Heterotrophic prokaryotic production (BP) was studied in the western tropical
South Pacific (WTSP) using the leucine technique, revealing spatial and
temporal variability within the region. Integrated over the euphotic zone, BP
ranged from 58 to 120 mg C m
Heterotrophic prokaryotes can process, on average, 50 % of the carbon (C) fixed by photosynthesis in many aquatic systems (Cole, 1988). Understanding the controls of heterotrophic bacterial production and respiration rates is fundamental for two major aspects of marine C cycling: (i) to explore the possible fate of primary production through the microbial food web, and (ii) to construct a metabolic balance based on C fluxes. To assess these two major features, bacterial carbon demand (BCD; i.e., the sum of heterotrophic bacterial production (BP) and bacterial respiration (BR)) is compared to primary production (PP). The metabolic state of the ocean, and in particular the status of net heterotrophy within oligotrophic systems, has been largely debated in the last decade (see for example the review in Duarte et al., 2013; Ducklow and Doney, 2013; Williams et al., 2013).
Review of integrated primary production rates published in the South
Pacific, PP fluxes in mg C m
The South Pacific Gyre (GY) is ultra-oligotrophic, and is characterized by
deep UV penetration, by deep chlorophyll maximum (dcm) depth as great as
200 m, and by a 0.1
The waters coming from the GY are essentially transported by the South
Equatorial Current toward the Melanesian archipelagos in the western tropical
South Pacific (WTSP). Interest in this region has increased due to field and
satellite observation showing intermittent phytoplankton blooms in the area
associated with
While the dynamics of heterotrophic prokaryotes coupling with primary
producers has been explored in many regions of the ocean, these processes
have not been studied in the WTSP. Because most oligotrophic oceans are
nitrogen limited, PP and N
Position of stations during the OUTPACE cruise. The white line shows the vessel route (data from the hull-mounted ADCP positioning system). In dark green WMA (Western Melanesian Archipelago) including SD1, 2, 3 and LDA; in light green, EMA: Eastern Melanesian Archipelago including SD6, 7, 9 and 10 and in blue WGY (Western Subtropical Gyre) including stations SD13, 14, 15 and LDC. Figure courtesy of Thibaut Wagener.
In this study, we examined the horizontal and vertical distributions of
heterotrophic prokaryotic production alongside photosynthetic rates,
N
The OUTPACE cruise (
At the SD stations, water samples used for measuring in situ simulated
primary production (PP
In addition to the measurements of chlorophyll
For chlorophyll
The overall correlation between in vivo fluorescence (Chl iv) and Chl
Bacterial production (BP,
Primary production (PP) and dissolved inorganic phosphate turnover times
(
Measurements of PP using the JGOFs protocol (in situ moored lines immerged
for 24 h from dawn-to-dawn, IPP
Bacterial growth efficiency (BGE) and DOC lability were estimated at the
three LD sites using dilution experiments with seawater sampled in the mixed
layer. The seawater used for these experiments was sampled from Niskin
bottles (9 m at LDA, 7 m at LDB and 16 m at LDC) from a CTD cast done at
12:00 local time on the first day of occupation at each LD site. A
Samples for DOC concentration were filtered through two precombusted (24 h,
450
Rates of dark community respiration (DCR) were used to estimate bacterial
growth efficiency (see the discussion). Briefly, DCR was estimated from
changes in the dissolved oxygen (O
Physical and biological characteristics of main biogeochemical areas
and long duration stations sampled during the OUTPACE cruise. Depth of the
dcm (deep chlorophyll maximum, based on vertical profiles of in vivo
fluorescence),
Enrichment experiments were performed along vertical profiles at the three LD
sites LDA, LDB and LDC. Seawater was sampled at 12:00 local time on day 2 of
occupation at each site (CTD cast numbers 33, 117 and 166, respectively).
Nutrients were added in 60 mL transparent polycarbonate bottles at a final
concentration of 1
Relationships between variables were established using model II Tessier
linear regressions, from log-transformed data. Multiple regressions were also
used to study the simultaneous effects of PP and N
Distribution of primary production
The longitudinal transect started northwest of New Caledonia, crossed the
Vanuatu and Fiji Arcs and finished inside the western part of the
ultra-oligotrophic South Pacific Gyre. It covered a vast region of the WTSP
and the main gradient of biogeochemical and biological properties between the
Melanesian Archipelago (MA) (stations SD1 to SD12 and LDA) and the western
part of the South Pacific Gyre (WGY) (SD13 to SD15 and LDC) separated by the
Tonga Volcanic Arc (Fig. 1). Temperature ranged from 19.7 to 30.2
The mixed layers for most of the cruise were
Maximum primary production rates reached 20.8 mg C m
Vertical distributions of phosphate turnover times
(
Integrated primary production (IPP
Evolution of surface PAR
DIP turnover times (
Evolution of surface PAR
Site LDA showed variable dcm depth during the occupation time, with patches
of in vivo fluorescence moving up and down the water column with time over a
band of 40 m height (dcm depth varied between 63 and 101 m, Table 2).
However, the dcm depth corresponded to a stable density horizon (
Evolution of surface PAR
Site LDB, sampled inside a high chlorophyll patch, showed maxima of in vivo
fluorescence between 10 and 77 m, the chlorophyll maximum depth showing a
significant linear deepening with time (10.4
Site LDC, typical of the ultra-oligotrophic WGY area, presented a deeper dcm
depth, ranging from 115 to 154 m, due to internal waves (Fig. 7). Similarly
to site LDA, the dcm depth corresponded to a stable density horizon (
Results of multiple regressions log BP
ns: not significant.
