Deep-Sea Research II 50 (2003) 605–617 The temporal dynamics ofthe flagellated and colonial stages of Phaeocystis antarctica in the Ross Sea Walker O. Smith Jr.a,*, Mark R. Dennettb, Sylvie Mathota, David A. Caronc a Virginia Institute of Marine Science, College of William and Mary, Gloucester Point, VA, 23062, USA b Biology Department, Woods Hole Oceanographic Institution, Woods Hole, MA, 02543, USA c Department of Biological Sciences, University of Southern California, Los Angeles, CA 90089-0371, USA Accepted 27 September 2002 Abstract Phaeocystis antarctica in the Ross Sea forms colonies, and blooms of colonial P. antarctica often occur over large areas in the southern Ross Sea. Sites where colonies occur often have significant vertical fluxes of carbon in the form of aggregated and flocculent material. P. antarctica also is a key component of the sulfur cycle; therefore, the species is critically important in many ofthe biogeochemical cycles in the Ross Sea. Despite this fundamental role, the lifehistory and temporal dynamics ofthis species are poorly known. This study investigated the contribution ofsolitary, flagellated forms and colonial cells of P. antarctica to phytoplankton abundance and autotrophic carbon, and the factors that might control the relative importance ofthese two morphological forms. Solitary P. antarctica cells numerically dominated the phytoplankton assemblage early in austral spring, although colony formation occurred almost immediately upon the onset ofnet population growth. The percentage ofsolitary cells relative to total cells (colonial+solitary) was high in early austral spring but decreased to a minimum during late spring; specifically, nearly 98% ofthe P. antarctica cells were in colonies in late spring, coinciding with the seasonal chlorophyll maximum. Significant phytoplankton mortality rates were positively correlated with high ratios ofsolitary: total P. antarctica cells. Highest mortality rates were observed during austral spring when solitary cells dominated the P. antarctica population. The abundance ofsolitary P. antarctica cells began to increase again during late summer, but colonial cell numbers were always greater than those ofsolitary cells during this period. This increase in the contribution ofsolitary forms during summer may have been a consequence ofmore severe micronutrient limitation forthe colonies (relative to solitary cells), lifehistory processes of P. antarctica, reduced microzooplankton grazing at that time, or a combination ofthese and other factors. We conclude that the relative abundance ofsolitary and colonial forms ofthis prymnesiophyte alga may be a consequence ofseasonal changes in these factors. The outcome of these interactions affects the contribution of this alga to the vertical flux of carbon and the degree to which P. antarctica participates in the microbial food web of the Ross Sea. r 2003 Elsevier Science Ltd. All rights reserved. 1. Introduction There has been the recognition during the past few decades that phytoplankton assemblage com- *Corresponding author. Tel.: +1-804-684-7709. position exerts a strong control over pelagic E-mail address: [email protected] (W.O. Smith Jr.). carbon transformations and biogeochemical cycles. 0967-0645/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved. PII: S 0967-0645(02)00586-6 606 W.O. Smith Jr. et al. / Deep-Sea Research II 50 (2003) 605–617 For example, the dominance ofprimary produc- Little is known concerning the environmental tion by small cyanobacteria in oligotrophic systems factors that regulate the dominance of solitary and results in the utilization and remineralization of colonial forms of the alga, although it has been much ofthe production within the euphotic zone suggested that inorganic nutrients and grazing by the microbial food web. In contrast, diatoms may influence the ratio ofthese forms. In the are largely consumed by mesozooplankton, and North Sea when phosphate is reduced to limiting thus contribute significantly to the vertical flux of concentrations, flagellated cells of P. pouchetii organic carbon by virtue oftheir incorporation appeared to be favored (Riegman et al., 1992). into sinking fecal material. Other forms, such as Similarly, solitary cells ofthe same species tend to some prymnesiophytes, overwhelmingly contribute be favored when ammonium is the dominant form to sulfur cycling by releasing large amounts of ofnitrogen available to the alga ( Lancelot et al., dimethylsulfide (DMS) to the water, which then 1998). Riegman and van Boekel (1996) concluded volatilizes and exchanges with the atmosphere. that colonial cells of P. globosa are more effective Coccolithophorids, another type ofprymnesio- competitors for nitrate and are favored at irra- phyte, form calcium carbonate