海洋浮游植物通过光合固碳(以初级生产力（PP）为指标)启动生物泵，而不同类群固碳速率及碳输出皆不同，故实现区分类群的初级生产力遥感，关乎碳收支与碳循环研究，意义重大。传统算法基于低反演准确度的叶绿素浓度，忽略浮游植物类群差异，导致遥感估算PP准确度低，难以反映实际的时空变化。浮游植物吸收系数遥感反演准确度高，且吸收与量子产率结合的固碳模型符合光合固碳机理，并可反映类群差异，故本项目提出吸收中心与分类群量子产率结合的固碳遥感新思路。首先基于浮游植物吸收光谱，研发可判别优势类群的半分析算法，并研究典型类群的量子产率特征，从而发展以吸收和分类群的量子产率为输入的PP模型。拟通过实验室培养与现场调查，覆盖近海至海盆不同生境，主要区分硅藻、甲藻、绿藻与蓝藻研究PP遥感机理与算法；并沿陆架-海盆断面，应用遥感数据检验新算法对初级生产力与类群动态变化的反映。成果将为碳循环与全球变化研究奠定核心技术基础。

Phytoplankton fix carbon via photosynthesis (measured as primary production, PP), and then initiate the biological pump. Different phytoplankton have varying photosynthesis and export rates, thus for the study of carbon budget and carbon cycle it is important to develop phytoplankton-resolved remote sensing algorithms. Traditional remote sensing models are centered on the concentration of chlorophyll and ignore differences in phytoplankton functional types, consequently the estimated PP from remote sensing data has a low accuracy which is inadequate to describe the temporal-spatial variations of PP. The remote sensing of phytoplankton absorption coefficient (aph) has a much higher accuracy than the remote sensing of chlorophyll concentration; also, a model through the combination of absorption and photosynthesis quantum yield is in accordance with photosynthesis. Therefore, this project proposes to develop phytoplankton-resolved-absorption-centered remote sensing model for PP. First, a semi-analytical algorithm will be developed to discriminate phytoplankton functional types (PFTs) based on aph spectrum; at the same time, the quantum yield of various typical PFTs will be characterized. Based on these, a PP model using aph and phytoplankton-resolved quantum yield as inputs will be developed. We will conduct both laboratory measurements and field surveys in China seas, covering both nearshore and offshore waters, focusing on diatom, dinoflagellates, green algae and cyanobacteria to study their photosynthesis mechanisms. Further, remote sensing data crossing coastal to oceanic waters will be selected to evaluate the developed algorithms on describing the dynamics of PFTs and PP in these environments. Outcomes of this project will not only greatly improve the accuracy of PP estimation, but also establish solid base in technology and data for studying carbon cycle and climate change.