Conclusions and directions for future research

Our experimental results must be considered in the context of seasonal changes in mean irradiance and dissolved iron concentrations for ocean surface waters in our study region. Smith and van Hilst (2003) estimate mean mixed-layer irradiances of 96 ± 58 iE m-2 s-1 and 180 ± 110 iE m-2 s-1 in the southern Ross Sea for the periods November

Fig. 4 Nitrate-specific growth rates (iN) of conlonial Phaeocystis in the incubation bottles between days 25 and 31, versus initial dissolved iron concentrations. The Monod hyperbola (dashed line) was fitted using an Eadie-Hofstee linear transformation (r2 = 0.85), and excludes the +1.8 nM Fe datum (in parentheses). The half-saturation constant for growth (K^) and maximum nitrate-specific growth rate (iNmax) are indicated

Fig. 4 Nitrate-specific growth rates (iN) of conlonial Phaeocystis in the incubation bottles between days 25 and 31, versus initial dissolved iron concentrations. The Monod hyperbola (dashed line) was fitted using an Eadie-Hofstee linear transformation (r2 = 0.85), and excludes the +1.8 nM Fe datum (in parentheses). The half-saturation constant for growth (K^) and maximum nitrate-specific growth rate (iNmax) are indicated

10-December 8, 1994 (mid spring-early summer) and December 20-January 15,1995 (mid summer), respectively. Over these same periods, dissolved iron concentrations in the upper water column averaged 1.0 and 0.23 nM, respectively (Sedwick et al. 2000). It is also important to acknowledge the antagonistic relationship between the iron and light requirements of phytoplankton, whereby cellular iron requirements are expected to decrease as mean irradiance increases (Raven 1990; Sunda and Huntsman 1997). Thus, in the Ross Sea polynya, we would expect the cellular iron requirements of colonial Phaeocystis to decrease during the growing season, as the surface mixed-layer shoals and mean irradiance increases, and as dissolved iron concentrations decrease.

The results of our two experiments are consistent with these seasonal changes in the availability of light and iron, given the interrelated influences of irradiance and iron availability on phytoplankton growth rate. The data from our laboratory dose-response iron-addition experiment indicate a relatively high iron requirement for colonial Phaeocystis at an irradiance of -20 iE m-2 s-1, a value that is representative of the mean irradiance in the mixed layer of the southern Ross Sea during early spring (Smith et al. 2000; Smith and van Hilst 2003; Hiscock 2004). At that time, dissolved Fe concentrations are likely to exceed our estimated half-saturation constant of 0.45 nM dissolved Fe (Sedwick et al. 2000; Coale et al. 2005), thus the growth of colonial P. antarctica should not be strongly limited by iron availability. However, we would expect colonial Phaeocystis to have significantly lower iron requirements at the higher irradiance levels used in our shipboard iron-light manipulation experiment, which was conducted in late December. In the southern Ross Sea during mid summer, integrated clear-sky irradiance at the sea surface estimated from the model of Gregg and Carder (1990) is around 640 iE m-2 s-1 (Hiscock 2004), thus we estimate irradiance on the order of 100 iE m-2 s-1 and 300 iE m-2 s-1 for our low-light and high-light treatments, respectively. The expectation of lower iron requirements of Phaeocystis under such irradiance levels (compared with the much lower irradiance of 20 iE m-2 s-1 used in our laboratory dose-response experiment) are borne out by the results of our shipboard experiment, since we observed no evidence for growth limitation or iron stress at the ambient dissolved Fe concentration of 0.38 nM.

Boyd (2002) speculated that the growth of Phaeocystis in the Ross Sea may be limited by iron availability from spring through late summer, and by low irradiance from autumn through early spring, with the potential for colimitation by iron and light during spring and early fall. We propose a slightly different scenario, whereby the impact of decreasing Fe availability during the spring is mitigated by the increase in irradiance, which significantly lowers the cellular iron requirements of Phaeocystis, thus allowing blooms to develop through the spring and into the summer. Eventually, however, dissolved Fe concentrations decrease to such low levels as to limit the growth rate of colonial Phaeocystis, even under relatively high mean irradiance, and at this point the blooms are terminated. To test this conceptual model requires an understanding of the effect of irradi-ance on the iron requirements of colonial P. antarctica. Towards this goal, we are currently undertaking further dose-response iron-addition experiments with our P. antarctica culture at higher levels of irradiance.

With regard to quantitative simulations of the Ross Sea ecosystem, we note that recent numerical models (Arrigo et al. 2003; Tagliabue and Arrigo 2005) have parameterized the growth of colonial Phaeocystis using a half-saturation constant of 0.01 nM for iron, which is more than an order of magnitude less than the KM value estimated from our laboratory experiment. Even allowing for a significant decrease in KM at the higher irradiance levels during late spring and into summer, it seems likely that the growth requirements of colonial Phaeocystis are closer to 0.1 nM dissolved Fe, similar to values estimated for Southern Ocean diatoms by Timmermans et al. (2001, 2004). The cellular C/Fe uptake ratio is another critical parameter for such numerical models, with a value of 450,000 (estimated from model diagnostics) used in the recent modelling study of Tagliabue and Arrigo (2005). The measurement of C/Fe uptake ratios in colonial Phae-ocystis represents a formidable methodological challenge, because dissolved Fe is likely to be partitioned into both the Phaeocystis cells and surrounding mucus, and the extent to which mucus-bound Fe is available to the cells is unknown (Schoemann et al. 2005). Techniques that have been used to remove extracellular Fe for cellular Fe uptake measurements (e.g., Hudson and Morel 1989; Tovar-Sanchez et al. 2003; Tang and Morel 2006) might not be applicable to colonial Phaeocystis, and sophisticated analytical tools such as the synchrotron X-ray fluorescence microprobe (Twining et al. 2003) may be required to provide accurate estimates of cellular C/Fe uptake ratios.

Acknowledgements The authors gratefully acknowledge the officers and crew of the RV Nathaniel B. Palmer, and Raytheon Polar Services Company personnel for their outstanding support in the field. We thank two anonymous reviewers for their helpful criticisms and suggestions. This work has benefited from discussions with Veronique Schoemann, Walker Smith, and Alessandro Tagliabue. Funding for this research was provided by US National Science Foundation grants OPP-0230559 (PNS), OPP-0230513 (GRD) and DGE-0139313 (NSG through M. Vansickle and G. Tempel).

Was this article helpful?

0 0

Post a comment