Biogeochemistry

DOI 10.1007/s10533-007-9090-6

REVIEW PAPER

Current understanding of Phaeocystis ecology and biogeochemistry, and perspectives for future research

Peter G. Verity · Corina P. Brussaard · Jens C. Nejstgaard · Maria A. van Leeuwe · Christiane Lancelot · Linda K. Medlin

Received: 1 June 2006 / Accepted: 13 September 2006© Springer Science+Business Media B.V. 2007

Abstract The phytoplankton genus Phaeocystishas well-documented, spatially and temporallyextensive blooms of gelatinous colonies; these areassociated with release of copious amounts ofdimethyl sulphide (an important climate-coolingaerosol) and alterations of material Xows amongtrophic levels and export from the upper ocean.A potentially salient property of the importance

of Phaeocystis in the marine ecosystem is itsphysiological capability to transform between sol-itary cell and gelatinous colonial life cycle stages, aprocess that changes organism biovolume by 6–9orders of magnitude, and which appears to beactivated or stimulated under certain circum-stances by chemical communication. Both life-cycle stages can exhibit rapid, phased ultradiangrowth. The colony skin apparently confers pro-tection against, or at least reduces losses to,smaller zooplankton grazers and perhaps viruses.There are indications that Phaeocystis utilizeschemistry and/or changes in size as defensesagainst predation, and its ability to create refugesfrom biological attack is known to stabilize preda-tor–prey dynamics in model systems. Thus the lifecycle form in which it occurs, and particularlyassociated interactions with viruses, determineswhether Phaeocystis production Xows through thetraditional “great Wsheries” food chain, the moreregenerative microbial food web, or is exportedfrom the mixed layer of the ocean.

Despite this plethora of information regard-ing the physiological ecology of Phaeocystis,fundamental interactions between life historytraits and system ecology are poorly understood.Research summarized here, and described in thevarious papers in this special issue, derives froma central question: how do physical (light, tem-perature, particle distributions, hydrodynamics),chemical (nutrient resources, infochemistry,

P. G. Verity (&)Skidaway Institute of Oceanography, 10 Ocean Science Circle, Savannah, GA, 31411, USAe-mail: Peter.Verity@skio.usg.edu

C. P. BrussaardRoyal Netherlands Institute for Sea Research, P.O. Box 59, 1790 AB Den Burg, The Netherlands

J. C. NejstgaardUNIFOB AS, Department of Biology, University of Bergen, Bergen High Technology Centre, 5020 Bergen, Norway

M. A. van LeeuweUniversity of Groningen, Biological Centre, P.O. Box 14, 9750 AA Haren, The Netherlands

C. LancelotEcologie des Systèmes Aquatiques, Université Libre de Bruxelles, Campus de la Plaine – C.P. 221, Bd. du Triomphe, 1050 Bruxelles, Belgium

L. K. MedlinAlfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, Germany

123

Biogeochemistry

allelopathy), biological (grazers, viruses, bacteria,other phytoplankton), and self-organizationalmechanisms (stability, indirect eVects) interactwith life-cycle transformations of Phaeocystis tomediate ecosystem patterns of trophic structure,biodiversity, and biogeochemical Xuxes? Ulti-mately the goal is to understand and thus predictwhy Phaeocystis occurs when and where it does,and the bio-feedbacks between this keystonespecies and the multitrophic level ecosystem.

Keywords Biocomplexity · Plankton life cycles · Phaeocystis · Viruses · Zooplankton

Introduction

The assembly of phytoplankton communities is acontinuous process in which instantaneous exter-nal assessments, e.g., species abundances, diver-sity, and bulk constituents, reXect the outcome ofa host of physical, chemical, and biological pro-cesses operating at a hierarchy of temporal andspatial scales. The absolute abundance of a giventaxon increases according to growth, immigration,physical concentration, and other mechanisms,whereas losses including grazing, lysis, sedimenta-tion, dilution, and emigration act to stabilize pop-ulation Xuctuations. In this manner, many speciesgenerally compete yet coexist in a water mass ofseemingly similar properties. That one speciesbecomes numerically dominant, much less formsa near monospeciWc bloom, is a remarkable phe-nomenon, especially given that many loss pro-cesses are density dependent in nature. To do soregularly, over large spatial and long temporalscales, is indicative of a life-history strategy moresuccessful than the competition. It may also indi-cate that one species is more genetically diversethan another, and that diversity enables it to existover these scales as one suite of genotypes givesway to another suite of genotypes to perpetuatethe bloom. From this perspective, each suite ofgenotypes conceivably thrives in a discrete set ofenvironmental parameters.

The genus Phaeocystis is one such taxon.Whereas it contains at least six species (Medlinand Zingone this volume; Rousseau et al. this vol-ume), it is ecologically acknowledged that the

blooms are typically caused by the two cold waterspecies, P. pouchetii (high boreal and arcticwaters) and P. antarctica (Southern Ocean), andby P. globosa (primarily cold north temperatewaters). What is unusual in marine ecosystems isthat such blooms are primarily a life cycle event:these Phaeocystis species transform from a tinysingle-celled morphotype into a colonial stage inwhich hundreds or thousands of cells are embed-ded within a gelatinous matrix and protectedbehind an elastic but solute-permeable commu-nity membrane. Something about this transforma-tion and/or the characteristics of the colony life-cycle stage (Lancelot and Rousseau 1994; Lance-lot et al. 1998) permit Phaeocystis to out-competeother phytoplankton, which belong to a variety ofalgal classes with diverse strategies for competingfor light and nutrients, but without a functionallycomparable gelatinous house. At suYciently highconcentrations, colony blooms have been associ-ated with a variety of ecosystem changes and neg-ative eVects on Wsheries and Wsh farming(Lancelot et al. 1987; Schoemann et al. 2005;Nejstgaard et al. this volume), such that blooms ofPhaeocystis colonies are considered as harmfulalgal blooms (HABs, e.g., Anderson et al. 1998,Veldhuis and Wassmann 2005). Less is knownabout the occurrence and signiWcance of solitarycell stages of Phaeocystis, primarily because thesesmall nanoplankton require speciWc careful meth-ods for quantiWcation. However, blooms of soli-tary cells have been reported on occasion, e.g.,Ratkova and Wassmann (2002) and Wassmannet al. (2005), and thus the temporal and spatialdistribution of Phaeocystis may have been under-sampled. Solitary cells may also be important toover-wintering strategies (see below) but relevantquantitative studies are lacking. Nevertheless,given that colonies derive from solitary cells,knowledge of their dynamics and mechanismsinXuencing their occurrence are important tounderstanding the success of this genus (Reigstadand Wassmann, this volume).

Thus the causes of life-cycle alterations inPhaeocystis may be under environmental con-trols, whether abiotic or biotic, and elucidation ofthese controls will shed light on the fundamentalprocesses that regulate ecosystem organizationand biogeochemical Xows. Research attempting

123

Biogeochemistry

to explain these life-cycle transformations fromthe perspective of resource limitation or physio-logical capability have been underway for almosta century, but have failed to provide conclusiveevidence of why Phaeocystis is so successful insequestering resources and dominating its com-petitors. Emphasis of studies on shifts in life cycledominance in situ has increasingly focused oninXuences by other organisms, either direct causessuch as viral- or grazer-mediated reduction of abun-dance of speciWc stages, or indirect causes such asstage transformation induced by allelopathy or info-chemicals. Because of the large change in eVectiveorganism size when it transforms from solitary cellsto colonies or vice-versa, Phaeocystis can functionin reality as a dual species: a nanoplanktonicsolitary cell, a micro- to macroplanktonic colony,or both concurrently. But whereas we understandsome of the evolutionary costs and beneWts ofthese two stages (see Verity and Medlin 2003;Veldhuis et al. 2005 for discussions), and some ofthe factors inXuencing biomass accumulation ofthese stages, we do not yet understand how theyinteract to make the keystone phytoplankton thatis Phaeocystis. If we could, the expected insightsinto planktonic ecosystem structure and functionmight be remarkable.

This was the rationale and motivation for theformation of the scientiWc committee on oceanresearch (SCOR) working group #120 (“MarinePhytoplankton and Global Climate Regulation: ThePhaeocystis spp. Cluster as a Model”). The work-ing group’s mission during the period 2000–2005was to deWne research needs on Phaeocystis, andto provide a logistical mechanism to integrateresults into a revised conceptual model of thistaxon’s role in the transfer of elements relevant toclimate regulation, such as carbon and sulfur, andthe biogeochemical processes that inXuence sucha role. These goals were a natural developmentfrom previous symposia on Phaeocystis held inBelgium (1991) and the Netherlands (1999), andresulting in special publications in the Journal ofMarine Systems (1994, Volume 5) and Journal ofSea Research (2000, Volume 43), respectively. Assuch, the Weld of interest was broad, encompass-ing genes to climate, with a goal of generatingsuYcient information to place Phaeocystis notonly in the context of biodiversity but also in the

framework of global international programs suchas the surface ocean lower atmosphere study(SOLAS: http://www.uea.ac.uk/env/solas) andintegrated marine biogeochemistry and ecosys-tem research (IMBER: http://www.igbp.kva.se/obe). The ultimate objective was to improve pre-dictability in the essentially chaotic planktonworld. The following section summarizes progressregarding key aspects of the Wtness, ecology, andbiogeochemistry of Phaeocystis that occurredduring the working group’s organizational life-time, and which are described in more detail inpapers in this special issue of Biogeochemistry.

New perspectives on Phaeocystis and its role in upper ocean biogeochemistry

Phaeocystis-dominated ecosystems include largeand biogeochemically important portions of theglobal ocean, which are trophically unusual andpoorly understood. Life-cycle alternations causedby system stresses on Phaeocystis, whether abioticor biotic, restructure the entire food web andchange ecosystem characteristics ranging fromnutrient utilization to trophic energy transfer tovertical export of biogeochemically salientelements. Evidence supports the notion thatPhaeocystis is so successful because, through acombination of life cycle stages, a range of colonysizes, and perhaps grazer-activated defenses, itcan reduce mortality more than competing phyto-plankton. Interactions between Phaeocystis andother organisms, especially zooplankton, mayinvolve sophisticated signaling, and attempts toaccurately model and predict the complexity ofthese ecosystems requires quantitative informa-tion on the feedbacks between Phaeocystis and itsenvironment. For these reasons, among many,Phaeocystis is a model organism from which tobegin the study of complexity in marine pelagicecosystems.

Taxonomy

The genus Phaeocystis is thought to have origi-nated 75 million years ago. There are currentlysix recognized species: P. antarctica Karsten,P. cordata Zingone et Chretiennot-Dinet, P. globosa

123

Biogeochemistry

ScherVel, P. jahnii Zingone, P. pouchetii (Hariot)Lagerheim, and P. scrobiculata Moestrup (Medlinet al. 1994; Zingone et al. 1999; Medlin andZingone, this volume; Rousseau et al. this volume),and several as yet undescribed species arethought to exist. However, one or more of thesemay represent species complexes. For example,the existence of diVerent internal transcribedspacer (ITS) alleles in a single strain of Phaeocystisglobosa was interpreted as evidence of hybridiza-tion between cryptic species within the P. globosamorphotype (Lange et al. 2002). Normally ITSregions should be homogenized by concertedevolution but this was not the case in P. globosa.In hybrids of land-plant species, both parentalcopies are present in the oVspring and a contin-ued crossing of the F1 generation will eventuallylead to the loss of one of the parental types.These regions were not evolving too rapidly forhomogenization to take place based on relativerates tests made on the 18S ribosomal deoxyribo-nucleic acid (rDNA; Lange et al. 1996). Crypticspecies within Phaeocystis globosa had beeninferred earlier in this species based on diVer-ences in DNA content determined by Xow cytom-etry among strains of the species (Vaulot et al.1994). Thus there seems to be substantial inter-and intraspeciWc variation in this genus.

