Synopsis of U.S. JGOFS Mini-workshop on Functional Groups
Rutgers University, January 18 -19, 1999

On January 18-19, 1999 approximately 30 scientists, including post-doctoral research associates and graduate students, met at Rutgers University to discuss how to represent key functional groups of phytoplankton in biogeochemical models. The  specific goals of the workshop were:

(1) To explore remotely sensed and/or in situ methods that can be used to produce a seasonal or monthly climatology of key "functional" groups of phytoplankton.
(2) To examine how functional groups can be mathematically represented in biogeochemical models.
(3) To develop informal international, working relationships amongst specialists in remote sensing, in situ optics,  phytoplankton ecophysiology, and biogeochemical modeling, both within and outside of JGOFS.
The first two related issues are born out of the realization, developed in several SMP proposals, that changes in biogeochemical fluxes are mediated by relatively few groups of phytoplankton, such as coccolithophorids, diatoms, and nitrogen fixing cyanobacteria.  It is hypothesized that the spatial and temporal distributions of the key functional groups are determined largely by mesoscale physical and chemical phenomena.  If this hypothesis is correct to a first order, then empirical relationships could be used to predict the statistical distributions of the functional groups given specific model fields.  The last issue is critical for the success of the first two goals, as the information required to develop a functional group model cannot be assimilated by one individual, but rather requires a concerted, long-term effort amongst colleagues who effectively work as a multidisciplinary team.

The first day of the workshop was largely devoted to remotely sensed (e.g. satellite-based) information, physiological models and their application to retrieve and predict key groups of marine phytoplankton.  An overview of the problems and potential approaches was given by Paul Falkowski, who discussed the concept of "functional groups" in the context of specific biogeochemical processes and the evolution of these processes over geological time.  Functional groups can embrace several taxonomic divisions, (e.g. in the case of photoautotrophy), or can be restricted to a single taxa of photoautotrophs (e.g. in the case of diazotrophy).  Descriptions of factors that select for given functional groups (e.g. "Margalef's mandala") will inevitably be generalizations for which many exceptions can be found.  Nonetheless, the fundamental concept of universal distribution and local selection remains; that is, in any given water mass there is a potential to find any functional group, however one group can dominate as a consequence of processes that occur locally.  Determining the local selective pressures from remotely sensed information appears tractable.

Presentations by Chris Brown and Ajit Subramaniam provided a framework for understanding the retrievals for coccolithophorids and the N2-fixing cyanobacteria in the genus Trichodesmium from ocean color imagery.  Algorithms for both functional groups are implemented ad hoc, and weekly and monthly maps for the global ocean can be generated.  The retrievals are based on relatively high concentrations of the functional groups in the upper ocean, but can be used to derive "climatological" maps for the groups.  Several geochemists, especially David Archer, were concerned that the maps for coccolithophorid distributions do not correspond to their perceptions of calcite production based on sedimentation patterns (e.g. there is a paucity of coccolothophorid "hits" in the central Pacific basins based on the satellite retrievals, but there is an abundance of carbonate in the underlying sediment).  Similarly, the sensitivity of the algorithm used to selectively retrieve Trichodesmium from SeaWiFS images does not closely correspond to the geochemical estimation of the spatial distribution of nitrogen fixation in the upper ocean.  There was active discussion concerning the effects of vertical structure, physiological status, and atmospheric interference in estimating the abundance of these two functional groups, however, it is generally agreed that the retrievals presently serve as a basis for developing simple "rules" for ascribing climatological distributions in geochemical models.

Mary Elena Carr described how compound remote sensing can be used to help develop climatological maps for functional groups.  Inclusion of SST, wind stress, and other physical variables can be used in conjunction with selective retrieval optical algorithms to further constrain functional groups in relation to local physical forcings.  Storm events and calm periods between storms can be extremely important in selecting for specific functional groups.  The potential for retrieving Phaeocystis, diatoms and other organisms was examined in the context of hyperspectral imagery.  Oscar Schofield suggested that hyperspectral data could be used to selectively retrieve specific groups, but such retrievals are not based solely on optics, but requires knowledge of the ecology of the target organisms.  This view was echoed by Norm Nelson and Ricardo Letelier, who provided data from BATS and HOT respectively to seasonal and annual changes in species compositions do not always conform to standard biological oceanographic paradigms.

Richard Geider and Bernd Kroon provided two different physiological models that can be used to predict specific growth rates of individual species, or functional groups, with some knowledge of pigment composition and other physiological state variables (e.g. cell carbon content).  While both models explictly focus on photosynthesis and growth, they can accommodate other processes, such as calcification, nitrogen fixation, etc.  Such physiological representations are not presently used in models of biogeochemical cycles, with a consequence that biological processes are generally represented as "black boxes" in biogeochemical models.  A summation of the presentations was presented by Egil Sakshaug.