tests, and hence diances above 50 mmol photons mÀ2 sÀ1. A recent exert control over the alkalinity and carbon cycles laboratory study of P. globosa found that colony when present. Thus, not all phytoplankton affect formation and survival were enhanced under elemental and energy flow equally, and present conditions ofenhanced microzooplankton grazing models are beginning to incorporate the critical (Jakobsen and Tang, 2002), confirming the role of roles ofindividual taxa in biogeochemical cycles grazing. However, no clear relationship between (Armstrong, 1999). environmental conditions and life cycle stage for The Southern Ocean has long been considered to P. antarctica can, at present, be determined. be a diatom-dominated system, and indeed many P. antarctica is critical to primary production, locations have substantial numbers ofthese forms elemental transformations and vertical flux in the (e.g., Wilson et al., 1986; Brown and Landry, Ross Sea. For example, massive blooms ofthis 2001). More recently it has been recognized that species occur in the southern Ross Sea, and they Phaeocystis antarctica, a prymnesiophyte with a have been reported to produce some ofthe highest relatively complex life cycle, is extremely important concentrations ofDMS observed in the ocean in some regions. P. antarctica forms large blooms (DiTullio and Smith, 1995; Kettle et al., 1999). in areas ofthe Southern Ocean such as Prydz Bay Furthermore, P. antarctica colonies appear to be (e.g., Davidson and Marchant, 1992) and the relatively ungrazed by microzooplankton in this southern Ross Sea (El-Sayed et al., 1983). Indeed, region (Caron et al., 2000), and thus its role in it is often the dominant species in the latter region carbon export and flux is substantial. Dunbar et al. (DiTullio and Smith, 1996; Arrigo et al., 1999; (1998) compared the vertical fluxes oftwo sites Mathot et al., 2000; Dennett et al., 2001). (one dominated by diatoms and the other domi- P. antarctica is one ofthe fourspecies of nated by P. antarctica) in the southern Ross Sea Phaeocystis that can occur either as solitary, and found that carbon flux rates were approxi- flagellated unicells or as colonies ofhundreds of mately equal (although silica flux rates were an non-flagellated cells (Rousseau et al., 1994; Lan- order ofmagnitude greater at the site dominated celot et al., 1998; Zingone et al., 1999). The by diatoms). The material collected in the traps at colonies are spherical to cylindrical in shape and the P. antarctica-dominated site was largely hollow, and the algal cells are embedded in an aggregates and amorphous organic material, im- organic matrix that is the basis ofthe colony like plying that the material was exported without those of P. globosa (Hamm et al., 1999). It has transformation by the food web. It has also been been reported that P. pouchetii cells are liberated suggested that P. antarctica contributes to sub- from the colonial matrix as the colonies sink in the stantial fluxes during periods ofrapid growth water column, resulting in the generation oflarge during austral spring (DiTullio et al., 2000), but numbers ofsolitary cells ( Wassmann et al., 1990). these flux events have not yet been confirmed by W.O. Smith Jr. et al. / Deep-Sea Research II 50 (2003) 605–617 607 sediment trap collections in the Ross Sea. Collec- B. Palmer. Cruises were completed in early spring tively, this information indicates that detailed (October–November 1996), in late spring (Novem- knowledge ofthis species is essential forunder- ber–December 1997), in summer (January–Febru- standing ecosystem function in this and other ary 1997) and autumn (March–April 1997). Data Antarctic coastal waters. from all of the cruises were merged to produce a This study investigated the contribution of temporal composite (Smith et al., 2000a, b). Much solitary and colonial P. antarctica cells to the ofthe sampling occurred along 76 1300S at a series overall temporal dynamics ofthe phytoplankton ofeight stations, each separated by ca. 60 km assemblages ofthe southern Ross Sea. We (Fig. 1). Ice concentrations were variable during hypothesized that the morphological form of the study with 100% ice cover being observed in the species was important in determining its fate early spring and in autumn and ice-free conditions in the planktonic food web, given that we know occurring during summer (Smith et al., 2000a). that solitary cells were actively grazed by micro- Water samples were collected from the upper zooplankton during spring, while rates ofphyto- 100 m using a trace-metal clean rosette or standard plankton mortality were low (mostly undetectable) Niskin bottles (with Teflon-coated internal when P. antarctica colonies dominated the phyto- springs) mounted on a rosette. Chlorophyll