Gaebler et al. (this volume) attempted to tracethe biogeographical history of Phaeocystis strainsin the Southern Ocean by quantifying the geneticdiversity within populations of P. antarctica fromeach of the three major gyres in the Antarcticcontinental waters and calculating the gene Xowbetween them. They described methods to quan-tify genetic diversity, and preliminary results forP. antarctica in comparison to two other colonialspecies: P. globosa and P. pouchetii. Resultsshowed diVerences in the AFLP patterns betweenisolates of P. antarctica from diVerent regions,and that a wide variety of microsatellite motifscould be obtained from the three Phaeocystis spe-cies. Initial results (Lange et al. 2002) support theITS analysis that the populations from the Ant-arctic circumpolar current (ACC) are seeding thecontinental gyres.

Lange et al. (2002) used sequence variation tocompare 22 isolates representing a global distri-bution of the genus. The most conserved 18S

rRNA analysis suggested that an undescribedunicellular Phaeocystis sp. (isolate PLY559) is asister taxon to the Mediterranean unicellularPhaeocystis jahnii; this clade branched prior tothe divergence of all other Phaeocystis species,including the colonial ones. Phaeocystis globosaand pouchetii have multiple diVerent ITS copies,suggestive of cryptic species that are still able tohybridize. A molecular clock was constructed thatestimates the divergence of the cold-water formsfrom the warm-water forms to be about 30 Myaand the divergence of P. antarctica and P. pouchetiito be about 15 Mya. Medlin and Zingone (thisvolume) have reviewed the morphologicalinformation on the validly described Phaeocystisspecies and presented some new information onundescribed species. The combined molecularand morphological data implied that the numberof species in the genus is still underestimated, andthat cryptic or pseudo-cryptic diversity requires asound assessment in future research on this genus.

Life cycles

The complex life cycle of Phaeocystis is the key tounderstanding its role in marine ecosystems(Verity et al. 1988b; Rousseau et al. 1994; Peperzaket al. 2000). However, despite almost a century ofstudy, the number and role of the diVerent celltypes involved in the Phaeocystis life cycle is stilluncertain. Rousseau et al. (this volume) provideda detailed review of the diVerent morphotypes ofthe six recognized Phaeocystis species, and foundfour diVerent cell types. There are two types ofscaly Xagellates: one produces star-forming Wla-ments and has been reported for all speciesexcept P. jahnii, while the other, lacking Wlamentsand stars, has been found in P. globosa and P. jahnii.The three colony-forming species, P. globosa,P. pouchetii, and P. antarctica, share three celltypes in common: a Xagellate with scales and Wla-ments, a colonial cell, and a Xagellate devoid ofscales and Wlaments. In contrast, only Xagellateswith scales and Wlaments have been observed inthe non-colony-forming species, P. scrobiculataand P. cordata. The authors provided evidencefor a haploid–diploid life cycle in P. globosa thatincludes sexuality during colony bloom forma-tion, and two types of vegetative reproduction.

123

Biogeochemistry

The high genetic diversity recovered in the pre-liminary genetic analysis done by Gaebler et al.(this volume) would suggest that sexual reproduc-tion is quite common.

DiVerent hypotheses have been proposed toexplain the selective pressures that led to the evo-lution of colony formation in Phaeocystis, includ-ing as a strategy to reduce population mortalityby grazing or viral lysis (Verity and Medlin 2003;Brussaard et al. this volume; Nejstgaard et al. thisvolume). If cells in colonies have lower per capitalosses than freely dispersed cells, and also haveequivalent or even higher intrinsic growth rates(Veldhuis et al. 2005), then colonies wouldappear to provide a distinct life-cycle advantage.Nevertheless, there remain numerous questionsregarding the ecological roles of the diVerent life-cycle stages, before their signiWcance can becompletely understood (Rousseau et al. thisvolume). However, there is universal agreementthat these life-cycle transitions may have largedirect and indirect eVects on trophic dynamicsand biogeochemistry of planktonic ecosystems. Incoccolithophorids, another member of the Pry-mnesiophyta, life-cycle changes are likely relatedto niche partitioning (Noel et al. 2004). The hap-loid holococcolith stage thrives in oligotrophic,deeper oceanic waters, whereas the diploid het-erococcolith stage lives in more-eutrophic coastalwater. To evaluate the hypothesis of the eVect oflife cycle transitions on trophic dynamics, a betterunderstanding of the occurrence, transitions, andcausal mechanisms of the life cycle is needed.

Yet another life-cycle issue is how Phaeocystissurvives the long polar winter, i.e., several monthsof darkness. It would be advantageous in such anenvironment for photosynthetic solitary cells tobe mixotrophic (capable of also ingesting bacteriaor other small particles), or for mixotrophy to beinduced by decreasing photoperiod or some othersignal of the impending months of darkness.Phagotrophy of bacteria and phytoplankton byother prymnesiophytes related to Phaeocystis hasbeen documented (Nygaard and Tobiesen 1993;Tillmann 1998), but to the authors’ knowledge itis unknown whether Phaeocystis has such capa-bility. It is known that with the induction of lowtemperature and low light, the expression ofphotosynthetic genes is increased in sea-diatoms

(Mock and Valentin 2004), but whether or notthis is a universal response of other polar phyto-plankton has not been investigated. Interestingly,Phaeocystis may serve as a source of photosyn-thetic potential for its own predators. A newundescribed dinoXagellate, related to Karenia andKarlodinium, has been found in the Antarcticwith kleptoplastids from Phaeocystis (Gast et al.2006). Cultures of this heterotrophic dinoXagel-late are maintained on single cells of Phaeocystis,from which it extracts its plastids by a yet asunknown mechanism to become photosynthetic.The dinoXagellate has also been seen insidePhaeocystis colonies, but it is unknown if thedinoXagellate prefers colonial or unicellularstages as its plastid source. The dinoXagellate isquite abundant in Antarctic waters, and this rela-tionship represents another unusual relationship inunderstanding the biological role of Phaeocystisand its predators.

To provide a framework for the details of cellbiology, genetics, and chemical signaling as con-trols on P. pouchetii life-cycle transitions, Whip-ple et al. (2005a) constructed a 15-compartmentconceptual life cycle that includes solitary celltypes and size classes of growing and senescentcolonies. Whipple et al. (2005b; this volume)described how this conceptual approach was usedto direct integrated research into various facets ofP. pouchetii ecology: appearance of colonies fromfjord sediments; in vitro colony division; in vitrochanges in colony shape of P. pouchetii associatedwith senescence; determining which P. pouchetiilife stage is vulnerable to viral infection and lysis;colony division rate and growth rates; enhance-ment by diatoms frustules of P. pouchetii colonyformation from solitary cells; and investigation ofmesozooplankton suppression and microzoo-plankton enhancement of P. globosa colony for-mation by grazer infochemicals.

Although life-cycle alternations may be the keyto unraveling the biofeedbacks that occurbetween Phaeocystis and the ecosystem, manyquestions remain unanswered. For example, het-eromorphic life cycles such as those exhibited byPhaeocystis function in macroalgae as adaptationsto reduce grazing pressure (Lubchenco and Cubit1980; Slocum 1980). Do life-cycle transformationssuch as development of motility and emigration

123

Biogeochemistry

of cells from colonies (Verity et al. 1988b) happenrandomly, are they driven only by physical (light,temperature) or chemical (nutrient) stimuli, orare they also under biological control such asgrazing (Tang 2003), or perhaps intra- or interspe-cies signaling? In any case, these transformationsare best viewed as responses to stressors [grazing,viruses, nutrient and light limitation, prolongeddarkness (polar winter)]. By changing its life-cycle morphology, Phaeocystis changes the sizestructure of the ecosystem and also essentiallydisperses oVspring to another future where thestressor may be avoided or reduced.

Environmental regulation

The distribution patterns of Phaeocystis speciescan partly be explained by its speciWc light andnutrient requirements, which are thoroughlyreviewed in Schoemann et al. (2005). The genusoccurs under a wide range of light intensities. Intemperate regions, the ability of Phaeocystis togrow at low light levels allows bloom develop-ment early in the season (Peperzak et al. 1998).Massive colony blooms in Antarctic watersearly in spring are only terminated once nutri-ents become depleted. In those areas, high bio-mass of Phaeocystis cells may also be foundunder the sea ice, where light intensities areoften <5 �mol quanta m¡2 s¡1 (van Hilst andSmith 2002).

The concentrations of the macronutrients PO4and NO3 largely control biomass development ofPhaeocystis in temperate areas (Lancelot et al.1987; Verity et al. 1988a; Veldhuis et al. 1991).Early studies demonstrated that this reXected theability of Phaeocystis to both utilize organicphosphate (Admiraal and Veldhuis 1987) and becompetitive at high nitrate levels (Lancelot et al.1998). Gieskes et al. (this volume) studied theextensive data set on Phaeocystis that is providedby the continuous plankton recorder (CPR), asampling device towed behind ships-of opportu-nity throughout the North Atlantic since 1948(Reid et al. 2003). Phaeocystis abundance hasmainly been restricted to neritic regions and lim-ited to springtime, with the southeastern NorthSea as a hotspot. Abundance and length of seasonaloccurrence of Phaeocystis in the North Sea was not

correlated with the increasing eutrophicationduring the period of data collection. This isapparently because Phaeocystis is not only inXu-enced by anthropogenic nutrient enrichment inriver discharge: nutrients in the Channel/Atlanticwater entering through the Dover Straits may beequally or more important, although this region isalso inXuenced by discharge from the Seine andSomme rivers. Only very close to the Dutch coastwas phytoplankton biomass related signiWcantlyto Rhine river run-oV in spring; elsewhere in theNorth Sea, Atlantic water incursion overruledanthropogenic nutrient enrichment. As discussedlater, the relative importance of Phaeocystisblooms compared to diatoms in the Belgiancoastal zone is related to the combined eVects ofriverine nutrient load and the North Atlanticoscillation (NAO: Breton et al. 2006).

In the Southern Ocean, algal growth is partlycontrolled by iron when this micronutrient isavailable in limiting concentrations, but variable-light climate also plays an important role. Thusa complex interaction between iron and lightlimitation has been described (Stefels and vanLeeuwe 1998; van Leeuwe and Stefels 1998; Arrigoand Tagliabue 2005). Recently, experiments weredesigned to study photoacclimation in P. antarcticaunder more natural light conditions (van Leeuweand Stefels, this issue). Cells were exposed to avariable light regime that mimicked vertical mix-ing of the water column. Iron limitation wasinvoked by adding ethylenediaminetetraacetic acid(EDTA) to natural seawater. Periodic exposureto high irradiance induced rapid conversion of thelight-harvesting pigment, diadinoxanthin, into thephotoprotective pigment, diatoxanthin. Underiron limitation, cells showed characteristics ofhigh light acclimation. An expansion of thephotoprotective pigment pool coincided withweaker diurnal patterns of pigment cycling. Theseresults supported prior conclusions that photoacc-limation can be successful under conditions ofiron limitation.

