The equatorial Pacific Ocean: Grazer-controlled phytoplankton populations in an iron-limited ecosystem1

Abstract

Experiments were conducted in the equatorial Pacific Ocean to assess the role of Fe and grazing in regulating use of N03-by the phytoplankton community. Nitrate uptake rates in situ were slow because NH,+ concentrations were inhibitory and because phytoplankton biomass was kept low by grazing. When feeding of grazers was artificially suppressed, phytoplankton net growth rate increased, biomass accumulated, and NO, -was consumed. Rapid rates of Fe uptake [40 pmol Fe (g Chl a)-' h-l] decreased by an order of magnitude in l-2 d after Fe was added, demonstrating that these rates were under physiological regulation and were elevated in response to low Fe concentrations. Addition of Fe increased carbon uptake and the short-term N-specific NO,-uptake rate by 2-9 times. These physiological stimulations were confined to large phytoplankton (> 3 pm), which thus must have been Fe-limited in situ. N03-uptake rate and biomass of small phytoplankton were unaffected by Fe enrichment. The results thus suggest that the low biomass, N03--rich condition of the equatorial Pacific Ocean exists because low Fe concentrations limit use of N03-by large phytoplankton and favor growth of small phytoplankton that are grazed efficiently and use NH,+ preferentially.The equatorial Pacific is one of the regions of open ocean that is comparatively rich in nitrate and phosphate and low in chlorophyll; other such areas are in the Antarctic and subarctic Pacific. Several explanations for this unusual condition have been offered. Two of these, grazing control and Fe limitation, were originally portrayed as antipodal hypotheses. We and others (Price et al. 1991; Chavez et al. 199 1; Cullen 199 1; Frost 199 1; Martin et al. 199 1; Miller et al. 199 1) have argued, however, that these explanations are more likely I Dedicated to the memory of John Martin. AcknowledgmentsFunding for this research was provided by grants from NSF, ONR, and NSERC.We are indebted to Ken Bruland for his invitation to participate on this cruise. Surface NO,-and dissolved Fe concentrations were provided by L. Miller, G. Smith, and K. Bruland. We thank L. Harris and W. G. Harrison for allowing us to use their Europa Analyzer; M. Wells and G. Smith for collecting clean water samples; M. Durand, J. Dusenbcrry and S. W. Chisholm for analyzing samples for Prochlorococcus and Synechococcus; and the officers and crew of the RV Moana Wave for smooth sailing. J. Cullen and an anonymous reviewer provided criticisms of the initial version of this manuscript.

Iron limits the cell division rate of Prochlorococcus in the eastern equatorial Pacific

“…This hypothesis, in its broadest form, states that the cell division rates of the small phytoplankton that dominate the indigenous community are relatively fast and are ''not strongly'' limited by iron (not to the extent that is Ͻ1/2 max ; Cullen 1995). These cells are dependent on recycled nutrients (Price et al 1994;Landry et al 1997), and any increases in number following iron addition are rapidly cropped by quickly growing microzooplankton grazers (Banse 1982;Landry et al 1997). In contrast, large cells with relatively small surface area to volume ratios are at a competitive disadvantage for NH and Fe assimilation be-ϩ 4 cause of diffusion limitation (Morel et al 1991a) and are iron limited.…”

“…The iron hypothesis explains these observations by maintaining that phytoplankton productivity and biomass are limited by iron (Martin et al 1991;Coale et al 1996b). Indeed, ambient iron levels are low in high-nutrient low-chlorophyll (HNLC) areas (Martin et al 1991;Coale et al 1996b), and the addition of iron to incubation bottles has been shown to increase net growth rates, chlorophyll yields, nitrate uptake, and nutrient consumption, particularly of the larger cells Martin et al 1991;Price et al 1991Price et al , 1994Takeda and Obata 1995;Fitzwater et al 1996). The ecumenical hypothesis (Morel et al 1991b;Price et al 1994;Cullen 1995) and the recent synthesis of data from the equatorial Pacific by Landry et al (1997) extend this explanation by invoking both iron limitation and grazing to explain the HNLC regions.…”

“…Indeed, ambient iron levels are low in high-nutrient low-chlorophyll (HNLC) areas (Martin et al 1991;Coale et al 1996b), and the addition of iron to incubation bottles has been shown to increase net growth rates, chlorophyll yields, nitrate uptake, and nutrient consumption, particularly of the larger cells Martin et al 1991;Price et al 1991Price et al , 1994Takeda and Obata 1995;Fitzwater et al 1996). The ecumenical hypothesis (Morel et al 1991b;Price et al 1994;Cullen 1995) and the recent synthesis of data from the equatorial Pacific by Landry et al (1997) extend this explanation by invoking both iron limitation and grazing to explain the HNLC regions. This hypothesis, in its broadest form, states that the cell division rates of the small phytoplankton that dominate the indigenous community are relatively fast and are ''not strongly'' limited by iron (not to the extent that is Ͻ1/2 max ; Cullen 1995).…”

“…These are relatively high values for a picoplankton-dominated community (Chisholm 1992;Cullen et al 1992), which suggests that these cells are growing at near maximal growth rates. Estimated ratios of new production to total primary production are low (McCarthy et al 1996), which indicates that most of this growth is fueled by regenerated nutrients, which are produced by the intensive grazing that keeps the picoplankton community in check (Wheeler and Kokkinakis 1990;Chavez et al 1991;Cullen et al 1992;Frost and Franzen 1992;Price et al 1994;Landry et al 1997;Loukos et al 1997).…”

“…In contrast, large cells with relatively small surface area to volume ratios are at a competitive disadvantage for NH and Fe assimilation be-ϩ 4 cause of diffusion limitation (Morel et al 1991a) and are iron limited. Therefore, transient iron enrichment significantly stimulates diatoms, which bloom because their mesozooplankton grazers cannot respond quickly enough to prevent an increase in cell numbers (Price et al 1994;Coale et al 1996b;Cavender-Bares et al 1999).…”