The global ocean can be divided into three contrasting regions with
respect to new production: regions where the stock of surface nitrate is
renewed each winter and depleted in the spring by biological utilization;
areas where high levels of nitrate persi st throughout the year; and large
regions in the oligotrophic gyres where nitrate stocks are permanently
depleted throughout the euphotic zone. The natural abundance of 15N
(d15N) in core top sediments appears to serve as an indicator of the
extent of NO
utilization in overlying surface waters [ Francois and
Altabet, 1994], and may help in understanding the history and distribution
of the relative strength of the nutrient supply mechanisms vs biological
utilization in each province.
NO
supplied by deep winter mixing triggers phytoplankton blooms driven by
new production in coastal and shelf regions [ Townsend et al., 1992;
Hansell et al., 1993] including the Southern Ocean [ Holm-Hansen
and Mitchell, 1991; Sullivan et al., 1993], marginal ice zones [
Smith, 1991] and in the north Atlantic [ Campbell and Aarup, 1992; Sambrotto et al.,
1993a; Takahashi et al., 1993]. Large-scale nutrient fluxes to the north
Atlantic supplied by the Gulf Stream [ Pelegri and Csanady, 1991] drive
the basin scale bloom revealed by Coastal Zone Color Scanner imagery. Blooms
are characterized by high rates of new production relative to the total
production (the f ratio; cf. [ Garside and Garside, 1993], and the
uncoupling of production and consumption processes [ Karl et al., 1991;
Banse, 1992; Dam et al., 1993] leading to episodic export of phytoplankton biomass
[ Honjo and Manganini, 1993; Ho and Marra, 1994]. Analysis of
ocean-wide ratios of nutrient regeneration indicate that organic matter reaching the
deep ocean is remineralized in fixed proportions [ Anderson and Sarmiento,
1994], suggesting that episodic export events might dominate the input of
organic matter to the deep sea. This view is not inconsistent with other
observations of primary production supported by nutrient utilization at
higher C:N ratios [ Laws, 1991; Banse, 1994] and export of
dissolved organic matter (see below) if it is assumed that the latter processes do not
export new production deeper than about 400 m. During the JGOFS North
Atlantic Bloom Experiment, new production was 5-8 mMol N m
day
[ Bender et al., 1992; Sambrotto et al, 1993a] but
only a small fraction was recovered in sediment traps [ Martin et al., 1993].
Large areas of the ocean in which surface nitrate stays high while
phytoplankton stocks are paradoxically low are termed ``high-nutrient,
low-chlorophyll'' or HNLC regions [ Cullen, 1991]. The subarctic north
Pacific and central equatorial Pacific are HNLC regions which have been
extensively studied in the past decade. Net oxygen production in the mixed
layer is becoming a useful and powerful tool for estimating new
production, partly because the contributions of physical and biological
processes to the O
budget can be discriminated with appropriate
tracers [ Emerson et al., 1991]. In the subarctic north Pacific, mass
balances of oxygen were used to estimate new production, and compared to
estimates from the 15NO
utilization rate, particulate export into
shallow sediment traps, and nitrogen mass balance [ Emerson et al., 1991,
1993a,b]. Different pairs of estimates differed by a factor of two or
more. At the current time, this level of uncertainty represents the state
of the art in estimating new product ion and export from the surface
layer. In general, new production is low in HNLC regions [ Dugdale et al.,
1992] but the reasons are still unclear. Intense grazing keeps
phytoplankton stocks low [ Frost, 1991; Frost and Franzen, 1992],
and ammonium excretion from the grazers inhibits nitrate uptake [ Wheeler
and Kokkinakis, 1990].
In the oligotrophic gyres, there is no measurable NO
at depths well below
the upper 100 m, and the mechanisms which supply NO
to the euphotic zone
and maintain new production remain unsolved. The two US JGOFS Time Series
stations located in the north Atlantic and Pacific gyres, are addressing
this problem. At Bermuda, CO
depletion by biological production in
spring-summer occurs in the virtual absence of any NO
, providing another
example of non-Redfield production [ US JGOFS, 1993; Keeling,
1993]. New production calculated by a variety of approaches including oxygen
mass balance [ Emerson et al., 1993b], greatly exceeded the annual export
caught in sediment traps, suggesting a major uncertainty in our capability to
close ocean carbon budgets. In the central north Pacific at the VERTEX
time series site (33N, 139W) new production was about 10% of the total
annual production and was balanced by export into shallow traps
[ Harrison et al., 1992]. Preliminary results of a mass balance of the oxygen field at
the Hawaii station suggest that sediment trap estimates of the particle
export do not balance the new production in the euphotic zone above
[ Emerson et al., 1993b]. Thus recent studies in all three `nitrate
provinces' of the global ocean suggest that sediment traps underestimate
the export required to balance the estimated new production, or that
export by other means than sinking particles must be factored into the
balance.
These studies show that new production continues at unequivocally
significant rates even in the most oligotrophic regions, in the apparent
absence of new NO
input from deep mixing. Some other sources have been
suggested. Buoyant mats of the diatom Rhizo solenia are enriched in NO
and might supply 50% of the annual N requirement to the euphotic zone
[ Villareal et al., 1993]. Atmospheric input of oxidized and reduced
nitrogen species to the global ocean total about. 20 x 10
gN annually
[ Duce et al., 1991]. This represents 1-2% of the global new production
(Table 1), suggesting that aerial deposition is not significant globally.
However in nutrient depleted waters of the central gyres, individual deposition
events could drive local blooms [ Michaels et al., 1993]. The air-sea
exchange of nitrogen deserves further study, and the balance is not always
clear. For example, the sea might be a net source of ammonia to the
atmosphere [ Zhuang and Huebert, 1994]. Atmospheric inputs of
micronutrients might also stimulate localized episodes of new production in
oligotrophic waters [ DiTullio and Laws, 1991]. The major beneficiaries of episodic
inputs of new nutrients, whether from above or below, may be large celled
diatoms with rapid growth and sinking rates. These cells respond rapidly
to nitrogen inputs, even at low light levels, leaving a chemical signature
in the form of increased oxygen and dissolved organic carbon (DOC), but
sink quickly, leaving little trace of their own biomass [ Goldman, 1993].