DiTullio et al. (this volume) studied the pig-ment composition of freshly isolated strains ofP. antarctica. The experiments were performedin EDTA-free media thus utilizing the naturalorganic Fe binding ligands present in theseawater, and applying ecologically relevant Fe

123

Biogeochemistry

concentrations (i.e., subnanomolar). Ratios of thepigments 19�-hexanoyloxyfucoxanthin: chloro-phyll a increased while ratios of fucoxanthin:chlorophyll a decreased under increasing Fe-lim-ited conditions. Thus DiTullio and colleaguesconWrmed the Wndings by van Leeuwe and Stefels.The ratios retrieved in the laboratory experi-ments mimic those typically measured inP. antarctica spring blooms in the Ross Sea, whichstresses the importance of Fe limitation in thisregion.

At low light intensities, the iron demand ofcells increases in relation with enlargement of thephotosynthetic apparatus (Raven 1990; van Leeuweand de Baar 2000). Sedwick et al. (this volume)studied the iron requirements of P. antarctica inthe laboratory, and recorded a relatively highiron requirement (a half-saturation constant of0.45 nM dissolved iron) at low light levels (at ca.20 �mol photons m¡2 s¡1). Their results showedthat dissolved iron availability plays a primaryrole in regulating blooms of colonial P. antarcticain the southern Ross Sea during summer. Sed-wick and colleagues also performed Weld studiesin the Ross Sea, which showed that in late springiron was still in ample supply and Phaeocystisgrowth was controlled by the ambient light con-ditions rather than by iron availability. Itappears that in situ cellular iron demand was<0.45 nM Fe. The authors related this low ironrequirement to the relatively high ambient lightlevels.

Annual patterns of P. antarctica in theSouthern Ocean may thus be determined by theinterplay between variable micronutrient concen-trations and irradiance availability. In contrast,patterns in high latitudes in the northernhemisphere, where iron is not typically limiting,are primarily determined by nutrient conditionsand large-scale hydrographical phenomena.

Characterization of organic matter and biodegradability

The gross biochemical composition of PhaeocystisdiVers to a certain extent from other phytoplankton.For example, colonies appear to have a speciWccarbohydrate signature, including a ratherunusual, or even unique, nuclear magnetic reso-

nance (NMR) spectra of the mucus (van Boekel1992; Janse et al. 1996). A combination of sterolsand fatty acids can be used as a biomarker ofPhaeocystis, and a set of fatty acids, sterols, andpigments is diagnostic for the class of algae towhich Phaeocystis belongs (Prymnesiophyceae),although individually they are widespread amongphytoplankton. However, the fatty acids 20:5w3(eicosapentaenoic acid, EPA) and 22:6w3(docosahexaenoic acid, DHA), which are essentialfor the growth of multicellular animals, occur inonly trace amounts in Phaeocystis, whereas theyare common in other phytoplankton (Nichols et al.1991). Moreover, based upon the fatty acid and lipidcontents, solitary cells are more nutritious thancolonies.

Phaeocystis is well known for producingextravagant amounts of particulate and dissolvedorganic carbon, especially during the colonialstage of its life cycle. Alderkamp et al. (2005; thisvolume) reviewed the characteristics and dynam-ics of the organic matter compartments producedby Phaeocystis. When nutrients become limitingand Phaeocystis blooms reach a stationary growthphase, excess energy is stored as carbohydratesthat leads to an increase in C/N and C/P ratios ofPhaeocystis organic material. At the end of abloom, deterioration of colonies, autolysis ofcells, grazing of colonies and cells, and lysis ofcells that are virally infected release dissolvedorganic matter (DOM) rich in glucan and muco-polysaccharides. Whereas this organic matter ispotentially readily degradable by heterotrophicbacteria (Brussaard et al. 2005b; Ruardij et al.2005), the high C/N and C/P ratios of Phaeocystisorganic matter may lead to nutrient limitation ofmicrobial degradation. Furthermore, carbohy-drate hydrogel formation may physically retardthe degradation potential of the Phaeocystisorganic material, and also accelerate its sedimen-tation. The production of dissolved organic mat-ter and subsequent formation of transparentexopolymers (TEP) may inXuence microbial pop-ulation dynamics directly through bacterial colo-nization and indirectly through scavenging ofpredators and viruses (Brussaard et al. 2005b).

The relationship between Phaeocystis photo-synthesis, DOC release, and microbial dynamicsis thus complex and likely mediated by nutrient

123

Biogeochemistry

availability (Thingstad and Billen 1994). Forexample, bacterial degradation of primary pro-duction in high-nutrient low-chlorophyll regionscan be limited by iron availability (Arrieta et al.2004). Iron supply can either directly stimulatemicrobial activity, or do so indirectly by enhanc-ing DOC production of phytoplankton upon theirrelease from iron limitation. Becquevort et al.(this volume) investigated indirect eVects of Fesupply on bacterial degradation by measuringbacterial response to organic substrates derivedfrom P. antarctica cultures set up in high and lowFe-amended treatments using Antarctic seawater.Iron addition inXuenced the relative abundanceof colonies versus single cells of Phaeocystis andcaused a decrease in the C/N ratio of Phaeocystismaterial. In turn this altered microbial commu-nity composition and cellular activity levels, suchthat more Phaeocystis-derived organic matter wasdegraded in high iron compared to low-iron bio-assays. The authors concluded that by increasingbacterial degradation and thereby preventing theaccumulation of dissolved organic carbon, thepositive eVect of Fe supply on the carbon biologi-cal pump via enhanced phytoplankton exportmay partly be counteracted.

Another aspect of Phaeocystis colonies that isunder-appreciated is that they provide sites fortemporary settlement of other plankton. Forexample, Sazhin et al. (this volume) enumeratedthe consortia of organisms associated with colo-nies during blooms of P. globosa in the easternEnglish Channel and P. pouchetii in mesocosms inwestern Norway. In both environments, massdevelopment of the small needle-shaped diatom,Pseudonitzschia species, occurred on Phaeocystiscolonies, and comprised up to 70% as much car-bon as contained in the Phaeocystis cells insidethe colonies at the end of the bloom. The authorsproposed that the diatoms use the organic sub-stance of the colonial matrix for growth.

Grazing

Despite almost a century of qualitative observa-tions and two decades of quantitative studies,there remains considerable uncertainty aboutthe extent to which Phaeocystis, especially colo-nies, provide a good food for, and are ingested

signiWcantly by, zooplankton (reviews by Weisseet al. 1994; Schoemann et al. 2005; Nejstgaardet al. this volume). Reported feeding rates arevery variable and span the full range from non-detectable to close to physiological maxima forthe respective grazer and prey size. A variety ofWltering or suspension-feeding organismsdecrease or stop their feeding when Phaeocystiscolonies are present, including zooplankton andvarious bivalves (Weisse et al. 1994; Smaal andTwisk 1997; Nejstgaard et al. this volume).Active rejection of colonies has also beenobserved (Kamermans 1994) and mechanical dis-turbance of the gills of bivalves and the mouth-parts of zooplankton caused by colony mucus hasbeen hypothesized. Widespread speculationattends the notion that Phaeocystis may beunpalatable, the wrong size, noxious, toxic, orcombinations of these (see Weisse et al. 1994 forearlier review), at least during some bloomphases when investigated in situ (Estep et al.1990). Colony blooms themselves are oVered asevidence of reduced mortality for whatever rea-son; in contrast, regions and seasons with regularblooms are also ecosystems with great Wsheries.Hamm et al. (1999) argued that the colony mem-brane may be largely responsible for minimizinga variety of loss processes including grazing,viruses, cell lysis, and sedimentation. In thisregard, intact small colonies of P. cf. pouchetiiwith living cells have been observed in copepodfecal pellets from the Barents Sea (P.G. Verity,E. Arashkevitch, P. Wassmann, unpubl.). Seurontet al. (this volume) oVered another mechanismthat protects Phaeocystis against grazing. Theyinvestigated the distribution patterns of chloro-phyll a and seawater excess viscosity over thecourse of a phytoplankton spring bloom domi-nated by Phaeocystis globosa. Positive and nega-tive correlations were measured during thebloom and after the formation of foam in theturbulent zone. The authors suggested that thebiologically induced increase in seawater viscos-ity might be a competitive advantage to Phaeo-cystis as a potential anti-predator strategy. Inaddition, through this unique environmentalengineering strategy, P. globosa can create afavorable, turbulent-free physical habitat thatprotects colony integrity.

123

Biogeochemistry

In marine systems, both proto- and metazoanzooplankton use species-speciWc contact chemo-reception and diVusable chemical cues in analyzingparticles they encounter (Friedman and Strickler1975; Spero 1985; Verity 1988). If grazing isselective and varies in response to prey ratio,abundance, or palatability, then investmentin mortality reduction by prey is rewarded.Considerable evidence illustrates that many proto-and metazoan zooplankton are indeed selectivefeeders (e.g., Alcaraz et al. 1980; Verity 1991;Sanders and Wickham 1993). Hence, theory wouldsuggest that phytoplankton such as Phaeocystiswould evolve defenses against predation. Butrelatively little is known about production ofsecondary chemicals or metabolites that functionas feeding deterrents in Phaeocystis as well as forother phytoplankton (see reviews by Pohnert2004; Yoshida et al. 2004). Phaeocystis has beenshown to have toxic eVects on several organismsand cell types (see Nejstgaard et al. this volumefor a review), and dimethylsulphoniopropionate(DMSP) and its cleavage products have been sug-gested to function as grazing deterrents in theclosely related Emiliania huxleyi (e.g., Wolfe2000; Strom et al. 2003). Further, lectins areknown to function as infochemicals signaling thepresence of certain prey to predators so that thepredators can recognize and select their prey item(Wooten and Roberts 2006). Phaeocystis isknown to possess lectins in the wall of its colonialstage (van Rijssel et al. 1997; Hamm et al. 1999),but perhaps these lectins function in the oppositeway by discouraging the predator rather than sig-naling its presence to the predator.

Phytoplankton vary widely in their physiologyand internal chemistry as a function of growthconditions, and may change their physiology con-siderably and irreversibly when isolated fromnature into culture conditions (Edvardsen andPaasche 1998). Likewise it is also known thatmany zooplankton may show strong variations infeeding rates in response to chemical and physio-logical cues (e.g., Dutz et al. 2005; Koski et al.2005; Dutz and Koski 2006; Long and Hay 2006).Nejstgaard et al. (this volume) critically reviewedthe evidence for and against Phaeocystis as anutritious ration for herbivores and omnivores.Based on available data, they concluded that much

of the reported variation in feeding on Phaeocystisespecially between Weld and laboratory studiesmay (a) be due to methodological problems(including use of nonspeciWc bulk pigments inWeld investigations, and zooplankton not havingalternative prey available in laboratory experi-ments), and (b) reXect physiological diVerencesbetween Phaeocystis species, strains, and celltypes within a strain, and/or induced changes inalgae during culture. This study showed that thismay have resulted in an overestimation of feedingrates in laboratory studies of as much as Wvefoldcompared to Weld estimates based on speciWc cellcounts. They further posited that colony size pro-vides a (partial) refuge from predation by smallcopepods (such as Temora spp. and Acartia spp.)at least for P. globosa, while colonies of P. pouchetiiappear less protected from large copepods(such as Calanus), and that chemical signalingbetween the predator and its prey may be oneof the major factors controlling grazing onPhaeocystis. From presently available data, it isunclear whether colonies of P. pouchetii are lessprotected from copepod grazing compared tosimilarly sized colonies of P. globosa due to diVer-ences in the colony physiology. This should betested in future experiments. For smaller colonies,size may also be less important for those proto-zooplankton whose feeding mechanisms permitengulfment of larger prey, e.g., heterotrophicdinoXagellates (Stelfox-Widdicombe et al. 2004).

