Assessing biases in computing size spectra of automatically classified zooplankton from imaging systems: A case study with the ZooScan integrated system

Abstract Body size constrains prey–predator interactions and physiology, therefore plankton size spectra have been appointed as synthetic descriptors of plankton community structure and functioning. Recently developed imaging systems and supervised classification tools provide size measurements of any object in situ or in net samples and automatically classify them into previously defined categories. But because the nature of objects detected by these imaging systems is diverse, from non-living detritus to organisms of different plankton taxa, and because the steps in the analysis could introduce specific biases, a careful analysis of such plankton size spectra is needed before going deeper into ecological considerations. Using a WP2 net time series, we propose a general framework to analyze and validate zooplankton size spectra collected with nets and analyzed with the ZooScan integrated system that includes supervised classification. Size spectra were controlled, at each step of the procedure, to assess the modification of their shape due to several possible biases: (i) the effect of objects touching each other during the image acquisition, (ii) the error of the automatic classification differing among size classes and (iii) the choice of model to estimate body biovolume.

[1]  S. Comeau,et al.  Impact of ocean acidification on a key Arctic pelagic mollusc ( Limacina helicina ) , 2009 .

[2]  R. W. Sheldon,et al.  ON THE OCCURRENCE AND FORMATION OF SMALL PARTICLES IN SEAWATER , 1967 .

[3]  Mark E. Baird,et al.  A size-resolved pelagic ecosystem model , 2007 .

[4]  Thomas M. Powell,et al.  Bottom–up and top–down forcing in a simple size-structured plankton dynamics model , 2008 .

[5]  Blaise Faugeras,et al.  Modeling environmental effects on the size-structured energy flow through marine ecosystems. Part 1: The model , 2007 .

[6]  T. Parsons.  The Use of Particle Size Spectra in Determining the Structure of a Plankton Community , 1969 .

[7]  Astthor Gislason,et al.  Comparison between automated analysis of zooplankton using ZooImage and traditional methodology , 2009 .

[8]  C. Duarte,et al.  Annual zooplankton succession in coastal NW Mediterranean waters: the importance of the smaller size fractions , 2001 .

[9]  A. Burd,et al.  Effect of net avoidance on estimates of diel vertical migration , 2004 .

[10]  I. Batista,et al.  Unusual early morning development of the equatorial anomaly in the Brazilian sector during the Halloween magnetic storm , 2006 .

[11]  J. Stockwell,et al.  Calibration of an optical plankton counter for use in fresh water , 1998 .

[12]  P. Utgoff,et al.  RAPID: Research on Automated Plankton Identification , 2007 .

[13]  Meng Zhou.  What determines the slope of a plankton biomass spectrum , 2006 .

[14]  Ali Abbas,et al.  A Clustering Approach for the Separation of Touching Edges in Particle Images , 2008 .

[15]  C. Duarte,et al.  Some aspects of the analysis of size spectra in aquatic ecology , 1997 .

[16]  F. Echevarría,et al.  Dealing with size-spectra: some conceptual and mathematical problems , 1994 .

[17]  P. Culverhouse,et al.  Do experts make mistakes? A comparison of human and machine identification of dinoflagellates , 2003 .

[18]  S. Souissi,et al.  North Atlantic climate and northwestern Mediterranean plankton variability , 2005 .

[19]  D. Marmorek,et al.  The use of zooplankton in a biomonitoring program to detect lake acidification and recovery , 1993 .

[20]  S. Hernández‐León,et al.  Zooplankton biomass estimation from digitized images: a comparison between subtropical and Antarctic organisms , 2009 .

[21]  A. D. Poularikas,et al.  Automated sizing, counting and identification of zooplankton by pattern recognition , 1984 .

[22]  K. Denman,et al.  Organisation in the pelagic ecosystem , 1977, Helgoländer wissenschaftliche Meeresuntersuchungen.

