Marine ecosystem community carbon and nutrient uptake stoichiometry under varying ocean acidiﬁcation during the PeECE III experiment

Inorganic carbon and nutrient biogeochemical responses were studied during the 2005 Pelagic Ecosystem CO 2 Enrichment (PeECE III) study. Inverse analysis of the temporal inorganic carbon dioxide system and nutrient variations was used to determine the net community stoichiometric uptake characteristics of a natural pelagic ecosystem 5 production perturbed over a range of pCO 2 scenarios (350, 700 and 1050 µ atm). Nutrient uptake showed no sensitivity to CO 2 treatment. There was enhanced carbon production relative to nutrient consumption in the higher CO 2 treatments which was positively correlated with the initial CO 2 concentration. There was no signiﬁcant calciﬁcation response to changing CO 2 in Emiliania huxleyi by the peak of the bloom and 10 all treatments exhibited low particulate inorganic carbon production ( ∼ 15 µ mol kg − 1 ). With insigniﬁcant air-sea CO 2 exchange across the treatments, the enhanced carbon uptake was due to increase organic carbon production. The inferred cumulative C:N:P stoichiometry of organic production increased with CO 2 treatment from 1:6.3:121 to 1:7.1:144 to 1:8.25:168 at the height of the bloom. This study discusses how ocean 15 acidiﬁcation may incur modiﬁcation to the stoichiometry of pelagic production and have consequences for ocean biogeochemical cycling.


Introduction
Consequent to the increase in the atmospheric load of carbon dioxide (CO 2 ), due to anthropogenic release, there has been an increase in the oceanic carbon reservoir 20 (Sabine et al., 2004). At the air-sea interface, the productive, euphotic surface ocean is a transient buffer in the process of ocean-atmosphere CO 2 equilibrium, a process retarded by the slow mixing of the surface waters with the intermediate and deep ocean, As such, the greatest changes to the CO 2 system are occurring in the surface waters. This build up of CO 2 is already altering the carbonate chemistry of the oceans and pro-the coming decades leads to modifications in physiological or morphological properties which may have consequences for ecological structure and biogeochemical cycling. The general assertion is that increasing CO 2 has deleterious effects on the growth and productivity of marine calcifiers (e.g. Riebesell et al., 2000;DeLille et al., 2005;Orr et al., 2005), although there are notable exceptions (Langer et al., 2006). Overconsump-10 tion of carbon, a common response to nutrient and environmental stress is magnified under high CO 2 conditions (Zondervan et al., 2002;Engel et al., 2005). Changing CO 2 in aquatic systems has been shown to influence phytoplankton species succession (Tortell et al., 2002). Changes to the nutritional quality (higher C:P), in response to increased CO 2 , of phytoplankton as a food source results in lower growth rate and 15 fecundity in zooplankton (Urabe et al., 2003) Efforts to understand potential consequences and feedbacks of increasing CO 2 have employed laboratory and mesocosm studies either at the individual species level or on natural and perturbed ecosystems DeLille et al., 2005). In this study, a natural ecosystem was perturbed with nutrients over a range of atmospheric EGU nipulated in order to obtain triplicates of three different pCO 2 concentrations (1x CO 2 (350 µatm), 2x CO 2 (700 µatm) and 3x CO 2 (1050 µatm)). Addition of fresh water to the upper 5.5m of the enclosures ensured the generation of a mixed layer separated from the underlying water by a salinity gradient of 1.5. Nitrate and Phosphate were added to the upper mixed layer resulting in initial respective concentrations of 16 and 5 0.8 µmol kg −1 . A comprehensive description of the mesocosm setup, CO 2 and nutrient perturbation and sampling strategy can be found in Schulz et al. (2007). Nutrient measurement methodology is also reported in Schulz et al. (2007).
