投稿中心

审稿中心

编辑中心

期刊出版

网站地图

友情链接

引用本文:潘佳敏,冯禹昊,谢平,方精云.基于小波-神经网络耦合模型对云南星云湖富营养化气象驱动因子的分析.湖泊科学,2021,33(2):428-438. DOI:
Pan Jiamin,Feng Yuhao,Xie Ping,Fang Jingyun.Meteorological driving factor of eutrophication in Lake Xingyun, Yunnan based on the Coupling Model of Wavelet analysis and Neural Network. J. Lake Sci.2021,33(2):428-438. DOI:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 162次   下载 83 本文二维码信息
码上扫一扫!
分享到: 微信 更多
基于小波-神经网络耦合模型对云南星云湖富营养化气象驱动因子的分析
潘佳敏1, 冯禹昊1, 谢平2, 方精云1
1.北京大学城市与环境学院, 北京 100871;2.云南大学生态与环境学院高原湖泊生态与治理研究院, 昆明 650091
摘要:
气象因子是影响湖泊富营养化的重要因素,而湖泊富营养化对人群健康、生态系统和社会经济等均有负面影响.本文基于统计资料及遥感数据,结合Morlet小波分析和BP多层前馈神经网络(BP神经网络)构建了不同时间尺度下的小波-神经网络耦合模型,分析了1986-2011年云南星云湖水华强度变化与月降雨量、月平均气温、月平均风速、月日照时数变化之间的关系,探究了影响湖泊富营养化的主导气象因子.结果表明:气象因子的波动周期是影响湖泊年内水华强度变化的重要因素;小波-神经网络耦合模型能有效提高数据拟合的精度,最优小波-神经网络耦合模型的拟合优度为0.605,高于BP神经网络的拟合优度0.292;小波-神经网络耦合模型能更有效地对星云湖富营养化程度进行分析和描述,其均方误差和相关系数均优于BP神经网络;根据最优小波-神经网络耦合模型下的各气象因子的平均影响值,可知月平均气温是影响星云湖富营养化的主导气象因子,其次是月降水率、月平均风速,最后是月日照时数.综上,小波-神经网络耦合模型相比BP神经网络对样本数据具有更好的适应性,拟合精度更高,能为星云湖的保护与富营养化的治理提供参考依据.
关键词:  富营养化  神经网络  小波分析  耦合模型  气象因子  星云湖
DOI:
分类号:
基金项目:国家自然科学基金项目(31988102)资助.
Meteorological driving factor of eutrophication in Lake Xingyun, Yunnan based on the Coupling Model of Wavelet analysis and Neural Network
Pan Jiamin1, Feng Yuhao1, Xie Ping2, Fang Jingyun1
1.College of Urban and Environmental Sciences, Peking University, Beijing 100871, P. R. China;2.Institute for Ecological Research and Pollution Control of Plateau Lakes, School of Ecology and Environmental Science, Yunnan University, Kunming 650091, P. R. China
Abstract:
There is a link between changes of climate and the degree of eutrophication of lakes. Eutrophication of lakes has a negative impact on human health, ecosystems, and socioeconomics. Based on statistical data and remote sensing data, coupling models of Morlet wavelet analysis and BP neural network in different time scales were used to analyze the eutrophication trends of Lake Xingyun in Yunnan from 1986 to 2011 and their multi-scale relationships with the climatic variables including monthly precipitation, monthly average temperature, monthly average wind speed, and monthly sunshine duration. The results show that the fluctuation period of climatic factors are important for affecting the monthly variation in the bloom intensity. The coupling models of Morlet wavelet analysis and BP neural network can effectively improve the accuracy of data fitting. Goodness of fit of the optimal coupling model is 0.605 which is higher than the BP neural network's goodness of fit (0.292). The optimal coupling model can analysis and describe eutrophication better than the BP neural network. The mean square error and the correlation coefficients of the optimal coupling model were better than those by the BP neural network. The monthly average temperature was the dominant climatic factor affecting the eutrophication of Lake Xingyun, followed by the monthly precipitation, monthly average wind speed and monthly sunshine duration in the optimal coupling model. In conclusion, we show that the coupling models of Morlet wavelet analysis and BP neural network, which has better adaptability to periodically changing sample data and higher prediction accuracy than the BP neural network, can provide reliable reference for the protection and eutrophication control of Lake Xingyun.
Key words:  Eutrophication  neural network  wavelet analysis  coupling of models  meteorological factor  Lake Xingyun
分享按钮