机器学习-多元线性回归预测铜的水生态基准

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (1010 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要以铜和我国代表性水生生物作为研究对象,在原有生物配位体模型(BLM)理论框架基础上,通过梯度提升决策树算法筛选和评估环境要素(硬度、pH值和溶解有机碳)的特征重要性,建立多变量耦合的多物种水生毒性预测模型;在此基础上对毒性预测值进行物种敏感度分布(SSD)分析,构建适合我国水环境特征的铜淡水生态基准预测模型.研究发现,基于3门5科水生毒性数据建立的三参数耦合预测模型的预测准确率(RFx,2.0)比BLM模型提升了42%;以Sigmoidal-logistic1模型拟合SSD曲线的效果最佳(0.922<R²<0.991,0.0267<RMSE<0.0767,P>0.05),进而计算出我国流域水环境中铜的短期水生态基准推荐阈值为0.07350~15.38µg/L;基于机器学习的特征重要性分析定量识别了DOC在金属水生态基准制定中的关键作用,也为众多环境影响因素的集约化处理提供直接证据.与以BLM模型为代表的国外技术相比,研究工作在开发适用于中国水环境特征和管理需求的金属“原位”水质基准制定技术方面做了有益尝试,不仅节约了环境监测和管理成本,也体现了水环境管理的区域化和精准化.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	杨晓玲
	王梦晓
	李晓娟
	袁雅文
	邵美晨
	穆云松
	白英臣
	吴丰昌

关键词 ：铜, DOC, 水质基准, 模型校正, 物种敏感度分布, 机器学习

Abstract：In this research, copper and representative aquatic organisms in China were investigated as a case study. Based on the theoretical framework of the biotic ligand model (BLM), the key environmental factors (hardness, pH and dissolved organic carbon) were screened by a gradient boosting decision tree algorithm, and multivariate coupled predictive models were established for predicting acute toxicities of different aquatic organisms. And then, the species sensitivity distribution (SSD) analysis was performed to predict the WQC of copper for protecting aquatic organisms, which was suitable for the characteristics of water environment in China. It was found that the prediction accuracy (RFx, 2.0) of a three-variable model based on aquatic toxicity data of 3phylum and 5families was 42% higher than that of the BLM. The SSD curves for the nine organisms were best fitted using a sigmoidal-logistic model (0.922<R²<0.991, 0.0267<RMSE<0.0767, P>0.05), and the threshold of short-term water ecological criteria of copper is recommended as 0.07350~15.38µg/L in the river basin of China. Based on the feature importance analysis from machine learning, the key role of DOC in the formulation of WQC for metals was quantitatively identified, and it also provided direct evidence for intensively treating multiple environmental factors. Compared with existing technologies including the BLM, our finding makes a beneficial attempt to develop an "in situ" WQC predictive model to meet water environment characteristics and management needs in China. It will reduce the costs for environmental monitoring and management, and enhance the regionalization and precision of water environment management.

Key words： copper DOC water quality criteria model correction species sensitivity distribution machine learning

收稿日期: 2023-12-27

PACS:

X52

基金资助:国家自然科学基金资助项目(42277235,41521003)

通讯作者: 穆云松,副教授,muyunsong@ruc.edu.cn E-mail: muyunsong@ruc.edu.cn

作者简介: 杨晓玲(1996-),女,云南昭通人,中国人民大学硕士研究生,主要研究方向为金属水质基准研究.发表论文1篇.yangxiaoling@ruc.edu.cn.

引用本文:

杨晓玲, 王梦晓, 李晓娟, 袁雅文, 邵美晨, 穆云松, 白英臣, 吴丰昌. 机器学习-多元线性回归预测铜的水生态基准[J]. 中国环境科学, 2024, 44(7): 3976-3985. YANG Xiao-ling, WANG Meng-xiao, LI Xiao-juan, YUAN Ya-wen, SHAO Mei-chen, MU Yun-song, BAI Ying-chen, WU Feng-chang. Predicting the water ecological criteria of copper using machine learning and multiple linear regression approaches. CHINA ENVIRONMENTAL SCIENCECE, 2024, 44(7): 3976-3985.

