Environmental Engineering Reference
In-Depth Information
15
N during passage through the water
column. If so, then much of the flux at 100 m must be rapidly remineralized,
making a minimal contribution to the deeper flux. Few organisms feed actively
on
Trichodesmium
[47, 48], which is known to have toxic (allelopathic) effects
on both phytoplankton and zooplankton [26, 28, 29, 30]. Allelopathic com-
pounds produced by
Trichodesmium
may also reduce the rate of degradation
and remineralization of sinking organic material derived from
Trichodesmium
blooms at the surface, which in turn would result in an increasing relative con-
tribution of
Trichodesmium
to the sinking flux with greater depth. The contrast
between the 100 m and 500 m traps would then imply that the deeper trap is
much more effectively sampling particles originating at the very surface, where
Trichodesmium
dominated both the biomass and the isotopic signature of the
suspended particles.
Our time series at 18˚N shows a similar decrease in the δ
ization of organic matter with a high δ
15
N of sinking
particles between the traps at 100 and 500 m. In this case, the material collected
by the trap at 100 m was enriched in
15
N by about 1 to 2‰relative to the particle
field in the upper 100 m of the water column.
Trichodesmium
was present in the
upper water column at this station, though not at concentrations approaching
those we saw at 10˚N (Fig. 3). We deployed only one trap at 100 m depth at
our short (2 day) time series at 3˚N. The sinking particles at this depth have
a δ
15
N very similar to the mean value for particles in the upper 100 m of the
water column.
We did not collect sinking particles at greater depth during this study, but
the data from our time series at 10˚N clearly show the potential for N
2
-fixation
in the surface layer to have a strong local impact on the δ
15
N of sinking
particles at depth. Interestingly, the diazotroph signature is expressed more
clearly at 500 m than at 100 m depth, likely reflecting differences in sinking
rate and extent of degradation of organic matter from different sources (i.e.,
Trichodesmium
vs. other primary producers). Taken together, the sediment trap
samples from our three time series stations suggest that the particles leaving
the surface mixed layer are imprinted with the mean isotopic composition of
the upper water column, and that a substantial fraction of the particle flux at
100 m is remineralized during transit through the upper part of the OMZ. Since
NH
4
+
doesn't accumulate in the water column, most or all of the remineralized
nitrogen must be nitrified, adding NO
3
−
with a relatively low δ
15
Ntothe
dissolved pool. In effect, the sinking flux of organic matter will tend to reduce
both the isotopic and the mass effects of denitrification on the NO
3
−
pool in
the upper part of the OMZ.
