Yuanli Qin , Li Zhang , Qiaolin Wang , Yuntao Song , Min Peng , Hangxin Cheng
{"title":"Driving factors of soil selenium accumulation in regional enrichment area at selenium-deficient soil belt of China: An enlightenment of Moran's index and machine learning","authors":"Yuanli Qin , Li Zhang , Qiaolin Wang , Yuntao Song , Min Peng , Hangxin Cheng","doi":"10.1016/j.gexplo.2025.107746","DOIUrl":null,"url":null,"abstract":"<div><div>Selenium (Se) is one of the most indispensable trace elements for human beings, known as the ‘longevity element’ in the 21st century. Due to the complexity of the global geological environment, the distribution of soil Se varies significantly across the world. Understanding the distribution pattern and driving factors of soil Se enrichment is crucial for land resource development and addressing “hidden hunger”, especially in regions where soil Se is abundant. Therefore, a soil geochemical survey was conducted in the Baoshan area of southwest China, which is located at the Se-deficient soil belt yet exhibited characteristics of soil Se enrichment. Combined with Moran's I analysis and machine learning (ML), the enrichment and distribution characteristics, and driving factors of soil Se in this region were revealed. Specifically, we quantify the feature importance of each key factors. Additionally, distribution of soil Se demonstrated strong spatial dependence. The results from Bivariate Moran's I analysis, back-propagation artificial neural networks (BP–ANN) and random forest (RF) models indicated that soil organic carbon (SOC), pH, manganese (Mn), weathering eluvial index (ba), land use and parent materials are the primary driving factors of soil Se retention at the regional scale in Baoshan. In contrast, climate and topography, which are common factors affecting Se distribution at global and continental scales, were found to have no effect on Se retention at this regional scale. Based on the RF model, the relative importance of key driving factors are SOC (24.5 %), land use (21.1 %), pH (15.5 %), ba (15.2 %), parent materials (12.0 %), and Mn (11.9 %). These findings provide valuable guidance for agricultural managements planning and for adjusting agricultural planting structures in marginal and soil Se-rich regions of the traditional Se-deficient soil belt in China.</div></div>","PeriodicalId":16336,"journal":{"name":"Journal of Geochemical Exploration","volume":"274 ","pages":"Article 107746"},"PeriodicalIF":3.4000,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Geochemical Exploration","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0375674225000780","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GEOCHEMISTRY & GEOPHYSICS","Score":null,"Total":0}
引用次数: 0
摘要
硒(Se)是人类最不可或缺的微量元素之一,在 21 世纪被称为 "长寿元素"。由于全球地质环境的复杂性,世界各地土壤中硒的分布差异很大。了解土壤 Se 富集的分布模式和驱动因素对于土地资源开发和解决 "隐性饥饿 "问题至关重要,尤其是在土壤 Se 丰富的地区。因此,我们在位于缺Se土壤带但呈现土壤Se富集特征的中国西南保山地区开展了土壤地球化学调查。结合莫兰 I 分析和机器学习(ML),揭示了该地区土壤硒的富集分布特征和驱动因素。具体而言,我们量化了各关键因素的特征重要性。此外,土壤硒的分布还表现出很强的空间依赖性。双变量莫兰I分析、反向传播人工神经网络(BP-ANN)和随机森林(RF)模型的结果表明,土壤有机碳(SOC)、pH值、锰(Mn)、风化冲积指数(ba)、土地利用和母质是保山区域尺度土壤Se保留的主要驱动因子。相比之下,气候和地形是全球和大陆尺度上影响硒分布的常见因素,但在该区域尺度上却对硒的保留没有影响。根据 RF 模型,主要驱动因素的相对重要性分别为 SOC(24.5%)、土地利用(21.1%)、pH 值(15.5%)、ba(15.2%)、母质(12.0%)和锰(11.9%)。这些研究结果为中国传统缺硒土壤带上的边缘富硒地区的农业管理规划和农业种植结构调整提供了宝贵的指导。
Driving factors of soil selenium accumulation in regional enrichment area at selenium-deficient soil belt of China: An enlightenment of Moran's index and machine learning
Selenium (Se) is one of the most indispensable trace elements for human beings, known as the ‘longevity element’ in the 21st century. Due to the complexity of the global geological environment, the distribution of soil Se varies significantly across the world. Understanding the distribution pattern and driving factors of soil Se enrichment is crucial for land resource development and addressing “hidden hunger”, especially in regions where soil Se is abundant. Therefore, a soil geochemical survey was conducted in the Baoshan area of southwest China, which is located at the Se-deficient soil belt yet exhibited characteristics of soil Se enrichment. Combined with Moran's I analysis and machine learning (ML), the enrichment and distribution characteristics, and driving factors of soil Se in this region were revealed. Specifically, we quantify the feature importance of each key factors. Additionally, distribution of soil Se demonstrated strong spatial dependence. The results from Bivariate Moran's I analysis, back-propagation artificial neural networks (BP–ANN) and random forest (RF) models indicated that soil organic carbon (SOC), pH, manganese (Mn), weathering eluvial index (ba), land use and parent materials are the primary driving factors of soil Se retention at the regional scale in Baoshan. In contrast, climate and topography, which are common factors affecting Se distribution at global and continental scales, were found to have no effect on Se retention at this regional scale. Based on the RF model, the relative importance of key driving factors are SOC (24.5 %), land use (21.1 %), pH (15.5 %), ba (15.2 %), parent materials (12.0 %), and Mn (11.9 %). These findings provide valuable guidance for agricultural managements planning and for adjusting agricultural planting structures in marginal and soil Se-rich regions of the traditional Se-deficient soil belt in China.
期刊介绍:
Journal of Geochemical Exploration is mostly dedicated to publication of original studies in exploration and environmental geochemistry and related topics.
Contributions considered of prevalent interest for the journal include researches based on the application of innovative methods to:
define the genesis and the evolution of mineral deposits including transfer of elements in large-scale mineralized areas.
analyze complex systems at the boundaries between bio-geochemistry, metal transport and mineral accumulation.
evaluate effects of historical mining activities on the surface environment.
trace pollutant sources and define their fate and transport models in the near-surface and surface environments involving solid, fluid and aerial matrices.
assess and quantify natural and technogenic radioactivity in the environment.
determine geochemical anomalies and set baseline reference values using compositional data analysis, multivariate statistics and geo-spatial analysis.
assess the impacts of anthropogenic contamination on ecosystems and human health at local and regional scale to prioritize and classify risks through deterministic and stochastic approaches.
Papers dedicated to the presentation of newly developed methods in analytical geochemistry to be applied in the field or in laboratory are also within the topics of interest for the journal.