Atlantic Water in the Arctic Ocean - Mechanisms and Impacts

Abstract

The Arctic Ocean plays a fundamental role in regulating Earth’s climate, and a changing Arctic will affect climate, weather, and life everywhere on the planet. Understanding the fundamental dynamics and mechanisms driving natural variability, and the effects of anthropogenic warming in the Arctic climate system is imperative to improve future climate predictions. Warm and saline Atlantic Water (AW) entering the region across the Greenland-Scotland Ridge is the primary heat source to the Arctic Ocean and plays an essential role in modulating the Arctic climate system. However, our knowledge is still insufficient to make skillful projections of future Arctic climate change with uncertainty levels similar to other regions. This thesis improves our understanding of the role of AW in the Arctic Ocean, focusing primarily on: its variations in the twentieth and twenty-first centuries; the underlying mechanisms governing this variability; and its proliferating regional impacts on sea ice, marine-terminating glaciers, and stratification.

First, we investigate the twentieth-century variability of AW heat transport through the gates of the Arctic Ocean. The analysis is based on a simulation from the global ocean-ice Norwegian Earth System Model (NorESM) supported by an extensive set of hydrographic observations dating back to 1900. We quantify prominent variability in both AW temperature and volume transport on near-decadal time scales, as well as significant positive trends in the most recent decades. Variations in volume transport were found to be linked to the wind forcing in the Nordic Seas and Subtropical North Atlantic, as manifested through the North Atlantic Oscillation, although the correlation is not constant over time and breaks down entirely in specific periods, such as the Early Twentieth Century Warming period. Variations in temperature are a combination of advected signals originating upstream and variations in atmospheric cooling over the Nordic Seas, which effectively dampen the AW heat anomalies along their path northward.

Secondly, we provide a further in-depth investigation of the relationship between the AW flow and wind forcing. Here, we analyze results from a coordinated wind perturbation experiment in a suite of nine different Arctic Ocean models, and calculate “Climate Response Functions” (CRFs) to isolate the effects of wind anomalies on AW circulation, sea ice, and hydrography. The CRFs show that anomalously strong/weak wind forcing over the Greenland Sea results in an intensification/weakening of the poleward AW flow and a reduction/increase in the Arctic sea ice cover. Despite biases in hydrograph, all models respond in a similar manner to the anomalous winds and show a near-linear relationship between AW volume and heat transport, surface heat loss, and sea ice extent in the Barents Sea. Historical reconstructions show that the largescale wind forcing alone can explain 50% of the AW flow variance, indicating potential for predictability.

Third, we focus on the export of meltwater from Upernavik Fjord in northwest Greenland as the combined result of melting caused by AW and the release of subglacial discharge at the fronts of marine-terminating glaciers. Using hydrographic observations collected between 2013 and 2019 we provide the first description of the hydrographic structure in Upernavik Fjord, explain the complex water mass transformation occurring in the fjord, and quantify the composition of the water mass exported from the fjord. We show that meltwater is heavily diluted and exported as “Glacially Modified Water” (GMW), which in summer is composed of 57.8 +/-8.1% AW, 41.0 +/-8.3% Polar Water, 1.0 +/-0.1% subglacial discharge, and 0.2 +/-0.2% submarine meltwater. Consistent with its composition, we show a close relationship between water mass properties on the continental shelf (AW and Polar Water) and the exported GMW properties, and estimate an exchange across the fjord mouth of 50 mSv. This study provides a first order parameterization for the exchange at the mouth of glacial fjords for large-scale ocean models.

Finally, we investigate changes in central Arctic Ocean stratification in the twentieth and twenty-first centuries. Observations show that from 1970 to 2017, the stratification in the Amerasian Basin has strengthened, whereas the stratification has weakened in parts of the Eurasian Basin. These contrasting results are due to competing effects of increasing AW influence (“Atlantification) and local freshening. Simulations from the Community Earth System Model Large Ensemble and a suite of nine CMIP6 models project that under a strong greenhouse-gas forcing scenario (RCP8.5/SSP585), the upper layers in the Amerasian Basin will become even more stratified in the future. In the Eurasian Basin, models show diverging results, with approximately half of the models projecting a strengthened stratification in the future and the other half projecting a weakened stratification. These differences are mainly a result of different balances between local processes and advected signals.

Combined, the four papers highlight the diverse yet significant role of AW in the Arctic environment and advance our knowledge of the broad-scale mechanisms governing AW variability and the impacts of AW on different components of the climate system. Our results provide a spatially and temporally inclusive progressed understanding of natural and anthropogenic climate change in the Arctic and ultimately contribute to improved projections of future Arctic climate change.