Samples for determining the carbon dioxide system were taken from seawater pumped from 1m depth in each of the enclosures. The partial pressure of carbon dioxide (pCO 2 ) was determined in air equilibrated with seawater pCO 2 using an infrared gas analyser (Li-Cor 6262) (Wanninkhof and Thoning, 1993). Gas calibration of the instrument against high quality air standards containing mixing ratios of 345, 415 and 1100 ppm enveloped the daily seawater measurement program. Following the pCO 2 measurements, using the same sampling methodology, samples for total alkalin-15 ity (A T ) and total dissolved inorganic carbon (C T ) were drawn into 500 ml borate bottles and immediately poisoned with HgCl 2 . A T was measured using Gran potentiometric titration (Gran, 1952) on a VINDTA system (Mintrop et al., 2000) with a precision of ≤4 µmol kg −1 . C T was determined using coulometric titration (Johnson et al., 1987) with a precision of ≤2 µmol kg −1 . For both A T and C T measurements, samples were fil-20 tered through GF/F filters placed in the sample inlet tubes to the respective instruments immediately prior to measurement. Calculation of additional carbon dioxide system variables used the CO2SYS program (Lewis and Wallace, 1998), from measured C T and A T and adopting the dissociation constants for carbonic acid (Dickson and Millero, 1987), boric acid (Dickson,25 1990a) and sulphuric acid (Dickson, 1990b) and the CO 2 solubility coefficient from Weiss (1974). Seawater pH is reported on the total hydrogen scale.
Particulate inorganic carbon (PIC) production was calculated from temporal changes (∆t) in total alkalinity with appropriate correction for alkalinity contributions from net 4634 EGU nitrate and phosphate consumption (Goldman and Brewer, 1980): In order to evaluate biological contributions to the inorganic carbon system it was necessary to first account for the exchange of CO 2 (CO 2(ex) ) between the seawater and the overlying mesocosm atmosphere. Gas exchange was calculated according to Delille et 5 al. (2005) further employing the chemical enhancement factors of Kuss and Schneider (2004). Net community production (NCP) was calculated from temporal changes in C T allowing for modifications due to PIC production or dissolution and net CO 2 gas exchange thus: Removal of the contributions of net calcification and air-sea CO 2 exchange gives the net perturbation of the inorganic carbon system from community biological activitythe sum of autotrophic and heterotrophic processes within the mixed layer. Comparison of the inferred net organic production and nutrient uptake rates gives the pelagic 15 community molar stoichiometry changes as a response to changing CO 2 .

Temporal evolution of the carbon dioxide system
In accordance with the agreement for the PeECE special issue, only treatment means are represented in this study and not the results from individual mesocosm bags. Daily 20 treatment means, with standard deviations, of seawater partial pressure of carbon dioxide (pCO 2 ), total inorganic carbon (C T ) and nutrient concentrations for the mixed layers 4635 EGU in the three perturbation scenarios are shown in Fig. 1. The concentrations for each variable (Day 1) illustrate only the response of the mesocosms to the initial nutrient and carbon dioxide modifications following the CO 2 charging and nutrient additions as detailed in Schulz et al. (2007) and, consequently, the starting date for the present study, when the mesocosms reached a "steady state" was determined to be day 2. 5 Once the scenario CO 2 concentrations had been reached in the seawater, only the atmospheric concentrations over the water were maintained at the prescribed treatment level and the seawater CO 2 system was allowed to respond to processes as in a naturally occurring plankton bloom. There is a clear treatment dependant response in pCO 2 (Fig. 1a). Under the 1x CO 2 scenario, pCO 2 dropped by 202 µatm by the end of the bloom stage (Day 12), whilst pCO 2 reductions were 335 and 570 µatm in the 2x and 3x CO 2 scenarios, respectively. Similarly, the C T reductions also show the same order of response ( Fig. 1b). Under the 1x CO 2 treatment, C T showed a reduction of 94 µmol kg −1 , increasing to 107 µmol kg −1 and to 118 µmol kg −1 as the CO 2 treatment increases. Initial concentrations of A T (Fig. 1c) were higher in the 3x CO 2 15 treatment corresponding with the higher initial salinity . Changes in A T throughout the bloom stage indicated a modest treatment response before Day 10 with reductions in A T of 21, 23.5 and 24.8 µmol kg −1 (1x CO 2 through 3x CO 2 treatments). Total alkalinity increased after the bloom peak in the 2x and 3x CO 2 treatments but remained constant in the 1x CO 2 treatment. 20 The absolute change in carbonate ion [CO 2− 3 ] concentrations, on the other hand, showed no treatment dependence and increased by 50, 52 and 49 µmol kg −1 in the low to high treatments, respectively (Fig. 2a). Correspondingly, as Ω is controlled mainly by [CO 2− 3 ], there were similarly increases in the calcite saturation state; 1.2, 1.27 and 1.19 (Fig. 2b). Seawater pH (pH T ) values, however, increased the most in the 3x CO 2 treat-25 ment from 7.64 to 7.96 (∆pH=0.32); under the 2x CO 2 scenario the pH increased from 7.81 to 8.07 (∆pH=0.26) and in the 1x CO 2 treatment from 8.11 to 8.27 (∆pH=0.16).
