Predicting nitrate exposure from groundwater wells using machine learning and meteorological conditions

IF 2.6 4区 环境科学与生态学 Q3 ENGINEERING, ENVIRONMENTAL Journal of The American Water Resources Association Pub Date : 2023-11-23 DOI:10.1111/1752-1688.13175
Randall Etheridge, Janire Pascual-Gonzalez, Jacob Hochard, Ariane L. Peralta, Thomas J. Vogel
{"title":"Predicting nitrate exposure from groundwater wells using machine learning and meteorological conditions","authors":"Randall Etheridge,&nbsp;Janire Pascual-Gonzalez,&nbsp;Jacob Hochard,&nbsp;Ariane L. Peralta,&nbsp;Thomas J. Vogel","doi":"10.1111/1752-1688.13175","DOIUrl":null,"url":null,"abstract":"<p>Private groundwater wells can be unmonitored sources of contaminated water that can harm human health. Developing models that predict exposure could allow residents to take action to reduce risk. Machine learning models have been successful in predicting nitrate contamination using geospatial information such as proximity to nitrate sources, but previous models have not considered meteorological factors that change temporally. In this study, we test random forest (regression and classification) and linear regression models to predict nitrate contamination using rainfall, temperature, and readily available soil parameters. We trained and tested models for (1) all of North Carolina, (2) each geographic region in North Carolina, (3) a three-county region with a high density of animal agriculture, and (4) a three-county region with a low density of animal agriculture. All regression models had poor predictive performance (<i>R</i><sup>2</sup> &lt; 0.09). The random forest classification model for the coastal plain showed fair agreement (Cohen's <i>κ</i> = 0.23) when trying to predict whether contamination occurred. All other classification models had slight or poor predictive performance. Our results show that temporal changes in rainfall and temperature, or in combination with soil data, are not enough to predict nitrate contamination in most areas of North Carolina. The low level of contamination (&lt;25%) measured during the study could have contributed to the poor performance of the models.</p>","PeriodicalId":17234,"journal":{"name":"Journal of The American Water Resources Association","volume":null,"pages":null},"PeriodicalIF":2.6000,"publicationDate":"2023-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/1752-1688.13175","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of The American Water Resources Association","FirstCategoryId":"93","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/1752-1688.13175","RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ENVIRONMENTAL","Score":null,"Total":0}
引用次数: 0

Abstract

Private groundwater wells can be unmonitored sources of contaminated water that can harm human health. Developing models that predict exposure could allow residents to take action to reduce risk. Machine learning models have been successful in predicting nitrate contamination using geospatial information such as proximity to nitrate sources, but previous models have not considered meteorological factors that change temporally. In this study, we test random forest (regression and classification) and linear regression models to predict nitrate contamination using rainfall, temperature, and readily available soil parameters. We trained and tested models for (1) all of North Carolina, (2) each geographic region in North Carolina, (3) a three-county region with a high density of animal agriculture, and (4) a three-county region with a low density of animal agriculture. All regression models had poor predictive performance (R2 < 0.09). The random forest classification model for the coastal plain showed fair agreement (Cohen's κ = 0.23) when trying to predict whether contamination occurred. All other classification models had slight or poor predictive performance. Our results show that temporal changes in rainfall and temperature, or in combination with soil data, are not enough to predict nitrate contamination in most areas of North Carolina. The low level of contamination (<25%) measured during the study could have contributed to the poor performance of the models.

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
利用机器学习和气象条件预测地下水井的硝酸盐暴露量
私人地下水井可能是不受监控的污染水源,会对人类健康造成危害。开发可预测暴露程度的模型可以让居民采取行动降低风险。机器学习模型已经成功地利用地理空间信息(如与硝酸盐来源的距离)来预测硝酸盐污染,但以前的模型没有考虑随时间变化的气象因素。在本研究中,我们测试了随机森林(回归和分类)和线性回归模型,以利用降雨、温度和现成的土壤参数预测硝酸盐污染。我们对以下地区的模型进行了训练和测试:(1) 整个北卡罗来纳州;(2) 北卡罗来纳州的每个地理区域;(3) 畜牧业密度较高的三个县;(4) 畜牧业密度较低的三个县。所有回归模型的预测性能都很差(R2 为 0.09)。在试图预测是否发生污染时,沿海平原的随机森林分类模型显示出相当的一致性(Cohen's κ = 0.23)。所有其他分类模型的预测效果都较差。我们的研究结果表明,降雨量和温度的时间变化,或与土壤数据相结合,不足以预测北卡罗来纳州大部分地区的硝酸盐污染情况。研究期间测得的污染水平较低(25%),这可能是模型性能较差的原因之一。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Journal of The American Water Resources Association
Journal of The American Water Resources Association 环境科学-地球科学综合
CiteScore
4.10
自引率
12.50%
发文量
100
审稿时长
3 months
期刊介绍: JAWRA seeks to be the preeminent scholarly publication on multidisciplinary water resources issues. JAWRA papers present ideas derived from multiple disciplines woven together to give insight into a critical water issue, or are based primarily upon a single discipline with important applications to other disciplines. Papers often cover the topics of recent AWRA conferences such as riparian ecology, geographic information systems, adaptive management, and water policy. JAWRA authors present work within their disciplinary fields to a broader audience. Our Associate Editors and reviewers reflect this diversity to ensure a knowledgeable and fair review of a broad range of topics. We particularly encourage submissions of papers which impart a ''take home message'' our readers can use.
期刊最新文献
Issue Information Issue Information Evaluation of reported and unreported water uses in various sectors of the Potomac basin for the year 2017 Rapid geomorphic assessment walkabouts as a tool for stream mitigation monitoring Sources of seasonal water supply forecast uncertainty during snow drought in the Sierra Nevada
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1