Analysis and modeling of the distribution of tracers of glacial melt in the Southern Ocean with focus on the Southeast Pacific sector

Abstract

In the Pacific sector of the Southern Ocean, the ice shelves and glaciers of the West Antarctic Ice Sheet (WAIS) are rapidly thinning, especially in the Amundsen Sea (AS) and Bellingshausen Sea (BS). The high basal melting of these relative small ice shelves is caused by warm Circum- polar Deep Water (CDW) that, based on observations, mainly intrudes via submarine glacial troughs located at the continental shelf break. When CDW reaches the grounding line of the fringing glaciers, strong basal melting supports a buoyant melt water plume. As the glacial melt becomes part of the shelf circulation, it may cause a freshening of the shelf water locally as well as remotely in the Ross Sea (RS). In the RS, on the other hand, shelf water salinity has declined by 0.03 per decade over the past 50 years. Thus, the objective of this doctoral thesis is to simu- late the spreading of glacial melt water using the Finite-Element Sea-ice/ice-shelf/Ocean Model (FESOM) and investigate whether the ice shelf melting in the AS is one of the main reasons for the observed shelf water freshening in the RS. However, there are two main difficulties for the simulation of the spreading of glacial melt water: (1) the number of observations on the Amundsen Sea continental shelf and thus our un- derstanding of CDW intrusions, ice shelf basal melting, and spreading of glacial melt water is limited and (2) none of the high-resolution global models resolving all small ice shelves around Antarctica is able to simulate CDW intrusions realistically. Thus, in this doctoral study the hy- drographic data of the cruise ANTXXVI/3 from the eastern AS continental shelf was analyzed, focusing on intrusions of CDW, basal melting of Pine Island Ice Shelf (PIIS), and spreading of glacial melt water. The data analysis was followed by several model sensitivity experiments using FESOM. They revealed that the model configuration such as horizontal resolution and atmospheric forcing is important for the reproduction of CDW intrusions. With these modifications, FESOM reproduces the bottom temperatures in the AS and BS and basal melt rates of AS and BS ice shelves much closer to reality. Therefore, I tested whether the observed freshening of the RS is a consequence of enhanced basal melting of AS ice shelves. Using several independent virtual passive tracers to identify pathways of the glacial melt, I found that the melt water from the ice shelves in the AS mainly flows towards the Ross Ice Shelf front. After 10 years of simulation, about half of the melt water in the RS originates from the Getz Ice Shelf. Further, I investigated the sensitivity of the melt water transport into the RS associated with the strength of the basal melt water flux. When this flux is increased by 30%, the transport of glacial melt into the RS nearly doubles, supporting the idea that the basal melting of AS and BS ice shelves is one of the main reasons for the observed freshening on the RS continental shelf. This may imply that the observed freshening of the RS in the late 1970's/early 1980's, prior to the remotely sensed increase in ice shelf mass losses, might have been caused by enhanced basal melting already in the 1960' s.