TEP and DOC production during a bloom experiment

TEP and DOC production during a bloom experiment

TEP and DOC production during a bloom experiment
with Emiliania huxleyi exposed to different CO2 concentration
Anja Engel1 , Bruno Delille², Stephan Jacquet³, Ulf Riebesell1 ,
Emma Rochelle-Newall4 , Anja Terbrüggen1 and Ingrid Zondervan1
1 Alfred
Wegener Institute for Polar- and Marine Research, 27515 Bremerhaven, Germany
2 Unité
3 Station
4 Laboratoire
d'Océanographie Chimique, Université de Liège, 4000 Liège, Belgium
INRA d'Hydrobiologie Lacustre, 74203 Thonon-les-Bains cedex, France
d'Océanographie de Villefranche sur Mer, 06234 Villefranche-sur-Mer, France
Keywords: Emiliana huxleyi, TEP, DOC, carbon overconsumption, CO2, Redfield
ratios
1
Acknowledgments
- will be written after revision-
Abstract
The role of transparent exopolymer particles (TEP) and dissolved organic carbon (DOC)
for organic carbon partitioning under different CO2 conditions was examined during a
mesocosm study with the coccolithophorid Emiliania huxleyi. Nine outdoor enclosures
(~11m³) were modified to simulate ‘glacial’, ‘present’ and ‘year 2100’ CO2 environments
and fertilized with nitrate and phosphate in order to favor bloom development of E. huxleyi.
Extracellular organic carbon in form of DOC and TEP was determined over a period of 19
days, covering the pre-bloom and bloom period of E. huxleyi. In all of the mesocosms, an
uptake of approximately 60% more dissolved inorganic carbon (DIC) than inferred from
nitrate utilization and Redfield stoichiometry occurred and was largely traced to the
particulate organic carbon (POC) pool. TEP concentration increased after nutrient
exhaustion and accumulated steadily until the end of the study. TEP concentration was
closely related to the abundance of E. huxleyi and accounted for approximately 33% of
POC increase. Highest TEP production rates per cell were observed in the ‘year 2100’ CO2
treatment. DOC concentration was highly variable over time and neither related to E.
huxleyi abundance nor to TEP concentration. DOC concentration increased significantly
over time in two of the ‘year 2100’ mesocosms, in one ‘present’ mesocosm, but in none of
the ‘glacial’ mesocosm. Although our results showed that the role of DOC and TEP for
carbon partitioning was fundamentally different during the E. huxleyi bloom, they indicate
that production of both, DOC and TEP, may be sensitive to CO2 concentration.
2
Introduction
An important mechanism for the regulation of atmospheric CO2 concentration is the
fixation of CO2 by marine phytoplankton and the subsequent export of the organically
bound carbon to the deeper ocean. Following the ideas of Redfield et al. (1963) and Eppley
and Peterson (1979), the magnitude of organic carbon export is expected to depend on the
availability of major nutritional elements in the surface ocean and can be estimated from
nitrate uptake, using a C:N ratio of 106:16. However, the applicability of the ‘Redfield’
ratio for calculation of oceanic carbon fluxes has been challenged recently, since the draw
down of dissolved inorganic carbon (DIC) exceeds the amount expected from nitrate
removal and Redfield stoichiometry seasonally (Sambrotto et al. 1993; Michaels et al.
1994, Marchal et al. 1996, Thomas et al. 1999, Körtzinger et al. 2001). This observation
was referred to as ‘carbon overconsumption’ by Toggweiler (1993). Since then a number of
hypotheses were raised to explain carbon overconsumption, including the underestimation
of new production due to unaccounted for biological N2-fixation (Michaels et al. 1996,
Hood et al. 2001), the temporary accumulation of carbon-rich dissolved organic matter
(DOM) (Kähler & Koeve, 2001), preferential nutrient recycling (Thomas et al. 1999) or the
formation of carbon rich extracellular particles, known as transparent exopolymer particles
(TEP) (Engel et al. 2002a). To which extent these processes are responsible for the excess
DIC uptake in the field has yet to be determined and may depend on the area or season
considered. A question of particular importance in this context is the fate of excess carbon,
i.e. to what extent it will be exported below the winter mixed layer and hence removed
from exchange with the atmosphere for more than one year. Thus, it is of special interest
whether a mechanism mediating carbon overconsumption has the capability to also account
for a deep carbon export.
3
Considering carbon cycling at the cellular level, it is well known that the uptake of carbon
continues when nutrient acquisition limits cell division but not primary production. One
consequence of this assimilation of excess carbon is the release of extracellular organic
matter (EOM) (Fig.1) (Fogg, 1966, Wood and Van Valen, 1990). Although the mechanisms
of EOM release has not been fully elucidated yet, it can be assumed that low molecular
weight (LMW) substances, such as monomer or oligomer sugars penetrate the cell
membrane by diffusion. The rate of this LMW-EOM leakage should depend on the
concentration gradient between the inner and outer cell. The release of high molecular
weight (HMW) substances by diffusion is not possible, and has to be accomplished by
active exudation. Polysaccharides, for example, are synthesized in the vesicles of the Golgi
apparatus and secreted to the outer cell by exocytosis (see review by Leppard 1995). Since
EOM release is a possibility for the cell to dispose of excess photosynthesates under
nutrient deplete conditions, EOM should not contain more than negligible amounts of the
limiting element (Wood and Valen 1990). In this respect EOM release can be viewed as a
cellular carbon overflow. Whether this cellular carbon overflow is responsible for carbon
overconsumption in the field has yet to be determined. It is well known that EOM release
by autotrophic cells is an important source for dissolved organic carbon (DOC) in the upper
ocean (Alluwihare et al. 1997) and the production of DOM with high C:N ratios has
frequently been observed (Williams 1995, Kähler and Koeve 2000, Søndergaard et al.
2000), supporting the idea that DOC production results from a cellular carbon overflow.
Yet, the deep export of DOC is principally restricted to subduction of surface waters, e.g.
by thermohaline ventilation. Because this process operates on long times scales, i.e. months
to years, much of the seasonal accumulated DOC will likely be degraded before it arrives at
greater depths.
4
The major fraction of HMW-EOM are polysaccharides (Benner 2002), some of them
contain acidic sugars that facilitate polysaccharide aggregation into particles, known as
transparent exopolymer particles (TEP) (Alldredge et al. 1993, Leppard 1995, Engel et al.
submitted). TEP are therefore naturally rich in carbon but poor in nitrogen (Engel &
Passow 2001, Mari et al. 2001). Especially, when nutrients become limiting, TEP occur in
phytoplankton cultures, during experimental phytoplankton blooms, as well as in natural
environments (see Passow 2002 for review) and are therefore regarded as a result of the
cellular carbon overflow (Engel 2002, Engel et al. 2002). Because they represent a fraction
of the particulate organic matter (POM), a relative increase of TEP can induce a shift in
POC:PON ratios during diatom blooms (Engel et al. 2002a). In contrast to DOC, TEP can
participate in particle mediated processes such as marine snow formation and sinking
(Alldredge et al. 1993, Passow et al. 2001) and have therefore the potential to account for a
deep export of carbon on relatively short time scales.