链接本文:

http://www.zghjkx.com.cn/CN/ 或 http://www.zghjkx.com.cn/CN/Y2024/V44/I7/3976

[1] Li B, Li H, Ren S, et al. Commodity supply risk assessment of China's copper industrial chain: The perspective of trade network [J]. Resources Policy, 2023,81:103297.
[2] Zhu Y J, Zhu X Y, Xu Q J, et al. Water quality criteria and ecological risk assessment for copper in Liaodong Bay, China [J]. Marine Pollution Bulletin, 2022,185:114164.
[3] 生态环境部.2022中国生态环境状况公报[R]. Ministry of Ecology and Environment of the People's Republic of China. Report on the state of the ecology and environment in China 2022[R]. Beijing: Ministry of Ecology and Environment of the People's Republic of China, 2023.
[4] 何佳,时迪,王贝贝,等.10种典型重金属在八大流域的生态风险及水质标准评价[J]. 中国环境科学, 2019,39(7):2970-2982. He J, Shi D, Wang B B, et al. Ecological risk assessment and water quality standard evaluation of 10 typical metals in eight basins in China [J]. China Environmental Science, 2019,39(7):2970-2982.
[5] Donnachie R L, Johnson A C, Moeckel C, et al. Using risk-ranking of metals to identify which poses the greatest threat to freshwater organisms in the UK [J]. Environmental Pollution, 2014,194:17-23.
[6] Fu Z Y, Wu F C, Chen L L, et al. Copper and zinc, but not other priority toxic metals, pose risks to native aquatic species in a large urban lake in Eastern China [J]. Environmental Pollution, 2016,219: 1069-1076.
[7] 宋颖.南四湖典型入湖河流表层沉积物中重金属的分布及生态风险[D]. 济南:山东大学, 2014. Song Y. Distribution and ecological risk assessment of heavy metals in surface sediments of typical inflow rivers in Nansi Lake [D]. Jinan: Shandong University, 2014.
[8] 牛红义,吴群河,陈新庚,等.珠江广州河段沉积物重金属潜在生物毒性风险[J]. 环境科学与技术, 2010,33(8):185-190. Niu H Y, Wu Q H, Chen X G, et al. Potential biological toxicity risk of heavy metals in the column sediments from Guangzhou Section of the Pearl River [J]. Environmental Science and Technology, 2010,33(8):185-190.
[9] 符志友,冯承莲,赵晓丽,等.我国流域水环境中铜、锌的生态风险及管理对策[J]. 环境工程, 2019,37(11):70-74. Fu Z Y, Feng C L, Zhao X L, et al. Ecological risks and management countermeasures of copper and zinc in water environment of Chian [J]. Environmental Engineering, 2019,37(11):70-74.
[10] 刘娜,李亚兵,刘红玲.多因子影响下铜的水质基准及生态风险[J]. 中国环境科学, 2022,42(7):3353-3361. Liu N, Li Y B, Liu H L. The water quality criteria and ecological risks of copper under the influence of multiple factors [J]. China Environmental Science, 2022,42(7):3353-3361.
[11] Li Y, Mu D, Wu H Q, et al. Derivation of copper water quality criteria in Bohai Bay for the protection of local aquatic life and the ecological risk assessment [J]. Marine Pollution Bulletin, 2023,190:114863.
[12] 陈曲,郭继香,孙乾耀,等.甲萘威的淡水水生生物水质基准研究[J]. 环境科学研究, 2016,29(1):8. Chen Q, Guo J X, Sun Q Y, et al. Aquatic life ambient freshwater quality criteria for Carbaryl in China [J]. Research of Environmental Sciences, 2016,29(1):1-8.
[13] Mehta R, Templeton D M, O’Brien P J. Mitochondrial involvement in genetically determined transition metal toxicity: II. Copper toxicity [J]. Chemico-Biological Interactions, 2006,163(1):77-85.
[14] Enochs B, Meindl G, Shidemantle G, et al. Short and long-term phytoremediation capacity of aquatic plants in Cu-polluted environments [J]. Heliyon, 2023,9(1):e12805.
