Accuracy of Empirical Satellite Algorithms for Mapping Phytoplankton Diagnostic Pigments in the Open Ocean: A Supervised Learning Perspective

Andy Stock,Ajit Subramaniam

doi:10.3389/fmars.2020.00599

Andy Stock, Ajit Subramaniam

Open Access

https://doi.org/10.3389/fmars.2020.00599

Copy DOI

Abstract

Monitoring phytoplankton community composition from space is an important challenge in ocean remote sensing. Researchers have proposed several algorithms for this purpose. However, the in-situ data used to train and validate such algorithms at the global scale are often clustered along ship cruise tracks and in some well-studied locations, whereas many large marine regions have no in-situ data at all. Furthermore, oceanographic variables are typically spatially auto-correlated. In this situation, the common practice of validating algorithms with randomly chosen held-out observations can underestimate errors. Based on a global database of in-situ HPLC data, we applied supervised learning methods to train and test empirical algorithms predicting the relative concentrations of eight diagnostic pigments that serve as biomarkers for different phytoplankton types. For each pigment, we trained three types of satellite algorithms distinguished by their input data: abundance-based (using only chlorophyll-a as input), spectral (using remote sensing reflectance), and ecological algorithms (combining reflectance and environmental variables). The algorithms were implemented as statistical models (smoothing splines, polynomials, random forests and boosted regression trees). To address clustering of data and spatial auto-correlation, we tested the algorithms by means of spatial block cross-validation. This provided a less confident picture of the potential for global mapping of diagnostic pigments and hence the associated phytoplankton types using existing satellite data than suggested by some previous research and a 5-fold cross-validation conducted for comparison. Of the eight diagnostic pigments, only two (fucoxanthin and zeaxanthin) could be predicted in marine regions that the algorithms were not trained in with considerably lower errors than a constant null model. Thus, global-scale algorithms based on existing, multi-spectral satellite data and commonly available environmental variables can estimate relative diagnostic pigment concentrations and hence distinguish phytoplankton types in some broad classes, but are likely inaccurate for some classes and in some marine regions. Overall, the ecological algorithms had the lowest prediction errors. Finally, our results suggest that more discussion of the best approaches for training and validating empirical satellite algorithms is needed if the in-situ data are unevenly distributed in the study region and spatially clustered.

Highlights

Accounting for almost half of the world’s net primary production (Field et al, 1998), phytoplankton are the foundation of the marine food web and have a major climate-regulating function through the ocean’s biological pump: the oceans have stored about 1/3 of historic anthropogenic CO2 emissions (Sabine et al, 2004; Gruber et al, 2019)
Spectral algorithms use remote sensing reflectances (Rrs) or normalized water-leaving radiances measured in different wavelength bands, as well as derived variables describing absorption and backscattering, to predict variables related to phytoplankton community composition
We re-calculated a look-up table (LUT) linking chlorophyll-a to Rrs based on the data in Table 1, but given the size of these global daily data, we built the look-up

Summary

Introduction

Accounting for almost half of the world’s net primary production (Field et al, 1998), phytoplankton are the foundation of the marine food web and have a major climate-regulating function through the ocean’s biological pump: the oceans have stored about 1/3 of historic anthropogenic CO2 emissions (Sabine et al, 2004; Gruber et al, 2019). Because different types of phytoplankton play different ecological and biogeochemical roles, the development of algorithms that can map phytoplankton community composition from space is an important challenge in satellite monitoring of the oceans (Bracher et al, 2017). Many recent algorithms instead distinguish between various phytoplankton functional types (PFTs; e.g., Alvain et al, 2008), phytoplankton size classes (PSCs; e.g., Brewin et al, 2010), or combine both classifications (e.g., Hirata et al, 2011). Spectral algorithms instead use variations in the optical properties of different phytoplankton types to distinguish them. Some spectral algorithms directly link satellite-measured remote sensing reflectance or normalized water-leaving radiance to PFTs, PSCs or pigments (e.g., Alvain et al, 2005, 2008; Li et al, 2013; Ben Mustapha et al, 2014). Ecological algorithms (e.g., Raitsos et al, 2008; Palacz et al, 2013; Hu et al, 2018a) combine spectral and environmental variables related to the mechanisms that affect the biogeography of different phytoplankton types, such as sea surface temperature (Rudorff and Kampel, 2012)

Methods

Results

Conclusion