In the ocean, EOM release was found to be related to primary production (Baines and Pace,
1991). Since primary production of marine phytoplankton is sensitive to CO2
concentrations (Rost et al. 2003), one might speculate that EOM release is not only
responsible for, but also mediates a CO2 effect on organic carbon production.
We
investigated these hypotheses during a mesocosm bloom study with Emiliana huxleyi,
exposed to three different CO2 concentrations, focusing on a) the temporal changes in EOM
concentration, respectively DOC and TEP, b) the role of DOC and TEP for storage of
excess carbon during an E. huxleyi bloom and c) the influence of seawater CO2
concentrations on DOC and TEP production.
5
Material & Methods
Set-up and sampling.
This study was conducted at the EU-Large Scale Facilities (LSF) in Bergen, Norway, as
part of the outdoor-mesocosm project ‘Biological responses to carbon dioxide-related
changes in seawater carbonate chemistry during a bloom of the coccolithophorid Emiliana
huxleyi’. A detailed description of the experimental set-up will be reported elsewhere
(Zondervan et al. in prep.). Briefly, nine polyethylene enclosures (~ 11 m3 , 4.5 m water
depth) were moored to a raft in a fjord (for more details see Williams & Egge, 1998). The
bags were filled with unfiltered, nutrient-poor, post-bloom fjord water, which was pumped
from 2m depth adjacent to the raft. The enclosures were covered by gas-tight tents made of
ETFE foil, which allowed for 95% light transmission of the complete spectrum of sunlight.
The atmospheric and seawater pCO2 were manipulated to achieve 3 different CO2 levels in
triplicate, corresponding to approximately year 2100, assuming the IPCC's 'business as
usual' scenario IS92a- (mesocosms 1-3), present (mesocosms 4-6) and glacial atmospheric
CO2 levels (mesocosms 7-9), respectively (Delille et al., in prep.). To promote the
development of a coccolithophorid bloom, nitrate and phosphate were added in a ratio of
30:1 yielding initial concentrations of 15 µmol L-1 NO3 and 0.5 µmol L-1 PO4. After nutrient
addition, the water was gently mixed by means of an airlift (for more details see Egge &
Asknes, 1992), using the same air as for gassing the tents. Over a period of three weeks
samples were taken daily from each mesocosm by gentle vacuum pumping of 20 L through
a siphon at 0.5 m depth. After day 16 large particle aggregates (> 0.5 cm; 'marine snow')
appeared in the mesocosms and were abundant enough to be collected manually with a
syringe in the upper 0.5m of the water column on day 17.
6
Biological and chemical analyses. Nitrate and nitrite were determined from GF/F-filtered
and poisoned (0.1% HgCl2) samples with an autoanalyser (AA II) at the home laboratory.
Phosphate and ammonium were measured on the day of sampling using the methods of
Koroleff & Grasshof (1983). Particulate organic carbon (POC) and particulate organic
nitrogen (PON) were determined by mass spectrometry (ANCA SL 20-20, Europa
Scientific) from 1 L (day 0-12) and 0.5 L (day 13-19) samples filtered gently (200 mbar)
through precombusted glass fiber filters (GF/F, Whatman). Particulate organic phosphorus
(POP) was determined colorimetrically (Koroleff & Grasshof, 1983) after persulfate
oxidation from 0.5-1.0 L samples filtered onto GF/F filters. All filters were prepared in
duplicates and stored at –20°C until analysis.
Samples for DOC analysis were collected in glass ampoules after filtration through
precombusted GF/F filters. The samples were poisoned with 85% H3PO4, flame sealed
immediately after collection and stored until measurement at 4°C in the dark. The DOC
analysis was performed using high temperature combustion on a Shimadzu TOC-5000 total
organic carbon (TOC) analyser. A four point calibration curve was constructed for each
measurement day using potassium phthalate standards prepared fresh in UV-treated Milli-Q
-1
water. The standards covered the range 0 to 200 µmol C L . In order to assess the
instrument blank we used two external standards (Certified Reference Standards, CRM`s)
obtained from the Hansell Laboratory, Bermuda Biological Station. The machine blank was
between 8 – 12 µmol L-1 C for all samples and was subtracted from the measurements. All
DOC concentrations reported are the average of three injections from each sample.
TEP are detected by staining with Alcian Blue (Fig. 2), a cationic copper phthalocyanine
dye that complexes carboxyl (–COO–) and half-ester sulfate (OSO3–) reactive groups of
7
acidic polysaccharides. The amount of Alcian Blue adsorption per sample volume is a
measure for TEP concentration and was determined colorimetrically according to Passow
and Alldredge (1995) from 50-100 ml samples filtered onto 0.4 µm Nuclepore filters. All
filters were prepared in triplicate. The carbon content of TEP was determined following the
approach of Engel and Passow (2001). Three aliquots of 5 L (pooled samples of
mesocosms 1-3, 4-6 and 7-9, respectively) were collected on 7 days throughout the bloom
and filtered through precombusted glass fiber filters. TEP were generated from the filtrate
during 24 h of circulation through a Tangential Flow Filtration (TFF) system with a 0.16
µm membrane. The fraction >0.16 µm was concentrated from 5 L to a final volume of 1- 2
L. The concentrated samples were analyzed for carbon, nitrogen and TEP concentration.
TEP were measured colorimetrically (Passow & Alldredge 1995) from 100-200 ml, with at
least 2 replicates each. POC and PON were determined from 0.8-1.6 L as described above.
All materials in contact with the sample were either autoclaved or acid (10% HCl) rinsed.
Blank glass fiber filters were prepared for each filtration series. In 10 samples the carbon
concentration was below the detection limit of 30 µg L
-1
. For the remainder of the
samples, the slope of the regression of POC concentration versus colorimetrically
determined TEP concentration (in µg Xanthan Equivalents (Xeq.) L-1 ) was calculated with:
[POC, µg ] = 0.39±0.08 [TEP, µg Xeq.] (r²=0.73, n=11, p<0.005). No correlation was
observed between TEP and PON, indicating that nitrogen is not a major element of TEP as
suggested before (Engel & Passow 2001).
Cell counts of E. huxleyi were performed with a FACSCalibur flow-cytometer (Becton
Dickinson) equipped with an air-cooled laser providing 15 mW at 488 nm and with a
standard filter set-up. The algae were analysed from fresh samples at high flow rate (~70 µl
min-1 ) with the addition of 1 µm-fluorescent beads (Molecular Probes). E. huxleyi cells
were discriminated on the basis of their Forward or Right Angle Light Scatter and
8
chlorophyll fluorescence. Listmode files were analysed using CYTOWIN (Vaulot 1989)
and WinMDI (version 2.7, Trotter).