[15] Gaetke L M, Chow-Johnson H S, Chow C K. Copper: Toxicological relevance and mechanisms [J]. Archives of Toxicology, 2014,88(11): 1929-1938.
[16] Gonzalez Alcocer A, Gopar Cuevas Y, Soto Dominguez A, et al. Combined chronic copper exposure and aging lead to neurotoxicity in vivo [J]. NeuroToxicology, 2023,95:181-192.
[17] Chupani L, Sjöberg V, Jass J, et al. Water hardness alters the gene expression response and copper toxicity in Daphnia magna [J]. Fishes, 2022,7(5):248.
[18] Lake Thompson I, Hofmann R. Effectiveness of a copper based molluscicide for controlling Dreissena adults [J]. Environmental Science: Water Research & Technology, 2019,5(4):693-703.
[19] Bui T K L, Do Hong L C, Dao T S, et al. Copper toxicity and the influence of water quality of Dongnai River and Mekong River waters on copper bioavailability and toxicity to three tropical species [J]. Chemosphere, 2016,144:872-878.
[20] Crémazy A, Braz-Mota S, Brix K V, et al. Investigating the mechanisms of dissolved organic matter protection against copper toxicity in fish of Amazon's black waters [J]. Science of The Total Environment, 2022,843:157032.
[21] Smith D S, Cooper C A, Wood C M. Measuring biotic ligand model (BLM) parameters in vitro: Copper and silver binding to Rainbow Trout gill cells as cultured epithelia or in suspension [J]. Environmental Science & Technology, 2017,51(3):1733-1741.
[22] Di Toro D M, Allen H E, Bergman H L, et al. Biotic ligand model of the acute toxicity of metals. 1. Technical basis [J]. Environ Toxicol Chem, 2001,20(10):2383-2396.
[23] ECCC. Federal environmental quality guidelines-Copper [R]. Canada: Government of Canada, 2021.
[24] USEPA. Aquatic life freshwater quality criteria-copper [R]. Washington, DC: U.S. Environmental Protection Agency Office of Water Office of Science and Technology, 2007.
[25] 王学东,马义兵,华珞,等.环境中金属生物有效性的预测模型——生物配体模型研究进展[J]. 生态毒理学报, 2006,(3):193-202. Wang X D, Ma Y B, Hua L, et al. Advances in biotic ligand model to predict the bioavailability of metals in environments [J]. Asian Journal of Ecotoxicology, 2006,(3):193-202.
[26] Brix K V, DeForest D K, Tear L, et al. Use of multiple linear regression models for setting water quality criteria for copper: A complementary approach to the biotic ligand model [J]. Environmental Science and Technology, 2017,51(9):5182-5192.
[27] Brix K V, Tear L, Santore R C, et al. Comparative performance of multiple linear regression and biotic ligand models for estimating the bioavailability of copper in freshwater [J]. Environ. Toxicol. Chem., 2021,40(6):1649-1661.
[28] 徐潇.铜对斜生栅藻急性毒性预测模型—生物配体模型的建立与验证[D]. 杭州:浙江工业大学, 2019. Xu X. Biotic ligand model development predicting acute Cu toxicity to the algae Scenedesmus obliquus: construction and validation [D]. Hangzhou: Zhejiang University of Technology, 2019.
[29] 李扬,牛永华,李会仙,等.基于生物配位体模型的汾河铜水质基准研究[J]. 环境工程技术学报, 2022,12(5):1711-1718. Li Y, Niu Y H, Li H X, et al. Study on water quality criteria of copper in the Fen River based on biotic ligand model [J]. Journal of Environmental Engineering Technology, 2022,12(5):1711-1718.
