A Deep Learning Method for Mapping Glacial Lakes from the Combined Use of Synthetic-Aperture Radar and Optical Satellite Images

Renzhe Wu,Wei Xiang,Jialun Cai,Yong Li,Bo Zhang,Xiaowen Wang,Guoxiang Liu,Rui Zhang

doi:10.3390/rs12244020

Abstract

Glacial lakes (GLs), a vital link between the hydrosphere and the cryosphere, participate in the local hydrological process, and their interannual dynamic evolution is an objective reflection and an indicator of regional climate change. The complex terrain and climatic conditions in mountainous areas where GLs are located make it difficult to employ conventional remote sensing observation means to obtain stable, accurate, and comprehensive observation data. In view of this situation, this study presents an algorithm with a high generalization ability established by optimizing and improving a deep learning (DL) semantic segmentation network model for extracting GL contours from combined synthetic-aperture radar (SAR) amplitude and multispectral imagery data. The aim is to use the high penetrability and all-weather advantages of SAR to reduce the effects of cloud cover as well as to integrate the multiscale and detail-oriented advantages of multispectral data to facilitate accurate, quantitative extraction of GL contours. The accuracy and reliability of the model and algorithm were examined by employing them to extract the contours of GLs in a large region of south-eastern Tibet from Landsat 8 optical remote sensing images and Sentinel-1A amplitude images. In this study, the contours of a total 8262 GLs in south-eastern Tibet were extracted. These GLs were distributed predominantly at altitudes of 4000–5500 m. Only 17.4% of these GLs were greater than 0.1 km2 in size, while a large number of small GLs made up the majority. Through analysis and validation, the proposed method was found highly capable of distinguishing rivers and lakes and able to effectively reduce the misidentification and extraction of rivers. With the DL model based on combined optical and SAR images, the intersection-over-union (IoU) score increased by 0.0212 (to 0.6207) on the validation set and by 0.038 (to 0.6397) on the prediction set. These validation data sufficiently demonstrate the efficacy of the model and algorithm. The technical means employed in this study as well as the results and data obtained can provide a reference for research and application expansion in related fields.

Highlights

Mountainous glacial lakes (GLs) are a major constituent of lake groups and are vitally important for research relating to water resources, cryospheric science, climate, and mountain disasters [1,2]
Where y is the label, yis the classification result produced by the softmax function, Pxi is the foreground probability of the labelled pixel, Pxis the foreground probability of the predicted pixel, α is the control parameter, S is generally set to be greater than 0 but smaller than 10−6 to ensure that T > 0, and L is the loss function corresponding to the deep learning (DL) computation
Gradual temperature increases and accelerated glacial melting have been seen in recent years in south-eastern Tibet [27,28], where the annual precipitation has reached 1000–3000 mm in the glacial zones

Summary

Introduction

Mountainous glacial lakes (GLs) are a major constituent of lake groups and are vitally important for research relating to water resources, cryospheric science, climate, and mountain disasters [1,2]. Many researchers in China and elsewhere have extensively investigated effective methods for extracting GL contours from remote sensing images. Good results have been achieved by employing spectral information index (e.g., the normalized difference water index (NDWI) and the modified NDWI (MNDWI)) threshold methods to extract the contours of bodies of water from multispectral remote sensing images [13,14]. Hanshaw et al used 158 multispectral satellite images spanning almost four decades, from 1975 to 2012, to obtain lake-area outlines for the understudied Cordillera Vilcanota region [23] These methods facilitate the extraction of GL contours and increase the understanding of the temporal and spatial development of GLs. extensive manual interpretation and editing remain a factor that circumscribes the development of GL monitoring.

U-Net Model Architecture

Principle and Algorithm Analysis

General Information about the Study Area

Data Processing