|
A Method for Identifying Flooded Areas Based on Change Detection Techniques and Google Earth Engine Cloud Processing Using Radar and Optical Images
|
Reza Saeed *   , Ali Mohammadzadeh |
|
|
|
Abstract: (5 Views) |
| Flooding is one of the most destructive and frequent natural disasters that occurs due to heavy rainfall, rising river levels, and various combined factors.This study presents a new multisource approach in remote sensing that combines multispectral optical images and weather-independent radar data (SAR), facilitating the automatic and accurate mapping of flood extent.By leveraging the cloud processing capabilities and near-real-time features of Google Earth Engine (GEE), this method enables rapid and extensive flood monitoring.In this study, using Sentinel-1 and Sentinel-2 satellite data, two selected regions in Spain and Mexico, a multi-temporal change detection and dynamic thresholding framework for SAR data was designed, and its accuracy was evaluated with flood extent maps extracted from Sentinel-2 optical images.The results showed that fixed thresholding for flood extraction from SAR data is not generalizable to all geographical areas, so an automatic sensitivity analysis was used to determine optimal thresholds suitable for each region.The accuracy assessment of this method showed that the degree of agreement between SAR data and optical images for Spain is between 82% and 85%, and for Mexico between 67% and 74%, with the advantage that SAR is capable of identifying flood-prone areas even under cloudy conditions.Additionally, the studies showed that CVA (Change Vector Analysis) performs better under various climatic conditions, different land uses, and diverse topographical features.This study shows that the combination of SAR and optical images within a multi-source and multi-temporal framework provides an accurate, rapid, and effective approach for monitoring extensive floods.The results of this research can provide reliable and timely information for policymakers, relief organizations, and crisis managers, playing a significant role in disaster management and mitigating the impacts of devastating floods. |
Article number: 5 |
| Keywords: Flood Detection, Crisis Management, Google Earth Engine, Change Detection |
|
|
Full-Text [PDF 2665 kb]
(3 Downloads)
|
Type of Study: Research |
Subject:
Photo&RS
Received: 2025/03/8
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