Allelopathy and toxicity

The eVects of anti-metabolites on phytoplanktonwere recognized over 40 years ago (Johnston1963). Marine plankton themselves also produceallelochemicals that have antagonistic eVects.Gauthier and Aubert (1981) presented severalexamples within and between bacteria, phyto-plankton, and zooplankton. Pratt (1966) reportedallelopathy between the diatom Skeletonemacostatum and the Xagellate Olisthodiscus luteus.Where investigated, release of allelochemicalsby marine microalgae has been observed(Uchida 1977; Chan et al. 1980; Van Alstyne1986; Targett and Ward 1991), although theidentity of the allelochemicals was seldomknown. Arguments have also been presented in

123

Biogeochemistry

support of contact inhibition in microalgae(Costas et al. 1993).

Phaeocystis has long been suspected to nega-tively impact co-occurring organisms, includingallelochemical and toxic eVects, and recently ithas been identiWed as a harmful algal bloom(HAB) species, at least in European waters(Veldhuis and Wassmann 2005). Van Rijssel et al.(this volume) used a bioassay to quantify thepotential impact of living P. pouchetii, collectedduring mesocosm blooms in western Norway, onother living cells. Red blood cells were used asmodel targets representing unprotected cells, andtheir lysis were quantiWed by monitoring dis-solved hemoglobin. Hemolytic activity was bestcorrelated with P. pouchetii abundance comparedto other phytoplankton, including diatoms, andwas enhanced by increased temperature and light.In comparison to other harmful algae, livingP. pouchetii colonies were highly hemolytic andthe mechanism was apparently quite diVerentfrom the hemolytic harmful algal bloom speciesstudied so far. Whereas it can be argued that redblood cells may not mimic the responses ofmarine phytoplankton, the lytic response of thechemicals to the membranes is likely the same.Long et al. (subm.) reached similar conclusionsthat Phaeocystis possesses allelopathic propertiesfrom experiments with both algal cultures andnatural plankton communities during Phaeocystisblooms in mesocosms. Results such as these sup-port the notion that Phaeocystis may use plantchemistry in its competitive interactions withother phytoplankton, as well as a means of adjust-ing to grazing pressure, as discussed earlier.These antagonistic eVects may help account forthe prodigious, near-monospeciWc colony bloomsregularly observed in high latitudes. It seemsincreasingly likely that “allelochemistry ... mayhelp explain complexity in biological systems”(Lewis 1986).

Viruses

Viruses are now universally recognized as poten-tially being a major contributor to phytoplanktonand bacterial mortality. Given a life cycle thatincludes solitary cell and gelatinous colony stages,and their role in food web dynamics and biogeo-

chemical Xuxes, viral infection of Phaeocystis mayaVect not only host population dynamics, but alsoecosystem structure and function. Viral infectionand lysis have been described in P. pouchetii andP. globosa (Bratbak et al. 1998a, b; Jacobsen2002; Brussaard et al. 2005a). Among otherissues, determining which stages of the life cycleare vulnerable to viral infection and lysis is crit-ical for understanding the eVects of viruses onPhaeocystis life-cycle dynamics and ecology.Brussaard et al. (this volume) reviewed the role ofviruses with respect to Phaeocystis bloom dynamics,and oVered a discussion on potential factors aVect-ing viral control. The colony morphotype providesprotection against infection (Brussard et al.2005a; Jacobsen et al. 2005, subm.), with theimportance of viral infection being inverselyrelated to colony size (Ruardij et al. 2005). Degra-dation of the colonial matrix and the resultingformation of transparent exopolymeric particles(TEP) also indirectly aVect the process of viralinfection by acting to scavenge viruses from thewater column (Brussaard et al. 2005b). Itbecomes more and more obvious that viral lysis ofPhaeocystis may be signiWcant to biogeochemicalprocessing in the water column by aVecting nutri-ent remineralization from Phaeocystis biomass,stimulating bacterial productivity, and increasingrelease of climatically active organic sulfur com-pounds.

Sedimentation and vertical export

Aggregation and accompanying sedimentation isa physical process that can provide a signiWcantfate for Phaeocystis blooms in situ, in addition totrophic losses caused by grazing and viral lysis.Understanding of vertical export of organic mate-rial or cells comprising Phaeocystis blooms thushelps to constrain water column biogeochemistryand reXects physical water column and pelagicfood web structures. For example the Xuxesthrough the water column not only depend ongrazing pressure per se but also the (meso-)zoo-plankton present. TEP formation and coagulationassociated with colony degradation may favorexport. The importance of sedimentation to thefate of blooms, and the signiWcance of verticalexport of POC associated with Phaeocystis

123

Biogeochemistry

blooms is hard to deWne for all species and cases,mainly due to the extraordinary plasticity in theecology and life cycle of the species (Wassmann1994; Schoemann et al. 2005).

Reigstad and Wassmann (this volume) havemade an attempt to quantify the various exportmechanisms and determined cell and colonysedimentation rates and their contribution to thevertical POC directly. Their compilation of data,ranging from polar to sub-Arctic and borealregions, suggests that the overall contribution ofPhaeocystis species to vertical export of biogenicmatter is not signiWcant. Colony blooms are oftenassociated with strong vertical export in the upperphotic zone, but between 40 m and 100 m, thecontribution of Phaeocystis cells to total exportdeclines to only »3% at 100 m depth. By compar-ison, vertical export of diatoms is signiWcantlylarger. Other processes that favor export ofPhaeocystis-derived carbon, such as grazing andfecal pellet production, downwelling, and sinkingmucilagenous matter, are diYcult to quantify, buteven considering colony mucilage, the contribu-tion of Phaeocystis apparently increases only slightlyto <5% of POC export.

Not withstanding these results, episodic sedi-mentation events may still be of local importance(e.g., DiTullio et al. 2000). Such events are hard tomonitor, e.g., they can be associated with grazingby highly mobile krill (Hamm et al. 2001), andother diYculties remain. Reigstad and Wassmann(this volume) note how diYcult it is to quantify thecarbon content of Phaeocystis colonies and alsorefer to the problems caused by fragmentation ofcolonies. Given the known links between Phaeo-cystis physiology (Brussaard et al. 2005b; Mariet al. 2005; Alderkamp et al., this volume), it seemslikely that bloom export is coupled to both nutrientregime and zooplankton dynamics. While N limita-tion would favor retention of fragmented colonymaterial in upper euphotic zone (Mari et al. 2005),P limitation may induce rapidly colony disruptionresulting in heavy and sticky mucus aggregates(thought to sink rapidly). Again, it is diYcult toquantify carbon export along this pathway, sincethis pool is not visible as pigments or cells usingmicroscopy, and perhaps not even picked up accu-rately by traditional POC measurements (Reigstadand Wassmann, this volume).

Additional information on sedimentation ratescan be derived from tracer studies. Schmidt et al.(this volume) and Lalande (2006) show, however,that interpretation of tracers is not straightfor-ward in the presence of Phaeocystis. Particulatethorium-234 (234Th) removal from the water col-umn and presence in sediment traps can be usedas indication of vertical export of biogenic matter(Buesseler et al. 2006). However, the occurrenceof Phaeocystis can temporarily delay the export ofparticulate 234Th from surface waters, e.g., sedi-ment trap POC Xuxes were much lower thanlarge-volume POC Xuxes in such waters (Lalande2006). Schmidt et al. (this volume) suggests thatlarge Xocculation negatively aVects the scaveng-ing of 234Th onto biological particles. When largecolonies make up a signiWcant fraction of the ver-tical Xux, 234Th clearly underestimated particleXuxes.

DMS(P) and the sulfur cycle

Spring blooms dominated by Phaeocystis havebeen associated with considerable production ofdimethyl sulphide (DMS). It is thought that theatmospheric oxidation products of DMS result incooling of the Earth’s atmosphere and therebyreduce the eVects of greenhouse gases such asCO2. Given the global importance of DMS for thebiogeochemical sulfur cycle and climate, theobserved correlations between Phaeocystisblooms and DMS in the atmosphere are signiW-cant (Turner et al. 1995; Kwint and Kramer 1996).The main direct source for DMS is dimethylsul-phoniopropionate (DMSP), which is produced byseveral phytoplankton including Emilianiahuxleyi and Phaeocystis spp. (Keller et al. 1989;Liss et al. 1994). DMSP has properties in commonwith the compatible solute glycine betaine, andproduction of DMSP and DMS may be related tophysiological stress (Stefels 2000; Sunda et al.2002). Phaeocystis has an extracellularly located,membrane-bound DMSP-lyase enzyme thatcleaves DMSP into equimolar amounts of DMSand acrylate (Stefels and Dijkhuizen 1996). It haslong been speculated that the production of acry-late may function as a microbial inhibitor(Sieburth 1960). But ambient concentrations ofacrylate during Phaeocystis blooms are inade-

123

Biogeochemistry

quate to expect bacterial inhibition. However,acrylate concentrations inside the colonial matrixare much higher (Noordkamp et al. 2000), provid-ing the potential to function as a grazing deter-rent. Wolfe et al. (1997) reported that strains ofthe related genus Emiliania, which contained lowlevels of DMSP-lyase were preferred by grazersover strains with high DMSP-lyase levels. Themechanism apparently was lyase-induced releaseof high concentrations of acrylate in grazer foodvacuoles. Given the very high concentrations ofacrylate in the colony (Noordkamp et al. 1998)and also high DMSP-lyase activity in cells (Stefelsand van Boekel 1993), DMS production and graz-ing-activated DMSP cleavage by Phaeocystis isthought to contribute to the avoidance of healthycolonies by protozoan and metazoan zooplankton(Wolfe et al. 2000). It has also recently beenshown that copepods react strongly to DMS in thefeeding current (Steinke et al. 2006), although it isnot clear whether this is a cue to trigger ingestionof the Phaeocystis cell or its micropredator by thecopepod (see discussion in Nejstgaard et al. thisvolume).

Stefels et al. (this volume) reviewed the func-tional relationships between physical and chemi-cal ecosystem parameters, biological productivity,and the production and consumption of DMSPand DMS. They made speciWc attempts to quan-tify these functional relationships, and describedthe various factors controlling DMSP production(species composition, abiotic factors), the conver-sion of DMSP to DMS (by algal and bacterialenzymes), the fate of DMSP-sulfur (e.g., grazing,microbial consumption, sedimentation), and fac-tors controlling DMS removal from the water col-umn (microbial consumption, photo-oxidation,emission to the atmosphere). The authors con-cluded with speciWc recommendations for futurestudies needed to improve accuracy of model pre-dictions linking ecosystem structure and functionto DMS dynamics. One of these recommenda-tions is the subdivision of the algal realm into sixgroups, one of which (PhytoXagellate B) is basedon the features of Phaeocystis. Although it cannotbe excluded that a few other prymnesiophyteswith high DMSP-lyase activity fall within thissame category, Phaeocystis is thought to be themain contributor. This outstanding position is

indicative of the global importance of this singlegenus. In this regard, DiTullio et al. (this volume)described rapid sedimentation of DMS andDMSP at the end of spring blooms of P. antarcticain the Ross Sea. Shenoy and Kumar (this volume)provided a detailed overview of spatial and tem-poral patterns of organic sulfur compounds in theIndian Ocean, where Phaeocystis globosa bloomshave recently been reported to occur (Madhupratapet al. 2000; Schoemann et al. 2005).

Modeling

The search for organizational pattern in planktonand its causes is not new. The realization is grow-ing that there is more to the organization ofnature than can be revealed in empirical data.Thus there must be theory, and there must bemodeling. The role of model simulation in under-standing system complexity and in decision-mak-ing is recognized even in the business world(Schrage and Peters 1999). Models oVer the abilityto test conceptual understanding of howcomponents of a given system are linked, and tosimulate complex interactions in a quantiWablerepeatable manner. Modeling, however, succeedsonly if new insights subsequently lead to valida-tion or rejection of hypotheses upon moredetailed examination. Thus information extractedfrom model runs is strongly linked to the chosenstructure (trophic resolution) of the model and tothe parameterizations of the interactions betweenthe components. When properly validated, mod-els can be useful in both hindcast and forecastmodes and used for guiding decision-making inenvironmental matters.