Statistical treatment of data
Average values are given by the statistical mean ( x ) and its standard variation (SD). In
order to determine the significance of a regression coefficient, i.e. the slope of a linear
relationship (d[y]/d[x]), a t-test according to Sachs (1974) was performed. Significance
was accepted for values of p<0.1 .
Results
General bloom development
At day 1 of the experiment, seawater CO2 concentrations were adjusted to 741±47 ppmV in
the 'year 2100' scenario of mesocosms 1-3, to 414±11 ppmV in the 'present' enclosures 4-6
and to 190±2.4 ppmV in the 'glacial' mesocosms 7-9. More detailed information about the
seawater carbonate chemistry during the experiment will be given elsewhere (Delille et al.
in prep.). Increase of biomass was detectable after day 3 in all of the mesocosms and after
day 10 the phytoplankton community was dominated by number by E. huxleyi. In opposite
to the simultaneously increase in biomass, inorganic nutrients declined rapidly. Phosphate
dropped from an initial concentration of ~ 0.5 µmol L-1 to below detection limit after day
10. Nitrate started at ~ 15.5 µmol L-1 and was not detectable after day 13. Ammonia was
undetectable during the whole study. Maximum concentrations of POP and PON were
observed on day 13 and 14, respectively, and declined steadily thereafter. Cell abundance
of E. huxleyi increased exponentially after day 3, but maximum cell concentrations varied
considerably between mesocosms (Zondervan et al., in prep.). Cell growth was on average
9
0.45 d-1, equivalent to a cell doubling every 2.2 days. This value is considerable lower than
maximum growth rates, observed for nutrient replete cells (µmax=0.76 d-1) (Riegmann et al.
2000) and indicates that E. huxleyi was growth limited during the bloom development.
Average cell abundance decreased after day 19, possibly due to viral lysis (Dellile et al., in
prep.). Therefore, only data before the collapse of the bloom (day 19) will be presented
here.
Transparent exopolymer particles (TEP)
Mean TEP concentration at the beginning of the study was 166±34.8 µg Xeq. L–1, stayed
rather constant until day 10 and increased rapidly thereafter in all of the mesocosms (Fig.
3a-c). Between mesocosms, variability of TEP production was small at the beginning of
the experiment, but high during the late bloom, yielding a wide range of TEP
concentrations on day 19. During the study, TEP accounted for an increase of POC
concentration on the order of 40.8±17.5 µmol L-1. TEP concentrations were closely related
to the cell abundance of E. huxleyi with d[TEP]/d[cell] yielding 32.3±0.92 pg Xeq. cell-1 or
1.05±0.03 pmol TEP-C cell-1 or 12.6±0.36 pg TEP-C cell-1 (Fig. 4). The production rate of
TEP-C per cell can therefore be estimated with 5.7 pg TEP-C cell-1 day-1.
Slight but significant differences in the ratio d[TEP]/d[cell] were observed between the
mesocosms. The highest increase of TEP concentration with cell abundance occurred in
the ‘year 2100’ mesocosms 1 and 3 (table 1). In the ‘present’ and ‘glacial’ treatments TEP
production per cell was lower and more uniform.
Dissolved organic carbon (DOC)
High day to day variations were observed for DOC concentrations in the course of the
experiment (Fig. 5a-c). A significant increase in DOC concentration with time (d[DOC]/dt,
10
t: days 1-18) was observed for the ‘year 2100’ mesocosms 1 (1.30±0.44 µmol L-1 d-1) and 2
(1.11 ±0.17 µmol L-1 d-1) and for the ‘present’ mesocosm 4 (1.27±0.30 µmol L-1 d-1). In the
others and in all of the ‘glacial’ mesocosms no significant increase of DOC occurred within
the 18 days of observation (no samples were taken on day 19), indicating that degradation
counterbalanced the production of DOC. DOC concentration was not related to the
abundance of E. huxleyi cells (p=0.13) (Fig. 6). Furthermore, no significant relationship
between TEP-C and DOC concentrations were observed (p=0.10).
Particulate carbon dynamics
During the first two weeks, changes in POC concentration were closely related to changes
in PON and POP concentrations, with d[POC]/d[PON] equal to 6.8±0.3 mol mol-1, a value
which is in accordance with the expected Redfield value of 6.6. In contrast, changes of
POC relative to POP were more than twice as high as the Redfield value, yielding
d[POC]/d[POP] equal to 338±13 mol mol-1, which underlines the exceptional ability of E.
huxleyi to grow with low cellular P (Riegman et al., 2000). At the same time, changes in
PON were tightly related to POP resulting in d[PON]/d [POP] of 48.0 ±1.3 mol mol-1. After
day 13, POC production was clearly decoupled from POP and PON. While POC
concentration more than doubled, PON and, even more pronounced, POP concentrations
decreased, leading to a steep rise of [POC]:[POP] and [POC]:[PON] concentration ratios
(Fig. 7a, b). The decoupling of POC from PON production coincided with DIN exhaustion
(Fig. 8), indicating that a large amount of carbon was assimilated under nutrient deplete
conditions and accumulated in the POC pool. Contribution of TEP-C to POC production
was roughly 16% (p<0.05) within the first 10 days of the study, but was responsible for
approximately 35% of the POC increase (p<0.001) after nutrient depletion (Fig. 9).
11
Marine snow formation
Amorphous marine snow appeared in the mesocosms on day 16 and was sampled on day
17. The amount of TEP within marine snow was high and TEP-C comprised about 38-55%
of POC. The [POC]:[PON] ratios were accordingly large and ranged between 9.9 and 35.
Microscopic examination showed that marine snow was mainly composed of a TEP matrix
with particles entangled. The [PON]:[POP] ratios of marine snow ranged between 59 and
119, which, compared to the range of 26-65 for [PON]:[POP] ratios of suspended particles,
indicate a preferential release of phosphorus from marine snow. This rapid enzymatic
degradation of POP in marine snow has been observed previously (e.g. Smith et al 1992,
Engel et al. 2002) and can be explained by increased activities of the ectoenzyme alkaline
phosphatase in marine snow (Smith et al 1992, Grossart & Simon 1998).
Discussion
TEP dynamics during the E. huxleyi bloom
Although there has been no documentation about TEP production by coccolithophorids
hitherto, the production, composition and release of an acidic polysaccharide is well
documented for E. huxleyi (De Jong et al. 1976, Van Emburg et al. 1986, Nanninga et al.
1986). Similar to TEP, this coccolithophorid polysaccharide (CP) was detected by staining
with Alcian Blue (De Jong et al. 1976, Fichtinger-Schepman et al, 1979). The composition
of CP includes mainly neutral sugars, such as manose, rhamnose and xylose. About 20% of
the total sugar content of CP is represented by D-galacturonic acid, an acidic sugar, which
can mediate the aggregation of CP chains, because the carboxyl group of one Dgalacturonic acid can align to another by divalent cation (Ca2+) bridging (Leppard 1995).
