climate variability on sub-ice shelf melt rates. However, the effects of future cool-ocean periods on ice sheet mass loss from the ASE have received less attention. Studies such as those by DeConto and Pollard (2016) and DeConto et al. (2021) highlight the long-term influence of delays in the reduction of melt rates on Antarctic sea level contribution.

This study employs ice sheet model sensitivity experiments using the BISICLES ice sheet model. The model solves a one-layer longitudinal stress balance approximation in two dimensions (L1L2) with adaptive mesh refinement. Simulations were performed for a regional ASE domain with topography and thickness from BedMachine v1, a constant surface mass balance field from Arthern et al. (2006), and a three-dimensional internal ice temperature field from Pattyn et al. (2010). Basal traction and ice viscosity coefficients were determined through an inverse procedure to ensure consistency with observations. A 181-member ensemble was generated, consisting of 180 simulations forced with perturbed sub-ice shelf basal melting and a control simulation. The 180 forced simulations were divided into 6 groups of 30, each with increased melt rate anomalies over varying durations (20–100 years) and magnitudes (5–30 m y⁻¹). Thirty simulations returned melt rates to present-day values (R0), while the remaining 150 simulations applied constant negative melt rate anomalies ranging from 0 to 5 m y⁻¹ to represent cooler ocean conditions. The ensemble was filtered to select simulations consistent with observed Antarctic mass loss from 2007-2022 derived from GRACE satellite data. The simulations were categorized as 'restored' melt simulations (returning to present-day levels) and 'reduced' melt simulations (further reductions below present-day levels). Ice streams were grouped into drainage basins (Pine Island Glacier, Thwaites Glacier, and Pope Smith and Kohler Glaciers) for analysis. Sub-ice shelf melting was based on an initial steady-state field from Cornford et al. (2015), with anomalies added and scaled according to observed patterns of basal melting. Melt rate reduction anomalies up to 5 m y⁻¹ were applied to explore the effects of local ocean cooling, aligning with observed variability in the region. The model outputs were analyzed to assess the impact of different melt rate scenarios on sea level equivalent contribution, grounding line position, and ice stream dynamics.

The restored melt simulations (R0) demonstrated that even after returning elevated melt rates to present-day levels after periods of enhanced melting (20-100 years), continued ice mass loss occurs, with the ASE contributing 44 to 130 mm of global sea level equivalent (SLE) after 100 years and 82-165 mm SLE by the end of the simulation (200 years). However, the rate of mass loss declines immediately upon reduction of melt rates to present-day values, indicating a sensitivity to basal melt rates. Grounding line positions show widespread retreat after 100 years, with the extent dictated by cumulative total melt. By the end of the simulation, grounding line positions converge towards topographically-controlled positions. Simulations with the greatest total melt showed re-advance of Pine Island, Pope Smith, and Kohler Glaciers. The reduced melt simulations (R1-R5) demonstrated that reducing present-day melt rates significantly limits total mass loss. A melt reduction of 2 m y⁻¹ below present-day is required to prevent ongoing mass loss for high-end simulations. Reducing melt rates from 0 to 5 m y⁻¹ below present-day lowered the upper-end SLE contribution by 60 mm (from 165 to 105 mm). This mitigation effect is most substantial for upper-end projections. Reduced melt rates promoted the re-advance of ASE grounding lines, with the spread in simulations decreasing as melt rates are reduced. While grounding lines advanced in these scenarios, the total mass deficit remained, ranging from 42 to 105 mm of SLE over 200 years for a 5 m y⁻¹ reduction. Substantial increases in accumulation were also required to completely offset sea level contribution. The Thwaites Glacier Basin showed the most significant mass loss and complex response to the forcing, exhibiting high sensitivity to both increased and decreased melt rates. The study showed that short periods of large increases in melting have a similar long-term sea level effect as long periods of small increases.

The findings demonstrate the significant long-term impacts of short-term increases in sub-ice shelf melt rates on future sea level rise. The results highlight the direct sensitivity of ice streams to basal melt rates and show that restoring melt rates to present-day levels after periods of enhanced melting leads to an immediate reduction in the rate of mass loss, although mass loss continues. The study's exploration of mitigated future change scenarios demonstrates that reducing melt rates below present-day levels significantly limits total mass loss and promotes grounding line re-advance. However, complete reversal of mass loss requires not only reduced melt rates but also substantial increases in accumulation. This work emphasizes the importance of minimizing ocean-driven melting through reduced emissions or geoengineering to prevent severe future sea-level rise and underscores the complex interplay between melt rates, accumulation, and grounding line dynamics in determining the future contribution of the ASE to sea level.

This research highlights the significant and long-lasting effects of even short-term increases in Antarctic sub-ice shelf melt rates on sea level rise. Minimizing ocean-driven melting through emissions reduction or geoengineering strategies is crucial for mitigating future sea level rise. While reducing melt rates can significantly limit ongoing retreat and encourage grounding line re-advance, achieving a complete reversal of Antarctica’s sea level contribution necessitates a substantial increase in accumulation over grounded ice in addition to reduced melt rates.

The study utilizes idealized simulations with simplified representations of ocean-ice interactions and climate variability. The constant surface mass balance field might not fully capture the complexities of future snow accumulation. The model's resolution and parameterizations could influence the results. Furthermore, the study's focus on the ASE limits the generalizability of its findings to other regions of Antarctica.