Life-cycle, food web and ecosystem modelshave all been formulated to facilitate understand-ing of the ecology of Phaeocystis and the impactof their blooms on marine food web structuresand biogeochemical cycles. Environmental appli-cation of these models included the control ofPhaeocystis blooms in eutrophic coastal waters(Gypens et al. 2004; Lancelot et al. 2006) andtheir contribution to the oceanic biological pumpof atmospheric CO2 (Gypens et al. 2004; Pasqueret al. 2005; Tagliabue and Arrigo 2005).

Conceptual models for the life cycle of Phaeo-cystis (Rousseau et al. 1994, this volume; Whipple

123

Biogeochemistry

et al. 2005a, b; this volume) have been essential toprogress in understanding Phaeocystis ecosystemsby providing the basis for subsequent experimen-tal studies and for best trophic resolution in math-ematical models. Mechanistic models includingdiVerent life forms of Phaeocystis as the dominantprimary producer and their various micro- andmeso-grazers have been developed to mimicPhaeocystis bloom dynamics in mesocosms andinvestigate the role of life history attributes at thepopulation level. These include several types offood web models such as individual-based modelsthat track cells and colonies as they grow anddivide (Canziani and Hallam 1996), food webmodels that investigate trophic interactions(Verity 2000), and a combination of these two(Ruardij et al. 2005). The latter model is themost sophisticated, describing the dynamics ofPhaeocystis free-living and colonial cells including50 colony classes (2–80,000 cells per colony).Simulations obtained with these diVerent modelssuggest critical processes determining Phaeocystislife form dominance and the exchanges betweenthem. These are the speciWc growth rate of colonycells compared to solitary cells, their light andnutrient controls (e.g., Ruardij et al. 2005), themechanisms controlling colony lysis, as well astop–down controls (including virus) speciWc to thediVerent Phaeocystis life forms.

Several ecosystem models with Phaeocystis anddiatoms as dominant primary producers but vari-ous trophic resolution, have been implemented todescribe seasonal cycles of ecosystem constituentsand elemental Xuxes in areas where Phaeocystisblooms have been recorded: the Barents Sea(Wassmann and Slagstad 1993), the southernNorth Sea (Gypens et al. 2004; Lancelot et al.2005; Lacroix et al. 2006), and the SouthernOcean (Arrigo et al. 2003; Pasquer et al. 2005;Tagliabue and Arrigo 2005, 2006).

Mechanisms controlling Phaeocystis domi-nance over diatoms in the iron-limited SouthernOcean, and the inXuence of ecosystem structureon oceanic capability of absorbing atmosphericCO2, were investigated by two ice-ocean ecosys-tem models applied to the Ross Sea: the coupledice atmosphere ocean (CIAO; Arrigo et al. 2003;Tagliabue and Arrigo 2005, 2006) and the seawtermicrobial community (SWAMCO-4; Pasquer

et al. 2005) models. These models diVered in theirtrophic resolution, notably in the description ofPhaeocystis life forms with only free-living cells inCIAO, while both colony forms and free-livingcells were explicitly described in SWAMCO-4.Simulations all suggested a similar substantialuptake of atmospheric CO2 associated withPhaeocystis blooms under control of light andiron availability, but explanatory mechanisms forPhaeocystis success were contrasting, due todiVerent Phaeocystis life-cycle resolutions. InCIAO simulations, the success of Phaeocystisover diatoms was explained by better photosyn-thetic performance of Phaeocystis cells at lowlight and low iron, while iron availability deter-mined the height of the bloom and the relatedCO2 sink (Tagliabue and Arrigo 2005). In con-trast, the explicit description of colony forms inSWAMCO-4 simulations emphasized the keyiron-sequestering role played by the Phaeocystiscolony polysaccharidic matrix, which depriveddiatoms of iron (Pasquer et al. 2005).

Mechanisms regulating the diatom–Phaeocystissuccession and the magnitude of colony blooms inthe eutrophic coastal waters of the eastern Chan-nel and Southern North Sea were investigatedwith the MIRO model implemented in a multi-box (Gypens et al. 2004; Lancelot et al. 2005,2006) and a tridimensional (Lacroix et al. 2006)structure. MIRO was very similar to SWAMCO-4,including two Phaeocystis stages (free-living cellsand colonies) as well as their interrelations, tro-phic fates, and associated carbon and nutrientXuxes. Interpretation of MIRO simulations sug-gested a key role of Phaeocystis colony aYnity forlow PO4 in determining the success and magni-tude of Phaeocystis blooms along the NO3 richcoastal zone, as already proposed from Weldobservations (Veldhuis and Admiraal 1987;Veldhuis et al. 1991). Historical reconstruction ofeutrophication in the area indicated that a well-balanced nitrate and phosphate enrichment asobserved between 1960 and 1989 was more bene-Wcial to diatoms than to Phaeocystis, while thecontemporary decrease in phosphate loadswhile high nitrate is maintained is beneWcial toPhaeocystis (Lancelot et al. 2006). The latterstudy concluded that future management of nutri-ent emission reduction aiming at decreasing

123

Biogeochemistry

Phaeocystis blooms in the southern North Seawithout impacting on diatoms should decreasenitrate loads. Further MIRO simulations showedthat: (i) a signiWcant sink of atmospheric CO2 wasassociated with Phaeocystis blooms in the southernNorth Sea (Gypens et al. 2004), (ii) the coastalecosystem had a low nutrient retention and elimi-nation capacity, (iii) the trophic eYciency of theplanktonic system was low, and (iv) both weremodulated by meteorological forcing (Lancelotet al. 2005).

Beside mechanistic models, inductive modelsbuilt on cause–eVect relationships such as Regres-sion Analysis (Breton et al. 2006) and fuzzy logic(Chen and Mynett 2004) have been developed asearly-warning tools for predicting the occurrence,timing, and magnitude of undesirable Phaeocystisblooms in the southern North Sea. Chen andMynett (2004) demonstrated that the integrationof a fuzzy cellular automata model in a three-dimensional numerical model describing the lightand nutrient Welds was a promising approach toforecast possible Phaeocystis blooms in Dutchcoastal waters. Statistical analysis of 14 years ofdiatom and Phaeocystis records in Belgiancoastal waters identiWed synergy between climatedriven by the North Atlantic oscillation (NAO)and human-induced river-based nitrate inputs(Breton et al. 2006), and predicted that elevatedPhaeocystis blooms during years of middle NAOwhile high NAO years would favor diatom blooms.

Concluding thoughts on the complexity of Phaeocystis ecosystems and recommendations for future research

Much of the research on this enigmatic phyto-plankton species cluster during the 20th centurywas conducted from two mutually reinforcingperspectives: that phytoplankton patterns in timeand space could be understood essentially from aperspective of physiological ecology, and thatplankton, being microscopic, were necessarilyrudimentary in their physiology, behavior, inter-actions, and integration of environmental stimuli.We now know better. For example, appreciationis spreading that evolutionary strategies whichresult in, e.g., 10% reduction in mortality are

every bit as eVective selective forces as those thatenhance growth by 10% (Verity and Smetacek1996). Single-celled plankton are no longerviewed as biologically simple creatures, merelybecause they bear little resemblance to complexhuman structures: auto- and heterotrophic unicel-lular plankton are fully capable of assessing theirenvironment with multiple sensors and commu-nicating with one another using a host of chem-icals. It should not be surprising therefore thatPhaeocystis is one of these complex organisms.

The papers deriving from the SCOR sympo-sium on Phaeocystis and the arguments mar-shaled here infer that it is essential to distinguishbetween factors aVecting production and standingstocks, in which the nature, type, and amount ofresources may be essential, versus factors aVect-ing time/space distributions, diversity, communitycomposition, and energy transfer, in which preda-tion, morphology, chemical ecology, and lifecycles may be more inXuential. Phaeocystis is oneof the few recognized key species or trophic engi-neers (sensu Lawton and Jones 1995) whose mor-phologies and transformations result in theircontributing the bulk of biogenic Xuxes or Wshproduction in speciWc ecosystems (Verity andSmetacek 1996). Population oscillations representcomplex ecosystem behavior inXuenced by com-ponent organisms, and evidence clearly indicatesthat Phaeocystis reorganizes these complex eco-systems (Fig. 1). Thus biocomplexity drives bio-geochemistry, but the fundamental mechanismsbehind these patterns are only now becomingappreciated. Through chemistry and/or size, thelife cycle of Phaeocystis Wnds refuge from preda-tion and viral infection, a process thought to stabi-lize ecosystems (see Strom and Loukos 1998).DiversiWcation at a given trophic level promotesdiversiWcation at the levels above (through morefood options) and below (through resource utili-zation), and is considered responsible for adap-tive radiation throughout the course of evolution(Stanley 1973; ButterWeld 1997). Interestingly, thestrategy for success apparently used by Phaeocys-tis (gelatinous morphology, heteromorphic lifecycle, size changes, and allelochemistry) is verysuccessful in freshwater ecosystems. Is thisbecause grazing pressure is greater in freshwatercompared to marine systems? Is Phaeocystis so

123

Biogeochemistry

ecologically successful because it is one of the fewmarine phytoplankton to combine these behav-ioral and physiological capabilities? Ultimately,taxa such as Phaeocystis may help unravelsimilarities and diVerences in the biocomplexityof freshwater versus marine pelagic ecosystems,and help us better understand and predictaquatic ecosystem structure and function.

Recent studies including those conductedunder the aegis of SCOR working group 120 haveadvanced understanding of Phaeocystis ecosys-tems, as reviewed above and described elsewherein this special issue. However, there remain sev-eral salient topics where data and informationneeds are foremost. For example, organelle-speciWc or complete genomes of several phyto-planktons have been sequenced (Prochlorococ-cus: Dufresne et al. 2003; Emiliania: Puerta et al.2004; Ostreococcus: Derelle et al. 2006) and thisinformation has identiWed metabolic pathwayspreviously unsuspected, e.g., a complete ureacycle in Thalassiosira pseudonana (Armbrustet al. 2004). Thus it would likely prove very usefulto sequence Phaeocystis in the future (see theJoint Genome Institute website http://genome.

jgi-psf.org/euk_home.html for eukaryotes plannedand already sequenced), and such knowledgewould enable answers to questions such as theidentity of genes responsible for life cycle trans-formations. In this regard, determination of pro-tocols and culture conditions under which eithergrowth stage can be maintained reliably in cul-ture, and induced to transform to the other stage,would permit a vast array of experiments aimedat quantifying evolutionary advantages and disad-vantages of each life-cycle stage, their susceptibil-ity to mortality, and role in biogeochemistry.Comparisons of genomic architecture would alsoenhance the basis for comparing adaptations inthe Weld. In general, quantitative understandingof controls and causes, and accurate prediction ofswitches between colonies and solitary cells,would advance the Weld tremendously.

Another important gap remains. Phaeocystis issuspected of producing large pools of dissolvedorganic matter, and the magnitude of this extra-cellular release likely vary with each life stage,their physiological status, and environmentalstresses. Moreover, its composition and labilityinXuences the microbial food web, production oftransparent exopolymers, and thus particle coagu-lation and sedimentation. Thus we need to learnmore about this pool, its sources, turnover, andsinks, to determine the speciWc role of Phaeocystisin biogeochemical cycles.

