Ping Xin, Charles Pesch, Trine Norgaard, Goswin Heckrath, Lis W. de Jonge, Bo V. Iversen
{"title":"The Linkage Between Near-Saturated Hydraulic Conductivity and Tritium Leaching","authors":"Ping Xin, Charles Pesch, Trine Norgaard, Goswin Heckrath, Lis W. de Jonge, Bo V. Iversen","doi":"10.1111/ejss.70121","DOIUrl":null,"url":null,"abstract":"<p>Macropore flow in structured soils is an important process determining the transport of water, contaminants, and nutrients in the soil. Therefore, we also expect a close connection between hydraulic conductivity (<i>k</i>(<i>h</i>)) near saturation and the potential of macropore flow. In combination with measurements of soil hydraulic properties (SHPs), tracer breakthrough characteristics can be used to get an insight into the understanding of macropore flow in structured soils. In this study, we aim to investigate if a direct link exists between tracer breakthrough characteristics and SHPs of structured soils, which may partly explain the dynamics and the spatial variation of solute transport in soils. We hypothesize that a direct relationship exists between the characteristics of breakthrough curves (BTCs) and the near-saturated <i>k</i>(<i>h</i>) of the soil. We used SHPs and tracer breakthrough characteristics for 71 undisturbed topsoil columns (20 cm height, 20 cm diameter) sampled from eight different sites in Denmark. We defined <i>k</i>[10] (near-saturated hydraulic conductivity) as <i>k</i>(<i>h</i>) at a matric potential (<i>h</i>) of −10 cm. On the same soil columns, based on the tracer breakthrough experiment, we calculated the 5%, 25%, and 50% arrival times (ATs) as the percentage of the cumulative relative mass of the tritium tracer leaching through the soil column. Linear mixed models (LMMs) effectively captured the linear relationships among variables. However, applying a machine learning method (Gradient Boosting Decision Trees, GBDT) further clarified the importance of predictors by capturing nonlinear threshold effects and key interactions among soil hydraulic properties. Although the overall predictive accuracy of GBDT was slightly lower compared to LMM, both methods consistently highlighted <i>k</i>[10] as the most influential predictor, emphasizing its key role in preferential flow dynamics. We conclude that linking SHPs with tracer breakthrough characteristics on large intact columns is highly useful for characterizing soil macropore functions.</p>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 3","pages":""},"PeriodicalIF":4.0000,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/ejss.70121","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"European Journal of Soil Science","FirstCategoryId":"97","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/ejss.70121","RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"SOIL SCIENCE","Score":null,"Total":0}
引用次数: 0
Abstract
Macropore flow in structured soils is an important process determining the transport of water, contaminants, and nutrients in the soil. Therefore, we also expect a close connection between hydraulic conductivity (k(h)) near saturation and the potential of macropore flow. In combination with measurements of soil hydraulic properties (SHPs), tracer breakthrough characteristics can be used to get an insight into the understanding of macropore flow in structured soils. In this study, we aim to investigate if a direct link exists between tracer breakthrough characteristics and SHPs of structured soils, which may partly explain the dynamics and the spatial variation of solute transport in soils. We hypothesize that a direct relationship exists between the characteristics of breakthrough curves (BTCs) and the near-saturated k(h) of the soil. We used SHPs and tracer breakthrough characteristics for 71 undisturbed topsoil columns (20 cm height, 20 cm diameter) sampled from eight different sites in Denmark. We defined k[10] (near-saturated hydraulic conductivity) as k(h) at a matric potential (h) of −10 cm. On the same soil columns, based on the tracer breakthrough experiment, we calculated the 5%, 25%, and 50% arrival times (ATs) as the percentage of the cumulative relative mass of the tritium tracer leaching through the soil column. Linear mixed models (LMMs) effectively captured the linear relationships among variables. However, applying a machine learning method (Gradient Boosting Decision Trees, GBDT) further clarified the importance of predictors by capturing nonlinear threshold effects and key interactions among soil hydraulic properties. Although the overall predictive accuracy of GBDT was slightly lower compared to LMM, both methods consistently highlighted k[10] as the most influential predictor, emphasizing its key role in preferential flow dynamics. We conclude that linking SHPs with tracer breakthrough characteristics on large intact columns is highly useful for characterizing soil macropore functions.
期刊介绍:
The EJSS is an international journal that publishes outstanding papers in soil science that advance the theoretical and mechanistic understanding of physical, chemical and biological processes and their interactions in soils acting from molecular to continental scales in natural and managed environments.