The Antarctic Ice Sheet plays a key role in our climate system by influencing current and future global mean sea level through the accumulation, deposition and discharge of ice. In the last four In the past few decades, Antarctic mass loss has increased sixfold, driven by melting ice shelves from below and the subsequent retreat of the West Antarctic grounding line, leading to accelerated ice discharge into the ocean. Future Antarctic mass balance simulations project mass losses of up to 150 cm of sea level equivalent per year by 2100.

Precipitation is critical for offsetting the vast majority of mass loss from Antarctica to the ocean each year (91 %). In the future, in medium-emission scenarios, net precipitation is projected to increase the Antarctic surface mass balance by 20–30 %. In many regions of the continent, up to 40 % of net precipitation and 70 % of precipitation variability are caused by synoptic extreme events.

One of the most intense poleward moisture transport mechanisms in the atmosphere is atmospheric rivers (ARs). Although they occur only 1 % of the time at a given location along the Antarctic coast, these narrow, elongated filaments of extreme water vapor transport are associated with 13 % of total annual precipitation and 35 % of interannual precipitation variability over Antarctica. Currently, ARs are dominated by snowfall on the Antarctic Ice Sheet, but they are also associated with most of the extreme surface melt events in West Antarctica and the Antarctic Peninsula, contributing to the destabilization of the supporting ice shelves.

A new study using a large set of simulations from the Community Earth System Model, version 2 (CESM2) climate model under the Shared Socio-economic Pathway 3-7.0 (SSP3-7.0) radiative scenario examines how Antarctic ARs and their impact on precipitation will change in the late 21st century (2066–2100) compared to the present (1980–2014). The study applies a polar-specific AR detection tool.

Key insights and sensitivity to detection methodology:

• Increase in atmospheric humidity: Over the Southern Ocean and Antarctica, there is an exponential increase in atmospheric moisture, which strengthens the pole-equatorial moisture gradient. Although the absolute increase in integrated water vapour (IWV) over the continent is small, the relative increase is the largest (1.5 times present at the end of the century). There is a large variability in this increase between the individual members of the simulation ensemble.
• Sensitivity to AR detection threshold: The future role of AR in the Antarctic climate system is extremely sensitive to the chosen detection method, especially whether the detection threshold takes into account the increase in atmospheric humidity caused by the Clausius-Clapeyron effect (warmer air can hold more moisture).
  • If the same AR detection threshold is applied to the future (2066–2100) as to the present (1980–2014), the AR frequency doublesThe integrated frequency of AR over the ice sheet will increase from 15.0 % to 31.7 %. Precipitation caused by AR will will increase 2.5 timesThe contribution of AR to total precipitation will increase to 24 %.
  • However, if the AR detection threshold is adjusted (scaled) to account for the increase in integrated water vapor in the future, the AR frequencies in the period 2066–2100 are comparable to the present. Precipitation caused by AR will increase only 1.25 times. The relative contribution of AR to total precipitation will remain at around 13 %.
• Regional changes and circulation: Although the scaled threshold shows an overall AR frequency similar to the present, there are regional variations. These are partly caused by changes in atmospheric circulation, such as eastward shift of the polar jet streamThese dynamic changes play a secondary role to the influence of humidity.
• Increase in rainfall associated with AR: Currently, ARs contribute to rainfall mainly in coastal Antarctica and on ice shelves. Using the current detection threshold, a significant increase in AR-related rainfall is projected in the future. Although rain is a small component of total precipitation in Antarctica, the increase in AR-related rainfall could have a significant impact on future stability of ice shelvesRain can increase the temperature of the snow (firn), reduce the surface albedo, deplete the air in the firn, and cause ice erosion, thus preparing the surface for widespread melting.

The difference between the results using the current threshold and the threshold scaled for future moisture increases is enormous. This difference shows how the sensitivity of AR detection to increases in atmospheric moisture affects the description of the future importance of ARs in the Antarctic climate system and their future contributions to the surface mass balance.

Quantifying future changes in extreme precipitation events, such as those associated with AR, is essential for refining estimates of global sea level rise in the 21st century and beyond. The choice of a future AR threshold is a decision dependent on the specific scientific purpose. Despite existing model biases in CESM2, robust increases in coastal precipitation in Antarctica at the end of the 21st century under high-emission scenarios, including a doubling or tripling of rainfall, are consistent with results from other models. This suggests that future increases in AR activity may initiate complex feedbacks between the atmosphere, ice, and ocean in Antarctica. Spring