[30] 王春艳.生物配体模型预测中国典型河流水体铜毒性及其水质基准指标应用研究[D]. 武汉:武汉大学, 2012. Wang C Y. Application research of Cu toxicity and WOC predicted by BLM in typical Chinese rivers [D]. Wuhan: Wuhan University, 2012.
[31] 王春艳,陈浩,安立会,等.BLM预测水中重金属生物有效性研究进展[J]. 环境科学与技术, 2011,34(8):75-80. Wang C Y, Chen H, An L H, et al. An updated review on biotic ligand model in predicting metal bioavailability in surface waters [J]. Environmental Scienceand Technology, 2011,34(8):75-80.
[32] 陈莎.澜沧江铜的水质基准与生态风险评价研究[D]. 昆明:昆明理工大学, 2014. Chen S. Water quality criteria and ecological risk assessment of copper in the Langcang River [D]. Kunming: Kunming University of Science and Technology, 2014.
[33] Zhang Y, Zang W, Qin L, et al. Water quality criteria for copper based on the BLM approach in the freshwater in China [J]. PLoS ONE, 2017,12(2):e0170105.
[34] Wu F C, Meng W, Zhao X L, et al. China embarking on development of its own national water quality criteria system [J]. Environmental Science and Technology, 2010,44(21):7992-7993.
[35] Feng C, Wu F, Zheng B, et al. Biotic ligand models for metals - A practical application in the revision of water quality standards in China [J]. Environmental Science and Technology, 2012,46(20):10877- 10878.
[36] Deng P, Hu X, Mu L. Machine learning provides opportunities to recognize greenhouse gas emissions from water at a large scale [J]. ACS Esand T Water, 2024,4(3):837-843.
[37] Krishnan R, Howard I S, Comber S, et al. In silico prediction of acute chemical toxicity of biocides in marine crustaceans using machine learning [J]. Science of The Total Environment, 2023,887:164072.
[38] Zhou Y, Wang Y, Peijnenburg W, et al. Using machine learning to predict adverse effects of metallic nanomaterials to various aquatic organisms [J]. Environmental Science and Technology, 2023,57(46): 17786-17795.
[39] Zhu J-J, Yang M, Ren Z J. Machine learning in environmental research: common pitfalls and best practices [J]. Environmental Science and Technology, 2023,57(46):17671-17689.
[40] Zhong S, Zhang K, Bagheri M, et al. Machine learning: New ideas and tools in environmental science and engineering [J]. Environmental Science and Technology, 2021,55(19):12741-12754.
[41] Fan J, Huang G, Chi M, et al. Prediction of chemical reproductive toxicity to aquatic species using a machine learning model: An application in an ecological risk assessment of the Yangtze River, China [J]. Science of The Total Environment, 2021,796:148901.
[42] 吴丰昌,冯承莲,曹宇静,等.我国铜的淡水生物水质基准研究[J]. 生态毒理学报, 2011,6(6):12. Wu F C, Feng C L, Cao Y J, et al. Aquatic ife ambient freshwater quality criteria for Copper in China [J]. Asian Journal of Ecotoxicology, 2011,6(6):12.
[43] 林颖,高俊敏,郭劲松,等.基于物种敏感度分布的典型抗生素的长期水质基准推导及其在生态风险评估中的应用[J]. 环境科学学报, 2023,43(3):503-515. Lin Y, Gao J M, Guo J S, et al. Long-term water quality criteria derivation of typical antibiotics based on species sensitivity distribution and its application to ecological risk assessment [J]. Acta Scientiae Circumstantiae, 2023,43(3):503-515.
