东篱科研大数据发现系统（DRDS）

位置：成果数据库 > 期刊 > 期刊详情页

太湖冬季有色可溶性有机物吸收荧光特性及遥感算法

期刊名称：湖泊科学
时间：0
页码：348-356
分类：P733.3[天文地球—物理海洋学;天文地球—海洋科学]
作者机构：[1]南京农业大学资源与环境科学学院,南京210095, [2]中国科学院南京地理与湖泊研究所湖泊与环境国家重点实验室,太湖湖泊生态系统研究站,南京210008, [3]河海大学环境科学与工程学院,南京210098
相关基金：中国科学院知识创新工程项目（KZCX1-YW-14 KZCX2-YW-QN312）; 国家自然科学基金项目（40971252 40825004 40730529）联合资助致谢：野外采样得到冯胜、王鑫、赵巧华、李俊生、张浩等同志的鼎力帮助,在此一并表示谢意.
相关项目：基于光谱吸收、三维荧光、碳同位素研究太湖CDOM组成特征及来源

关键词：有色可溶性有机物, 荧光, 遥感反射率, BP神经网络模型, 太湖, Chromophoric dissolved organic matter, fluorescence, remote sensing reflectance, BP neural network model, Lake Taihu

中文摘要：

基于2006年和2007年1月两次太湖采样,对50个点位的有色可溶性有机物（CDOM）光谱吸收、荧光、溶解性有机碳（DOC）浓度及遥感反射率进行测定与分析,探讨冬季太湖CDOM的吸收荧光特性及空间分布,建立CDOM吸收系数的遥感反演算法.结果表明,太湖冬季CDOM在355nm处吸收系数a（355）变化范围和均值分别为1.83-7.34m-1、3.37±1.01m-1,相应的荧光及DOC浓度变化范围、均值分别为9.79-29.18N.FL.U、13.4±3.37N.FL.U;4.61-10.45mg/L、6.37±1.24mg/L.CDOM吸收系数、CDOM荧光值、DOC浓度三者呈显著正相关.空间分布上,两次调查均显示CDOM吸收系数、CDOM荧光值、DOC浓度呈现出明显的南低北高,最大值都出现在太湖北部的藻型湖区梅梁湾内,最小值则在东太湖和贡湖湾2个草型湖区.通过单波段、一阶微分和BP神经网络模型3种不同CDOM反演方法精度的分析、比较发现,BP神经网络模型反演结果最好,模型验证的相对均方根误差和平均相对误差分别为14.9%、11.7%,可以用于冬季太湖CDOM吸收系数a（355）的遥感估算.

英文摘要：

Based on two investigations with 100 sampling sites in Lake Taihu in January,2006 and 2007,the characteristics of spectral absorption and fluorescence,spatial distribution,and the retrieval model of chromophoric dissolved organic matter（CDOM） were studied.The ranges and mean values of CDOM absorption coefficient at 355nm a（355）,fluorescence normalized Fn（355） and dissolved organic carbon（DOC） concentration were 1.83-7.34,3.37±1.01 m-1;9.79-29.18,13.4±3.37 N.FL.U;and 4.61-10.45,6.37±1.24 mg/L,respectively.Significant positive correlations between a（355） and DOC,a（355） and Fn（355） were found.Spatially,two surveys have shown that the higher values of a（355）,Fn（355）,DOC concentration were found in Meiliang Bay and lower values were found in East Lake Taihu and Gonghu Bay.Overall,a（355）,Fn（355）,and DOC concentration were significantly higher in two transects in northern lake regions than those in other transects in southern lake regions.The results showed that BP neural network model was superior to a single band model and the first order differential model for CDOM absorption estimation.The relative root mean square error（RRMSE） and mean relative error（MRE） of BP neural network model were 14.9% and 11.7%,respectively,based on an independent validation dataset including 25 samples.Thus,BP neural network model could be better used to estimate CDOM absorption in Lake Taihu.

同期刊论文项目