Coupling a fjord circulation model with subglacial discharge to a glacier dynamics model with calving, through the estimation of submarine melting at the glacier front

Resumo

Around 70 % of the total freshwater volume on Earth is stored in the cryosphere. Among its components, there are more than 200 thousand glaciers distributed worldwide. The smaller size of the glaciers compared to the ice sheets of Antarctica and Greenland make them more responsive to atmospheric and/or oceanic forcing, in particular to global warming. The loss of mass experienced by these glaciers contributes 25-30% to the current observed sea level rise, with 10-30% of such loss associated with frontal ablation from tidewater glaciers through calving and submarine melting. It has been shown that the ocean can be a key factor in the acceleration, thinning and retreat of the outlet glaciers in the periphery of Greenland and Antarctica, promoting and accelerating the mass loss from these large ice sheets. However, these findings are relatively recent and our knowledge is still in a premature stage. The objective of this thesis is to gain knowledge on the physical processes involved in the glacier-ocean interaction and their feedback on glacier dynamics. To achieve it, we have developed a numerical coupled glacier-ocean model that allows us to study the influence that submarine melting exerts on glacier dynamics, calving and front position evolution. Likewise, we have built a simple plume model that parameterizes the buoyant plume generated at the glacier-fjord interface as a result of sudden discharges of meltwater through subglacial discharge channels. Using this plume model, we have studied the effect that surface meltwater exerts on fjord stratification, analyzing its impacts from physical, biogeochemical and ecological perspectives. Finally, we have built a computationally less demanding version of the coupled model, the glacier-plume model, aimed to simulate the medium/long-term response of glacier-fjord systems. To carry out our investigations, we have used observations of two glacier-fjord systems, one in Svalbard and the other in Greenland. The results of our model simulations suggest that both submarine melting and crevasse hydrofracturing exert important controls on seasonal frontal ablation, with submarine melting alone not being sufficient for reproducing the observed patterns of seasonal glacier front retreat. Both submarine melt and calving rates are of the same order of magnitude, highlighting the strong link between both mechanisms. Our model results also indicate that changes in submarine melting lag meltwater production by 4-5 weeks, indicating the complexity of the intra and subglacial drainage network. We have also proved the high sensitivity of submarine melting to the intraseasonal evolution of subglacial discharge and fjord temperature, increasing by up to three orders of magnitude from the beginning to the end of the melt season. Due to this dependency, we believe that it is important to properly constraint predictive models, in order to obtain more accurate results. Glacier-plume and glacier-fjord coupled models differed in submarine melt rates (up to 30 % higher for the glacier-plume model) and also produced distinct melt-undercutting front shapes, which had an effect on the net stress fields near the glacier front. The quasi-linear front shape of the glacier-plume model promoted higher calving rates than the quasi-parabolic front shape of the glacier-fjord model, although both models predicted similar front positions. From the plume-parameterization model alone, we have found that increased fjord stratification (from accumulation of surface meltwater inputs) exerts a dominant control on plume vertical extent, even preventing the plume from surfacing. Under projections of increased surface melting, the appearance of plumes at the fjord surface could become less common due to the increased fjord stratification, so plume monitoring and remote tracking might become harder to achieve. Overall, and given that the glacier-plume model diminished the computational time by a factor larger than 50, we think that our glacier-plume coupled model is a suitable candidate for future projections of tidewater glacier evolution, as long as we use appropriate constraints on subglacial discharge fluxes and ambient fjord temperatures.