[23]  J. Gasol,et al.  Seasonal variations in size structure and procaryotic dominance in sulfurous Lake Cisó , 1991 .

[24]  Karen Manríquez,et al.  The influence of coastal upwelling on the mesozooplankton community structure in the coastal zone off Central/Southern Chile as assessed by automated image analysis , 2009 .

[25]  D. S. Glazier,et al.  Beyond the ‘3/4‐power law’: variation in the intra‐and interspecific scaling of metabolic rate in animals , 2005, Biological reviews of the Cambridge Philosophical Society.

[26]  Gabriel Gorsky,et al.  The Autonomous Image Analyzer - enumeration, measurement and identification of marine phytoplankton , 1989 .

[27]  W. G. Sprules.  Nonmetric multidimensional scaling analyses of temporal variation in the structure of limnetic zooplankton communities , 1980, Hydrobiologia.

[28]  Pierre Geurts,et al.  Supervised learning with decision tree-based methods in computational and systems biology. , 2009, Molecular bioSystems.

[29]  K. Banse.  Zooplankton: Pivotal role in the control of ocean production I. Biomass and production , 1995 .

[30]  D. Boix,et al.  Size and species diversity of zooplankton communities in fluctuating Mediterranean salt marshes , 2006 .

[31]  D. Boix,et al.  Zooplankton structure and dynamics in permanent and temporary Mediterranean salt marshes: taxon-based and size-based approaches , 2005 .

[32]  A. King,et al.  Using an optical plankton counter to determine the size distributions of preserved zooplankton samples , 1999 .

[33]  G. Beaugrand,et al.  Spatial dependence of calanoid copepod diversity in the North Atlantic Ocean , 2002 .

[34]  A. M. John,et al.  THE RELATIONSHIP BETWEEN SIZE OF NET USED AND ESTIMATES OF ZOOPLANKTON DIVERSITY1 , 1966 .

[35]  Manfred Rolke,et al.  Size structure analysis of zooplankton samples by means of an automated image analyzing system , 1984 .

[36]  H. Ducklow,et al.  A nitrogen-based model of plankton dynamics in the oceanic mixed layer , 1990 .

[37]  E. Eyto,et al.  Assessing the status of shallow lakes using an additive model of biomass size spectra , 2007 .

[38]  B. Planque,et al.  Long-term time series in Calanus finmarchicus abundance - A question of space? , 1997 .

[39]  M. Heath.  Size spectrum dynamics and the planktonic ecosystem of Loch Linnhe , 1995 .

[40]  Philippe Grosjean,et al.  Spring zooplankton distribution in the Bay of Biscay from 1998 to 2006 in relation with anchovy recruitment , 2008 .

[41]  P. Wiebe,et al.  From the Hensen net toward four-dimensional biological oceanography , 2003 .

[42]  Marcello Vichi,et al.  A generalized model of pelagic biogeochemistry for the global ocean ecosystem. Part I: Theory , 2007 .

[43]  Stewart W. Wilson,et al.  Noname manuscript No. (will be inserted by the editor) Learning Classifier Systems: A Survey , 2022 .

[44]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[45]  Phil F. Culverhouse,et al.  Automatic image analysis of plankton: future perspectives , 2006 .

[46]  Philippe Grosjean,et al.  Enumeration, measurement, and identification of net zooplankton samples using the ZOOSCAN digital imaging system , 2004 .

[47]  T. Platt,et al.  Patterns of biomass-size spectra from oligotrophic waters of the Northwest Atlantic [review article] , 2003 .

[48]  J. Gattuso,et al.  Community metabolism and air-sea C02 fluxes in a coral reef ecosystem (Moorea, French Polynesia) , 1993 .

[49]  Marc Picheral,et al.  Digital zooplankton image analysis using the ZooScan integrated system , 2010 .

[50]  Alex W. Herman,et al.  Design and calibration of a new optical plankton counter capable of sizing small zooplankton , 1992 .

