水土保持研究2025，Vol.32Issue(3)：63-71,9.DOI:10.13869/j.cnki.rswc.2025.03.022
松嫩平原西部草地土壤有机质含量预测
Prediction of soil organic matter content in the grassland of the western Songnen Plain
摘要
Abstract
[Objective]This study aims to monitor and predict changes in grassland soil organic matter(SOM)and explore the mechanisms by which climate conditions,topographic factors,and soil physicochemical properties influence SOM content,providing a scientific basis for soil management and agricultural production in specific regions.[Methods]Using the grasslands in the western Songnen Plain as the study area,the predictive performance of the Random Forest,Neural Network,and Gradient Boosting Tree models was evaluated using the Google Earth Engine(GEE)cloud platform and Sentinel-2A multispectral remote sensing images,coupled with meteorological and topographic data.The optimal model was selected to generate a SOM distribution map for the western Songnen Plain grasslands and analyzed factors influencing the distribution of SOM.[Results](1)The Neural Network model based on Sentinel-2 data was highly applicable in predicting SOM content in the western Songnen Plain grasslands.(2)The spatial distribution of SOM content in the western Songnen Plain grasslands exhibited a pattern of higher values in the northeast and low values in the southwest,with SOM content ranging from 0.016 to 62.058 g/kg and an average value of 15.216 g/kg,which was considered moderate.(3)Temperature,latitude,and soil salinization were identified as significant factors influencing SOM content.[Conclusion]The spatial distribution of SOM shows a trend of higher values in the northeast and lower values in the southwest,indicating that areas with lower average annual temperatures,higher latitudes,and moderate electrical conductivity(EC)have significantly higher SOM content than other regions.关键词
土壤有机质/草地/GEE云平台/机器学习/松嫩平原西部Key words
soil organic matter/grassland/GEE cloud platform/machine learning/western Songnen Plain分类
农业科技引用本文复制引用
薄延素,李昊明,王葛霏,温惠清,石振宇,李晓燕..松嫩平原西部草地土壤有机质含量预测[J].水土保持研究,2025,32(3):63-71,9.基金项目
国家自然科学基金"综合植被与土壤信息的松嫩平原西部草地退化遥感评价模型研究"(42171328) （42171328）