[44] 乔宇,闫振飞,冯承莲,等.几种典型模型在物种敏感度分布中的应用和差异分析[J]. 环境工程, 2021,39(10):85-92,109. Qiao Y, Yan Z F, Feng C L, et al. Applications and differences analysis of several typical models in species sensitivity distribution [J]. Environmental Engineering, 2021,39(10):85-92,109.
[45] 邵美晨,杨晓玲,王梦晓,等.我国流域铜的长期水质基准预测模型研究——MLR vs. BLM [J]. 环境科学研究, 2023,36(6):1236-1244. Shao M C, Yang X L, Wang M X, et al. Predictive model for setting long-term water quality criteria of copper in Chinese river basins: MLR vs. BLM [J]. Research of Environmental Sciences, 2023,36(6): 1236-12344.
[46] 王春艳,陈浩,郑丙辉,等.应用生物配体模型研究湘江水体中铜的生物有效性[J]. 生态毒理学报, 2013,8(6):998-1004. Wang C Y, Chen H, Zheng B H, et al. Study on bioavailability of Cu to Medaka in the Xiangjiang river with biotic ligand model [J]. Asian Journal of Ecotoxicology, 2013,8(6):998-1004.
[47] Niyogi S, Wood C M. Biotic ligand model, a flexible tool for developing site-specific water quality guidelines for metals [J]. Environ Sci Technol, 2004,38(23):6177-6192.
[48] Di Toro D M, Allen H E, Bergman H L, et al. Biotic ligand model of the acute toxicity of metals. 1. Technical basis [J]. Environ Toxicol Chem, 2001,20(10):2383-2396.
[49] Meyer J S, Traudt E M, Ranville J F. Is the factor-of-2rule broadly applicable for evaluating the prediction accuracy of metal-toxicity models? [J]. Bull Environ Contam Toxicol, 2018,100(1):64-68.
[50] Garman E R, Meyer J S, Bergeron C M, et al. Validation of bioavailability-based toxicity models for metals [J]. Environmental Toxicology Chemistry, 2020,39(1):101-117.
[51] 孔祥臻,何伟,秦宁,等.重金属对淡水生物生态风险的物种敏感性分布评估[J]. 中国环境科学, 2011,31(9):1555-1562. Kong X Z, He W, Qin N, et al. Assessing acute ecological risks of heavy metals to freshwater organisms by species sensitivity distributions [J]. China Environmental Science, 2011,31(9):1555- 1562.
[52] 赵芊渊,侯俊,王超,等.应用概率物种敏感度分布法研究太湖铜水生生物水质基准[J]. 生态毒理学报, 2015,10(1):191-203. Zhao Q Y, Hou J, Wang C, et al. Deriving aquatic water quality criteria for heavy metals in Taihu Lake by probabilistic species sensitivity distribution [J]. Asian Journal of Ecotoxicology, 2015,10(6):121-128.
[53] Zhao Y, Xu M, Liu Q, et al. Study of heavy metal pollution, ecological risk and source apportionment in the surface water and sediments of the Jiangsu coastal region, China: A case study of the Sheyang Estuary [J]. Marine Pollution Bulletin, 2018,137:601-609.
[54] Santore R C, Mathew R, Paquin P R, et al. Application of the biotic ligand model to predicting zinc toxicity to rainbow trout, fathead minnow, and Daphnia magna [J]. Comparative Biochemistry Physiology Part C: Toxicology Pharmacology, 2002,133(1):271-285.
[55] 胡释尹,李非里,方小满.溶解性有机质对自然水体中重金属生物有效性评价的影响[J]. 环境科学与技术, 2016,39(1):6. Hu S Y, Li F L, Fang X M. Effect of dissolved organic matter in evaluating heavy metals bioavailability in natural water [J]. Environmental Science and Technology, 2016,39(1):6.
[56] Ryan A C, Tomasso J R, Klaine S J. Influence of pH, hardness, dissolved organic carbon concentration, and dissolved organic matter source on the acute toxicity of copper to Daphnia magna in soft waters: implications for the biotic ligand model [J]. Environmental Toxicology Chemistry, 2010,28(8):1663-1670.