A Caltech Library Service

Inter-comparison of phytoplankton functional type phenology metrics derived from ocean color algorithms and Earth System Models

Kostadinov, Tihomir S. and Cabré, Anna and Vedantham, Harish and Marinova, Irina and Bracher, Astrid and Brewin, Robert J. W. and Bricaud, Annick and Hirata, Takafumi and Hirawake, Toru and Hardman-Mountford, Nick J. and Mouw, Colleen and Roy, Shovonlal and Uitz, Julia (2017) Inter-comparison of phytoplankton functional type phenology metrics derived from ocean color algorithms and Earth System Models. Remote Sensing of Environment, 190 . pp. 162-177. ISSN 0034-4257.

[img] PDF (Appendix A. Supplementary Material) - Supplemental Material
See Usage Policy.


Use this Persistent URL to link to this item:


Ocean color remote sensing of chlorophyll concentration has revolutionized our understanding of the biology of the oceans. However, a comprehensive understanding of the structure and function of oceanic ecosystems requires the characterization of the spatio-temporal variability of various phytoplankton functional types (PFTs), which have differing biogeochemical roles. Thus, recent bio-optical algorithm developments have focused on retrieval of various PFTs. It is important to validate and inter-compare the existing PFT algorithms; however direct comparison of retrieved variables is non-trivial because in those algorithms PFTs are defined differently. Thus, it is more plausible and potentially more informative to focus on emergent properties of PFTs, such as phenology. Furthermore, ocean color satellite PFT data sets can play a pivotal role in informing and/or validating the biogeochemical routines of Earth System Models. Here, the phenological characteristics of 10 PFT satellite algorithms and 7 latest-generation climate models from the Coupled Model Inter-comparison Project (CMIP5) are inter-compared as part of the International Satellite PFT Algorithm Inter-comparison Project. The comparison is based on monthly satellite data (mostly SeaWiFS) for the 2003–2007 period. The phenological analysis is based on the fraction of microplankton or a similar variable for the satellite algorithms and on the carbon biomass due to diatoms for the climate models. The seasonal cycle is estimated on a per-pixel basis as a sum of sinusoidal harmonics, derived from the Discrete Fourier Transform of the variable time series. Peak analysis is then applied to the estimated seasonal signal and the following phenological parameters are quantified for each satellite algorithm and climate model: seasonal amplitude, percent seasonal variance, month of maximum, and bloom duration. Secondary/double blooms occur in many areas and are also quantified. The algorithms and the models are quantitatively compared based on these emergent phenological parameters. Results indicate that while algorithms agree to a first order on a global scale, large differences among them exist; differences are analyzed in detail for two Longhurst regions in the North Atlantic: North Atlantic Drift Region (NADR) and North Atlantic Subtropical Gyre West (NASW). Seasonal cycles explain the most variance in zonal bands in the seasonally-stratified subtropics at about 30° latitude in the satellite PFT data. The CMIP5 models do not reproduce this pattern, exhibiting higher seasonality in mid and high-latitudes and generally much more spatially homogeneous patterns in phenological indices compared to satellite data. Satellite data indicate a complex structure of double blooms in the Equatorial region and mid-latitudes, and single blooms on the poleward edges of the subtropical gyres. In contrast, the CMIP5 models show single annual blooms over most of the ocean except for the Equatorial band and Arabian Sea.

Item Type:Article
Related URLs:
URLURL TypeDescription
Vedantham, Harish0000-0002-0872-181X
Additional Information:© 2016 Elsevier Inc. Received 16 December 2015. Received in revised form 9 November 2016. Accepted 17 November 2016. Available online 28 December 2016. This work was performed with funding from NASA Ocean Biology and Biogeochemistry Program (grant #NNX13AC92G to Irina Marinov and Tihomir S. Kostadinov). The contribution of A. Bracher, R. Brewin and A. Bricaud was partly funded via the ESA SEOM SY-4Sci Synergy project SynSenPFT. We thank Tilman Dinter and Bernard Gentili for help with the PhytoDOAS and CB06 algorithm processing, respectively, Amane Fujiwara for leading development of the FUJI11 algorithm, and Aurea Ciotti for leading the development of the CB06 algorithm. We thank David Shields (specifically for producing Supplement Fig. S11C and processing BATS data, and for work on the Chl gyre contour lines), Svetlana Milutinović, and Danica Fine for providing help in the completion of this work. We thank Libe Washburn for FFT processing advice. We thank Jordan Rosenthal for his compass plot labeling script (used here with modifications in Fig. 7). All data processing, analysis and plotting for the phenological analysis in this work was done in MATLAB®. We additionally acknowledge all the satellite algorithm providers (and their funding agencies) for their support and providing their data. The coastlines displayed on maps were extracted with the NOAA/NGDC GEODAS-NG software using the L1 layer of the GSHHG v2.2.3 (Wessel and Smith, 1996) coastline data set. We acknowledge the World Climate Research Programme's Working Group on Coupled Modelling, which is responsible for CMIP, and we thank the climate modeling groups (listed in Table S1 of this paper) for producing and making available their model output. For CMIP the U.S. Department of Energy's Program for Climate Model Diagnosis and Inter-comparison provides coordinating support and led development of software infrastructure in partnership with the Global Organization for Earth System Science Portals. We are grateful to three anonymous reviewers and the editor whose comments and suggestions improved the quality of this manuscript.
Funding AgencyGrant Number
European Space Agency (ESA)SynSenPFT
Subject Keywords:Phytoplankton bloom; Phenology; Phytoplankton functional types; Microplankton; Ocean color algorithms; Inter-comparison; CMIP5 Earth System Models; Discrete Fourier Transform
Record Number:CaltechAUTHORS:20170407-093456585
Persistent URL:
Official Citation:Tihomir S. Kostadinov, Anna Cabré, Harish Vedantham, Irina Marinov, Astrid Bracher, Robert J.W. Brewin, Annick Bricaud, Takafumi Hirata, Toru Hirawake, Nick J. Hardman-Mountford, Colleen Mouw, Shovonlal Roy, Julia Uitz, Inter-comparison of phytoplankton functional type phenology metrics derived from ocean color algorithms and Earth System Models, Remote Sensing of Environment, Volume 190, 1 March 2017, Pages 162-177, ISSN 0034-4257, (
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:75831
Deposited By: Ruth Sustaita
Deposited On:07 Apr 2017 16:56
Last Modified:03 Oct 2019 16:54

Repository Staff Only: item control page