A method for spatial correlations between groundwater vulnerabilities and pollution sources based on machine learning and cluster analysis

BIAN Haobo; CHEN Jian; YANG Ruijie; ZHOU Rui; ZHANG Tao; HUANG Guoxin

doi:10.13205/j.hjgc.202512013

Volume 43 Issue 12

Dec. 2025

Turn off MathJax

Article Contents

Article Navigation > ENVIRONMENTAL ENGINEERING > 2025 > 43(12): 112-120

BIAN Haobo, CHEN Jian, YANG Ruijie, ZHOU Rui, ZHANG Tao, HUANG Guoxin. A method for spatial correlations between groundwater vulnerabilities and pollution sources based on machine learning and cluster analysis[J]. ENVIRONMENTAL ENGINEERING , 2025, 43(12): 112-120. doi: 10.13205/j.hjgc.202512013

Citation:

BIAN Haobo, CHEN Jian, YANG Ruijie, ZHOU Rui, ZHANG Tao, HUANG Guoxin. A method for spatial correlations between groundwater vulnerabilities and pollution sources based on machine learning and cluster analysis[J]. ENVIRONMENTAL ENGINEERING , 2025, 43(12): 112-120. doi: 10.13205/j.hjgc.202512013

Citation:

BIAN Haobo, CHEN Jian, YANG Ruijie, ZHOU Rui, ZHANG Tao, HUANG Guoxin. A method for spatial correlations between groundwater vulnerabilities and pollution sources based on machine learning and cluster analysis[J]. ENVIRONMENTAL ENGINEERING , 2025, 43(12): 112-120. doi: 10.13205/j.hjgc.202512013

PDF( 1645 KB)

A method for spatial correlations between groundwater vulnerabilities and pollution sources based on machine learning and cluster analysis

doi: 10.13205/j.hjgc.202512013

1. College of New Energy and Environment, Jilin University, Changchun 130021, China;
2. Chinese Academy of Environmental Planning, Beijing 100041, China;
3. Shaoguan Ecological Environmental Monitoring Station, Shaoguan 512000, China;
4. Solids Waste and Chemicals Management Center, Ministry of Ecology and Environment, Beijing 100029, China

Received Date: 2024-12-02
Accepted Date: 2025-01-12
Rev Recd Date: 2024-12-28

Available Online: 2026-01-09

Abstract

Abstract

The efficacy of groundwater pollution risk control is often limited by a lack of information on spatial correlations between groundwater vulnerabilities and pollution sources. To overcome this limitation， a method was established to reveal the spatial correlations between groundwater vulnerabilities and industrial pollution sources in an industrialized city of Guangdong Province， China， using a combination of genetic algorithm （GA）， back propagation neural network （BPNN）， kernel density estimation （KDE）， and bivariate local Moran’s I （BLMI）. The subjectivity in the indicator weighting of a DRASTICL model was successfully reduced by GA-BPNN， and then the groundwater vulnerability map was created by the GA-BPNN-DRASTICL model. A spatial distribution map of the industrial pollution source distribution was produced by KDE. The spatial clustering map between groundwater vulnerabilities and industrial pollution sources was generated by BLMI， explicitly showing their distribution characteristics and implying that specific measures should be taken to control risks of groundwater pollution in different parts of the study area. The results showed that the best accuracy of the GA-BPNN algorithm was obtained， with the training function of trainlm， the neuron number of 6 in the hidden layer， the learning rate of 0.1， the population size of 40， the crossover rate of 0.6， and the mutation rate of 0.01. The best indicator weights for depth to the water table， net recharge， aquifer medium， soil medium， topography， impact of vadose zone， hydraulic conductivity， and land use were 2.84， 5.27， 0.84， 2.20， 2.36， 6.58， 1.21， and 6.70， respectively. The high and very high vulnerability classes were concentrated in the central and southern parts of the study area. The largest hotspot of the industrial pollution sources was located in the midwest part， and the high-high areas were mainly distributed in the central， northeast， and southeast parts.
- groundwater vulnerabilities,
- industrial pollution sources,
- genetic algorithm,
- back propagation neural network,
- bivariate local Moran’s I

