The stacking method enhances machine learning models for monitoring and understanding regional soil available nitrogen variations in croplands
文献类型：期刊论文
| 作者 | Lei, Sihong1,3; Shao, Mingan1,2,3; Jia, Xiaoxu1,3; Zhu, Zhaocen2; Zhao, Chunlei1,3 |
| 刊名 | SOIL & TILLAGE RESEARCH
 |
| 出版日期 | 2026-02-01 |
| 卷号 | 256页码:106880 |
| 关键词 | Agricultural ecosystem Machine learning Model ensemble Quantitative prediction |
| ISSN号 | 0167-1987 |
| DOI | 10.1016/j.still.2025.106880 |
| 产权排序 | 1 |
| 文献子类 | Article |
| 英文摘要 | Soil available nitrogen (AN) is crucial for crop growth, grain yield, and sustainable agricultural management. The Guanzhong Plain (GP) is an important grain production area in the Yellow River basin of China with intensive agricultural activities for over 2000 years and excess nitrate loading. To predict the spatial distribution of AN in the root zone (0-100 cm), 124 soil samples were collected via borehole drilling, followed by lab analysis and AN prediction model development (machine learning models, MLMs and ensemble models, EMs). The results indicated that nitrate (NO3--N) and ammonia (NH4+-N) contents declined with increasing depth, with significantly higher values in the upper 40 cm. NH4+-N contents were lower and relatively stable across soil layers. EMs outperformed MLMs, with the stacking method performing better and improving averaged R2, RMSE, and MAE by 10.48 %, 4.93 %, and 6.99 % for NO3--N prediction and 6.75 %, 9.41 %, and 8.94 % for NH4+-N prediction. Soil variables were most critical for NO3--N prediction, contributing 46 % of the relative importance, followed by topography (22 %) and climate (17 %). NH4+-N predictors were dominated by topographic variables, accounting for 51 %. These findings highlight the distinct roles of soil and topography in regulating nitrogen dynamics, with soil properties controlling nitrification and leaching processes for NO3--N and topography influencing water redistribution and retention for NH4+-N. This study provides references to precise fertilizer management and non point source pollution control in GP. It also underscores the potential of ensemble models, particularly stacking, in improving AN prediction accuracy across agroecosystems. |
| URL标识 | 查看原文 |
| WOS关键词 | NITRATE ; CARBON ; CHINA |
| WOS研究方向 | Agriculture |
| 语种 | 英语 |
| WOS记录号 | WOS:001577479700001 |
| 出版者 | ELSEVIER |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/216111]  |
| 专题 | 黄河三角洲现代农业工程实验室_外文论文 |
| 通讯作者 | Zhao, Chunlei |
| 作者单位 | 1.Univ Chinese Acad Sci, Coll Resources & Environm, Beijing 100190, Peoples R China; 2.Northwest A&F Univ, Coll Nat Resources & Environm, Yangling 712100, Peoples R China 3.Chinese Acad Sci, Key Lab Ecosyst Network Observat & Modeling, Inst Geog Sci & Nat Resources Res, Beijing 100101, Peoples R China; |
| 推荐引用方式 GB/T 7714 | Lei, Sihong,Shao, Mingan,Jia, Xiaoxu,et al. The stacking method enhances machine learning models for monitoring and understanding regional soil available nitrogen variations in croplands[J]. SOIL & TILLAGE RESEARCH,2026,256:106880. |
| APA | Lei, Sihong,Shao, Mingan,Jia, Xiaoxu,Zhu, Zhaocen,&Zhao, Chunlei.(2026).The stacking method enhances machine learning models for monitoring and understanding regional soil available nitrogen variations in croplands.SOIL & TILLAGE RESEARCH,256,106880. |
| MLA | Lei, Sihong,et al."The stacking method enhances machine learning models for monitoring and understanding regional soil available nitrogen variations in croplands".SOIL & TILLAGE RESEARCH 256(2026):106880. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。