
Detection of Inverted Channels Using Sentinel-1 Data and Sentinel-2 Data on Google Earth Engine: A Comparative Analysis
Author(s) -
Xuhua Weng
Publication year - 2025
Publication title -
ieee access
Language(s) - English
Resource type - Magazines
SCImago Journal Rank - 0.587
H-Index - 127
eISSN - 2169-3536
DOI - 10.1109/access.2025.3591178
Subject(s) - aerospace , bioengineering , communication, networking and broadcast technologies , components, circuits, devices and systems , computing and processing , engineered materials, dielectrics and plasmas , engineering profession , fields, waves and electromagnetics , general topics for engineers , geoscience , nuclear engineering , photonics and electrooptics , power, energy and industry applications , robotics and control systems , signal processing and analysis , transportation
Inverted fluvial channels are significant geomorphological features, particularly in arid regions, offering insights into past hydrological regimes and paleo-climatic conditions. Accurate mapping of these features is crucial for understanding landscape evolution in arid regions. However, it is challenging to detect these inverted channels over large areas based on traditional field methods. This study develops and evaluates a workflow for inverted channel detection by comparing the performance of a Sentinel-1 Synthetic Aperture Radar (SAR) only approach with a synergistic data fusion approach. The first method relies on thresholding the backscatter intensity of Sentinel-1 imagery. The second method is to fuse Sentinel-1 SAR data and Sentinel-2 multispectral data for classification using Random Forest (RF) classifier to generate classification probability maps, which are then used for channel detection. The methodologies are tested and validated across four distinct study sites. The results demonstrate the superiority of the data fusion approach. Accuracy assessments show that the integration of Sentinel-2 data improve the classification performance, with Overall Accuracy (OA) increasing from a range of 0.82-0.86 to 0.84-0.90 and Kappa coefficients rising from 0.66-0.77 to 0.69-0.79. Receiver operating characteristic (ROC) analysis further confirms this enhancement, with an increase in the area under the curve (AUC). Visual comparison also shows that the fusion approach produces a more coherent and complete channel network, overcoming the inherent problems of the SAR-only approach. We conclude that the synergy between Sentinel-1’s sensitivity to surface roughness and Sentinel-2’s spectral information provides a more comprehensive characterization of the landscape, leading to a highly accurate and robust delineation of inverted channels. The proposed workflow provides an efficient method for detecting inverted channel, offering a valuable tool for geomorphological research, landscape evolution studies, and hydrological exploration in arid environments.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom