Prediction and mapping soil organic carbon in Jianghan Plain by machine learning
Prediction and mapping soil organic carbon in Jianghan Plain by machine learning
About this item
Full title
Author / Creator
Shen Jiali , Chen Songchao , Hu Bifeng and Li, Shuo
Publisher
Tianjin: Journal of Agricultural Resources and Environment (JARE)
Journal title
Language
Chinese
Formats
Publication information
Publisher
Tianjin: Journal of Agricultural Resources and Environment (JARE)
Subjects
More information
Scope and Contents
Contents
土壤有机碳（SOC）不仅显著响应于表层，而且随深度的变化呈现不同的响应。江汉平原作为长江经济带农田生态系统的重要组成部分，其SOC垂向分布状况仍有待考察。本研究收集了2009—2012年湖北省土系调查的66个剖面数据，基于9个环境因子为协变量使用随机森林法构建 0~30、>30~60、>60~100 cm土层深度的 SOC含量预测模型，绘制了 30 m空间分辨率的 SOC含量分布图，并估算了 SOC 储量。结果表明，SOC 含量随土层深度的增加而减少，总体呈中东部高、西部低的特征。模型对表层（0~30 cm）SOC的预测精度最高（R2=0.45，RMSE=3.28 g·kg-1），温度、黏粒含量和降水对模型重要性居前三。江汉平原 1 m深 SOC储量为 183.75 Tg，>30~100 c...
Alternative Titles
Full title
Prediction and mapping soil organic carbon in Jianghan Plain by machine learning
Authors, Artists and Contributors
Author / Creator
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_proquest_journals_2820324830
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2820324830
Other Identifiers
ISSN
2095-6819
E-ISSN
2095-6819
DOI
10.13254/j.jare.2022.0583
