Retrieving the vertical distribution of chlorophyll a concentration and phytoplankton community composition from in situ fluorescence profiles: A method based on a neural network with potential for global‐scale applications

Abstract

A neural network-based method is developed to assess the vertical distribution of (1) chlorophyll a concentration ([Chl]) and (2) phytoplankton community size indices (i.e., microphytoplankton, nanophytoplankton, and picophytoplankton) from in situ vertical profiles of chlorophyll fluorescence. This method (FLA-VOR for Fluorescence to Algal communities Vertical distribution in the Oceanic Realm) uses as input only the shape of the fluorescence profile associated with its acquisition date and geo-location. The neural network is trained and validated using a large database including 896 concomitant in situ vertical profiles of HighPerformance Liquid Chromatography (HPLC) pigments and fluorescence. These profiles were collected during 22 oceanographic cruises representative of the global ocean in terms of trophic and oceanographic conditions, making our method applicable to most oceanic waters. FLAVOR is validated with respect to the retrieval of both [Chl] and phytoplankton size indices using an independent in situ data set and appears to be relatively robust spatially and temporally. To illustrate the potential of the method, we applied it to in situ measurements of the BATS (Bermuda Atlantic Time Series Study) site and produce monthly climatologies of [Chl] and associated phytoplankton size indices. The resulting climatologies appear very promising compared to climatologies based on available in situ HPLC data. With the increasing availability of spatially and temporally wellresolved data sets of chlorophyll fluorescence, one possible global-scale application of FLAVOR could be to develop 3-D and even 4-D climatologies of [Chl] and associated composition of phytoplankton communities. The Matlab and R codes of the proposed algorithm are provided as supporting information.

Reviews and syntheses: parameter identification in marine planktonic ecosystem modelling

“…Their quality-controlled data comprise profiles of total Chl a concentration together with some additional estimates of the relative contributions from pico, nano, and micro phytoplankton. The employed relationship between the relative size distributions and total Chl a concentration was derived from an extensive analysis of highperformance liquid chromatography pigment data in combination with Chl a fluorescence measurements (Sauzède et al, 2015a). The consideration of these profile data will possibly facilitate the estimation of photo-acclimation parameters in particular, and of phytoplankton growth parameters in general.…”

Vertical distribution of chlorophyll <I>a</I> concentration and phytoplankton community composition from in situ fluorescence profiles: a first database for the global ocean

“…In order to assess the vertical distribution of the total chlorophyll a concentration (hereafter, [TChl]) and the chlorophyll a concentration associated to each phytoplankton size index (hereafter, [microChl], [nanoChl] and [picoChl] for microphytoplankton, nanophytoplankton and picophytoplankton respectively), the FLAVOR method (Sauzède et al, 2015a) is applied to each chlorophyll fluorescence profile, satisfying the quality control procedure (see Sect. 2.2).…”

“…2.2). In summary, FLAVOR is a neural network-based method which uses (1) (1) (Sauzède et al, 2015a). Admittedly, the FLAVOR method has some limitations.…”

“…FLAVOR (Fluorescence to Algal communities Vertical distribution in the Oceanic Realm) is a method developed to transform and combine large numbers of fluorescence profiles from various sampling sensors and platforms (Sauzède et al, 2015a). This neural network-based method generates vertical distributions of (1) chlorophyll a concentration and (2) phytoplankton community size indices (i.e., microphytoplankton, nanophytoplankton and picophytoplankton) based on the shape of in situ fluorescence profiles (i.e., normalized profiles) and the day and location of acquisition.…”

Effects of temperature on the optical properties ofMicrocystis aeruginosaandScenedesmus obliquus

“…Determination of in vivo chlorophyll fluorescence is simple, fast, and reliable and is widely used to identify phytoplankton community composition and estimate primary production (Gilbert et al 2000;Zhang et al 2008;Garrido et al 2013;Sauz ede et al 2015). For example, Gilbert et al (2000) used in vivo chlorophyll fluorescence data to estimate primary production with results that were slightly overestimated compared to those derived from 14 C incorporation.…”