Inversion of Phytoplankton Pigment Vertical Profiles from Satellite Data Using Machine Learning

Observing the vertical dynamic of phytoplankton in the water column is essential to understand the evolution of the ocean primary productivity under climate change and the efficiency of the CO2 biological pump. This is usually made through in-situ measurements. In this paper, we propose a machine learning methodology to infer the vertical distribution of phytoplankton pigments from surface satellite observations, allowing their global estimation with a high spatial and temporal resolution. After imputing missing values through iterative completion Self-Organizing Maps, smoothing and reducing the vertical distributions through principal component analysis, we used a Self-Organizing Map to cluster the reduced profiles with satellite observations. These referent vector clusters were then used to invert the vertical profiles of phytoplankton pigments. The methodology was trained and validated on the MAREDAT dataset and tested on the Tara Oceans dataset. The different regression coefficients R2 between observed and estimated vertical profiles of pigment concentration are, on average, greater than 0.7. We could expect to monitor the vertical distribution of phytoplankton types in the global ocean.

[1]  Sylvie Thiria,et al.  Reconstruction of satellite chlorophyll images under heavy cloud coverage using a neural classification method , 2013 .

[2]  Walker O. Smith,et al.  The MAREDAT global database of high performance liquid chromatography marine pigment measurements , 2012 .

[3]  Sylvie Thiria,et al.  Completion of a Sparse GLIDER Database Using Multi-iterative Self-Organizing Maps (ITCOMP SOM) , 2015, ICCS.

[4]  R. Korb,et al.  SeaWiFS in the southern ocean: spatial and temporal variability in phytoplankton biomass around South Georgia , 2004 .

[5]  Yonggang Liu,et al.  Ocean Circulation in the Western Gulf of Mexico Using Self‐Organizing Maps , 2019, Journal of Geophysical Research: Oceans.

[6]  Christopher Chapman,et al.  Reconstruction of Subsurface Velocities From Satellite Observations Using Iterative Self-Organizing Maps , 2016, IEEE Geoscience and Remote Sensing Letters.

[7]  J. Ruiz,et al.  Consequences of a future climatic scenario for the anchovy fishery in the Alboran Sea (SW Mediterranean): A modeling study , 2014 .

[8]  Peter Cornillon,et al.  The Past, Present, and Future of the AVHRR Pathfinder SST Program , 2010 .

[9]  Julia Uitz,et al.  Assessing Pigment-Based Phytoplankton Community Distributions in the Red Sea , 2017, Front. Mar. Sci..

[10]  Peer Bork,et al.  Open science resources for the discovery and analysis of Tara Oceans data , 2015, Scientific Data.

[11]  S. W. Jeffrey.  Algal Pigment Systems , 1980 .

[12]  X. Irigoien,et al.  Microbial planktonic communities in the Red Sea: high levels of spatial and temporal variability shaped by nutrient availability and turbulence , 2017, Scientific Reports.

[13]  S. Thiria,et al.  Inferring the seasonal evolution of phytoplankton groups in the Senegalo‐Mauritanian upwelling region from satellite ocean‐color spectral measurements , 2015 .

[14]  M. Gehlen,et al.  Twenty-first century ocean warming, acidification, deoxygenation, and upper-ocean nutrient and primary production decline from CMIP6 model projections , 2020, Biogeosciences.

[15]  R. Bidigare,et al.  Temporal variability of phytoplankton community structure based on pigment analysis , 1993 .

[16]  A. Mangin,et al.  High Resolution Water Column Phytoplankton Composition Across the Atlantic Ocean From Ship-Towed Vertical Undulating Radiometry , 2020, Frontiers in Marine Science.

[17]  T. Ishimaru,et al.  A phytoplankton absorption-based primary productivity model for remote sensing in the Southern Ocean , 2011, Polar Biology.

[18]  E. S. Chalhoub,et al.  Chlorophyll profile estimation in ocean waters by a set of artificial neural networks , 2015 .

[19]  H. Claustre,et al.  Vertical distribution of phytoplankton communities in open ocean: An assessment based on surface chlorophyll , 2006 .

[20]  A. Zingone,et al.  Plankton in the open Mediterranean Sea: a review , 2009 .

[21]  M. Kahru,et al.  Blending of ocean colour algorithms applied to the Southern Ocean , 2010 .

[22]  R. Guillard,et al.  Synechococcus spp. as likely zeaxanthin‐dominant ultraphytoplankton in the North Atlantic1 , 1985 .

[23]  Andrew J. Watson,et al.  Ecosystem dynamics based on plankton functional types for global ocean biogeochemistry models , 2005 .

[24]  Jan Kaiser,et al.  Phytoplankton spring bloom initiation: The impact of atmospheric forcing and light in the temperate North Atlantic Ocean , 2019, Progress in Oceanography.

[25]  S. Thiria,et al.  Estimation of Secondary Phytoplankton Pigments From Satellite Observations Using Self‐Organizing Maps (SOMs) , 2019, Journal of Geophysical Research: Oceans.

