Driving Forces of Circum-Antarctic Glacier and Ice Shelf Front Retreat over the Last Two Decades

Abstract

Abstract. The safety band of Antarctica consisting of floating glacier tongues and ice shelves buttresses ice discharge of the Antarctic Ice Sheet. Recent disintegration events of ice shelves and glacier retreat indicate a weakening of this important safety band. Predicting calving front retreat is a real challenge due to complex ice dynamics in a data-scarce environment being unique for each ice shelf and glacier. We explore to what extent easy to access remote sensing and modelling data can help to define environmental conditions leading to calving front retreat. For the first time, we present a circum-Antarctic record of glacier and ice shelf front retreat over the last two decades in combination with environmental variables such as air temperature, sea ice days, snowmelt, sea surface temperature and wind direction. We find that the Antarctic ice sheet area shrank 29,618 ± 29 km2 in extent between 1997–2008 and gained an area of 7,108 ± 144.4 km2 between 2009 and 2018. Retreat concentrated along the Antarctic Peninsula and West Antarctica including the biggest ice shelves Ross and Ronne. Glacier and ice shelf retreat comes along with one or several changes in environmental variables. Decreasing sea ice days, intense snow melt, weakening easterlies and relative changes in sea surface temperature were identified as enabling factors for retreat. In contrast, relative increases in air temperature did not correlate with calving front retreat. To better understand drivers of glacier and ice shelf retreat it is of high importance to analyse the magnitude of basal melt through the intrusion of warm Circumpolar Deep Water (CDW) driven by strengthening westerlies and to further assess surface hydrology processes such as meltwater ponding, runoff and lake drainage.

Object Detection and Image Segmentation with Deep Learning on Earth Observation Data: A Review—Part II: Applications

“…This proof-of-concept study develops a deep-learning-based method, which can monitor the Antarctic coastline, by segmentation of Sentinel-1 data in different locations. This CNN-based deep-learning model was recently incorporated into a set of methods to answer geoscientific research questions on a large scale [310]. Such a large-scale application of radar data with deep learning was found to be missing until now in Earth observation [18].…”

Automated Extraction of Annual Erosion Rates for Arctic Permafrost Coasts Using Sentinel-1, Deep Learning, and Change Vector Analysis

“…This study exploited the capabilities of DL and, in particular, of the U-Net architecture for extracting the Arctic coastline with high resolution and accuracy. The successful implementation of the U-Net framework for detecting coastlines based on SAR data in Antarctic environments was hereby already demonstrated in previous works [40][41][42]. In this study, nine different U-Net architectures were employed to generate binary classification maps that differentiate between sea and land area (including inland lakes and rivers).…”

“…One major goal of this study is to exploit the segmentation capabilities of DL in order to create a high-quality Arctic coastline product, which will act as a reference for the analysis of coastal erosion and build-up. For this purpose, a CNN-based U-Net architecture was employed, which proved to be highly capable in the context of detecting coastlines using SAR imagery [39][40][41][42]. A brief overview of the CNN-based U-Net structure and the hyper-parameters used within this study will be given here.…”

A Novel Method for Automated Supraglacial Lake Mapping in Antarctica Using Sentinel-1 SAR Imagery and Deep Learning

“…For removal of false lake classifications seaward of the Antarctic coastline, a Sentinel-1 coastline product [34,49] was additionally integrated into post-classification. The circum-Antarctic coastline was derived from Sentinel-1 data covering the period June to August 2018 using an automated processing pipeline specifically developed for semantic segmentation of the Antarctic calving front based on a modified U-Net [34].…”