Geometric and Atmospheric Controls on Warm Water Access to Glacial Fjords

The melting of the Greenland Ice Sheet (GrIS) has the potential to contribute over 7 m to sea level and has recently accelerated, primarily due to the warming of ocean currents abutting many of Greenland’s fjords. A strong overturning circulation driven by near-glacier buoyancy forcing transports heat towards the glaciers and meltwater away and is postulated to control the ocean-driven melt of Greenland’s glaciers. However, it remains unclear what factors control the exchange rate between fjords and the open ocean; proposed candidates include buoyancy forcing at the surface and at depth, wind stresses, coastal and shelf currents, bathymetry, and mixing processes. This gap in understanding hinders our ability to estimate the ocean’s current and future contribution to glacial melt, since we cannot accurately translate offshore variations in heat content to glacial melt rates. To address this, we conduct idealized numerical simulations using an isopycnal model with domains consisting of continental shelves and fjords representative of those in Greenland and the West Antarctic Peninsula. Our results show the strength of the overturning circulation (renewal of fjord waters) is controlled by three factors: the eddy transport from the open ocean to the fjord mouth, the strength of the water mass transformation via plume entrainment at the fjord head, and geometric/dynamic controls at the fjord mouth that act to constrain the overturning. Our results highlight a comprehensive and previously unrecognized role for geometry in hydraulically controlling access of warm water to ice-shelf faces in Greenland; even in the absence of a sill at the fjord mouth, warm water access may be hydraulically controlled for sufficiently narrow fjord widths. We also show that the fjord’s bulk vorticity balance sets the steady-state horizontal recirculation within the fjord and influences the strength of the coastal current external to the fjord. We present dynamical theories to explain and justify these numerical results. Our results suggest that that previous approaches to translating offshore stratification to fjord heads are likely to lead to substantial errors in estimates of ocean-driven melt. We discuss potential improvements to these approaches based on insights from our model experiments.