Future needs are not limited to empirical stud-ies, and in fact substantial progress likely will notbe achieved without continued model develop-ment. These include the need for the larger mod-eling community to share and combine models,to agree upon minimum formats or base modelsfor widespread use, and to use commonparameterizations. Where datasets permit,model intercomparisons should be conducted.Experimentalists clearly need to be intimatelyinvolved in the entire process, from model con-ceptualization to simulations, especially regard-ing Phaeocystis (Whipple et al. 2005a). Whileempiricists and modelers often speak diVerentlanguages (Anderson 2005; Flynn 2005), theeVort to design and implement such models canintegrate research and teach us more about eco-systems than the outcomes or predictions of themodels themselves.

Fig. 1 A diorama of a Phaeocystis ecosystem, in whichphysical, chemical, geological, and biological componentsinteract to varying degrees to inXuence food web structure,biogeochemical Xows, and ecosystem function. For simplic-ity, only well-documented components are illustrated. Cur-rent extent of our knowledge base is qualitatively related tofont size. Arrow magnitude and direction indicate majorknown Xows; increasingly smaller and more numerous ar-rows imply unknown connections, the importance of indi-rect Xows among ecosystem compartments, and overallsystem complexity

123

Biogeochemistry

Acknowledgments This special issue of Biogeochemistryrepresents the culmination of extensive eVorts by many sci-entists involved in various aspects of Phaeocystis research.Integration of their activities would not have been possiblewithout the support of the scientiWc committee on ocean re-search (SCOR) (www.jhu.edu/scor, accessed 6/30/06).SCOR working group #120 was devoted to Phaeocystisstudies; chair and co-chair were W.W.C. Gieskes and S.Belviso, respectively. We thank E. Urban at SCOR andthe organizers and participants of the Phaeocystis work-shops held in 2002 (University of East Anglia, UK), 2004(Savannah, GA, USA), and 2005 (University of Groningen,The Netherlands). Participation by the senior author in theresearch and SCOR activities reported here was providedby USA National Science Foundation grant OPP-00-83381and Department of Energy grant FG02-98ER62531. Wethank G. Malin, J. Stefels, and W.W.C. Gieskes forimprovements to an earlier draft, and A. Boyette for draft-ing the Wgure.

References

Admiraal W, Veldhuis MJW (1987) Determination of nu-cleosides and nucleotides in seawater by HPLC: appli-cation to phosphatase activity in cultures of the algaPhaeocystis pouchetii. Mar Ecol Prog Ser 36:277–285

Alcaraz M, PaVenhöfer G-A, Strickler JR (1980) Catchingthe algae: a Wrst account of visual observations of Wlter-feeding copepods. In: Kerfoot WC (ed) Evolution andecology of zooplankton communities. Univ. Press ofNew England, Hanover, NH, pp 241–248

Alderkamp A-C, Buma AGJ, van Rijssel M The carbohy-drates of Phaeocystis and their degradation in themicrobial food web. Biogeochemistry. doi:10.1007/s10533-007-9078-2

Alderkamp A-C, Nejstgaard JC, Verity PG, Zirbel MJ, Saz-hin AF, van Rijssel M (2006) Dynamics in carbohydratecomposition of Phaeocystis pouchetii colonies duringspring blooms in mesocosms. J Sea Res 55:169–181

Anderson TR (2005) Plankton functional type modeling:running before we can walk? J Plankton Res 27:1073–1081

Anderson DM, Cembella AD, HallegraeV GM (1998)Physiological ecology of harmful algal blooms. Spring-er-Verlag, Berlin

Armbrust EV et al (2004) The genome of the diatom Tha-lassiosira pseudonana: ecology, evolution, and metab-olism. Science 306:79–86

Arrieta JM, Weinbauer MG, Lute C, Herndl GJ (2004) Re-sponse of bacterioplankton to iron fertilization in theSouthern Ocean. Limnol Oceanogr 49:799–808

Arrigo KR, Tagliabue A (2005) Iron in the Ross Sea, PartII: impact of discrete iron addition strategies. J Geo-phys Res 110, C03010. DOI:10.1029/2004JC002568

Arrigo KR, Worthen DL, Robenson DH (2003) A coupledocean-ecosystem model of the Ross Sea: 2. Iron regu-lation of phytoplankton taxonomic variability and pri-mary production. J Geophys Res 108(C7):3231.DOI:10:1029/2001JC000856

Becquevort S, Lancelot C, Schoemann V Experimentalstudy on the role of Fe in the bacterial degradation oforganic matter derived from Phaeocystis antarctica.Biogeochemistry. doi:10.1007/s10533-007-9079-1

Bratbak G, Jacobsen A, Heldal M (1998a) Viral lysis ofPhaeocystis pouchetii and bacterial secondary produc-tion. Aquat Microb Ecol 16:11–16

Bratbak G, Jacobsen A, Heldal M, Nagasaki K, ThingstadF (1998b) Virus production in Phaeocystis pouchetiiand its relation to host cell growth and nutrition.Aquat Microb Ecol 16:1–9

Breton E, Rousseau V, Parent J-Y, Ozer J, Lancelot C(2006) Hydroclimatic modulation of diatom/Phaeo-cystis blooms in nutrient-enriched Belgian coastal wa-ters (North Sea). Limnol Oceanogr 51:1401–1409

Brussaard CPD, Bratbak G, Baudoux A-C, Ruardij PPhaeocystis and its interaction with viruses. Biogeo-chemistry. doi:10.1007/s10533-007-9096-0

Brussaard CPD, Kuipers B, Veldhuis MJW (2005a) A mes-ocosm study of Phaeocystis globosa populationdynamics. I. Regulatory role of viruses in bloom con-trol. Harmful Algae 4:859–874

Brussaard CPD, Mari X, Van Bleijswijk JDL, VeldhuisMJW (2005b) A mesocosm study of Phaeocystis glob-osa population dynamics. II. SigniWcance for themicrobial community. Harmful Algae 4:875–893

Buesseler KO, Benitez-Nelson CR, Moran SB, Burd A,Charette M, Cochran JK, Coppola L, Fisher NS, Fowl-er SW, Gardner WD, Guo LD, Gustafsson O, Lam-borg C, Masque P, Miquel JC, Passow U, Santschi PH,Savoye N, Stewart G, Trull T (2006) An assessment ofparticulate organic carbon to thorium-234 ratios in theocean and their impact on the application of 234Th asa POC Xux proxy. Mar Chem 100:213–233

ButterWeld NJ (1997) Plankton ecology and the Protero-zoic–Phanerozoic transition. Paleobiology 23:247–262

Canziani GA, Hallam TG (1996) A mathematical modelfor a Phaeocystis sp. dominated plankton communitydynamics. I. The basic model. Nonlin World 3:19–76

Chan AT, Andersen RJ, LeBlanc MJ, Harrison PJ (1980)Algal plating as a tool for investigating allelopathyamong marine microalgae. Mar Biol 59:7–13

Chen Q, Mynett A (2004) Predicting algal blooms in theDutch coast by integrated numerical and fuzzy cellularautomata. In: Liong S-Y, Phoon K-K, Babovic V (eds)Proceedings of the 6th international conference on hy-droinformatics. World ScientiWc Publishing Company,Singapore, pp 502–510

Costas E, Aguilera A, Gonzalez-Gil S, Lopez-Rodas V(1993) Contact inhibition: also a control for cell prolif-eration in unicellular algae? Biol Bull 184:1–5

Derelle E et al (2006) Genome analysis of the smallest free-living eukaryote Ostreococcus tauri unveils manyunique features. Proc Natl Acad Sci 103:11647–11652

DiTullio GR, Garcia N, Riseman SF, Sedwick PN EVects ofiron concentration on the pigment composition ofPhaeocystis antarctica in the Ross Sea. Biogeochemis-try. doi:10.1007/s10533-007-9080-8

DiTullio GR, Grebmeier JM, Arrigo KR, Lizotte MP,Robinson DH, Leventer A, Barry JB, VanWoert ML,Dunbar RB (2000) Rapid and early export of Phaeo-

123

Biogeochemistry

cystis antarctica blooms in the Ross Sea, Antarctica.Nature 404:595–598

Dufresne A et al (2003) Genome sequence of the cyano-bacterium Prochlorococcus marinus SS120, a nearlyminimal oxyphototrophic genome. Proc Natl Acad Sci100:10020–10025

Dutz J, Klein Breteler WCM, Kramer G (2005) Inhibitionof copepod feeding by exudates and transparent exop-lymer particles (TEP) derived from a Phaeocystisglobosa dominated phytoplankton community. Harm-ful Algae 4:929–940

Dutz J, Koski M (2006) Low grazing vulnerability in Xagel-lated, solitary cells of the prymnesiophyte Phaeocystisglobosa. Limnol Oceanogr 51:1230–1238

Edvardsen B, Paasche E (1998) Bloom dynamics and phys-iology of Prymnesium and Chrysochromulina. In:Anderson DM, Cembella AD, Hallegraef GM (eds)The physiological ecology of harmful algal blooms.Springer Verlag, Heidelberg, pp 193–208

Estep KW, Nejstgaard JC, Skjoldal HR, Rey F (1990) Pre-dation by copepods upon natural populations of Phae-ocystis pouchetii as a function of the physiological stateof the prey. Mar Ecol Prog Ser 67:235–249

Flynn KJ (2005) Castles built on sand: dysfunctionality inplankton models and the inadequacy of dialogue be-tween biologists and modelers. J Plankton Res27:1205–1210

Friedman MM, Strickler JR (1975) Chemoreception andfeeding in the calanoid copepods. Proc Nat Acad SciUSA 72:4185–4188

Gaebler S, Hayes PK, Medlin LK Methods used to revealgenetic diversity in the colony forming prymnesio-phyte Phaeocystis antarctica – preliminary results. Bio-geochemistry. doi:10.1007/s10533-007-9084-4

Gast RJ, Moran D, Dennett MR, Beaudoin DJ, Caron DA(2006) Studies in protistan diversity in Antarctica. Ab-stract 57th Annual meeting of the International Soci-ety of Protozoologists, June 20–24 (p 24). Lisbon,Portugal

Gauthier MJ, Aubert M (1981) Chemical telemediators inthe marine environment. In: Duursma EK, Dawson E(eds) Marine organic chemistry: evolution, composi-tion, interactions, and chemistry of organic matter inseawater. Elsevier, Amsterdam, pp 225–257

Gieskes WWC, Leterme SC, Peletier H, Edwards M, ReidPC Annual variation of Phaeocystis colonies Atlantic-wide since 1948, and interpretation of long-termchanges in ‘Phaeocystis hotspot’ North Sea. Biogeo-chemistry. doi:10.1007/s10533-007-9082-6

Gypens N, Lancelot C, Borges AV (2004) Carbon dynam-ics and CO2 air-sea exchanges in the eutrophied coast-al waters of the Southern Bight of the North Sea: amodelling study. Biogeosciences 1:147–157

Hamm CE, Simson DA, Merkel R, Smetacek V (1999) Col-onies of Phaeocystis globosa are protected by a thinbut tough skin. Mar Ecol Prog Ser 187:101–111

Hamm C, Reigstad M, Riser CW, Muhlebach A, Wass-mann P (2001) On the trophic fate of Phaeocystispouchetii. VII. Sterols and fatty acids reveal sedimen-tation of Phaeocystis—derived organic matter via krillfecal strings. Mar Ecol Prog Ser 209:55–69

Jacobsen A (2002) Morphology, relative DNA content,and hypothetical life cycle of Phaeocystis pouchetii(Prymnesiophyceae); with special emphasis on theXagellated cell type. Sarsia 87:338–349

Jacobsen A, Martinez-Martinez J, Verity P, Frischer ME,Sandaa R-A, Larsen A (2005) Are colonies or colonialcells of Phaeocystis pouchetii (Prymnesiophyceae) sus-ceptible to virus infection? American Society of Lim-nology and Oceanography, Summer Meeting, June19–24, Santiago de Compostela, Spain

Jacobsen A, Larsen A, Martínez-Martínez J, Frischer ME,Verity PG Are colonies and colonial cells of Phaeocys-tis pouchetii (Haptophyta) susceptible to viral infec-tion? Aquat Microb Ecol (Subm.)