12
CP has been isolated from coccoliths of E. huxleyi (De Jong et al. 1976), but also found as
an dissolved polysaccharide in culture media (Naninga et al. 1996). It is assumed that CP
plays a role in the biomineralisation process and probably also in the agglutination of
coccoliths in the coccosphere (Van Emburg et al. 1986). From microscopy of TEP filters,
we, in fact, noticed that in addition to TEP the surface of E. huxleyi coccoliths was stained
by Alcian Blue (Fig. 2). Since the colorimetric method can not discriminate between acidic
polysaccharides contained in TEP or in CP coating the coccoliths, the latter will be
included in the result. In order to estimate the contribution of this coating CP to TEP
concentration, we determined the Alcian Blue adsorption and cell number in five samples
prepared by diluting a nutrient replete E. huxleyi culture with 0.2 µm filtered artificial
seawater. Alcian Blue adsorption of cells was equivalent to 2.59±0.40 pg Xeq. Cell-1 (Fig.
10) and hence small compared to the TEP to Cell ratio within the mesocosm study. Thus
the strong relationship between the abundance of E. huxleyi and TEP concentration during
the mesocosm study indicated a tight coupling between the release of dissolved CP and
TEP formation. This interpretation is supported by model calculations showing that the
dynamics of dissolved polysaccharide and TEP concentrations during this study follow
aggregation kinetics (Engel et al. submitted). It is striking that the amount of TEP-C
produced per cell (12.6±3.6 pg TEP-C cell-1) can exceeded the amount of organic carbon
stored intracellular, e.g. 7.83±1.26 pg C cell-1 as determined by Riegman et al. (2000).
Nanninga et al. (1996) measured the release of 0.56 pg CP cell-1 in a culture of calcifying E.
huxleyi, which is equivalent to 0.22 pg C cell-1 if a carbon content of CP of 39%
(weight:weight) (Fichtinger-Schepman et al. 1979) is assumed. Hence, the authors
concluded that the release of CP contribute little to organic carbon production during an E.
huxleyi bloom. However, the authors calculated the production rate only for dissolved CP
that was released by nutrient saturated cells, even though they observed that CP production
increased after cell division terminated. Our study showed that a much larger fraction of CP
13
accumulated as particulate material, i.e. TEP. We therefore conclude that the amount of
organic carbon released by E. huxleyi in form of polysaccharides can be of high importance
for organic carbon cycling and may, under the condition of nutrient limitation, even exceed
POC production by cell growth.
It is long-standing knowledge that the concentration of particulate carbohydrates greatly
increases in the course of phytoplankton blooms (Mc Allister et al. 1961, Barlow 1982).
However, these changes were attributed to an intracellular increase of sugars, hitherto. This
study showed that the extracellular formation of TEP may explain much of this particulate
carbohydrate increase, at least during E. huxleyi blooms.
DOC dynamics during the E. huxleyi bloom
DOC in seawater is produced by various mechanisms, including exudation by
photoautotrophic cells (Fogg 1983), enzymatic solubilization of particles (Cho and Azam,
1988, Karner and Herndl, 1992), cell lysis (Fuhrmann, 1999) or sloppy feeding by metazoa
(Copping and Lorenzen, 1980, Nagata 2000). In cell cultures it has been observed that
exponentially growing E. huxleyi produce about 0.12 pmol DOC cell-1 day-1 (Biddanda and
Benner, 1997), which would have resulted in a daily production of approximately 2.5 µmol
DOC L-1 between day 12 and day 17 during this study. However, we did not find a
relationship between cell abundance and DOC production, but a high variability of DOC
concentrations over time, indicating that production and consumption of DOC was tightly
coupled during all stages of the E. huxleyi bloom. The largest fraction of DOC released by
E. huxleyi are carbohydrates (Biddanda and Benner, 1997). Because a large fraction of
released polysaccharides accumulated in TEP, we assume that the changes in DOC
concentration rather reflected the turn-over of labile, LMW carbohydrates and a slow
14
accumulation of refractory substances.
Influence of TEP and DOC on carbon dynamics
Carbon uptake and flow to the various pools of organic carbon was examined during the
bloom of E. huxleyi and is summarized in table 2. Comparison of observed carbon flows
with those expected from changes of nitrogen concentration and Redfield stoichiometry
allows the assessment of ‘carbon overconsumption’ during the bloom and the relative
importance of the different carbon pools for the storage of excess carbon.
As derived from the decrease in DIC concentrations (data will be given by Zondervan et
al.) about 60% ‘carbon overconsumption’ occurred within the first 18 days of the study.
The largest fraction of the excess carbon was traced in the POC pool. Approximately 4.6%
of the excess carbon accumulated in the DOM pool, which is an upper estimate, since DON
was not determined. Carbon contained in TEP explained the fate of 15.8% of
overconsumed carbon and was responsible for more than 50% of excess POC. These results
are in accordance with the observations of Engel et al. (2002), who observed that the largest
fraction of excess carbon was channeled to TEP during a mesocosm diatom bloom.
Compared to the mesocosm study of Engel et al. (2002), this study further showed that a
significant production of TEP and relatively high contributions of TEP-C to POC
concentrations are not exclusive to eutrophic systems with high phytoplankton biomass but
can occur in low nutrient environments, which are more comparable to oceanic conditions.
This conclusion is supported by observations from the Baltic Sea, where the relative
contribution of TEP-C to POC was observed to be higher in the nutrient poor central Baltic
Sea during summer, than at the coastal Baltic Sea in the course of the spring bloom (Engel
et al., 2002b).
15
Almost half of the carbon and nitrogen, which had been taken up during the present study,
were not recovered in the POM and DOC pool. We assume that this loss of organic matter
was most likely due to sedimentation of particles, which was also indicated by a huge
amount of ‘sediment’ pumped from the bottom of the enclosures at the end of the
experiment. The loss of carbon relative to the loss of nitrogen, made up for a C:N ratio of
13.6, which is an lower estimate, since some of the missing nitrogen may have entered the
DON pool. Accordingly, we can estimate a minimum ‘sedimentation’ of excess carbon of
~25%. This clearly shows that high C:N ratios of sinking material during a phytoplankton
bloom can be explained by a selective enrichment of carbon, as a consequence of ‘carbon
overconsumption’. High C:N ratios of the lost particles were in agreement with high C:N
ratios of marine snow, observed during this study. As for marine snow, we can assume that
TEP was partly responsible for the excess carbon in the sinking matter. However, C:N
ratios of POM, from which TEP-C was subtracted, were considerably above the Redfield
ratio also. This may be explained by intracellular carbon storage, which was observed for
E. huxleyi in culture experiments (Riegman et al. 2000). Another process potentially
responsible for the formation of POM with high C:N ratios would be the preferential
remineralisation of nitrogen from detrital particles. Because nitrogen regeneration supports
primary production simultaneously with nitrate, an underestimation of regenerated nitrogen
could be responsible for ‘carbon overconsumption’, provided that the organic carbon
produced in this way would enter the export flux (Thomas et al. 1999). Thus, TEP
production and preferential remineralisation of PON in the euphotic layer potentially have
the same impact of POC:PON ratios and on ‘carbon overconsumption’. We can safely
assume that neither process will occur exclusively and this study indicates that the
influence of TEP production and preferential nitrogen recycling on POC:PON ratios may
even be of equal importance. However, it is the special property of TEP to accelerate
marine snow formation and therewith to provide a vehicle for a deep export of particulate
16
matter.