[51]  M. Alcaraz,et al.  Estimating zooplankton biomass through image analysis , 2003 .

[52]  Gilles Blanchard,et al.  How wrong can we get? A review of machine learning approaches and error bars. , 2009, Combinatorial chemistry & high throughput screening.

[53]  Sandra Brucet Balmaña.  ZOOPLANKTON STRUCTURE AND DYNAMICS IN MEDITERRANEAN MARSHES (EMPORDÀ WETLANDS): A SIZE-BASED APPROACH , 2003 .

[54]  R. W. Sheldon,et al.  The Size Distribution of Particles in the OCEAN1 , 1972 .

[55]  J. H. Nichols,et al.  Mesh selection of copepodite and nauplius stages of four calanoid copepod species , 1991 .

[56]  J. Kelly,et al.  Evaluation of Optically Acquired Zooplankton Size-Spectrum Data as a Potential Tool for Assessment of Condition in the Great Lakes , 2005, Environmental management.

[57]  Drake Circus.  A test model for optical plankton counter (OPC) coincidence and a comparison of OPC-derived and conventional measures of plankton abundance , 2000 .

[58]  J J Vaquero,et al.  Applying watershed algorithms to the segmentation of clustered nuclei. , 1998, Cytometry.

[59]  Ricco Rakotomalala.  TANAGRA, une plate-forme d'expérimentation pour la fouille de données , 2005, Monde des Util. Anal. Données.

[60]  J. G. Field,et al.  The size-based dynamics of plankton food webs. I. A simulation model of carbon and nitrogen flows , 1991 .

[61]  J. A. Lozano,et al.  Optimizing the number of classes in automated zooplankton classification , 2009 .

[62]  J. Egozcue,et al.  A nonparametric method for the measurement of size diversity with emphasis on data standardization , 2008 .

[63]  S. Weisberg,et al.  A comparison of plastic and plankton in the north Pacific central gyre. , 2001, Marine pollution bulletin.

[64]  Jacquelyn L. Anderson,et al.  Effect of towing speed on retention of zooplankton in bongo nets , 2007 .

[65]  Magnus L. Johnson,et al.  Swimming capacity and pleopod beat rate as a function of sex, size and moult stage in Northern krill Meganyctiphanes norvegica , 2003 .

[66]  R. Armstrong.  A hybrid spectral representation of phytoplankton growth and zooplankton response: The control rod model of plankton interaction , 2003 .

[67]  D. Robins,et al.  Is Oithona the most important copepod in the world's oceans? , 2001 .

[68]  R. Hopcroft,et al.  Assessment of ZooImage as a tool for the classification of zooplankton , 2008 .

[69]  M. M. Mullin,et al.  Relation between biomass and body weight of plankton in a steady state oceanic ecosystem1 , 1986 .

[70]  S. Hernández‐León,et al.  Zooplankton biomass estimated from digitalized images in Antarctic waters: A calibration exercise , 2006 .

[71]  V. Smetácek.  Role of sinking in diatom life-history cycles: ecological, evolutionary and geological significance , 1985 .

[72]  K. Gorbatenko,et al.  Comparing the catch efficiency with different types of plankton nets in the high production zones of the Pacific Ocean , 2007 .

[73]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[74]  G. Hays,et al.  Climate change and marine plankton. , 2005, Trends in ecology & evolution.

[75]  P. Laval.  Un modéle mathématique de l'évitement d'un filet à plancton, son application pratique, et sa vérification indirecte en recourant au parasitisme de l'amphipode hypéride Vibilia armata Bovallius , 1974 .

[76]  Factors affecting the performance of the optical plankton counter in large lakes: Insights from Lake Michigan and laboratory studies , 2006 .

[77]  A. Fleminger,et al.  AVOIDANCE OF TOWED NETS BY ZOOPLANKTON1 , 1965 .