Advancing ecological restoration: A novel 3D interpolation method for assessing ammonia-nitrogen pollution in rare earth mining areas
文献类型：期刊论文
| 作者 | Nie, Shengdong4; Li, Hengkai4; Li, Ziyang2; Tao, Huan3; Wang, Guanshi4; Zhou, Yanbing1 |
| 刊名 | EXPERT SYSTEMS WITH APPLICATIONS
 |
| 出版日期 | 2025-06-01 |
| 卷号 | 276页码:127192 |
| 关键词 | Ammonia nitrogen Hierarchical modeling Geostatistics Bayesian prediction Three-dimensional spatial interpolation |
| ISSN号 | 0957-4174 |
| DOI | 10.1016/j.eswa.2025.127192 |
| 产权排序 | 2 |
| 文献子类 | Article |
| 英文摘要 | The in situ leaching method for ion adsorption of rare earth ores often leads to significant soil and water pollution, of which ammonia nitrogen (AN) pollution is one of the main problems. The three-dimensional spatial distribution of AN pollution displays weak spatial autocorrelation and strong spatial heterogeneity. Traditional geostatistical methods obscure to accurately capture these characteristics due to challenges in effectively utilizing sparse drilling sample data. In this paper, we propose a three-dimensional interpolation (Hierarchical Bayesian Model with Mean Surface and Stratified Non-homogeneity, HBM-MSN) method for ammonia-nitrogen pollution distribution in mining areas based on a hierarchical Bayesian framework. The method captures local spatial structures by integrating a hierarchical Bayesian model with a stochastic Gaussian process, enhancing the robustness of semi-variogram parameter fitting in areas with sparse data. A case study involving the three-dimensional spatial distribution of AN pollution in a rare earth industrial park in China demonstrates the superiority of HBM-MSN over traditional methods. Performance was assessed using leave-one-out cross-validation with mean absolute error (MAE) and root mean square error (RMSE). The results indicated that AN distribution in the mining area exhibited significant stratified heterogeneity (q = 0.365, P < 0.01). Compared to the 3D-OK and 3D-MSN models, HBM-MSN reduced the mean absolute error (MAE) by 1.29 mg/kg and 0.13 mg/kg, and the root mean square error (RMSE) by 1.28 mg/kg and 0.73 mg/kg, respectively. This study advances the field of three-dimensional modeling of AN pollution in rare earth mining areas and provides methodological support for ecological restoration, management, and risk assessment in these regions. |
| URL标识 | 查看原文 |
| WOS研究方向 | Computer Science ; Engineering ; Operations Research & Management Science |
| 语种 | 英语 |
| WOS记录号 | WOS:001447081600001 |
| 出版者 | PERGAMON-ELSEVIER SCIENCE LTD |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/213211]  |
| 专题 | 陆地表层格局与模拟院重点实验室_外文论文 |
| 通讯作者 | Li, Hengkai; Tao, Huan |
| 作者单位 | 1.Beijing Acad Agr & Forestry Sci, Informat Technol Res Ctr, Beijing 100097, Peoples R China 2.Beijing Forestry Univ, Sch Technol, Beijing 100083, Peoples R China; 3.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Key Lab Land Surface Pattern & Simulat, Beijing 100101, Peoples R China; 4.Jiangxi Univ Sci & Technolog, Jiangxi Prov Key Lab Water Ecol Conservat Headwate, 1958 Ke jia Rd, Ganzhou 341000, Peoples R China; |
| 推荐引用方式 GB/T 7714 | Nie, Shengdong,Li, Hengkai,Li, Ziyang,et al. Advancing ecological restoration: A novel 3D interpolation method for assessing ammonia-nitrogen pollution in rare earth mining areas[J]. EXPERT SYSTEMS WITH APPLICATIONS,2025,276:127192. |
| APA | Nie, Shengdong,Li, Hengkai,Li, Ziyang,Tao, Huan,Wang, Guanshi,&Zhou, Yanbing.(2025).Advancing ecological restoration: A novel 3D interpolation method for assessing ammonia-nitrogen pollution in rare earth mining areas.EXPERT SYSTEMS WITH APPLICATIONS,276,127192. |
| MLA | Nie, Shengdong,et al."Advancing ecological restoration: A novel 3D interpolation method for assessing ammonia-nitrogen pollution in rare earth mining areas".EXPERT SYSTEMS WITH APPLICATIONS 276(2025):127192. |
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。