Sensitivity of TEP and DOC production to CO2 concentration
There are almost no previous studies about the direct effects of CO2 on carbon exudation,
extracellular carbon dynamics or concentrations. In incubation experiments with natural
plankton harvested from the central Baltic Sea, Engel (2002) observed that final TEP
concentrations were related to initial seawater CO2 concentration. Similar results were
obtained during incubation experiments with monospecific cultures of T. weissflogii and a
non-calcifying strain of E. huxleyi (Heemann 2002, Heemann et al., in prep.).
During this study TEP production rates normalized to cell number were significantly higher
in two of the three high CO2 mesocosms, than in the ‘present’ and ‘glacial’ CO2treatments. This indicates that a direct effect of CO2 on polysaccharide exudation and TEP
formation, as suggested from fully enclosed systems and culture experiments, may also
emerge in more complex plankton communities, like in the mesocosms. Release of
carbohydrates from autotrophic cells contributes also to the build up of the DOC pool, and
the observation that DOC increased in the ‘year 2100’ and ‘present’ treatment, but not in
the ‘glacial’ treatment may point to a stimulating CO2 effect on phytoplankton exudation
also. The hypothesis that in particular carbon release by marine phytoplankton is sensitive
to CO2 concentration can be supported by theoretical arguments. First, as primary
production is the first order process to control organic carbon release under nutrient
limiting conditions, factors influencing primary production will consequently affect organic
carbon release. For marine phytoplankton, an influence of CO2 on primary production is
well known (Riebesell et al. 1993, Rost et al. 2003). Second, the onset of organic carbon
release, e.g. at the end of a phytoplankton bloom, is the time when seawater CO2
17
concentration has already been reduced through phytoplankton growth. Many bloom
forming phytoplankton species are capable to enhance CO2 supply by carbon
concentrations mechanisms (CCM) (Raven 1991) and therewith saturate primary
production even at low CO2 concentrations in the ocean; i.e. 8-22 µmol L-1 (Goerike & Fry
1994). This active regulation of carbon uptake results in an apparent insensitivity of
primary production to CO2 concentrations at oceanic conditions (Goldman 1999). However,
CCMs may be down-regulated at times of nutrient exhaustion, because they require the
biosynthesis of enzymes and depend on ATP supply (Beardall & Giordano 2002). In this
case, primary production could be CO2 limited even at oceanic CO2 concentrations,
because the major carboxylating enzyme RubisCo, has a low affinity for CO2 (Riebesell et
al. 1993). This would explain a direct relationship of TEP production and diffusion
controlled CO2 uptake rates as suggested by Engel (2002). Hence, as the timing of carbon
overflow potentially coincides with reduced CO2 concentration and a low CO2 uptake
capacity of the phytoplankton cell, changes in seawater CO2 may directly influence primary
production rates and in consequence carbon release.
However, as far as total concentration of TEP and DOC are concerned, we could not
determine any significant difference between the CO2 treatments during this study.
Seemingly, the standing stocks of TEP and DOC were determined by mechanisms other
than production. For DOC, it is known that total concentration strongly depends on the
response of the microbial food web. This comprises a plethora of possible predator- prey
interactions, which may not at all be influenced by CO2 concentration. The concentration of
TEP may be ‘top down’ controlled either, but the role of heterotrophs for the fate of TEP is
not well understood. Obernosterer and Herndl (1995) showed that exopolymers released by
phytoplankton under phosphate limitation are rather resistant to bacterial decomposition. In
a similar way TEP produced under nutrient deficiency in the mesocosm may also escape
18
bacterial degradation. The role of zooplankton grazing on TEP production is under debate.
Passow and Alldredge (1999) documented that euphausids graze on TEP; Prieto et al.
(2000), however, showed that copepods may not. On the other hand, a well known process
for the fate of TEP is the aggregation of TEP with solid particles (Logan et al. 1995, Engel
2002), such as organisms and debris, and the subsequent sinking to deeper waters
(Alldredge et al. 1993, Passow et al. 2001). Since marine snow formation was observed in
the course of this study, it is suggested that aggregation had the major regulating effect on
TEP concentration and cell abundance during this study. Thus, the total amount of TEP
produced during the bloom may have been influenced by CO2 but may not have been
detectable from suspended TEP concentration, because of aggregate sedimentation. One
consequence would be that export of POC would as well become sensitive to CO2 . In order
to evaluate, if this hypothesis is correct and applicable to the ocean, a series of additional
studies need to be performed, which should elucidate the fate of TEP and DOC more
properly. Additionally, more general questions will have to be solved, which address the
temporal and regional importance of TEP production for carbon cycling and carbon export
in the ocean. Also the role of bacteria for TEP degradation needs to be elucidated, since
degradation may be an important loss process for TEP, specifically in ocean environments,
where carbon rather than nitrogen or phosphorous are the limiting elements. Nevertheless,
our results indicate that production of TEP by marine phytoplankton provides a potential
link between CO2 sensitive carbon assimilation and sequestration of ‘excess carbon’ to the
deep ocean.
19
References
Alldredge, A. L., Passow, U. & Logan, B. E. (1993) The abundance and significance of a
class of large, transparent organic particles in the ocean. Deep-Sea Research 40, 11311140.
Alluwihare, L. I., D. J. Repeta and R. F. Chen. 1997. A major biopolymeric component of
dissolved organic carbon in surface seawater. Nature 387: 166-169.
Baines, S. B. and M. L. Pace. 1991. The production of dissolved organic matter by
phytoplankton and its importance to bacteria: patterns across marine and freshwater
systems. Limnol. Oceanogr. 36: 1078-1090.
Barlow, R. G. (1982). Phytoplankton ecology in the Southern Benguela Current. II. Carbon
assimilation patterns. J. Exp. Mar. Biol. Ecol.. 63, 3, pp. 229-238,
Beardall, J. and Giordano, M. (2002). Ecological implications of microbial and
cyanobacterial CCMs and their regulation. Functional Plant Biol. 29, 213: 335-347.
Benner, R. (2002). Chemical composition and reactivity. In: Biogeochemistry of marine
dissolved organic matter. Eds. Hansell, D. A. and Carlson, C. pp.59-90. Academic Press.
Elsevier Science
Biddandha, B. and R., Benner. 1997. Carbon, nitrogen, and carbohydrate fluxes during the
production of particulate dissolved organic matter by marine phytoplankton. Limnol.
Oceanogr. 42: 506-518.
