Using empirical orthogonal functions derived from remote-sensing reflectance for the prediction of phytoplankton pigment concentrations

Ocean Science

T.Dinter¹  A.Bracher¹  F.Steinmetz¹  M. H.Taylor¹  R.Röttgers¹  B.Taylor¹ 

DOI  :  10.5194/os-11-139-2015

学科分类：海洋学与技术

来源: Copernicus Publications

【 摘 要 】

The composition and abundance of algal pigments provide information onphytoplankton community characteristics such as photoacclimation, overallbiomass and taxonomic composition. In particular, pigments play a majorrole in photoprotection and in the light-driven part of photosynthesis. Mostphytoplankton pigments can be measured by high-performance liquidchromatography (HPLC) techniques applied to filtered water samples. Thismethod, as well as other laboratory analyses, is time consuming andtherefore limits the number of samples that can be processed in a giventime. In order to receive information on phytoplankton pigment compositionwith a higher temporal and spatial resolution, we have developed a method toassess pigment concentrations from continuous optical measurements. Themethod applies an empirical orthogonal function (EOF) analysis to remote-sensing reflectance data derived from ship-based hyperspectral underwaterradiometry and from multispectral satellite data (using the Medium Resolution Imaging Spectrometer – MERIS – Polymerproduct developed by Steinmetz et al., 2011) measured in the Atlantic Ocean.Subsequently we developed multiple linear regression models with measured(collocated) pigment concentrations as the response variable and EOFloadings as predictor variables. The model results show that surfaceconcentrations of a suite of pigments and pigment groups can be wellpredicted from the ship-based reflectance measurements, even when only amultispectral resolution is chosen (i.e., eight bands, similar to those usedby MERIS). Based on the MERIS reflectance data, concentrations of total andmonovinyl chlorophyll a and the groups of photoprotective and photosyntheticcarotenoids can be predicted with high quality. As a demonstration of theutility of the approach, the fitted model based on satellite reflectancedata as input was applied to 1 month of MERIS Polymer data to predict theconcentration of those pigment groups for the whole eastern tropicalAtlantic area. Bootstrapping explorations of cross-validation error indicatethat the method can produce reliable predictions with relatively small datasets (e.g., < 50 collocated values of reflectance and pigmentconcentration). The method allows for the derivation of time series fromcontinuous reflectance data of various pigment groups at various regions,which can be used to study variability and change of phytoplanktoncomposition and photophysiology.

【 授权许可】

Unknown