Glaciology - Greenland Ice Sheet (3) Tidewater Glaciers

Sample question: How may climate change and extreme weather events affect marine-terminating glaciers in Greenland over the coming decades? (2023-7)

By Fiona Fang May 2026

Ocean:

a) terminus [1] large scale ocean current pattern, such as the warm Atlantic Water, controls the distribution of subsurface heat available to fjords. More heat lead to more more melt, thereby enhanced calving and thinning of ice tongue. Warmer ocean also reduce sea ice and weaken mélange. Mélange pushes against the calving front and reduces the ability for ice bergs to rotate, fracture and detach (Amundson et al., 2010); weaker mélange therefore alsp increases the calving rate. [2] Using 226 marine-terminating glaciers, Wood et al 2021 identify 74 glaciers in deep fjords where Atlantic Water controlled 49% of mass loss, and retreat occurred when warming increased undercutting by 48%. [3] Chudley et al. (2023) show that Steenstrup retreated about 7 km and thinned about 20% responding to the >2 °C anomaly in deeper Atlantic Water in 2018, highlighting that seemingly stable glacier can retreat quickly if ocean forcing changes.

b) Inland [1] The impact of ocean forcing on tidewater glacier is not only on the terminus. The changes in stress balance can also propagate further inland. For example, after the 2012 calving event, Petermann glacier accelerated by about 10% between winter 2011/12 and winter 2016/17. This is explained by the reduced lateral resistance as the ice tongue shortened and lost contact with fjord sidewall (Ruckmap et al., 2019). This acceleration means more ice is discharged into the ocean and therefore enhance sea level rise. [2] But loss of ice tongues does not always lead to dynamical response of inland ice, especially if they do not actively buttress the glacier: Hill et al (2018) while the 18 major northern Greenland outlet glaciers they studied were retreating during 1995–2015 (all retreating faster than at any time since 1948), grounded glaciers (with no ice tongues) accelerated and thinned after retreat, while most glaciers with floating ice tongues were dynamically insensitive to terminues retreat or even ice-tongue collapse. This is because these tongues are mechanically weak or poorly coupled to the grounded trunk.

Atmosphere:

a) Ice flow [1 driving stress]: Warmer air increases surface melt and runoff, so the glacier loses mass from above. Since the thinning usually concentrate near low elevation, this steepens the surface slope upstream. Since the driving stress τd=ρigHsinα, steeper surface profile increases driving stress, leading to acceleration and therefore retreat (Felikson et al 2017). [2 basal drag] if surface melt reaches the bed through moulins or crevasses, it can lead to acceleration by increasing basal water pressure. GPS data by Sugiyama et al (2025) showed that short-term speed-up events at the front of Bowdoin Glacier (NW Greenland) that were associated with heavy rain, and longer-term (week to month) variations were correlated with air temperature.**

b) submarine melt: [1] atmospheric forcing can also impact submarine melting, as highlighted by Slatter and Straneo (2022). The physical chain here is that, if surface melt enters the subglacial drainage, it can be released at the grounding line as a high energy plume; plume turbulence excites a near-glacier ocean circulation, which in turn the transfer of heat from ocean to ice and thus enhance submarine melt. [2] Slatter and Straneo parameterized submarine melt rate with the subglacial discharge and ocean thermal forcing raised to the power 0.31 and 1.18. (Slatter and Straneo, 2022). [3] They also highlighted the spatial heterogeneity of dominant driving force: variability in submarine melting in south Greenland was indeed governed by the ocean, while the atmosphere dominated in the northwest.

Topography

a) Fjord bathymetry acts as first-order filter on ocean forcing. [1] This explains the contrasting behevaiour of neighbouring glaciers, such as Tracy vs Heilprin (Porter et al., 2014). Tracy has a much deeper terminus than Heilprin, and is therefore more exposed to warmer subsurface water and undergoes much stronger submarine melting and thinning. [2] Simulation by Harger et al (2023) shows that, under identical external ocean boundary conditions, local fjord processes can modify GL thermal forcing by ~ 2.9°C. Therefore, bathymetric obstruction is the primary control on near-glacier thermal forcing.

b) Fjord geometry also matters, particularly the terminus width: [1] Enderlin et al (2013) shoes that wider terminus makes the glacier more sensitive to climate forcing, because same sidewall shear is spread across a larger ice width, and therefore lateral drag per unit ice area decreases. [2] Idealised experiments by Åkesson et al. (2018) show that under identical climate forcing, changing only fjord width can alter GL retreat by tens of kilometres.

c) Bed topography at terminus also matters. [1] Retreat over a retrograde bed can be self-sustaining, since retreat into deeper water increases ice thickness at the terminus or grounding line, enhances calving and/or ice flux, which leads to further thinning and retreat. In contrast, a prograde bed, or a topographic high, can limit ice loss: [2] for example, Ua model by Hill et al (2021) shows that the ice loss from Peterman glacier is likely limited (with SL contribution < 1mm) even under enhanced future basal melting and calving, likely explained by the topographic high around 12 km inland of the present GL. [3] Difference in bed topography can explain the distinctive behaviour of neighbouring glaciers. For example, both Umiamiko Isbræ and Ingia Isbræ began retreat around 2001, but Umiamiko restabilised on the prograde side of a large bed bump around 2010, while Ingia, which sits on a flatter bed, continued retreating (Catania and Felikson, 2022).

d) Bed topography upglacier: While terminus controls whether retreat begin or not, dynamic significance of that perturbation depends on how far that it travels inland. Here, bed topography upglacier matters. Steep bed reaches, or knickpoints, for example, can limit the efficiency of diffusion thinning. Felikson et al (2021) found that knickpoints exist beneath the majority of outlet glaciers but they are less steep in regions of gentle bed topography, giving them the potential to diffuse the thinning far inland.

---

*a side point on moraine: The existence of terminal moraine on glacier, for example, can provide extra basal resistance and limit water access to terminus at depth. This reduces buoyancy effect and therefore limit calving. If the moraine slope is retrograde, self-sustaining retreat can happen (Catania et al., 2018). This ice-sediment coupling is thought to explain the tidewater glacier cycle with no need of any external forcing (Brinkeroff, 2017).

**a side note on ice rheology, interesting but quite hard to fit in… ice rheology: refreezing of melt release latent heat, soften ice and increase velocity (Poinar et al 2016).