FullText(HTML)

References(38)

References

[1]	HE Z D，YE S，YANG H，et al. Evolution of groundwater quality in Jinding Industrial Park of Zhuhai city and its implications for water pollution control［J］. Environmental Monitoring in China，2024，40（2）：63-73. 何志东，叶珊，杨虹，等. 珠海金鼎工业园地下水水质演变及其对水污染防控的提示［J］. 中国环境监测，2024，40（2）：63-73.
[2]	MENG X S，WU M M，CHEN H H，et al. Analysis of typical pollutants properties and control measures of groundwater in representative iron& steel complexes［J］. Environmental Engineering，2019，37（12）：90-97. 孟祥帅，吴萌萌，陈鸿汉，等. 我国典型钢铁企业地下水污染特征及防治对策分析［J］. 环境工程，2019，37（12）：90-97.
[3]	LIU Z W，FAN S K，ZHANG M. Environmental risk analysis of soils and ground water in a typical Pb-Zn mine［J］. Nonferrous Metals（Extractive Metallurgy），2023（1）：88-91. 刘紫薇，范书凯，张萌. 典型铅锌矿山周边土壤和地下水环境风险分析［J］. 有色金属（冶炼部分），2023（1）：88-91.
[4]	CHEN Y D，LI Y Y，YE Z，et al. Groundwater vulnerability assessment in alluvial-proluvial fan area based on the improved DRASTIC model［J］. Environmental Science& Technology，2021，44（12）：194-202. 陈钰頔，李妍颖，叶忠，等. 基于改进 DRASTIC 模型的冲洪积扇地下水脆弱性评价［J］. 环境科学与技术，2021，44（12）：194-202.
[5]	CAI X H，KANG C X. Hydrochemical characteristics and water quality evaluation of groundwater in lower reaches of the Qiantang River［J］. Yangtze River，2023，54（6）：27-33. 蔡小虎，康丛轩. 钱塘江下游滨江地区地下水水化学特征与水质评价［J］. 人民长江，2023，54（6）：27-33.
[6]	ZHANG H，DU X Y，GAO F，et al. Groundwater pollution source identification by combination of PMF model and stable isotope technology［J］. Environmental Science，2022，43（8）：4054-4063. 张涵，杜昕宇，高菲，等. 联合PMF模型与稳定同位素的地下水污染溯源［J］. 环境科学，2022，43（8）：4054-4063.
[7]	HOU Y H，DOU X Y，BAO S L，et al. Assessment of groundwater contamination risk based on improved DRASTIC model［J］. Acta Scientiae Circumstantiae，2024，44（2）：227-236. 侯宇徽，窦筱艳，保善磊，等. 基于DRASTIC模型优化的地下水污染风险评价研究［J］. 环境科学学报，2024，44（2）：227-236.
[8]	XIAO R，CAO W，LIU Y，et al. The impacts of landscape patterns spatio-temporal changes on land surface temperature from a multi-scale perspective：a case study of the Yangtze River Delta［J］. Science of the Total Environment，2022，821：153381.
[9]	LIU K，CAI H S，ZHANG X L，et al. Spatial correlation and influencing factors of soil selenium and heavy metal content in Yuanzhou district［J］. Southwest China Journal of Agricultural Sciences，2023，36（7）：1492-1503. 刘珂，蔡海生，张学玲，等. 袁州区土壤硒与重金属含量空间关联及其影响因素分析［J］. 西南农业学报，2023，36（7）：1492-1503.
[10]	LYU S S，LI W，SU W C，et al. Ecological space zoning based on synergies and trade-offs of mountain，water，forest，farmland，lake and grassland［J］. Journal of Irrigation and Drainage，2024，43（9）：88-94. 吕思思，李威，苏维词，等. 基于山水林田湖草协同/权衡的生态管控分区研究［J］. 灌溉排水学报，2024，43（9）：88-94.
[11]	MENDIETA-MENDOZA A，HANSON R T，RENTERIA-VILLALOBOS M. Potential adverse impacts on vulnerability and availability of groundwater from climate-change and land use［J］. Journal of Hydrology，2021，594：125978.
[12]	VOUTCHKOVA D，SCHULLEHNER J，RASMUSSEN P，et al. A high-resolution nitrate vulnerability assessment of sandy aquifers（DRASTIC-N）［J］. Journal of Environmental Management，2021，277：111330.
