Ice-ocean interaction at Greenland's peripheral glaciers

Peripheral glaciers (PG) represent only a small part of the surface and ice volume of Greenland, but disproportionately affect the mass change of the Greenland Ice Sheet. They also contribute a particularly sensitive part of Greenland's freshwater balance. While there are relatively robust estimates of the surface mass balance of the PG (often referred to as the "climatic" mass balance, since these are interactions between ice and the atmosphere), there is no estimation for the role of the ocean-ice interaction for the mass change of Greenlandic PG. We do not only lack a quantitative estimation of the PG's contribution to the overall mass balance, but also a qualitative understanding of the relevance of the processes contributing to the mass change. Since the spatial distribution of the PG is significantly different from the distribution of the outlet glaciers of the Greenland ice sheet, it can be assumed that the freshwater originating from PG is responsible for other interactions between ocean and ice compared to the meltwater of the ice sheet. In any case, especially in modeling, due to the differences in scale between PG and the ice sheet, a separate treatment is required to account for the share of PG in Greenland's water balance.

Contacts: Dr.-Ing Muhammad Shafequee, Dr. Marco Möller & Prof. Dr. Ben Marzeion

Mean annual ice discharge from the peripheral tidewater glaciers of Greenland in the period 1995 to 2015 (a) and mean annual ice discharge in the period 2015 to 2018 displayed as change relative to the former period (b).

What was the main question you aimed to answer during the first project phase? 

In the first project phase we intended to improve our understanding of the processes that mainly influence the behavior of the PG. We aimed at developing a parameterization of the ice-ocean interactions specifically designed for PGs of Greenland and at implementing this parameterization into an existing glacier model in order to yield the first model-based, complete (i.e. also considering the calving of the PG) estimate of the mass changes of the PGs of Greenland.

Which methods do you use to answer your research question? 

For our modeling studies of the peripheral glaciers we use the Open Global Glacier Model (OGGM). The model is open source and enables the calculation of mass balances and resulting geometry changes of single glaciers, but was explicitly designed to be used in regional- or global-scale modeling. For determinations of ice discharge from the peripheral tidewater glaciers we also rely on remote sensing methods for extraction of glacier flow velocities at the calving fronts.

What were your main results?

Within the project, the Open Global Glacier Model (OGGM) has been further developed to yield realistic ice thickness estimates also in areas close to the calving fronts of tidewater glaciers. Moreover, a method suitable for implementation in OGGM which is capable of modelling the mass loss due to tidewater glacier retreat has been identified and tested. The ice thickness estimation has been calibrated specifically for the peripheral tidewater glaciers of Greenland. This made it possible to perform the first ever calculation of recent ice discharge from these glaciers. A conservative estimate involving remotely-sensed glacier flow velocities yielded 5.4±1.6 Gt per year. To compile this estimate, a separate study, which provides the first quantification of the extent of shelf ice areas at Flade Isblink ice cap and of the ice discharge feeding them, became necessary. Taken together, for the peripheral glaciers of Greenland a mean annual mass loss of 34.7±24.8 Gt has been calculated for the period 1995 to 2015.

What are your goals for GROCE-2?

During the second project phase, we will determine the sensitivity of glacier projections on forcing data of different spatial resolutions. We will force OGGM with both global regional data obtained through dynamical downscaling. Moreover, we will develop transfer functions between the global and regional scales supposed to replace the numerically costly dynamical downscaling. We will answer the question whether reliable projections of peripheral glaciers require a complex model setup including dynamical downscaling, or to which degree it is possible to obtain similar results using global scale forcing data and applying region-specific scale-transitioning approaches. Finally, we will produce CMIP6-based projections of the peripheral glaciers.