Improvement to the PhytoDOAS method for identification of coccolithophores using hyper-spectral satellite data

[摘要] The goal of this study was to improve PhytoDOAS, which is a new retrievalmethod for quantitative identification of major phytoplankton functionaltypes (PFTs) using hyper-spectral satellite data. PhytoDOAS is an extensionof the Differential Optical Absorption Spectroscopy (DOAS, a method fordetection of atmospheric trace gases), developed for remote identification ofoceanic phytoplankton groups. Thus far, PhytoDOAS has been successfullyexploited to identify cyanobacteria and diatoms over the global ocean fromSCIAMACHY (SCanning Imaging Absorption spectroMeter for AtmosphericCHartographY) hyper-spectral data. This study aimed to improve PhytoDOASfor remote identification of coccolithophores, another functional group ofphytoplankton. The main challenge for retrieving more PFTs by PhytoDOAS is toovercome the correlation effects between different PFT absorption spectra.Different PFTs are composed of different types and amounts of pigments, butalso have pigments in common, e.g. chl a, causing correlation effectsin the usual performance of the PhytoDOAS retrieval. Two ideas have beenimplemented to improve PhytoDOAS for the PFT retrieval of more phytoplanktongroups. Firstly, using the fourth-derivative spectroscopy, the peak positionsof the main pigment components in each absorption spectrum have been derived.After comparing the corresponding results of major PFTs, the optimizedfit-window for the PhytoDOAS retrieval of each PFT was determined. Secondly,based on the results from derivative spectroscopy, a simultaneous fit ofPhytoDOAS has been proposed and tested for a selected set of PFTs(coccolithophores, diatoms and dinoflagellates) within an optimizedfit-window, proven by spectral orthogonality tests. The method was thenapplied to the processing of SCIAMACHY data over the year 2005. Comparisonsof the PhytoDOAS coccolithophore retrievals in 2005 with othercoccolithophore-related data showed similar patterns in their seasonaldistributions, especially in the North Atlantic and the Arctic Sea. Theseasonal patterns of the PhytoDOAS coccolithophores indicated very goodagreement with the coccolithophore modeled data from the NASA OceanBiochemical Model (NOBM), as well as with the global distributions ofparticulate inorganic carbon (PIC), provided by MODIS (MODerate resolutionImaging Spectroradiometer)-Aqua level-3 products. Moreover, regarding the factthat coccolithophores belong to the group of haptophytes, the PhytoDOASseasonal coccolithophores showed good agreement with the global distributionof haptophytes, derived from synoptic pigment relationships applied toSeaWiFS chl a. As a case study, the simultaneous mode of PhytoDOAS has beenapplied to SCIAMACHY data for detecting a coccolithophore bloom which wasconsistent with the MODIS RGB image and the MODIS PIC map of the bloom,indicating the functionality of the method also in short-term retrievals.

[学科分类] 海洋学与技术