Abstract:
Trichodesmium, a genus of diazotrophic cyanobacteria, is an important
contributor to the marine nitrogen (N) and carbon (C) cycles. The extent to which
Trichodesmium dinitrogen (N2) fixation contributes to the marine N cycle has been
modeled based on abundance data and rate estimates from surface populations. However,
recent data show that Trichodesmium populations have a broad vertical distribution. The
presence of previously unaccounted for subsurface populations suggests that past
estimates of the contribution of new N by Trichodesmium to the North Atlantic may be
artificially low. Herein, culture and field studies were combined to examine trends in N2
fixation in discrete surface and subsurface Trichodesmium populations in the western
North Atlantic. Surface populations were dominated by the raft colony morphology of
Trichodesmium and surface N2 fixation rates ranged from (33 to 156 μmol h-1 mol C-1).
Subsurface populations were dominated by the puff colony morphology. Subsurface N2
fixation was typically detectable, but consistently lower than surface population rates (9
to 88 μmol h-1 mol C-1). In an analysis of the entire field dataset, N2 fixation rates varied non-linearly as a function of in situ irradiance. This trend in N2 fixation versus in situ
irradiance is consistent with field and culture observations in the literature (Bell et al.,
2005; Capone et al., 2005), however other models that predict N2 fixation based on light
predict higher subsurface N2 fixation than what was detected in this study. In culture, N2
fixation in Trichodesmium was proportional to light level over the range of irradiances
tested (10 to 70 μmol quanta m-2 s-1) and over long and short time scales, suggesting
subtle changes in the light field could depress subsurface N2 fixation. Since the
subsurface samples were dominated by the puff colony morphology, it is unclear if the
subsurface N2 fixation rates are the result of the intrinsic responses of different species of
Trichodesmium, or light driven population segregation within a single species, among
other possibilities including the effects of temperature and nutrient availability.
Regardless, the subsurface rates presented herein indicate that N2 fixation by subsurface
populations represents an undersampled source of new N to the western north Atlantic.
This result is consistent with the findings of Davis and McGillicuddy (2006), who
suggest that subsurface populations of Trichodesmium increase the average N2 fixation
rate in the North Atlantic by 2.9 to 3.3 times over estimates based solely on surface
estimates (Davis and McGillicuddy, 2006).