Log–log relationships between volumetric rates of heterotrophic
prokaryotic production (BP) and primary production (PP,
There are several limitations when comparing PP
Log–log relationships between BP and PP
Integrated N
Results of multiple regressions log
ns: not significant.
Enrichment experiments. Initial conditions illustrated by vertical
profiles (0–200 m) of in vivo fluorescence, BP, nutrients (nitrate
(NO
We also examined relationships between
Results of biodegradation experiments. Growth rates determined from BP data, degradation rates computed from DOC data and BGE computed from Eq. (1).
In the three biodegradation experiments starting on day 1 at each LD site
using sub-surface waters, BP increased significantly, with growth rates
(determined from exponential phase of BP increase) ranging from 0.08 to
0.14 h
Conditions prevailing before enrichments are presented together with
enrichment factors obtained at the different depths tested as vertical
distributions of in vivo fluorescence, nutrients and BP sampled from a CTD
cast starting at 12:00 on day 2 of each LD site (Fig. 9). Nitrate
concentrations were below the detection limits of standard methods in upper
layers. The depth of the nitracline varied with the dcm depth at LDA and LDC
(100 m at LDA, 135 m at LDC), but not at LDB (a large peak of chlorophyll
was observed within 20–70 m with a nitracline at 100 m). Slight peaks of
nitrite also occurred in the vicinity of the nitracline. Phosphate
concentrations exhibited more contrasted vertical profiles than did nitrate:
DIP concentrations were greater than 100 nM in the surface layers of LDC,
but presented a phosphacline shallower than the nitracline at LDA and LDB,
with DIP reaching concentrations below the analytical detection limits in the
mixed layer (i.e., < 50 nM; see Moutin et al., 2018, for more
details on nutrient distribution). As DOC was not sampled on day 2, DOC data
are presented for the whole site instead. DOC peaked near the surface at site
LDA (77
At site LDA, nitrogen was the first factor stimulating BP down to 100 m
depth, which corresponded to the
dcm depth and a nitrite regeneration layer. Although significant at 9, 24, 35
and 100 m in depth (Mann–Whitney test,
At site LDB, between the surface and 42 m, both nitrogen alone and phosphate
alone stimulated BP to a larger extent than at site LDA, but only by a factor
At site LDC, BP reacted mostly to glucose alone, with enhancement factors
increasing from
Here, we provide a unique, coherent dataset with simultaneous estimates of
PP, BP,
Stations in the western part of the transect along the Melanesian Archipelago
(MA) generally displayed greater fluxes of PP, BP and N
Previous in situ measurements of primary production in the tropical South
Pacific, not directly focusing on coastal areas or within upwelling areas in
the east, are scarce (Table 1). These daily particulate primary production
rates, based on the
Although the metabolic state of oligotrophic oceans is still controversial
(Duarte et al., 2013; Ducklow and Donney, 2013; Williams et al., 2013; Serret
et al., 2015; Letscher et al., 2017), a consensus emerges that
Simultaneous estimates of PP, BP and N
It is known that the in vitro
Bacterial growth efficiencies (BGE) obtained from biodegradation experiments
ranged from 6 to 12 %, with a small labile fraction of DOC (only
2–5 % of biodegradable DOC in 10 days). Thus, the bulk DOC was mainly
refractory, although DOC concentration was high in the surface (Moutin et
al., 2018). Large stocks of DOC, with C
In order to better explain the variability of BGE measurements, we also
estimated this parameter indirectly, using simultaneously measured dark
community respiration (DCR) and BP data. We converted DCR to carbon units
assuming a respiratory quotient RQ
Distribution of integrated bacterial carbon demand corrected for
Although bias introduced when converting hourly to daily BP rates was not
studied here, we used a dataset obtained in the South Pacific Gyre (Van
Wambeke et al., 2008) to estimate conversion errors. During the BIOSOPE
cruise, vertical profiles of BP were acquired every 3 up to 72 h, using the
leucine technique within the euphotic zone at three selected sites using a
Lagrangian sampling strategy. For the three series of profiles, standard
deviations of BPI with time were 13 % (
Finally, we considered the ability of
Such comparisons between GPP and BCD would be more complex if the short-term
temporal variability in conversion factors was considered. For example, site
LDB illustrates how rapidly these relative fluxes changed during the collapse
of the bloom. LDB was located inside a massive chlorophyll patch, which had
been drifting eastwards for several months (de Verneil et al., 2017) and
which collapsed at the time LDB was sampled. Considering the decreasing
Chl
Phylogenetic analyses of the functional gene
Regardless of the intermediary processes relating both fluxes, the amount of
N
Consequently, most N
Below the surface layers, where
Finally, within the WGY area, where
Our results provide a unique set of simultaneous measurements of BP, PP and
N
All data and metadata are available at the French INSU/CNRS
LEFE CYBER database (scientific coordinator: Hervé Claustre; data manager
and webmaster: Catherine Schmechtig) at the following web address:
The authors declare that they have no conflict of interest.
This article is part of the special issue “Interactions between
planktonic organisms and biogeochemical cycles across trophic and N
We thank Sandra Helias, Olivier Grosso, and Mathieu Caffin for their support
in providing nutrient and N