[13]	ELZAIN H E，CHUNG S Y，SENAPATHI V，et al. Comparative study of machine learning models for evaluating groundwater vulnerability to nitrate contamination［J］. Ecotoxicology and Environmental Safety，2022，229：113061.
[14]	TORKASHVAND M，NESHAT A，JAVADI S，et al. New hybrid evolutionary algorithm for optimizing index-based groundwater vulnerability assessment method［J］. Journal of Hydrology，2021，598：126446.
[15]	IJLIL S，ESSAHLAOUI A，MOHAJANE M，et al. Machine learning algorithms for modeling and mapping of groundwater pollution risk：a study to reach water security and sustainable development（SDG）goals in a Mediterranean aquifer system［J］. Remote Sensing，2022，14（10）：2379.
[16]	DONG Y，ZHOU W，WANG X，et al. A new assessment method for the vulnerability of confined water：W-F&PNN method［J］. Journal of Hydrology，2020，590：125217.
[17]	ZHANG T，WANG X H，BI E P，et al. Groundwater vulnerability evaluation and risk control in a certain area in northern Guangdong province based on a BP neural network［J］. Environmental Engineering，2023，41（12）：270-277. 张涛，王夏晖，毕二平，等. 基于BP神经网络的粤北某地区地下水脆弱性评价及其风险管控［J］. 环境工程，2023，41（12）：270-277.
[18]	CHU S，CHENG L，CHENG J，et al. Shallow water bathymetry based on a back propagation neural network and ensemble learning using multispectral satellite imagery［J］. Acta Oceanologica Sinica，2023，42（5）：154-165.
[19]	YANG J，HU Y，ZHANG K，WU Y. An improved evolution algorithm using population competition genetic algorithm and self-correction BP neural network based on fitness landscape［J］. Soft Computing，2021，25：1751-1776.
[20]	HE Y，WANG H，WANG H，et al. Meteorology and topographic influences on nocturnal ozone increase during the summertime over Shaoguan，China［J］. Atmospheric Environment，2021，256：118459.
[21]	WANG X，ZENG X，LIU C，et al. Heavy metal contaminations in soil-rice system：source identification in relation to a sulfur-rich coal burning power plant in Northern Guangdong Province，China［J］. Environmental Monitoring and Assessment，2016，188（8）：460.
[22]	CHEN X C. Exploring the construction of an evaluation system for the effectiveness of ecological restoration in mines in Shaoguan City［J］. World Nonferrous Metals，2024（16）：130-132. 陈贤超. 探讨韶关市矿山生态修复成效评估体系构建［J］. 世界有色金属，2024（16）：130-132.
[23]	Shaoguan Municipal Bureau of Statistics，Survey Office of The National Bureau of Statistics in Guangzhou. Statistical Communiqué of Shaoguan on the 2023 National Economic and Social Development［EB/OL］.（2024-04-15）［ 2025-11-12］. https：//www.sg.gov.cn/zw/sjfb/tjgb/content/post_2622025.html. 韶关市统计局，国家统计局广州调查队. 2023年韶关市国民经济和社会发展统计公报［EB/OL］.（2024-04-15）［ 2025-11-12］. https：//www.sg.gov.cn/zw/sjfb/tjgb/content/post_2622025.html.
[24]	ALLER L. DRASTIC：a standardized system for evaluating ground water pollution potential using hydrogeologic settings［Z］. Washington D C：United States Environmental Protection Agency，1985：99-102.
[25]	YANG J，TANG Z，JIAO T，et al. Combining AHP and genetic algorithms approaches to modify DRASTIC model to assess groundwater vulnerability：a case study from Jianghan plain，China［J］. Environmental Earth Sciences，2017，76（12）：426.
