(image source NASA)

Phytoplankton Dynamics

To bloom or not to bloom 
that is the question


In collaboration with:
K. H. Wiltshire, M. Boersma
(Biologische Anstalt HelgolandAlfred-Wegener-Institut für Polar- und Meeresforschung)

W. Greve
(Deutsches Zentrum für Marine BiodiversitätsforschungSenckenberg Forschungsinstitut und Naturmuseum)

P. C. Reid
(Sir Alister Hardy Foundation for Ocean Science (SAHFOS), Plymouth, UK)

U. Feudel
(ICBM)


Facts about phytoplankton:

Phytoplankton are tiny floating plants (algae) that live in the ocean and in lakes. In the process of photosynthesis, phytoplankton produce half of the world's oxygen. Moreover, by primary production, death and sinking they effectively transport carbon from the ocean's surface layer to marine sediments, a process by which phytoplankton exert a global-scale influence on climate (carbon dioxide and the greenhouse effect). Phytoplankton constitute the bottom level of aquatic foodwebs. There are many species of phytoplankton that can be distinguished by their morphology.

Under certain conditions algal species can grow abundantly in marine and limnic environments. The phases of accelerated growth, decelerated stagnation and rapid decline of cell counts together establish an algal bloom. Algal blooms are an essential component of biological productivity in aquatic communities and abundance as well as timing in the annual cycle are important factors for a proper functioning of the marine foodweb. Phytoplankton depend upon sunlight, water, and nutrients to grow and survive. Competition for common resources couples different species as do grazing by zooplankton or predators belonging to higher trophic levels. Dominance and succession of species is controlled by habitat specific conditions and susceptible to fluctuating or systematically changing environmental conditions.

Data Resources:

The Helgoland Roads (HR) Data, records of roughly 300 algal species (cell counts/litre), nutrient concentrations (ammonium, nitrate, nitrite, phosphate, silicate), and physical quantities (water temperature, salinity, Secchi depth, wind, etc.), constitute an exceptional data pool not only for its density (workdaily records) but also for its long-term character (1962 until present). The data is collected and hosted by the Biological Institute on Helgoland (BAH) and shared with collaborating researchers via the Publishing Network for Geoscientific & Environmental Data PANGAEA.
    As from 1975, Wulf Greve started a zoolankton time series at Helgoland Roads. This involves bi-weekly sampling, at the same site as the phytoplankton and nutrient samples of the daily time series. Analyses of the zooplankton data are also carried out to species level.
    In addition to these single-site databases with a comparatively high temporal resolution there exists the high spatial-resolution Continuous Plankton Recorder (CPR) database, a synoptic plankton-monitoring programme operated by the Sir Alister Hardy Foundation for Ocean Science (SAHFOS), resident in Plymouth, UK.

A common analyses of both phytoplankton and zooplankton time series and a combination of high temporal (HR) and spatial (CPR) resolution data is one of the central issues of our subproject "Match/mismatch of zooplankton-phytoplankton interactions, based on existing long-term information in the North Sea" as part of the DFG-Priority Programme AQUASHIFT.

Research Questions:

Changes Induced by Climate Variability

An increase of sea surface temperature of 1.13° C over the last 40 years (since 1962) and less frequent extremes in winter have recently been shown (see below figure).
temperature
As bifurcation theory teaches, systematic shifts of environmental conditions like global warming may cause catastrophic ecological responses. One of such scenarios is linked with the match/mismatch hypothesis (Hjort-Cushing) that relates survival of fish to the match between the time of larval occurrence and that of  the production of their food. In fact, more or less pronounced shifts of algal blooms are revealed by the Helogoland Roads time series of selected algae, e.g. for Ceratium lineatum shown in the figure below.
Ceratium lineatum
These observations raise the question whether dramatic changes of the marine ecological system of the North Sea might be expected.


Dominance and Succession Patterns

There is a high inter-annual variability of species specific peak values that can vary over two to three orders of magnitude. The dominance and succession of species within and across years reveals a complex pattern that still awaits explanations in the context of  evironmental conditions, nutrients and competition.
dominance & succession


Trigger Mechanisms of Algal Blooms

From recorded cell counts of monitored species it is possible to extract the onset of  an occuring principal bloom. Pretreatment of raw data series involves interpolation and appropriate filtering - the technical problem is to accomplish smoothing without significantly shifting the bloom start and the setup of an adequate criterion for the bloom onset (a criterion that is generally accepted within the marine biology community is unknown). After pinpointing the bloom start the immediate prehistory is investigated to detect trigger mechanisms: the method used is similar to reverse correlation analysis and "spike-trigered averaging" as in neurophysiological research of spike discharge events and, in analogy, might be termed "bloom-triggered averaging".

In the framework of process-oriented modelling there are several variants that generate algal blooms, e.g. the Truscott-Brindley model that formulates the phytoplankton-zooplankton interaction in the form of an excitable dynamics (again known from models of neuronal activity): overcritical deviations from the fixed-point, for instance by depressing the zooplankton concentration below the stationary value, elicit a phytoplankton bloom that is followed by a zooplankton bloom and a return (relaxation) to the fixed-point (see figure below).

Truscott-Brindley model

In the context of North Sea research and the Helgoland Roads Data it is interesting to extend existing analyses to a dynamics that includes the annual cycle and/or the effect of fluctuating parameters (multiplicative noise).

References:

Additional Links



           
Revised 18.10.06