Seasonal Tidewater Glacier Terminus Oscillations Bias Multi‐Decadal Projections of Ice Mass Change. Issue 2 (12th February 2022)

• Record Type: Journal Article
• Authors: Felikson, D.; Nowicki, S.; Nias, I.; Morlighem, M.; Seroussi, H.
• Is Part Of: Journal of geophysical research. Volume 127:Issue 2(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-02-12
• Subjects: Geomorphology -- Periodicals; 551.3
• DOI: 10.1029/2021JF006249
• Languages: English
• ISSNs: 2169-9003

Abstract

Abstract: Numerical, process‐based simulations of tidewater glacier evolution are necessary to project future sea‐level change under various climate scenarios. Previous work has shown that nonlinearities in tidewater glacier and ice stream dynamics can lead to biases in simulated ice mass change in the presence of noisy forcings. Ice sheet modeling projections that are used in the IPCC Assessment Report 6 (AR6) utilize atmospheric and oceanic forcings at annual temporal resolution, omitting any higher frequency forcings. Here, we quantify the effect of seasonal (<1 year) tidewater glacier terminus oscillations on decadal‐scale (30 years) mass change. We use an idealized geometry to mimic realistic tidewater glacier geometries and investigate the impact of the magnitude of seasonal oscillations, bed slope at the glacier terminus, and basal friction law. We find that omitting seasonal terminus motion results in biased mass change projections, with up to an 39% underestimate or 25% overestimate of mass loss when seasonality is neglected, depending on bed slope and magnitude of oscillations. The bias is most sensitive to the magnitude of the seasonal terminus oscillations and exhibits very little sensitivity to choice of friction law. Although omitting seasonal terminus motion may bias mass change projections of individual glaciers, further work is needed to investigate the impact on ice sheet mass change projections. In order to achieve this, seasonality in atmospheric and oceanic forcings must be adequately represented and observations of seasonal terminus positions and tidewater glacier thickness changes must be acquired to evaluate numerical models. Plain Language Summary: Computer models are required to predict how glaciers will evolve under future climate warming. Past studies have shown that rapid changes in external variables that affect glaciers can lead to a permanent shift in their state. However, not all computer models take these rapid changes into account. For example, model predictions of the ice sheets that are used in the IPCC Assessment Report 6 (AR6) leave out seasonal changes of glaciers. In this paper, we set up a computer model to resemble a typical glacier and we run the model by forcing the glacier to retreat either with or without seasonal terminus movement. Our results reveal that leaving out seasonality can cause either an overestimate or underestimate of mass loss, depending on the slope of the bed and the amount of seasonal terminus movement. We find that the overestimate or underestimate is most sensitive to the amount of seasonal advance and retreat and least sensitive to the friction law. Our results show that computer models must take the seasonal changes into account in order to make accurate predictions and to avoid underestimating or overestimating mass loss of glaciers in the future. Key Points: Seasonal terminus oscillations induce a systematic bias in multi‐decadal mass loss of retreating glaciers Sign of mass loss bias depends on bed slope and magnitude of oscillations in ice front position Mass loss bias is more sensitive to bed slope at the terminus than to sliding law