20
Broerse A. T. C., Ziveri P. and S. Honjo. 2000. Coccolithophore (-CaCO3) flux in the Sea
of Okhotsk: seasonality, settling and alteration processes. Mar. Micropal. 39: 179-200.
Cho, B. C. and Azam, F. (1988). Major role of bacteria in biogeochemical fluxes in the
ocean’s interior. Nature 332: 441-443.
Copping, A. E. and Lorenzen, C.J. (1980). Carbon budgetof marine phytoplanktonherbivore system with carbon-14 as tracer. Limnol. Oceanogr. 25:873-882.
De Jong, E. W., Bosch, L. and P. Westbroek (1976). Isolation and Characterization of a
Ca2+- binding Polysacchride associated with Coccoliths of Emiliania huxleyi (Lohmann)
Kamptner. Eur. J. Biochem. 70, 611-621.
De Jong E., van Rens, L., Westbroek P. and L. Bosch. 1979. Biocalcification by the marine
alga Emiliania huxleyi (Lohmann) Kamptner. Eur. J. Biochem. 99: 559-567.
Delille , B., Harlay, J., Chou, L., Frankignoulle, M., Jacquet, S., Riebesell, U., Wollast, R.,
Zondervan, I., and J.-P. Gattuso. (in prep.) Calcification of Emiliania huxleyi under
different atmospheric pCO2 in a mesocosm experiment. Global Biogeochem. Cycles.
Egge J. K. And D. L. Aksnes. 1992. Silicate as regulating nutrient in phytoplankton
competition. Mar. Ecol. Prog. Ser. 83: 281-289.
Engel, A. 2000. The role of transparent exopolymer particles (TEP) in the increase in
apparent particles stickiness (α) during the decline of a diatom bloom. J. Plankton Res., 22:
21
485-497
Engel, A. and U. Passow. 2001. The carbon and nitrogen content of transparent exopolymer
particles (TEP) derived from diatom exudates. Mar. Ecol. Prog. Ser., 219, 1-10..19
Engel, A. 2002. Direct relationship between CO2-uptake and Transparent Exopolymer
Particles (TEP) Production in Natural Phytoplankton. J. of Plankton Res., 24, 1, 49-53.
Engel, A., Goldthwait, S., Passow, U. and A. Alldredge. 2002a. Temporal decoupling of
carbon and nitrogen dynamics in a mesocosm diatom bloom. Limnol. Oceanogr., 47: 753761.
Engel A., Meyerhöfer M. and K. v. Bröckel. 2002b. Chemical and Biological Composition
of Suspended Particles and Aggregates in the Baltic Sea in Summer (1999). Est. Coast.
Shelf Sci. 55, 729–741.
Engel A., Thoms S., Riebesell U., Rochelle-Newall, E. and I. Zondervan. The Marine
Polysaccharides Sink. In prep..
Fogg, G. E. (1966). The extracellular products of algae. Oceanogr. Mar. Biol. Annu. Rev. 4,
195-212.
Fuhrmann, J. A. (1999). Marine viruses and their biogeochemical and ecological effects.
Nature 399: 541-548.
Goerike, R. and B. Fry (1994). Variations of marine plankton δ13C with latitude,
22
temperature, and dissolved CO2 in the world ocean. Global Biogeochem. Cycl. 8, 1: 85-90.
Goldman, J. C. (1999). Inorganic carbon availability and the growth of large marine
diatoms. Mar. Ecol. Prog. Ser. 180: 81-91.
Grossart, H-P. and H. Ploug. 2001. Microbial degradation of organic carbon and nitrogen
on diatom aggregates. Limnol. & Oceanogr. 46: 267-277.
Grossart, H-P. & Simon, M. (1998) Bacterial colonization and microbial decomposition of
limnetic organic aggregates (lake snow). Aquatic Marine Ecology 15,127-140.
Heemann,
C.
(2002).
Phytoplanktonexsudation
in
Abhängigkeit
der
Meerwasserkarbonatchemie. Master thesis, Universität Oldenburg.
Honjo, S. 1976. Coccoliths: production, transportation and sedimentation. Mar.
Micropaleontol. 1: 65-79.
Hood, R. R., Bates, N. R., Capone, D. G. and D. B. Olson (2001). Modeling the effect of
nitrogen fixation on carbon and nitrogen fluxes at BATS. Deep-Sea Res. II, 48: 1609-1648.
Kähler, P., Koeve, W. 2001. Marine dissolved organic matter: can its C:N ratio explain
carbon overconsumption? Deep Sea Research I, 48: 49-62.
Karner, M. and Herndl, G. J. (1992). Extracellular enzymatic activity and secondary
production in free-living and marine snow-associated bacteria. Mar. Biol. 113:341-347.
23
Körtzinger A., Koeve, W., Kähler P., L. Mintrop. 2001. C:N ratios in the mixed layer
during the productive season in the northeast Atlantic Ocean. Deep-Sea Research I, 48,
661-688.
Koroleff, F. and K. Grasshof. 1983. Determination of nutrients. In: Grasshof, K., Erhardt,
M. and Kremling, K. (eds.) (1983) Methods of seawater analyses 2, 125-188. Verlag
Chemie.
Logan, B. E., Passow, U., Alldredge, A. L., Grossart, H.-P. and Simon, M. (1995). Rapid
formation and sedimentation of large aggregates is predictable from coagulation rates (halflives) of transparent exopolymer particles (TEP). Deep-Sea Res. II, 42, 1, 203-214.
Leppard G G (1995) The characterization of algal and microbial mucilages and their
aggregates in aquatic ecosystems. Sci Total Environ 165: 103-131
Marchal, O., P. Monfray and N. R. Bates. 1996. Spring-summer imbalance of dissolved
inorganic carbon in the mixed layer of the northwestern Sargasso Sea. Tellus 48B: 115-134.
Mari X., Beauvais S., Lemee R., and M. L. Pedrotti. 2001. Non-Redfield C: N ratio of
transparent exopolymeric particles in the northwestern Mediterranean Sea. Limnol.
Oceanogr., 46: 1831–1836.
McAllister, C. D., Parsons, T. R., Stephens, K. and J. D. H. Strickland (1961).
Measurements of primary production in coastal sea water using a large volume plastic
sphere. Limnol. Oceanogr. 5, 237-258.
24
Michaels, A. F., N. R. Bates, K. O. Buesseler, C. A. Carlson and A. H. Knap. 1994.
Carbon-cycle imbalance in the Sargasso Sea. Nature 372: 537-540.
Michaels, A. F., Olson, D., Sarmiento, J. L., Ammerman, J. W., Fanning, K., Jahnke, R.,
Knap, A. H., Lipschultz, F. and J. M. Prospero (1996). Inputs, losses and transformation of
nitrogen and phosphorous in the pelagic North Atlantic Ocean. Biogeochemistry 35, 181226.
Nagata, T. (2000). Production mechanisms of dissolved matter. In: Microbial Ecology of
the oceans. Ed.: D.L. Kirchmann, pp. 121-152. Wiley-Liss, New York.
