Arctic sea ice loss continues to serve as a strong gauge of climate change. It is the component of the Earth system that is responding most visibly and rapidly to a warming climate. The implications of a shrinking sea ice cover include changes in physical processes like deep water formation and the reflection of solar radiation, and alterations to the way of live for humans and animals that depend on the ice in their daily lives. Here I evaluate long term trends in sea ice coverage in the Greenland Sea and Irminger Basin from 1979 to 2018. While in the Arctic Basin the recession of summer sea ice is more pronounced, it is shown that in the Greenland Sea the declining winter sea ice maximum is more pronounced than the summertime reduction. The strongest signature of this robust trend is the disappearance in 2004 of a sea ice feature called the Odden Ice Tongue that is characterized by local freezing and ice formation and to a lesser extent by the advection of sea ice. A budget constructed from sea ice concentration and velocity estimates from the National Snow and Ice Data Center, and sea ice thickness estimates from the University of Washington’s Pan-Arctic Ice Ocean Modeling and Assimilation System indicates that the area of sea ice transported into the Greenland Sea from the Arctic has gone largely unchanged since measurements began in late 1978. Despite this, the volume of sea ice flowing out of the Arctic has decreased 11% when compared to the 1979-2004 mean due to a significant thinning of sea ice.

In the last 15 years the average winter buildup of sea ice volume in the Greenland Sea is 16% smaller than the same winter accumulation from 1979 to 2004. The volume of sea ice that is advected into the Greenland Sea, from Fram Strait, is approximately twice as large as the change in volume of sea ice in the area over the course of a typical winter, indicating that half of the advected sea ice melts over the course of the winter.

The East Greenland Current (EGC) and the East Greenland Coastal Current (EGCC) are part of the North Atlantic Ocean circulation system that carries cold fresh melting water from the Arctic southward to the Subpolar North Atlantic region; hence, determining the variability and the drivers of these currents can help scientists to develop a better understanding of the circulation and climate in the Northern Hemisphere. In this study, we compare velocity from Acoustic Doppler Current Profilers (ADCP) data (2016-2020) from the Overturning in the Subpolar North Atlantic Program (OSNAP) moorings deployed near the Cape Farewell, at the southern tip of Greenland, with the geostrophic velocity derived from satellite altimeter data from AVISO that measures the sea level anomalies (SLA). The goal is to determine which part of the moored observed variability can be derived from the altimeter data. It is found that the seasonal variability observed via the two methods is similar. Similarly, the magnitude of across flow velocity anomaly of the two data sets are the same, but the along flow velocity anomaly computed from the satellite altimeter data is slightly smaller than that of the mooring data. Overall, our results suggest that the satellite altimeter is a complementary tool for ocean circulation observation at high latitudes where moorings are not deployed.

Mass loss of the Greenland Ice Sheet is expected to accelerate in the 21st century in response to both a warming atmosphere and ocean, with consequences for sea level rise, polar ecosystems and potentially the global overturning circulation. Glacial fjords connect Greenland’s marine-terminating glaciers with the continental shelf, and fjord circulation plays a critical role in modulating the import of heat from the ocean and the export of freshwater from the ice sheet. Understanding fjord dynamics is crucial to predicting the cryosphere and ocean response to a changing climate. However, representing glacial fjord dynamics in climate models is an ongoing challenge because fjord circulation is complex and sensitive to glacial forcing that is poorly understood. Additionally, there are limited observations available for constraining models and theory. This dissertation aims to improve our understanding of fjord dynamics, focusing on key aspects (heat variability, freshwater residence time, and fjord exchange) which need to be included in glacial fjord parameterizations.

We use three approaches combining novel observations, idealized, modeling and numerical simulations to investigate the dynamics of fjord circulation at different spatial scales. First, we investigate the heat content variability in the fjord using acoustic travel time (Chapter 2). We demonstrate that acoustic travel time can be used to model fjord stratification during winter months and monitor heat content variability at synoptic and seasonal timescales. Secondly, we use a combination of in situ observations and an idealized box model to evaluate freshwater residence time in a west Greenland Fjord (Chapter 3). We find that meltwater from the ice sheet is mixed downward across multiple layers near the glacier terminus resulting in freshwater storage and a delay in freshwater export from the fjord. Finally we analyze a multi-year realistically forced numerical simulation of Sermilik Fjord in southeast Greenland and identify the impact of shelf and glacial forcing on fjord exchange (Chapter 4). We show that the glacial-driven circulation is more efficient at renewing the fjord and that the sign of the exchange flow is related to the along-shelf wind stress. This dissertation strengthens our understanding of the fundamental connections between oceans and glaciers, and will lead to improved representation of ice-ocean interactions in climate models.

The convergence of freshwater from polar and subpolar waters influences nutrient (nitrate, phosphate, and silicate) concentrations on the southeastern Greenland shelf. Interannual variability of nutrient distributions from the shoreline to 150 km offshore were determined by using hydrographic and nutrient measurements from June to September of 2002 to 2016 to produce high-resolution transects. Although no significant trend was observed during the period analyzed, in-situ observations from 1991 to 2018 revealed considerable interannual variability and that nutrient concentrations in polar-origin waters (POLW) were two to three times less than those of Atlantic-origin (AW). In POLW, the mean nitrate, phosphate, and silicate concentrations (in µmol/kg) were 5.02, 0.51, and 3.10, respectively, compared to that of AW, with means of 15.16, 0.98, and 7.80, respectively over the same area. Waters of Pacific-origin, transported through the Arctic Ocean circulation and western Fram Strait, were observed furthest inshore in southeastern Greenland from 1997 to 2018 with increased fractions of Pacific Water concentration in 2004 (0.15) and 2018 (0.16). From relationships observed between nutrients, nitrate was identified as the least biologically-available nutrient, followed by phosphate, and then, silicate, which concurred with previous studies in the North Atlantic region. The accepted global stochiometric relationships for N:P, N:Si, and Si:P are 16, 1.07, and 15, respectively. As expected, results differed slightly throughout the cross section, as the area from the shoreline to 72 km exhibited ratios of 17.71, 1.66, and 9.70, respectively, while the area from 72 km to 150 displayed ratios of 13.56, 1.00, and 10.14, respectively.