Janse I, van Rijssel M, van Hall PJ, Gerwig GJ, GottschalJC, Prins RA (1996) The storage glucan of Phaeocystisglobosa (Prymnesiophyceae) cells. J Phycol 32:382–387

Johnston R (1963) Antimetabolites as an aid to the study ofphytoplankton nutrition. J Mar Biol Assoc UK43:409–425

Kamermans P (1994) Nutritional value of solitary cells andcolonies of Phaeocystis sp. for the bivalve Macomabalthica (L.). Ophelia 39:35–44

Keller MD, Bellows WK, Guillard RRL (1989) DimethylsulWde production in marine phytoplankton. In: Saltz-man ES, Cooper WJ (eds) Biogenic sulfur in the envi-ronment. American Chemical Society, Wash DC, pp167–182

Koski M, Dutz J, Klein-Breteler WCM (2005) Selectivegrazing of Temora longicornis in diVerent stages of aPhaeocystis globosa bloom – a mesocosm study.Harmful Algae 4:915–927

Kwint RLJ, Kramer KJM (1996) Annual cycle of the pro-duction and fate of DMS and DMSP in a marine coast-al system. Mar Ecol Prog Ser 134:217–224

Lacroix G, Ruddick R, Park Y, Gypens N, Lancelot C(2007) Validation of the 3D biogeochemical modelMIRO&CO with Weld nutrient and phytoplanktondata and MERIS-derived surface chlorophyll a imag-es. J Mar Syst 64:66–88

Lalande C (2006) Vertical export of biogenic in the Barentsand Chukchi Seas. PhD. Thesis, University of Knox-ville, Tennessee, USA

Lancelot C, Rousseau V (1994) Ecology of Phaeocystis: thekey role of colony forms. In: Green JC, LeadbeaterBSC (eds) The haptophyte algae. Clarendon Press,Oxford, pp 229–245

Lancelot C, Billen G, Sournia A, Weisse T, Colijn F, Vel-dhuis M, Davies A, Wassman P (1987) Phaeocystisblooms and nutrient enrichment in the continentalcoastal zones of the North Sea. Ambio 16:38–46

Lancelot C, Gypens N, Billen G, Garnier J, Roubeix V(2007) Testing an integrated river-ocean mathematicaltool for linking marine eutrophication to land use: thePhaeocystis dominated Belgian coastal zone (South-ern North Sea) over the past 50 years. J Mar Syst64:216–228

Lancelot C, Keller MD, Rousseau V, Smith WO Jr, MathotS (1998) Autecology of the marine haptophyte Phaeo-cystis sp. In: Anderson DM, Cembella AD, HallegraeVGM (eds) Physiological ecology of harmful algal

123

Biogeochemistry

blooms, NATO ASI series, vol G41. Springer-Verlag,Berlin, pp 209–224

Lancelot C, Spitz Y, Gypens N, Ruddick K, Becquevort S,Rousseau V, Lacroix G, Billen G (2005) Modellingdiatom–Phaeocystis blooms and nutrient cycles in theSouthern Bight of the North Sea: the MIRO model.Mar Ecol Prog Ser 289:63–78

Lange M, Chen Y-Q, Medlin LK (2002) Molecular geneticdelineation of Phaeocystis species (Prymnesiophy-ceae) using coding and non-coding regions of nuclearand plastid genomes. Eur J Phycol 37:77–92

Lange M, Guillou L, Vaulot D, Simon N, Amann RI, Lud-wig W, Medlin LK (1996) IdentiWcation of the classPrymnesiophyceae and the genus Phaeocystis withribosomal RNA-targeted nucleic acid probes detectedby Xow cytometry. J Phycol 32:858–868

Lawton JH, Jones CG (1995) Linking species and ecosys-tems: organisms as ecosystem engineers. In: Jones CG,Lawton JH (eds) Linking species and ecosystems.Chapman and Hall, NY, pp 141–150

Lewis WM Jr (1986) Evolutionary interpretations of allel-ochemical interactions in phytoplankton algae. AmNat 127:184–194

Liss PS, Malin G, Turner SM, Holligan PM (1994) Dimeth-yl sulWde and Phaeocystis: a review. J Mar Syst 5:41–53

Long JD, Hay ME (2006) When intraspeciWc exceeds inter-speciWc variance: eVects of phytoplankton morphologyand growth phase on copepod feeding and Wtness.Limnol Oceanogr 51:988–996

Long JD, Anderson JT, Nejstgaard JC, Verity PG, Hay MEAllelopathy of a bloom-froming marine phytoplank-ton, Phaeocystis, in mesocosm blooms and laboratorycultures. Aquat Microb Ecol (Subm.)

Lubchenco J, Cubit J (1980) Heteromorphic life historiesof certain marine algae as adaptations to variations inherbivory. Ecology 61:676–687

Madhupratap M, Sawant S, Gauns M (2000) A Wrst reporton a bloom of the marine prymnesiophycean, Phaeo-cystis globosa, from the Arabian Sea. Ocean Acta23:83–90

Mari X, Rassoulzadegan F, Brussaard CPD, Wassmann P(2005) Dynamics of transparent exopolymeric parti-cles (TEP) production by Phaeocystis globosa underN- or P-limitation: a controlling factor of the export/retention balance. Harmful Algae 4:895–914

Medlin LK, Lange M, Baumann MEM (1994) GeneticdiVerentiation among three colony-forming species ofPhaeocystis: further evidence for the phylogeny of thePrymnesiophyta. Phycologia 33:199–212

Medlin L, Zingone A A Review: the genus Phaeocystis and itsspecies. Biogeochemistry. doi:10.1007/s10533-007-9087-1

Mock T, Valentin K (2004) Photosynthesis and cold accli-mation: molecular evidence from a polar diatom. JPhycol 40:732–741

Nejstgaard JC, Tang KW, Steinke M, Dutz J, Koski M,Antajan E, Long JD Zooplankton grazing on Phaeo-cystis: a quantitative review and future challenges.Biogeochemistry (in press)

Nichols PD, Skerrat JH, Davidson A, Burton H, McMee-kin TA (1991) Lipids of cultured Phaeocystis pouch-etii: signatures for food web, biogeochemical and

environmental studies in Antarctica and the SouthernOcean. Phytochem 30:3209–3214

Noel MH, Kawachi M, Inouye I (2004) Induced dimorp-hid life cycle of a coccolithophorid Calytrosphaerasphaeroidea (Prymnesiophyceae). J Phycol 40:112–129

Noordkamp DJB, Gieskes WWC, Gottschal JC, ForneyLJ, van Rijssel M (2000) Acrylate in Phaeocystis colo-nies does not aVect the surrounding bacteria. J Sea Res43:287–296

Noordkamp DJB, Schotten M, Gieskes WWC, Forney LJ,Gottschal JC, van Rijssel M (1998) High acrylate con-centrations in the mucus of Phaeocystis globosa colo-nies. Aquat Microb Ecol 16:45–52

Nygaard K, Tobiesen A (1993) Bacterivory in algae: a sur-vival strategy during nutrient limitation. Limnol Ocea-nogr 38:273–279

Pasquer B, Laruelle G, Becquevort S, Schoemann V, Go-osse H, Lancelot C (2005) Linking ocean biogeochem-ical cycles and ecosystem structure and function:results of the complex Swamco-4 model. J Sea Res53:93–108

Peperzak L, Colijn F, Gieskes WWC, Peeters JCH (1998)Development of the diatom–Phaeocystis spring bloomin the Dutch coastal zone of the North Sea: the siliconversus the daily irradiance threshold hypothesis. JPlankton Res 20:517–537

Peperzak L, Duin RMN, Colijn F, Gieskes WWC (2000)Growth and mortality of Xagellates and non-Xagellatecells of Phaeocystis globosa (Prymnesiophyceae). JPlankton Res 22:107–119

Pohnert G (2004) Chemical defense strategies of marineorganisms. In: Schulz SE (ed) Topics in current chem-istry. Springer-Verlag GmBH, Berlin, pp 179–219

Pratt DM (1966) Competition between Skeletonema costa-tum and Olisthodiscus luteus in Narragansett Bay andin culture. Limnol Oceanogr 11:447–455

Puerta MVS, BachvaroV TR, Delwiche CF (2004) Thecomplete mitochondrial genome sequence of the ha-ptophyte Emiliania huxleyi and its relation to hetero-konts. DNA Res 11:1–10

Ratkova T, Wassmann P (2002) Seasonal variation andspatial distribution of phyto- and protozooplankton inthe central Barents Sea. J Mar Syst 38:47–75

Raven JA (1990) Predictions of Mn and Fe use eYcienciesof phototrophic growth as a function of light availabil-ity for growth and of C assimilation pathway. NewPhytol 116:1–18

Reid PC, Colebrook JM, Matthews JBL, Aiken J (2003)Continuous plankton recorder: concepts and history,from plankton indicator to undulating recorders. ProgOceanogr 58:117–173

Reigstad M, Wassmann P Does Phaeocystis spp. contributesigniWcantly to vertical export of biogenic matter? Bio-geochemistry. doi:10.1007/s10533-007-9093-3

Rousseau V, Chrétiennot-Dinet M-J, Jacobsen A, Verity P,Whipple S The life cycle of Phaeocystis: state of knowl-edge and presumptive role in ecology. Biogeochemis-try doi:10.1007/s10533-007-9085-3

Rousseau V, Vaulot D, Casotti R, Cariou V, Lenz J,Gunkel JJ, Baumann M (1994) The life cycle of

123

Biogeochemistry

Phaeocystis (Prymnesiophyceae): evidence andhypotheses. J Mar Syst 5:23–39

Ruardij P, Veldhuis MJW, Brussaard CPD (2005) Model-ing the bloom dynamics of the polymorphic phyto-plankter Phaeocystis globosa: impact of grazers andviruses. Harmful Algae 4:941–963

Sanders RW, Wickham SA (1993) Planktonic protozoa andmetazoa: predation, food quality, and population con-trol. Mar Microb Food Webs 7:197–223

Sazhin AF, Felipe Artigas L, Nejstgaard JC, Frischer MEColonization of Phaeocystis species by pennate dia-toms and other protists: an important contribution tocolony biomass. Biogeochemistry. doi:10.1007/s10533-007-9086-2

Schmidt S, Belviso S, Wassmann P, Thouzeau G, Stefels J,Reigstad M Vernal sedimentation trends in north Nor-wegian fjords: temporary anomaly in 234Th particulateXuxes related to Phaeocystis pouchetii proliferation.Biogeochemistry. doi:10.1007/s10533-007-9094-2

Schoemann V, Becquevort S, Stefels J, Rousseau V, Lance-lot C (2005) Phaeocystis blooms in the global oceanand their controlling mechanisms: a review. J Sea Res53:43–66

Schrage M, Peters T (1999) Serious play: how the world’sbest companies simulate to innovate. Harvard Busi-ness School Press, Cambridge, MD