Nanninga, H. J., Ringenaldus P. and P. Westbroek. 1996. Immunological quantification of a
polysaccharide formed by Emilinia huxleyi. J. of Mar. Sys. 9: 67-74.
Obernosterer I. & Herndl G. J. (1995) Phytoplankton extracellular release and bacterial
growth: dependence on inorganic N.P ratio. Marine Ecology Progress Series 116, 247-257.
Passow, U. and A. L. Alldredge. 1995. A dye binding assay for the spectrophotometric
measurement of transparent exopolymer particles (TEP) in the ocean. Limnol. & Oceanogr.
40: 1326-1335.
Passow, U. and A. Alldredge (1999). Do transparent exopolymer particles (TEP) inhibit
grazing by the euphausiid Euphausia pacifica? J. Plankton Res. 21, 2203-2217.
Passow, U., Shipe, R. F., Murray, A. Pak, D. K., Brzezinski, M. A. and A. L. Alldredge
(2001). Origin of transparent exopolymer particles 8TEP) and their role in the
25
sedimentation of particulate matter. Continental Shelf res. 21, 327-346.
Passow, U. (2002). Transparent exopolymer particles (TEP) in aquatic environments. Prog.
In Oceanogr. 55: 287-333
Prieto, L. Sommer, F., Stibor, H. and W. Koeve (2001). Effects of planktonic copepods on
transparent exopolymer particles (TEP) abundance and size spectra. J. Plankton Res.
32:515-525.
Raven, J. A. (1991). Physiology of inorganic C acquisition and implications for resource
use efficiency by marine phytoplankton: relation to increased CO2 and temperature. Plant.
Cell and Environment, 14: 779-794.
Redfield, A. C., Ketchum, B. M. and F. A. Richards (1963). The influence of organism on
the composition of sea-water. Hill, M. N. (ed.) (1963). The Sea., 2. Edt. Wiley, 26-77.
Riebesell, U., Wolf-Gladrow, D. A. and Smetacek, V. (1993) Carbon dioxide limitation of marine
phytoplankton growth rates. Nature, 361, 249-251.
Riegman R., Stolte W., Noordeloos A. A. M. and D. Slezak. 2000. Nutrient uptake and
alkaline phosphatase (APase) activity of Emiliania huxleyi (Prymnesiophyceae) during
growth under N and P limitation in continuous cultures. J. Phycol. 36: 87-96..21
Rost, B., Riebesell, U., Burkhardt, S., Sültemeyer, D.(2003). Carbon acquisition of bloomforming marine phytoplankton, Limnol. & Oceanogr., 48, 55-67.
26
Sachs, L. (1973). Angewandte Statistik. pp.:329-333. Springer Verlag Berlin Heidelberg
New York.
Sambrotto, R.N., G. Savidge, C. Robinson, P. Boyd, T. Takahashi, D.M. Karl, C. Langdon,
D. Chipman, J. Marra and L. Codespoti. 1993. Elevated consumption of carbon relative to
nitrogen in the surface ocean. Nature 363: 248-250.
Søndergaard, M., Williams, P. J. le B., Cauwet, G., Riemann, B., Robinson, C., Terzic, S.,
Woodward, E. M. S. and J. Worm (2000). Net accumulation and flux of dissolved organic
carbon and dissolved organic nitrogen in marine plankton communities. Limnol. Oceanogr.,
45: 1097–1111.
Smith, D. C., Simon, M., Alldredge, A. L. and F. Azam. 1992. Intense hydrolytic enzyme
activity on marine aggregates and implications for rapid particle dissolution. Nature 359,
139- 141.
Thomas, H., Ittekot, V., Osterroth, C. & Schneider, B. (1999) Preferential recycling of
nutrients- the ocean’s way to increase new production and to pass nutrient limitation?
Limnology & Oceanography 44, 1999-2004.
Toggweiler, J. R. 1993. Carbon overconsumption. Nature 363: 210–211.
Van Emburg, P. R., De Jong, E. W. and W. T. H. Daems. 1986. Immunochemical
localization of a polysaccharide from biomineral structures (coccoliths) of Emiliania
huxleyi. J. Ultrast. Mol. Structure Res. 94: 246-259.
27
Vaulot D. 1989. CytoPC: processing software for flow cytometric data. Signal Noise. 2: 8.
Williams P. J. le B. and J. K. Egge. 1998. The management and behaviour of the
mesocosms. Est. Coast. Shelf Sci. 46 : 3-14.
Wood, M. A. & Van Valen, L. M. Paradox lost? On the release of energy-rich compounds
by phytoplankton. Mar. Microb. Food Webs 4,103–116. (1990).
Williams P J leB (1995). Evidence for the seasonal accumulation of carbon-rich dissolved
organic material, its scale in comparison with changes in particulate material and the
consequential effect on net C/N assimilation ratios. Mar Chem 51: 17-29.
Williams, P.J. LeB. and
J.K. Egge, (1998). The management and behaviour of the
mesocosms. Estuar. Coast. Shelf Sci. 46: 3-14.
Zondervan, I., Aerts, K., Beaufort, L., Benthien, A., Chou, L., Dellile, B., Engel, A.,
Gattuso, J.-P., Harlay, J., Heemann, C., Hoffmann, L., Jacquet, S., Nejstgaard, J., Pizay,
M.-D. Riebesell, U., Rochelle-Newall, E., Schneider, U., Terbrüggen, A., Wollast, R. (in
prep.) Biological responses to CO2-related changes in seawater carbonate chemistry during
a bloom of Emiliania huxleyi - a mesocosm study.
28
Figure legends
Fig. 1: Conceptual model showing various pathways of organic matter released by an
autotrophic cell. Under the condition of nutrient exhaustion a fraction of the carbon
accumulating intracellular is released from the cell by leakage and exudation processes.
Depending on the quality, the extracellular organic carbon can enter the microbial food web
or aggregate into particles, such as TEP. For further description, see text.
Fig. 2 a,b: Microscopic view on material sampled from the mesocosm on a 0.4 µm membrane
filter and stained with the polysaccharide specific dye Alcian Blue. A) The bold, circular cells
of E. huxleyi (arrow) are silhouetted against TEP, which typically appear as blue stained
particles of fractal structure. Because the coccoliths of E. huxleyi are coated by an acidic
polysaccharide the cells surface is stained with Alican Blue also. B) Large web-like TEP
appeared later during the E. huxleyi bloom (day 19), often entangled with cells and detritus.
Fig. 3a-c: Increase of TEP concentration during the bloom experiment in a) the ‘year 2100’
(mesocosm: 1-3), b) the ‘present’ (mesocosm: 4-6) and c) the ‘glacial’ treatment (mesocosm:
7-9). Given are the colorimetrically determined TEP concentrations in µg Xanthan
equivalents (µg Xeq.) L-1.
Fig. 4: The concentration of carbon contained in TEP (TEP-C) was significantly related to
the abundance of E. huxleyi (p<0.001).