[26]  A. Morel,et al.  Surface pigments, algal biomass profiles, and potential production of the euphotic layer: Relationships reinvestigated in view of remote‐sensing applications , 1989 .

[27]  Teuvo Kohonen,et al.  Essentials of the self-organizing map , 2013, Neural Networks.

[28]  François-Marie Bréon,et al.  Remote sensing of phytoplankton groups in case 1 waters from global SeaWiFS imagery , 2005 .

[29]  B. Mitchell,et al.  Bio-optical properties of Antarctic Peninsula waters: differentiation from temperate ocean models , 1991 .

[30]  J. Huisman,et al.  Scientists’ warning to humanity: microorganisms and climate change , 2019, Nature Reviews Microbiology.

[31]  Hervé Claustre,et al.  Phytoplankton pigment distribution in relation to upper thermocline circulation in the eastern Mediterranean Sea during winter , 2001 .

[32]  S. Thiria,et al.  Phytoplankton Diversity in the Mediterranean Sea From Satellite Data Using Self‐Organizing Maps , 2019, Journal of Geophysical Research: Oceans.

[33]  Jim Aiken,et al.  An absorption model to determine phytoplankton size classes from satellite ocean colour , 2008 .

[34]  B. Mitchell,et al.  A chlorophyll-dependent semianalytical reflectance model derived from field measurements of absorption and backscattering coefficients within the Southern Ocean , 2001 .

[35]  P. Tett,et al.  Regional variations in bio-optical properties of the surface waters in the Southern Ocean , 1994, Antarctic Science.

[36]  Rosalia Santoleri,et al.  Modelling the Vertical Distribution of Phytoplankton Biomass in the Mediterranean Sea from Satellite Data: A Neural Network Approach , 2018, Remote. Sens..

[37]  J. Gattuso,et al.  Future biological and ecosystem impacts of ocean acidification and their socioeconomic-policy implications , 2012 .

[38]  Sylvie Thiria,et al.  The Sicily Channel surface circulation revisited using a neural clustering analysis of a high‐resolution simulation , 2016 .

[39]  Cédric Jamet,et al.  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 , 2015 .

[40]  B. Mitchell,et al.  Light limitation of phytoplankton biomass and macronutrient utilization in the Southern Ocean , 1991 .

[41]  Y. Yamanaka,et al.  Synoptic relationships between surface Chlorophyll- a and diagnostic pigments specific to phytoplankton functional types , 2011 .

[42]  Jean-Marc Nicolas,et al.  Seasonal and interannual variability of ocean color and composition of phytoplankton communities in the North Atlantic, equatorial Pacific and South Pacific , 2004 .

[43]  S. Wright,et al.  Fucoxanthin pigment markers of marine phytoplankton analysed by HPLC and HPTLC , 1987 .

[44]  W. Gregg,et al.  Decadal trends in global pelagic ocean chlorophyll: A new assessment integrating multiple satellites, in situ data, and models , 2014, Journal of geophysical research. Oceans.

[45]  B. Mitchell.  Predictive bio-optical relationships for polar oceans and marginal ice zones , 1992 .

[46]  A. Belgrano,et al.  Non-linear dynamics in marine-phytoplankton population systems , 2004 .

[47]  Joachim Denzler,et al.  Deep learning and process understanding for data-driven Earth system science , 2019, Nature.

[48]  Tapani Raiko,et al.  Tkk Reports in Information and Computer Science Practical Approaches to Principal Component Analysis in the Presence of Missing Values Tkk Reports in Information and Computer Science Practical Approaches to Principal Component Analysis in the Presence of Missing Values , 2022 .

[49]  H. Dierssen,et al.  Bio‐optical properties and remote sensing ocean color algorithms for Antarctic Peninsula waters , 2000 .

[50]  A. D. Barton,et al.  Physical controls of variability in North Atlantic phytoplankton communities , 2015 .

[51]  S. Jeffrey,et al.  Chlorophyllase distribution in ten classes of phytoplankton: a problem for chlorophyll analysis , 1987 .

[52]  Julie M. Roessig,et al.  Effects of global climate change on marine and estuarine fishes and fisheries , 2004, Reviews in Fish Biology and Fisheries.

[53]  Christopher D G Harley,et al.  The impacts of climate change in coastal marine systems. , 2006, Ecology letters.

[54]  Fouad Badran,et al.  Retrieving the evolution of vertical profiles of Chlorophyll-a from satellite observations using Hidden Markov Models and Self-Organizing Topological Maps , 2015 .

[55]  Robert J. W. Brewin,et al.  Obtaining Phytoplankton Diversity from Ocean Color: A Scientific Roadmap for Future Development , 2017, Front. Mar. Sci..

[56]  Kevin R. Arrigo,et al.  Primary production in Southern Ocean waters , 1998 .

[57]  B. Mitchell,et al.  Observations of modeling of the Antartic phytoplankton crop in relation to mixing depth , 1991 .

[58]  Rosalia Santoleri,et al.  An Artificial Neural Network to Infer the Mediterranean 3D Chlorophyll-a and Temperature Fields from Remote Sensing Observations , 2020, Remote. Sens..