Extracting Glacier Calving Fronts by Deep Learning: The Benefit of Multispectral, Topographic, and Textural Input Features

Publikation: Beitrag in FachzeitschriftForschungsartikelBeigetragenBegutachtung

Beitragende

  • Erik Loebel - , Professur für Geodätische Erdsystemforschung (Autor:in)
  • Mirko Scheinert - , Professur für Geodätische Erdsystemforschung (Autor:in)
  • Martin Horwath - , Professur für Geodätische Erdsystemforschung (Autor:in)
  • Konrad Heidler - , Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR) Standort Oberpfaffenhofen, Technische Universität München (Autor:in)
  • Julia Christmann - , Alfred-Wegener-Institut, Helmholtz-Zentrum für Polar- und Meeresforschung, Rheinland-Pfälzische Technische Universität Kaiserslautern-Landau (Autor:in)
  • Long Duc Phan - , Alfred-Wegener-Institut, Helmholtz-Zentrum für Polar- und Meeresforschung (Autor:in)
  • Angelika Humbert - , Alfred-Wegener-Institut, Helmholtz-Zentrum für Polar- und Meeresforschung, Universität Bremen (Autor:in)
  • Xiao Xiang Zhu - , Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR) Standort Oberpfaffenhofen, Technische Universität München (Autor:in)

Abstract

An accurate parameterization of glacier calving is essential for understanding glacier dynamics and constraining ice-sheet models. The increasing availability and quality of remote sensing imagery open the prospect of a continuous and precise mapping of relevant parameters, such as calving front locations. However, it also calls for automated and scalable analysis strategies. Deep neural networks provide powerful tools for processing large quantities of remote sensing data. In this contribution, we assess the benefit of diverse input data for calving front extraction. In particular, we focus on Landsat-8 imagery supplementing single-band inputs with multispectral data, topography, and textural information. We assess the benefit of these three datasets using a dropped-variable approach. The associated reference dataset comprises 728 manually delineated calving front positions of 23 Greenland and two Antarctic outlet glaciers from 2013 to 2021. Resulting feature importance emphasizes both the potential integrating additional input information as well as the significance of their thoughtful selection. We advocate utilizing multispectral features as their integration leads generally to more accurate predictions compared with conventional single-band inputs. This is especially prevalent for challenging ice mélange and illumination conditions. In contrast, the application of both textural and topographic inputs cannot be recommended without reservation, since they may lead to model overfitting. The results of this assessment are not only relevant for advancing automated calving front extraction but also for a wider range of glaciology-related land surface classification tasks using deep neural networks.

Details

OriginalspracheEnglisch
Aufsatznummer4306112
Seitenumfang12
FachzeitschriftIEEE Transactions on Geoscience and Remote Sensing
Jahrgang60
PublikationsstatusVeröffentlicht - 21 Sept. 2022
Peer-Review-StatusJa

Externe IDs

Scopus 85139407127
dblp journals/tgrs/LoebelSHHCPHZ22
Mendeley 3f3eb172-8e89-324f-a3fb-9d10eb5e942c
ORCID /0000-0001-5797-244X/work/142246563
ORCID /0000-0002-0892-8941/work/142248921
ORCID /0000-0001-9874-9295/work/142255131

Schlagworte

Fächergruppen, Lehr- und Forschungsbereiche, Fachgebiete nach Destatis

Schlagwörter

  • Deep learning, Fernerkundung, Gletschermonitoring, Grönland

Bibliotheksschlagworte