Fig. 5a-c: Changes of DOC concentration in the course of the experiment in a) the ‘year 2100’
treatment (mesocosm: 1-3), with significant increase of DOC in the mesocosms 1(p<0.1) and
2 (p<0.001), b) the ‘present’ treatment (mesocosm: 4-6), with significant increase of DOC in
mesocosm 4 (p<0.01) and c) the ‘glacial’ treatment (mesocosm: 7-9).
Fig. 6: No significant relationship between DOC concentration and cell abundance of E.
huxleyi was observed (n=70, r²=0.17).
Fig. 7a,b: Increase of molar [POC]:[PON] (a) and [POC]:[POP] (b) ratios towards the end of
the E. huxleyi bloom. Ratios are mean values of nine mesocosm, error bars indicate ± 1SD.
Fig. 8: During the experiment, a decoupling of POC (solid triangles) from PON concentration
occurred, when dissolved inorganic nitrogen (DIN) concentrations (open circles) felt under
the detection limit. Until DIN exhaustion, changes in POC concentration were directly related
to changes in PON concentration (p<0.001).
Fig. 9: TEP concentrations relative to POC concentrations, before (open circles) and after the
onset of nutrient (PO4) depletion on day 11 (solid circles). After day 11 changes in TEP
concentration were directly related to changes in POC concentration (p<0.001).
Fig. 10: A direct relationship between Alcian Blue adsorption and abundance of E. huxleyi
cells, harvested from a nutrient replete cell culture (n=10, r²=0.94). No significant y-offset
was determined. Adsorption of Alcian Blue onto the cells is caused by an acidic
polysaccharide that covers the coccoliths of E. huxleyi (see Figure 1), but is small relative to
the Alcian Blue adsorption of TEP observed during the mesocosms study.
DOM
LMW Substances
DIN
1000 Da
0.7 µm
Modification
Production
Leakage
Microbial Food Web
LMW DOM
Uptake
Ccell
Production
&
Degradation
COO-
COO-
Exudation
Polysaccharides
Transparent Exopolymer Particles
Aggregation
COOCa2+
COO-
DIP
Figure 1
COO-
C:P
>>106
0.4 µm
Refractory DOM
COO-
DIC
HMW Substances
COO-
C:N
>>6.6
POM
Ca2+
COO-
a
50µm
b
100µm
Figure 2a,b
2400
a
-1
TEP (µg Xeq. L )
2000
Mesocosm 1
Mesocosm 2
Mesocosm 3
1600
1200
800
400
0
2400
b
-1
TEP (µg Xeq. L )
2000
Mesocosm 4
Mesocosm 5
Mesocosm 6
1600
1200
800
400
0
2400
c
-1
TEP (µg Xeq. L )
2000
Mesocosm 7
Mesocosm 8
Mesocosm 9
1600
1200
800
400
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19
day
Figure 3a-c
60
a
-1
TEP-C (µmol l )
50
40
30
20
-6
[TEP-C]=1.05x10 [cells]+5.51
r²=0.91 n=135
10
0
0
10
20
30
40
6 -1
E. huxleyi cells (x10 l )
Figure 4
50
60
150
a
Mesocosm 1
Mesocosm 2
Mesocosm 3
-1
DOC (µmol L )
140
130
120
110
100
90
150
b
Mesocosm 4
Mesocosm 5
Mesocosm 6
-1
DOC (µmol L )
140
130
120
110
100
90
150
c
Mesocosm 7
Mesocosm 8
Mesocosm 9
-1
DOC (µmol L )
140
130
120
110
100
90
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19
day
Figure 5a-c
180
b
-1
DOC (µmol L )
160
140
120
100
80
60
0
10
20
30
40
6
50
-1
E. huxleyi cells (x10 L )
Figure 6
60
POC:POP
1500
1200
a
900
600
300
0
24
POC:PON
20
b
16
12
8
4
0
1c
3
5
7
9
11
13
15
17
19
day
Figure 7 a,b
18
140
16
120
-1
100
12
10
80
8
60
-1
POC (µmol L )
DIN (µmol L )
14
6
∆POC:∆PON=6.8 +/- 0.3
r²=0.98
4
40
20
2
0
0
0
2
4
6
8
-1
PON (µmol L )
Figure 8
10
12
80
70
-1
TEP-C (µmol l )
60
50
40
30
20
∆[TEP-C] = 0.35 ∆[POC]
r²=0.80 n=62
10
0
0
50
100
150
-1
POC (µmol l )
Figure 9
200
250
600
Alcian Blue adsorption
-1
(µg Xeq. l )
500
400
300
200
100
0
0
50
100
150
200
6 -1
E. huxleyi cells (x10 l )
Fig. 10
250
Table 1. Parameter values for the linear regressions between TEP concentration and E. huxleyi
calculated for each of the mesocosms. The analysis include data from days 3-17, n=15 for
each mesocosm, SD: standard deviation.
Regression Statistics : [TEP, µg Xeq. L-1]= a x10-6 [cell, L-1]+ b
CO2Mesocosm
No.
treatment
r²
p<
a±SD
b±SD
‘Year 2100’
‘Present’
‘Glacial’
1
53.1±3.53
149.6±18.6
0.95
0.001
2
28.2±1.56
167.8±24.4
0.96
0.001
3
49.9±2.43
164.5±24.7
0.97
0.001
4
33.2±2.73
142.4±48.0
0.92
0.001
5
29.0±2.40
189.1±49.5
0.92
0.001
6
39.1±3.21
161.5±31.2
0.92
0.001
7
35.1±2.04
103.6±29.6
0.96
0.001
8
33.8±4.03
144.4±43.6
0.84
0.001
9
31.8±1.18
117.6±24.7
0.98
0.001
Table 2: ‘Carbon overconsumption’ calculated as carbon taken up, or contained in the
different elemental pools in excess to the expected value (DINx6.6) during the mesocosm
study (days 1-18). Data are mean values calculated from all mesocosms except #8, where no
data for DOC were available for day 18. The value (%) ‘Excess C’ indicates the fraction of
excess carbon relative to the total C uptake. Therefore, excess carbon in the POC, DOC and
‘Loss’ fractions sum up to the amount of excess carbon uptake. It was assumed that N is not
an inherent component of TEP, because no N was determined in the acidic polysaccharides
released by E. huxleyi; n.d.: not determined.
∆N
Days 1-18
Uptake
∆C
µmol L-1 µmol L-1
C:N
∆N x 6.6
Mol:Mol
µmol L-1
µmol L-1
%
Excess C
15.2
248.1
16.3
100.3
147.8
59.6
POM
6.4
117.2
18.3
42.2
74.9
30.2
DOM
n. d.
11.5
n.d.
n.d.
≤11.5
≤4.6
TEP
-
39.2
-
-
39.2
15.8
(POM-TEP)
6.4
78.0
12.2
42.2
35.8
14.4
Calculated Loss
≤8.8
119.4
≥13.6
58.1
≥61.3
≥24.7
Partitioning