Particulate inorganic carbon (PIC) production showed a similar response under all CO 2 treatments (Fig. 3a)

EGU
Emiliania Huxleyi (Paulino et al., 2007) and PIC production increased to a maximum of 15 µmol kg −1 on Day 10 corresponding to the start of the demise of the Emiliania Huxleyi population (Paulino et al., 2007). Cumulative PIC remained constant in the 1xCO 2 treatment until the end of the experiment suggesting no significant calcification or dissolution. However, in the higher CO 2 treatments cumulative PIC decreased slowly 5 due to either increased dissolution or input of A T from below .
Net community production (NCP) (Fig. 3b), the net organic carbon production, contributed the most to the changes in inorganic carbon dioxide shown in Fig. 1b with contributions from PIC and air-sea CO 2 exchange (not shown) of only 15-20% to the C T changes. The cumulative NCP has a maximum on Day 12 and shows a clear treat-10 ment response with maximum production, relative to Day 2. The 3x CO 2 treatment produced 110 µmol kg −1 , falling to 96µmol.kg-1 and 80 µmol kg −1 in the 2x CO 2 and 1x CO 2 treatments. After Day 12 there is a fall in calculated NCP. It was not possible to ascertain the relative influence of respiration and the response to a storm -mixing of water, higher in CT, from below the halocline , followed by further 15 treatment divergence in NCP until the end of the experiment -again favouring the 3x CO 2 treatment.

Temporal evolution of nutrients
The treatment dependant changes in the major nutrients nitrate, phosphate and silicate are shown in Fig. 1c-e. Both nutrient uptake and the timing of nutrient minima showed 20 no significant dependency on CO 2 treatment although there were inter-treatment variations in the planktonic assemblages (Paulino et al., 2007). Silicate was not added during the experimental set-up, however a residual concentration of about 3 µmol kg −1 was inherited from the fjord water used to fill the mesocosm enclosures. Silicate concentrations dropped sharply due to diatom uptake and reached a minimum on day 9- 25 10. Phosphate concentrations showed only modest reductions until day 5 after which rapid uptake led to a minimum on day 11. From the onset of the experiment, concentrations of nitrate diminished at a steady rate until Day 12 then declined slowly until the 4637 uptake of nitrate to phosphate relative to the Redfield ratio (16:1) (Redfield et al., 1963).
