摘要
以研制使用单色发光二极管为光源的浮游植物荧光自动监测仪为目标,针对12种我国近海常见浮游植物种,选取12个激发波长点(400、430、450、460、470、480、490、510、525、550、570、590nm)的浮游植物活体叶绿素荧光激发光谱作为特征谱,建立多元线性回归模型,采用非负最小二乘法加以解析,实现了各门类浮游植物的识别测定,特别是硅藻和甲藻的识别测定。所测样品中,单门类浮游植物样品共有79个,识别正确率为96%,回收率≥68%,其中85%的浮游植物样品回收率≥80%;多门类浮游植物的混合样品共有17个,识别正确率为76%,回收率≥74%。
In order to develop an in situ algae fluorescence auto-analyzer, the discriminating technology for phytoplankton populations was established by in vivo chlorophyll fluorescence excitation spectra composed of 12 excitation wavelengths （400, 430, 450, 460, 470, 480, 490, 510, 525, 550, 570, and 590 nm） of the algae occurring frequently in the coastal waters of China and a multivariate linear regression model. The linear regression model was solved by non-negative least squares. Some samples were tested. For 79 single division algal samples, 96% samples were accurately discriminated, with recovery efficiency above 68%, and the recovery efficiency of 85% samples was above 80%. For 17 mixed division algal samples, 76% samples were accurately determinated, with recovery efficiency above 74%.
出处
《热带海洋学报》
CAS
CSCD
北大核心
2008年第5期24-29,共6页
Journal of Tropical Oceanography
基金
国家自然科学基金项目(40706036)
863国际合作项目(2006AA09Z178)
关键词
光谱学
浮游植物
活体叶绿素荧光
识别
非负最小二乘
spectroscopy
phytoplankton
in vivo chlorophyll fluorescence
discrimination
non-negativeleast square