Sedwick PN, Garcia N, Riseman SF, Marsay CM, DiTullioGR Evidence for high iron requirements of colonialPhaeocystis antarctica in the Ross Sea. Biogeochemis-try. doi:10.1007/s10533-007-9081-7

Seuront L, Lacheze C, Doubell MJ, Seymour JR, MitchellJG The inXuence of Phaeocystis globosa bloomdynamics on microscale spatial patterns of phyto-plankton biomass and bulk-phase seawater viscosity.Biogeochemistry. doi:10.1007/s10533-007-9097-z

Shenoy DM, Dileep Kumar M Variability in abundanceand Xuxes of dimethyl sulphide in the Indian Ocean.Biogeochemistry. doi:10.1007/s10533-007-9092-4

Sieburth JMcN (1960) Acrylic acid, an “antibiotic” princi-ple in Phaeocystis blooms in Antarctic waters. Science132:676–677

Slocum CJ (1980) DiVerential susceptibility to grazers intwo phases of an intertidal alga: advantages of hetero-morphic generations. J Exp Mar Biol Ecol 46:99–110

Smaal AC, Twisk F (1997) Filtration and absorption ofPhaeocystis cf. globosa by the mussel Mytilus edulis L.J Exp Mar Biol Ecol 209:33–46

Spero HJ (1985) Chemosensory capabilities in the phago-trophic dinoXagellate Gymnodinium fungiforme. JPhycol 21:181–184

Stanley SM (1973) An ecological theory for the sudden ori-gin of multicellular life in the late Precambrian. ProcNat Acad Sci USA 70:1486–1489

Stefels J (2000) Physiological aspects of the production andconversion of DMSP in marine algae and higherplants. J Sea Res 43:183–197

Stefels J, Dijkhuizen L (1996) Characteristics of DMSP-lyase in Phaeocystis sp. (Prymnesiophyceae). MarEcol Prog Ser 131:307–313

Stefels J, Van Boekel WHM (1993) Production of DMSfrom dissolved DMSP in axenic cultures of the marine

phytoplankton species Phaeocystis sp. Mar Ecol ProgSer 97:11–18

Stefels J, van Leeuwe MA (1998) EVects of iron and lightstress on the biogeochemical composition of AntarcticPhaeocystis sp. (Prymnesiophyceae). I. IntracellularDMSP concentrations. J Phycol 34:486–495

Stefels J, Steinke M, Turner S, Malin G, Belviso S Environ-mental constraints on the production of the climati-cally active gas dimethylsulphide (DMS) andimplications for ecosystem modeling. Biogeochemis-try. doi:10.1007/s10533-007-9091-5

Steinke M, Stefels J, Stamhuis E (2006) Dimethyl sulWdetriggers search behavior in copepods. Limnol Ocea-nogr 51:1925–1930

Stelfox-Widdicombe CE, Archer SD, Burkill PH, Stefels J(2004) Microzooplankton grazing in Phaeocystis anddiatom-dominated waters in the southern North Sea inspring. J Sea Res 51:37–51

Strom SL, Loukos H (1998) Selective feeding by protozoa:model and experimental behaviors and their conse-quences for population stability. J Plankton Res20:831–846

Strom S, Wolfe G, Holmes J, Stecher H, Shimeneck C,Lambert S, Moreno E (2003) Chemical defense inmicroplankton. I. Feeding and growth rates of hetero-trophic protests on the DMS-producing phytoplankterEmiliania huxleyi. Limnol Oceanogr 48:217–229

Sunda W, Kieber DJ, Kiene RP, Huntsman S (2002) Anantioxidant function for DMSP and DMS in marine al-gae. Nature 418:317–320

Tagliabue A, Arrigo KR (2005) Iron in the Ross Sea: 1. Im-pact on CO2 Xuxes via variation in phytoplanktonfunctional group and non-RedWeld stoichiometry. JGeophys Res 110, C03009. DOI: 10.1029/2004JC002531

Tagliabue A, Arrigo KR (2006) Processes governing thesupply of iron to phytoplankton in stratiWed seas. JGeophys Res 111, C06019. DOI: 10.1029/2005JC003363

Tang KW (2003) Grazing and colony size development inPhaeocystis globosa (Prymnesiophyceae): the role of achemical signal. J Plankton Res 25:831–842

Targett NM, Ward JE (1991) Bioactive microalgal metabo-lites: mediation of subtle ecological interactions inphytophagous suspension-feeding marine inverte-brates. Bioorganic Mar Chem 4:91–118

Thingstad F, Billen G (1994) Microbial degradation ofPhaeocystis material in the water column. J Mar Syst5:55–66

Tillmann U (1998) Phagotrophy by a plastidic haptophyte,Prymnesium patelliferum. Aquat Microb Ecol 14:155–160

Turner SM, Nightingale PD, Broadgate W, Liss PS (1995)The distribution of dimethyl sulWde and dimethylsul-phoproprionate in Antarctic waters and sea ice. Deep-Sea Res II 42:1059–1080

Uchida T (1977) Excretion of a diatom inhibitory substanceby Prorocentrum micans Ehrenberg. Jap J Ecol 27:1–4

Van Alstyne KL (1986) EVects of phytoplankton taste andsmell on feeding behavior of the copepod Centropageshamatus. Mar Ecol Prog Ser 34:187–190

123

Biogeochemistry

van Boekel WHM (1992) Phaeocystis colony mucus com-ponents and the importance of calcium ions for stabil-ity. Mar Ecol Prog Ser 87:301–305

van Hilst CM, Smith WO Jr (2002) Photosynthesis/irradi-ance relationships in the Ross Sea, Antarctica, andtheir control by phytoplankton assemblage composi-tion and environmental factors. Mar Ecol Prog Ser226:1–12

van Leeuwe MA, de Baar HJW (2000) Photoacclimationby the Antarctic Xagellate Pyramimonas sp. (Prasino-phyceae) in response to iron limitation. Eur J Phycol35:295–303

van Leeuwe MA, Stefels J (1998) EVects of iron and lightstress on the biogeochemical composition of AntarcticPhaeocystis sp. (Prymnesiophyceae). II. Pigment com-position. J Phycol 34:496–503

van Leeuwe MA, Stefels J Photosynthetic responses inPhaeocystis antarctica towards varying light and ironconditions. Biogeochemistry. doi:10.1007/s10533-007-9083-5

van Rijssel M, Alderkamp A-C, Nejstgaard JC, Sazhin AF,Verity PG Haemolytic activity of living Phaeocystispouchetii during mesocosm blooms. Biogeochemistry.doi:10.1007/s10533-007-9095-1

van Rijssel M, Hamm CE, Gieskes WWC (1997) Phaeocys-tis globosa (Prymnesiophyceae) colonies: hollowstructures built with small amounts of polysaccharides.Eur J Phycol 32:185–192

Vaulot D, Birrien J-L, Marie D, Casotti R, Veldhuis MJW,Kraaij GW, Chretiennot-Dinet M-J (1994) Morphol-ogy, ploidy, pigment composition, and genome size ofcultured strains of Phaeocystis (Prymnesiophyceae). JPhycol 30:1022–1035

Veldhuis MJW, Admiraal W (1987) InXuence of phosphatedepletion on the growth and colony formation ofPhaeocystis pouchetii. Mar Biol 95:47–54

Veldhuis MJW, Wassmann P (2005) Bloom dynamics andbiological control of a high biomass HAB species inEuropean coastal waters: a Phaeocystis case study.Harmful Algae 4:805–809

Veldhuis MJW, Brussaard CPD, Noordeloos AAM (2005)Living in a Phaeocystis colony: a way to be a successfulalgal species. Harmful Algae 4:841–858

Veldhuis MJW, Colijn F, Admiraal W (1991) Phosphateutilization in Phaeocystis pouchetii (Haptophyceae).Mar Biol 12:53–62

Verity PG (1988) Chemosensory behavior in marine plank-tonic ciliates. Bull Mar Sci 43:772–782

Verity PG (1991) Feeding in planktonic protozoans: evi-dence for non-random acquisition of prey. Mar MicrobFood Webs 5:69–76

Verity PG (2000) Grazing experiments and model simula-tions of the role of zooplankton in Phaeocystis foodwebs. J Sea Res 43:317–343

Verity PG, Medlin LK (2003) Observations on colony for-mation by the cosmopolitan phytoplankton genusPhaeocystis. J Mar Syst 43:153–164

Verity PG, Smetacek V (1996) Organism life cycles, preda-tion, and the structure of marine pelagic ecosystems.Mar Ecol Prog Ser 130:277–293

Verity PG, Villareal TA, Smayda TJ (1988a) Ecologicalinvestigations of blooms of colonial Phaeocystispouchetii. I. Abundance, biochemical composition,and metabolic rates. J Plankton Res 10:219–248

Verity PG, Villareal TA, Smayda TJ (1988b) Ecologicalinvestigations of blooms of colonial Phaeocystispouchetii. II. The role of life cycle phenomena inbloom termination. J Plankton Res 10:749–766

Wassmann P (1994) SigniWcance of sedimentation for thetermination of Phaeocystis blooms. J. Mar Syst 5:81–100

Wassmann P, Ratkova T, Reigstad M (2005) The contribu-tion of single and colonial cells of Phaeocystis pouch-etii to spring and summer blooms in the north-easternNorth Atlantic. Harmful Algae 4:823–840

Wassmann P, Slagstad D (1993) Seasonal and annualdynamics of particulate carbon Xux in the Barents Sea,a model approach. Polar Biol 13:363–372

Weisse T, Tande K, Verity P, Hansen F, Gieskes W (1994)The trophic signiWcance of Phaeocystis blooms. J MarSyst 5:67–79

Whipple SJ, Patten BC, Verity PG (2005a) Life cycle of themarine alga Phaeocystis: a conceptual model to sum-marize literature and guide research. J Mar Syst 57:83–110

Whipple SJ, Patten BC, Verity PG (2005b) Colony growthand evidence for colony multiplication in Phaeocystispouchetii (Prymnesiophyceae) isolated from meso-cosm blooms. J Plankton Res 27:495–501

Whipple SJ, Patten BC, Verity PG, Nejstgaard JC, LongJD, Anderson JT, Jacobsen A, Larsen A, Martinez-Martinez J, Borrett SR. Gaining integrated under-standing of Phaeocystis spp. (Prymnesiophyceae)through model-driven laboratory and mesocosm stud-ies. Biogeochemistry. doi:10.1007/s10533-007-9089-2

Wolfe GV (2000) The chemical defense ecology of marineunicellular plankton: constraints, mechanisms, and im-pacts. Biol Bull 198:225–244

Wolfe GV, Levasseur M, Cantin G, Michaud S (2000)DMSP and DMS dynamics and microzooplanktongrazing in the Labrador Sea: application of the dilutiontechnique. Deep-Sea Res I 47:2243–2264

Wolfe GV, Steinke M, Kirst GO (1997) Grazing-activatedchemical defense in a unicellular marine alga. Nature387:894–897

Wooten EC, Roberts EC (2006) Biochemical recognitionof prey by planktonic protozoa. Abstract 57th Annualmeeting of the International Society of Protozoolo-gists, June 20–24, 2006 (p 57). Lisbon, Portugal

Yoshida T, Hairston NG, Ellner SP (2004) Evolutionarytrade-oV between defense against grazing and compet-itive ability in a simple unicellular alga, Chlorella vul-garis. Proc Royal Soc Lond Ser B, Biol Sci 271:1947–1953

Zingone A, Chretiennot-Dinet M-J, Lange M, Medlin L(1999) Morphological and genetic characterization ofPhaeocystis cordata and P. jahnii (Prymnesiophyceae),two new species from the Mediterranean Sea. J Phycol35:1322–1337

123