In the initial stage of the bloom, relative phosphate uptake increases rapidly on Day 6. Thereafter, the system behaves very close to Redfieldian until the end of the bloom when N/P rose slowly to a final cumulative value of 21 in all treatments. Nitrate uptake is high compared to silicate uptake until Day 4, after which the diatom 10 bloom started (Paulino et al., 2007) and silicate was consumed at about 1:1 with nitrate until silicate reached a first minimum on Day 7. There followed a small peak in silicate concentration on Day 8 (Fig. 1f) after which silicate became depleted on Day 9-10. In contrast to the nutrient stoichiometric ratios, there are significant treatment dependant differences between the carbon to nitrate (C/N) (Fig. 4c, g) and carbon to 15 phosphate (C/P) ratios (Fig. 4d, h). Overconsumption of carbon results in a higher C/N under higher CO 2 exposures. Cumulative C/N uptake was higher in the 3x CO 2 scenario throughout the entire bloom, quickly reaching a peak of 7.5-8.25 by Day 6 and remaining at this level until Day 12. The pre-bloom responses in the 1x and 2x CO 2 treatments were similar until day 5 after which the 2x CO 2 treatment increased 20 C/N uptake to 7.1. A similar order of response was seen in the C/P ratio with the 3x treatment showing a greater carbon overconsumption throughout the experiment. In contrast to the C/N response, the C/P increased significantly in the post-bloom phase.

Modification of the carbonate system
As changes to A T (including contributions from nutrient alkalinity) were similar and low and gas exchange minimal in all treatments, modifications to C T were mainly due to organic carbon production. As such, the greatest changes were related to the treat-5 ment dependant carbon overconsumption with increasing uptake with increasing initial pCO 2 . The effect on pCO 2 due to the greater carbon uptake is exacerbated due the low buffer capacity of the high CO 2 seawater resulting in a rapid reduction of pCO 2 in the 3x CO 2 treatment. The opposite response is seen with pH T with the largest increases found in the 3x CO 2 scenario. Due to the weakly buffered system in a future 10 ocean, pelagic ecosystems will undergo greater seasonal changes in their ambient CO 2 fields. Increasing CO 2 will reach a point where changes to the carbonate system will move outside the contemporary "carbonate system envelope": in this experiment, at no point did any of the calculated carbonate system variables in the 2x CO 2 and 3x CO 2 treatments overlap with the range of the 1x CO 2 scenario.

Calcification
There has been shown to be a response to the saturation state of CaCO 3 (Ω calcite ) in seawater and the growth and health of many calcifying marine organisms (e.g. Reibesell et al., 2000;Langdon et al., 2003;Kleypass et al., 2006;DeLille et al., 2005;Langer et al., 2006). The dominant calcifier during the experiment was the 20 prymesiophyte Emiliania Huxleyi which is common to the Norwegian fjords and open waters of the Norwegian and Barents Sea. Calcification in Emiliania Huxleyii has been shown to be controlled by light, nutrient (especially phosphate) and carbonate ion concentrations and calcite saturation state (see review in Zondervan, 2007). The photon flux density (PFD) concentrations throughout the experiment (Schulz et al., ) al-al. (2005, showed that calcification was delayed by 1 day during a 2x CO 2 treatment compared to the contemporary treatment. This study does not find any relationship between [CO and Ω calcite values showed marked inter-treatment variations throughout the experiment there was no difference in cumulative net community calcification by the peak of the bloom. This varies from the results of DeLille et al. (2005) who reported a 40% reduction in PIC production between 1x CO 2 and 2x CO 2 treatments. Conversely, this study does not show the promotion of higher calcification with increasing CO 2 as found by Langer et al. (2006). The concentrations of Emiliania Huxleyii are much lower in this study (Paulino et al., 2007) than in Delille et al. (2005) experiment. It may be that the 15 subtle inter-treatment differences in PIC production seen before Day 12 (Fig. 3a) would magnify to be significant within a plankton bloom dominated by Emiliania Huxleyii.