Biogeochemical modeling efforts were presented by Scott Doney, Ray Najjar and Dave Archer.  Scott explained the computational constraints on model structure for global ocean processes, and described how model fields were generated that can be potentially used to provide input variables for funcational group models.  Ray gave an overview of biogeochemical model hierarchy that clearly described the basic tenets geochemists use in developing models (e.g. mass balance).  Dave described his attempts to explain changes in atmospheric CO2 concentrations during glacial/interglacial transitions and the importance of understanding the factors controlling ratio of opaline silica to carbonate fluxes (i.e. diatom/coccolithophorid ratios) in the upper water column.  Jorge Sarmiento explained how, regardless of what happens to the biological pump, ocean physical and chemical characteristics will change as a consequence of global warming.  Although it is generally believed that such changes will lead to alterations in the biological "structure" of marine food webs, predicting the changes remain elusive.  Rob Armstrong presented a model that predicts size structure of phytoplankton communities based on nutrient supplies and grazing rates.  Falkowski suggested implementing such models in a data assimilation program, and will work with Rob to test the model assumptions in the LEO program this summer.

Ina Tegen described the potentials and limitations of remotely sensed data, in situ measurements and mathematical models to describe aerosol fluxes of elements such as iron to the oceans.  The presentation led to active discussion about integration of such model fields in biogeochemical and functional group models.

The conclusion of the workshop led to a positive feeling that interactions between SMP researchers and a broader oceanographic and atmospheric community would be extremely productive.  Key issues addressed were a need to:

(1) Develop simple, empirical climatologies of key functional groups using, in so far as possible, pigment data from JGOFS as well as direct observations of organisms when available.
(2) More carefully and methodically examine physical and chemical features when specific functional groups are observed in an attempt to develop a more realistic ecological description of niches. These efforts might be aided by further autecological studies in the laboratory or in mesocosms.
(3) Develop remotely sensed algorithms and in situ analytical tools for identifying key functional groups in present and proposed observing systems.
(4) Develop high resolution, mesoscale models that predict functional group occurrences to test our knowledge of functional group niches.
(5) Develop predictive capabilities for key functional groups based on climate change scenarios in coupled atmosphere-ocean models.


List of participants:
 
Name Phone Fax E-mail
David Archer 773-702-0823 773-702-9505 d-archer@uchicago.edu
Rob Armstrong 609-258-5260 609-258-2850 raa@splash.princeton.edu
Bob Arnone 228-688-5265 228-688-4149 arnone@nrlssc.navy.mil
Michael Behrenfeld 732-932-7889 732-932-3036 behren@ahab.rutgers.edu
Trisha Bergmann 732-932-6555 ext.532 732-932-8578 bergmann@arctic.rutgers.edu
Ilana Berman-Frank 732-932-3497 732-932-3036 irfrank@imcs.rutgers.edu
Paul Bissett 813-273-3407 ext.2 813-273-4161 pbissett@marine.usf.edu
Chris Brown 301-763-8102 ext.158 301-763-8020 chrisb@orbit.nesdis.noaa.gov
Mary-Elena Carr 818-354-5097 818-393-6720 mec@pacific.jpl.nasa.gov
Jay Cullen 732-932-6555 ext.263 732-932-8578 cullen@imcs.rutgers.edu
Scott Doney 303-497-1639 303-497-1700 doney@ucar.edu
Paul Falkowski 732-932-6555 ext.370 732-932-8578 falko@imcs.rutgers.edu
Richard Geider +44-1752-633100 +44-1752-633102 RDG@wpo.nerc.ac.uk
Joe Grzymski 732-932-6555 ext.222 732-932-8578 joeg@imcs.rutgers.edu
Anna Hilting 252-729-1488 252-504-7548 ahilting@duke.edu
Uwe Kils 609-748-9693 732-932-1821 uwekils@aol.com
Dorota Kolber 732-932-6555 ext.244 732-932-8578 dorota@warrior.rutgers.edu
Zbigniew Kolber 732-932-6555 ext.233 732-932-8578 zkolber@imcs.rutgers.edu
Bernd Kroon 4+49-(0)471-4831-810 4+49-(0)471-4831-425 bkroon@awi-bremerhaven.de
Ricardo Letelier 541-737-3890 541-737-2064 letelier@oce.orst.edu
Anna Matteoda 732-932-6555 ext.542 732-932-1821 anna@arctic.rutgers.edu
Jim Miller 732-932-6555 ext.545 732-932-1821 miller@arctic.rutgers.edu
Mark Moline 805-756-2948 805-756-1419 mmoline@calpoly.edu
Norm Nelson 441-297-1880 ext.307 441-297-8143 norm@bbsr.edu
Emmeline Romana 732-932-6555 ext. 244 732-932-8578 romana@imcs.rutgers.edu
Egil Sakshaug +47-73-591580 +47-73-591597 Egil.Sakshaug@vm.unit.no
Jorge Sarmiento 609-258-6585 609-258-2850 jls@splash.princeton.edu
Oscar Scholfield 732-932-6555 ext.548 732-932-8578 oscar@ahab.rutgers.edu
Rob Sherrell 732-932-6555 ext.252 732-932-8578 sherrell@imcs.rutgers.edu
Ajit Subramaniam 410-326-7201 410-326-7341 subraman@cbl.umces.edu
Ina Tegen 212-678-5573 212-678-5552 itegen@giss.nasa.gov
Sasha Tozzi 732-932-6555 ext.244 732-932-8578 sasha@alpha.szu.it

Paul Falkowski
Environmental Biophysics and Molecular Ecology Program
Rutgers University
71 Dudley Road
New Brunswick, NJ 08901-8521
Tel: (732) 932-6555, x 370
Fax: (732) 932-8578
falko@imcs.rutgers.edu
http://www.marine.rutgers.edu/ebme/index.html
http://marine.rutgers.edu/opp