Center for Technology in Water and Wastewater, School of Civil and Environmental Engineering, University of Technology Sydney, Sydney, NSW 2007, Australia.
Department of Earth and Environmental Sciences, Macquarie University, North Ryde, NSW 2109, Australia; Astron Environmental Services, 129 Royal Street, East Perth, Western Australia 6004, Australia.
Sci Total Environ. 2022 Jan 15;804:150187. doi: 10.1016/j.scitotenv.2021.150187. Epub 2021 Sep 8.
Monitoring agricultural soil organic carbon (SOC) has played an essential role in sustainable agricultural management. Precise and robust prediction of SOC greatly contributes to carbon neutrality in the agricultural industry. To create more knowledge regarding the ability of remote sensing to monitor carbon soil, this research devises a state-of-the-art low cost machine learning model for quantifying agricultural soil carbon using active and ensemble-based decision tree learning combined with multi-sensor data fusion at a national and world scale. This work explores the use of Sentinel-1 (S1) C-band dual polarimetric synthetic aperture radar (SAR), Sentinel-2 (S2) multispectral data, and an innovative machine learning (ML) approach using an integration of active learning for land-use mapping and advanced Extreme Gradient Boosting (XGBoost) for robustness of the SOC estimates. The collected soil samples from a field survey in Western Australia were used for the model validation. The indicators including the coefficient of determination (R) and root - mean - square - error (RMSE) were applied to evaluate the model's performance. A numerous features computed from optical and SAR data fusion were employed to build and test the proposed model performance. The effectiveness of the proposed machine learning model was assessed by comparing with the two well-known algorithms such as Random Forests (RF) and Support Vector Machine (SVM) to predict agricultural SOC. Results suggest that a combination of S1 and S2 sensors could effectively estimate SOC in farming areas by using ML techniques. Satisfactory accuracy of the proposed XGBoost with optimal features was achieved the highest performance (R = 0.870; RMSE = 1.818 tonC/ha) which outperformed RF and SVM. Thus, multi-sensor data fusion combined with the XGBoost lead to the best prediction results for agricultural SOC at 10 m spatial resolution. In short, this new approach could significantly contribute to various agricultural SOC retrieval studies globally.
监测农业土壤有机碳（SOC）在可持续农业管理中发挥了重要作用。精确而稳健的 SOC 预测对农业行业的碳中和有很大贡献。为了在遥感监测碳土壤方面创造更多的知识，本研究设计了一种最先进的低成本机器学习模型，用于在国家和全球范围内利用主动和基于集成的决策树学习以及多传感器数据融合来量化农业土壤碳。这项工作探讨了利用 Sentinel-1（S1）C 波段双极化合成孔径雷达（SAR）、Sentinel-2（S2）多光谱数据以及一种创新的机器学习（ML）方法，该方法使用主动学习进行土地利用制图和高级极端梯度提升（XGBoost）的稳健性来集成 SOC 估计。从西澳大利亚野外调查中收集的土壤样本用于模型验证。使用决定系数（R）和均方根误差（RMSE）等指标来评估模型的性能。融合光学和 SAR 数据后计算了许多特征，用于构建和测试所提出的模型性能。通过与两种知名算法（随机森林（RF）和支持向量机（SVM））比较，评估所提出的机器学习模型的有效性，以预测农业 SOC。结果表明，通过使用 ML 技术，S1 和 S2 传感器的组合可以有效地估计农业区的 SOC。最优特征的 XGBoost 具有令人满意的准确性，实现了最高性能（R=0.870；RMSE=1.818 吨 C/公顷），优于 RF 和 SVM。因此，多传感器数据融合与 XGBoost 相结合，可以在 10 m 空间分辨率下实现农业 SOC 的最佳预测结果。总之，这种新方法可以为全球各种农业 SOC 检索研究做出重大贡献。
