摘要
研究湖泊水体DOM(溶解性有机质)特征及其与复合环境要素的响应关系,对水生态系统保护及生物地球化学循环研究具有重要意义.以内蒙古达里诺尔湖(简称“达里湖”)为研究区,利用紫外-可见光谱技术(UV-vis)和人工神经网络模型(ANN)方法,开展达里湖水体DOM特征识别及其水质响应关系的研究.结果表明:①达里湖水环境污染程度较为严重,水体呈富营养化趋势.DOM吸收系数〔α(355)〕表明,夏季湖体DOM浓度较高.②建立了α(355)与达里湖水体ρ(TN)、ρ(TP)、pH和ρ(Chla)的人工神经网络模型,该模型验证期和测试期决定系数(R 2)分别为0.78和0.84,该模型具有较好的DOM模拟和预测能力.③人工神经网络模型参数敏感性分析结果表明,α(355)对ρ(Chla)变化最敏感,敏感性水平为31.95%;其次为pH和ρ(TN),α(355)对二者变化的敏感性水平分别为28.53%和25.54%;α(355)对ρ(TP)变化敏感性最低,敏感性水平为8.16%.研究显示,达里湖DOM对富营养化的发生具有较显著的表征影响,人工神经网络模型可为富营养化预测提供科学参考.
Studying the relationship between the dissolved organic matter(DOM)characteristics of lakes and the complex environmental elements for DOM is of great significance for aquatic ecosystem protection and biogeochemical cycles.This article took Dali-Nor Lake(referred to as‘Dali Lake’),an important migratory bird distribution center in northern China,as a research area,and used modern acquisition methods such as artificial intelligence unmanned ships,and ultraviolet-visible spectroscopy(UV-vis)and artificial neural network model(ANN)methods to study the response relationship between DOM feature recognition and water quality in the lake water.The results showed that:(1)The water environment of Dali Lake was seriously polluted,and the water body was eutrophication.The DOM absorption coefficient showed that the DOM concentration in the lake was higher in summer.(2)The nonlinear response relationship betweenα(355)andρ(TN),ρ(TP),pH andρ(Chla)was established.The coefficient of determination(R 2)during the validation period and the testing period was 0.78 and 0.84,respectively,and the model had good DOM simulation and prediction capabilities.(3)The parameter sensitivity analysis results showed thatα(355)was the most sensitive to changes inρ(Chla),with a sensitivity level of 31.95%;the sensitity levels ofα(355)to pH was 28.53%;and the sensitity levels ofα(355)toρ(TN)was 25.54%;Forρ(TP),the change sensitivity was the lowest,with a sensitivity level of 8.16%.This study has shown that Dali Lake DOM has a significant effect on the occurrence of eutrophication,which can provide a scientific reference for the prediction of eutrophication.The artificial neural network model can provide a scientific reference for eutrophication prediction.
作者
孙伟
夏瑞
王晓
王璐
陈焰
马淑芹
张远
胡泓
SUN Wei;XIA Rui;WANG Xiao;WANG Lu;CHEN Yan;MA Shuqin;ZHANG Yuan;HU Hong(State Key Laboratory of Environmental Criteria and Risk Assessment,Chinese Research Academy of Environmental Sciences,Beijing 100012,China;College of Environmental Science and Engineering,Ocean University of China,Qingdao 266100,China)
出处
《环境科学研究》
EI
CAS
CSCD
北大核心
2020年第11期2458-2466,共9页
Research of Environmental Sciences
基金
国家水体污染控制与治理科技重大专项(No.2017ZX07301-001)
长江生态环境保护修复联合研究项目(第一期)(No.2019-LHYJ-01-0103,2019-LHYJ-01-0102)。
