Coexisting picoplankton experience different relative grazing pressures across an ocean productivity gradient

Abstract

Picophytoplankton populations [Prochlorococcus, Synechococcus (SYN), and picoeukaryotes] are dominant primary producers in the open ocean and projected to become more important with climate change. Their fates can vary, however, with microbial food web complexities. In the California Current Ecosystem, picophytoplankton biomass and abundance peak in waters of intermediate productivity and decrease at higher production. Using experimental data from eight cruises crossing the pronounced CCE trophic gradient, we tested the hypothesis that these declines are driven by intensified grazing on heterotrophic bacteria (HBAC) passed to similarly sized picophytoplankton via shared predators. Results confirm previously observed distributions as well as significant increases in bacterial abundance, cell growth, and grazing mortality with primary production. Mortalities of picophytoplankton, however, diverge from the bacterial mortality trend such that relative grazing rates on SYN compared to HBAC decline by 12-fold between low and high productivity waters. The large shifts in mortality rate ratios for coexisting populations are not explained by size variability but rather suggest high selectivity of grazer assemblages or tightly coupled tradeoffs in microbial growth advantages and grazing vulnerabilities. These findings challenge the long-held view that protistan grazing mainly determines overall biomass of microbial communities while viruses uniquely regulate diversity by "killing the winners".

Keywords: Synechococcus; grazing mortality; growth rate; heterotrophic bacteria; shared predation.

Fig. 1.

Relationships between nitrate concentration and temperature (A), primary production and nitrate (B) and primary production and chlorophyll a (C) during CCE process cruises in the southern CCE. Data are the averages of 3 to 4 depths of sampling in the upper euphotic zone (upper 26 m on average) at each experimental location; uncertainties are SEM values. Regression slopes with P values ≤ 0.05 are significant. Full regression statistics are in SI Appendix, Table S1.

Fig. 2.

Relationships of picoplankton population abundances (A–D) and cell growth rates to measured PP (E–H) for CCE Process cruises. HBAC = heterotrophic bacteria; PRO = Prochlorococcus; SYN = Synechococcus; PEUK = photosynthetic picoeukaryotes. Cruise abbreviations correspond to Table 1. Data are mean ± SEM estimates for the upper euphotic zone. For multiple panel regressions, P values are color-coded to the lines plotted: PP < 100 (black), PP > 10 (red), and PP > 100 (blue). Full regression statistics are in SI Appendix, Table S1.

Fig. 3.

Relationships of picoplankton grazing mortality (A–D) and mortality ratios (E–G) to measured PP in the southern CCE. Cruises and populations as defined in Fig. 2 and Table 1. Regression slopes with P values ≤ 0.05 are significant. Full regression statistics are in SI Appendix, Table S1.

Fig. 4.

Potential mechanisms to explain observed shifts in grazing mortality on phototrophic picoplankton (SYN and PEUK) relative to heterotrophic HBAC across trophic gradients in the CCE. (Upper) Bead-normalized FALS used as an index of HBAC cell size (panel A) and mean size ratios of SYN:HBAC and PEUK:HBAC (panels B and C). (Lower) Relationships of grazing mortality ratios SYN:HBAC (panel D) and PEUK:HBAC (panel E) with relative prey abundances of phototrophic and heterotrophic cells. Full regression statistics are in SI Appendix, Table S1.