[26]	WANG J N，WANG Y Y，HE S L，et al. Optimized BP neural network model based on improved genetic algorithm for soil moisture prediction［J］. Computer Systems& Applications，2022，31（2）：273-278. 王佳楠，王玉莹，何淑林，等. 基于改进遗传算法优化BP神经网络的土壤湿度预测模型［J］. 计算机系统应用，2022，31（2）：273-278.
[27]	SHI Y，REN C，YAN Z，et al. High spatial-temporal resolution estimation of ground-based global navigation satellite system interferometric reflectometry（GNSS-IR）soil moisture using the genetic algorithm back propagation（GA-BP）neural network［J］. ISPRS International Journal of Geo-Information，2021，10（9）：623.
[28]	DENG Y，ZHOU X，SHEN J，et al. New methods based on back propagation（BP）and radial basis function（RBF）artificial neural networks（ANNs）for predicting the occurrence of haloketones in tap water［J］. Science of the Total Environment，2021，772：145534.
[29]	PYO J C，HONG S M，KWON Y S，et al. Estimation of heavy metals using deep neural network with visible and infrared spectroscopy of soil［J］. Science of the Total Environment，2020，741：140162.
[30]	YANG C，SUN J Y. Evaluation and spatial evolution characteristics of high-quality economic development of Dongting Lake eco-economic zone［J］. Journal of Central South University of Forestry& Technology（Social Sciences），2024，18（3）：36-47. 杨灿，孙靖怡. 洞庭湖生态经济区经济高质量发展评估与空间演变特征［J］. 中南林业科技大学学报（社会科学版），2024，18（3）：36-47.
[31]	HUANG G，WANG X，CHEN D，et al. A hybrid data-driven framework for diagnosing contributing factors for soil heavy metal contaminations using machine learning and spatial clustering analysis［J］. Journal of Hazardous Materials，2022，437：129324.
[32]	WU S，ZHOU S，BAO H，et al. Improving risk management by using the spatial interaction relationship of heavy metals and PAHs in urban soil［J］. Journal of Hazardous Materials，2019，364：108-116.
[33]	LI Y，ZHANG N，ZHANG Y N，et al. A novel salp swarm algorithm-neural network model for predicting weak singular integrals of boundary elements［J］. Journal of Shandong University（Engineering Science），2023，53（6）：8-15. 李源，张妮，张艳娜，等. 用于预测边界元弱奇异积分的新型樽海鞘-神经网络模型［J］. 山东大学学报（工学版），2023，53（6）：8-15.
[34]	WANG Z，ZHANG J，WANG J，et al. A back propagation neural network based optimizing model of space-based large mirror structure［J］. Optik，2019，179：780-786.
[35]	QIAN W，WANG Y F，WANG C，et al. Joint estimation of SoH-SoC for lithium battery based on BP neural network and H infinity filter［J］. Chinese Journal of Scientific Instrument，2024，45（6）：307-319. 钱伟，王亚丰，王晨，等. 基于 BP 神经网络与 H∞滤波的锂电池 SoH-SoC 联合估计研究［J］. 仪器仪表学报，2024，45（6）：307-319.
[36]	WEI S A，CHEN Y P，LI J，et al. Path planning of complex surface inspection based on improved adaptive genetic algorithm［J］. Tool Engineering，2024，58（11）：153-160. 韦思安，陈岳坪，李杰，等. 基于改进自适应遗传算法的复杂曲面检测路径规划［J］. 工具技术，2024，58（11）：153-160.
[37]	SHU H，XIONG P. Reallocation planning of urban industrial land for structure optimization and emission reduction：a practical analysis of urban agglomeration in China's Yangtze River Delta［J］. Land Use Policy，2019，81：604-623.
[38]	AMARE A. Corporate environmental responsibility in Ethiopia：a case study of the Akaki River Basin［J］. Ecosystem Health and Sustainability，2019，5（1）：57-66.