Carbon consumption and nutrient stoichiometry
Elemental stoichiometry of biological production in the surface ocean controls, to a great extent, nutrient balance and cycling of the global ocean. Ocean nutrient stoi-20 chiometry is controlled on short timescales by the resource allocation in marine organisms (e.g. Redfield, 1958;Klausmeier et al., 2004) and on longer scales by continental nutrient supply (Broecker, 1982). The degree of efficiency of the connectivity between the surface and interior ocean is also controlled by the stoichiometry of sedimentary In this work, molar stoichiometric nutrient uptake ratios are inferred from the uptake of dis- 25 solved inorganic nutrients and thus reflect the community balance at a particular time. This method, therefore, cannot discriminate species specific uptake as suggested by Klausmeier et al. (2004)  EGU nutrient balance under differing CO 2 regimes. Overconsumption of carbon relative to nutrient supply has been reported in several studies of the planktonic response to increased CO 2 (Riebesell et al., 1993Banse, 1994;Delille at al., 2005;Engel et al., 2005) The increase in carbon uptake is usually seen at the end of the bloom suggested to result from TEP production in 5 response to nutrient stress (Engel et al., 2002(Engel et al., , 2004. Hein and Sand-Jensen reported an immediate (2h) increase in primary production in relation to increased CO 2 concentrations. Changes in dissolved aqueous CO 2 may determine the phytoplankton cell size distribution (Engel et al., 2007). Egge et al. (2007) also reported increased primary production, based on in situ 14 C incubations, in this study in the higher CO 2 treatments towards the peak of the bloom. However, the overconsumption derived from chemical uptake estimations is seen from the onset of the experiment and is proportional to the initial treatment CO 2 concentration (see also Riebesell et al., 2007). As  report no changes to the stoichiometry of dissolved and particulate organic matter, this implies that there was most likely an increase in the carbon exported from 15 the mixed layer relative to nutrient concentration in the high CO 2 treatments. 4.4 Consequences for ecosystem functioning, ocean biogeochemical cycling and atmospheric CO 2 control The elemental stoichiometry of biological export has an important role in controlling atmospheric CO 2 concentrations (Broecker, 1982;Volk and Hoffart, 1985;Omta et al., 20 2006). From first principles, the stoichiometry of osmotroph uptake is transferred to the standing stock of pelagic production. Non-redfield signatures of biological production have been documented in the surface ocean (Sambrotto et al., 1993;Anderson and Sarmiento, 1994;Falck and Anderson, 2005;Koeve, 2006) and at depth (Kortzinger et al., 2001;Pahlow and Riebesell, 2000). This study has indicated an increase in the 25 carbon:nutrient stoichiometry of pelagic osmotroph productivity in a high CO 2 world. If the PeECE results are representative of global productivity, with an efficient transfer of this carbon overconsumption to depth, the increase in atmospheric CO 2 increase by the 4641 (CaCO 3 :POC) and has been used to test the sensitivity of changes to the carbonate pump on atmospheric CO 2 (Heinze, 2003;Ridgwell et al., 2007). Previous studies have identified a reduction in the calcification rate of pelagic calcifiers (summarised in Ridgwell et al., 2007). Calcification was insensitive to CO 2 level in this study and Schulz et al. (2007) show that there was increased export of organic carbon from the 10 mixed layer under the 3x CO 2 treatment, suggesting a lowering of the ballast effect. This would suggest that the atmospheric control proposed by Riebesell et al. (2007) would be diminished if there were no compensatory processes to offset the reductions in the ballast effect. However, Engel et al. (2007) show that there was an increase in the cell size during the PeECE II experiment. This cell size change with increases in 15 TEP production and, thus potentially higher aggregation of particulates shown to be a response of similar pelagic plankton communities (Engel et al, 2002;2004), may be a mechanism to increase sedimentation rates in a high CO 2 environments. Carbon overconsumption by osmotrophs changes the nutritional quality of food for zooplankton. Lower growth, productivity and fecundity has been documented in zoo-20 plankton fed on phytoplankton grown under high CO 2 (Sterner and Elser, 2002;Urabe et al., 2003,;Anderson et al., 2005). Carotenuto et al. (2007) hypothesise that the high C:N uptake during the PeECE III study reduced recruitment of zooplankton nauplii. Changes to the ecological structure of pelagic community will also control the export of stoichiometric signals through changes to the size structure and ballast of sedimenting 25 material.
There exists now a large array of experimental data, from laboratory and mesocosm studies of individual species and ecosystem studies, on ecological and biogeochemical responses to ocean acidification. The requirement now is for detailed statistical meta-4642