Estimating optical vegetation indices and biophysical variables for temperate forests with Sentinel-1 SAR data using machine learning techniques: A case study for Czechia
作者: Daniel Paluba, Bertrand Le Saux, Přemysl Stych
分类: stat.ML, cs.LG, stat.AP
发布日期: 2023-11-13 (更新: 2024-08-27)
备注: Revised version of the preprint, based on comments from the reviewers. Full research article. 23 pages, 10 figures, 7 tables
DOI: 10.1080/20964471.2025.2459300
💡 一句话要点
利用SAR数据和机器学习估算森林光学植被指数以解决光学数据局限性问题
🎯 匹配领域: 支柱二:RL算法与架构 (RL & Architecture) 支柱四:生成式动作 (Generative Motion)
关键词: 光学植被指数 合成孔径雷达 机器学习 森林监测 多模态数据 环境监测 捷克 生态系统
📋 核心要点
- 现有的光学卫星数据在森林监测中受到大气影响的限制,无法提供稳定的时间序列数据。
- 本研究通过机器学习技术,利用SAR数据估算光学植被指数,克服了光学数据的局限性。
- 实验结果显示,使用SAR数据的估算方法在准确性和误差上优于传统光学方法,能够实现高频次的森林变化监测。
📝 摘要(中文)
当前用于监测森林生态系统的光学植被指数(VIs)已得到广泛应用,但受限于大气影响如云层。合成孔径雷达(SAR)数据能够穿透云层,提供全天候的森林监测。本研究旨在通过机器学习技术,利用SAR数据作为替代,估算森林的光学VIs。研究聚焦于捷克的健康和受损温带森林,使用Sentinel-1 SAR和Sentinel-2多光谱数据生成的地面真实标签,创建了多模态时间序列数据集。结果表明,传统机器学习算法RFR和XGBoost在所有VIs上表现优于AutoML方法,准确率高达70-86%,平均绝对误差(MAE)在0.055-0.29之间。该方法能够以高精度检测森林的突变变化。
🔬 方法详解
问题定义:本研究旨在解决光学卫星数据在森林监测中因大气影响而导致的局限性,尤其是云层对数据获取的干扰。
核心思路:通过利用合成孔径雷达(SAR)数据,结合机器学习技术,估算森林的光学植被指数(VIs),以实现全天候和高频次的森林监测。
技术框架:研究采用了多模态时间序列数据集,整合了Sentinel-1和Sentinel-2数据,以及数字高程模型(DEM)、气象和土地覆盖数据,构建了一个全面的监测框架。
关键创新:本研究的创新点在于首次将SAR数据与机器学习结合,用于估算光学VIs,突破了传统光学数据的限制,提供了新的监测手段。
关键设计:在模型选择中,采用了传统机器学习算法RFR和XGBoost,并与AutoML方法进行比较,优化了参数设置以提高模型的准确性,损失函数和特征选择也经过精心设计以适应SAR数据的特性。
📊 实验亮点
实验结果显示,传统机器学习算法RFR和XGBoost在估算光学植被指数时表现优异,准确率达到70-86%,平均绝对误差(MAE)在0.055-0.29之间,显著优于AutoML方法。这表明SAR数据在森林监测中的有效性和潜力。
🎯 应用场景
该研究的成果可广泛应用于森林生态监测、环境保护和资源管理等领域。通过提供高频次和高精度的森林变化监测手段,能够为政策制定和生态恢复提供科学依据,具有重要的实际价值和潜在影响。
📄 摘要(原文)
Current optical vegetation indices (VIs) for monitoring forest ecosystems are well established and widely used in various applications, but can be limited by atmospheric effects such as clouds. In contrast, synthetic aperture radar (SAR) data can offer insightful and systematic forest monitoring with complete time series (TS) due to signal penetration through clouds and day and night image acquisitions. This study aims to address the limitations of optical satellite data by using SAR data as an alternative for estimating optical VIs for forests through machine learning (ML). While this approach is less direct and likely only feasible through the power of ML, it raises the scientific question of whether enough relevant information is contained in the SAR signal to accurately estimate VIs. This work covers the estimation of TS of four VIs (LAI, FAPAR, EVI and NDVI) using multitemporal Sentinel-1 SAR and ancillary data. The study focused on both healthy and disturbed temperate forest areas in Czechia for the year 2021, while ground truth labels generated from Sentinel-2 multispectral data. This was enabled by creating a paired multi-modal TS dataset in Google Earth Engine (GEE), including temporally and spatially aligned Sentinel-1, Sentinel-2, DEM, weather and land cover datasets. The inclusion of DEM-derived auxiliary features and additional meteorological information, further improved the results. In the comparison of ML models, the traditional ML algorithms, RFR and XGBoost slightly outperformed the AutoML approach, auto-sklearn, for all VIs, achieving high accuracies ($R^2$ between 70-86%) and low errors (0.055-0.29 of MAE). In general, up to 240 measurements per year and a spatial resolution of 20 m can be achieved using estimated SAR-based VIs with high accuracy. A great advantage of the SAR-based VI is the ability to detect abrupt forest changes with sub-weekly temporal accuracy.