The response of seagrass ( Posidonia oceanica ) meadow metabolism to CO 2 levels and hydrodynamic exchange determined with aquatic eddy covariance

We investigated light, water velocity, and CO2 as drivers of primary production in Mediterranean seagrass (Posidonia oceanica) meadows and neighboring bare sands using the aquatic eddy covariance technique. Study locations included an open10 water meadow and a nearshore meadow, the nearshore meadow being exposed to greater hydrodynamic exchange. A third meadow was located at a CO2 vent. We found that, despite the oligotrophic environment, the meadows had a remarkably high metabolic activity, up to 20 times higher than the surrounding sands. They were strongly autotrophic, with net production half of gross primary production. Thus, P. oceanica meadows are oases of productivity in an unproductive environment. Secondly, we found that turbulent oxygen fluxes above the meadow can be significantly higher in the afternoon than in the morning at the same light 15 levels. This hysteresis can be explained by the replenishment of nighttime-depleted oxygen within the meadow during the morning. Oxygen depletion and replenishment within the meadow do not contribute to turbulent O2 flux. The hysteresis disappeared when fluxes were corrected for the O2 storage within the meadow and, consequently, accurate metabolic rate measurements require measurements of meadow oxygen content. We further argue that oxygen-depleted waters in the meadow provide a source of CO2 and inorganic nutrients for fixation, especially in the morning. Contrary to expectation, meadow metabolic activity at the CO2 vent 20 was lower than at the other sites, with negligible net primary production.

the same depths in open-water and nearshore environments. The nearshore meadow was exposed to greater water velocities and waves. Waves particularly enhance hydrodynamic mixing within seagrass canopies (Koch and Gust 1999;Hansen and Reidenbach 2017). A meadow at a CO2 vent was included to test the hypothesis that an enhanced CO2 supply enhances seagrass primary production. We also compared seagrass meadow productivity to that of surrounding sands. were at 13 m depth and known to persist for over 20 years. Percent areal coverage, estimated from plan-view underwater images, was high (95-99%) at both meadows (Table 1). Water velocities measured 0.3 m above the open-water meadow (mean of 1.2 cm s -1 ) were lower than at the nearshore meadow (mean of 2.6 cm s -1 ; Table 1). Oxygen fluxes were also measured over bare sands adjacent to the nearshore meadow.

Flux measurements
The eddy covariance technique utilizes high frequency measurements of solute concentration and water velocity at a single point above the habitat of interest to quantify solute fluxes over 10 m 2 footprint located directly upstream (Berg et al., 2003;Berg et al., 2007). Water velocity was determined at 16 Hz with an acoustic Doppler velocimeter (Nortek). Oxygen concentration 100 was determined at 5 Hz with high speed (t90 = 0.25s) micro-optode sensors of 50 or 430 µm in diameter (PyroScience). Their illumination, and measurement of their fluorescence, were managed by an oxygen meter (PyroScience) in a submersible housing.
Optodes have been successfully used for eddy covariance measurements previously (Chipman et al., 2012;Berg et al., 2016) and they lack stirring sensitivity (Holtappels et al., 2015). Optodes were calibrated in air-saturated and anoxic water before and after each deployment. The instruments were mounted to a tripod frame with narrow legs (4 cm in diameter) spaced 1.2 m apart. The 105 sensors were aligned at a 40-degree angle to the velocimeter, and the tip located 1 cm from the velocimeter measuring volume.
The legs of the frame were adjusted to allow for measurements at a height of 0.25 m over sands and 0.3 m above the top of seagrass canopies.
Multi-parameter probes that measured O2 concentration and pH (RBR) were mounted near the top and bottom of the instrument frame. Probes were used to record the vertical O2 concentration gradient in order to identify oxic stratification, which 110 can cause spurious O2 fluxes (Holtappels et al., 2013). The pH sensors were calibrated to NBS buffers. Observed pH values were converted to total scale by subtracting 0.13 pH units, according to Zeebe and Wolf-Gladrow (2001). Irradiance was quantified with HOBO Pendant sensors (Onset Computer Corp.). Because these luminance sensors are sensitive from 300 -1200 nm, we crosscalibrated them to quantify the photosynthetically active radiation (PAR, 400-700 nm) using a LI-192 sensor (Li-COR) placed at the same location underwater (Long et al., 2012). The instrument frame was deployed by divers at the research sites and was 115 aligned to minimize tilt. The time series of mean O2 concentration variation recorded by the fast optodes reproduced those recorded by the O2 logger. Thus, fluctuations recorded by the fast optode were accurate.

Flux calculations
The velocity and oxygen time series, recorded at 16 Hz by the ADV, were downsampled to the optode sampling frequency of 5 Hz and were processed similarly to the procedure described by Holtappels et al., (2013). The tilt of the ADV was corrected using 120 the planar fit method by Wilczak et al., (2001). A running average with a window of 300 s was subtracted from the time series to calculate the fluctuating vertical velocity ( ′ ) and O2 concentration ( '). Subsequently, the time series of ′ and ' were crosscorrelated for 1-hour bursts allowing stepwise time shifts to a maximum 4 s (McGinnis et al., 2008). The median time shift was 0.8 s. The time shift with the highest cross correlation coefficient was used to calculate the flux. The mean turbulent O2 flux in the vertical direction was calculated following Eq. (1): 125 where the overbar indicates averaging. Instantaneous fluxes ( ' ') were added over time to calculate cumulative fluxes. Particle collision with the tip of O2 minisensors can cause erratic fluxes (e.g., Lorrai et al., 2010). We examined the dataset for spikes in the O2 data that co-occurred with abrupt changes in calculated fluxes. Few hour-long averaging intervals (4%) were affected and they were excluded from further analyses. 130 In a further step, we accounted for oxygen storage in the water layer between the measurement volume of the eddy covariance system and the sediment surface. The diurnal variations in mean O2 concentration within this layer do not contribute to oxygen fluxes above the meadow but are nevertheless caused by photosynthesis and respiration. In a first approximation, to account for this O2 storage and quantify biotic production, we assumed that the O2 concentrations at the measurement volume represented the O2 concentrations in the entire bottom layer. This assumes that there was no vertical oxygen gradient in the bottom layer. We 135 made this correction according to Rheuban et al., (2014b) following Eq. (2): where EC Flux is the eddy covariance flux, h is the eddy covariance measuring height above the bed, and ΔC Δ is the change in the mean O2 concentration at the measurement volume over time for each 1h burst. From these corrected fluxes we infer benthic O2 production and consumption. Negative production represents O2 consumption. 140 The corrected O2 fluxes, quantified over hourly intervals, were grouped into light and dark (<1% irradiance). These were used to calculate respiration (R), gross primary production (GPP), and net ecosystem metabolism (NEM) in an approach similar to Hume et al., (2011), and reported here in mmol O2 m -2 d -1 . To accommodate datasets that lacked complete daytime or nighttime observations, metabolic fluxes were calculated following Eqs.
where Fluxdark is a nighttime flux of 1 h duration, Fluxlight is a daytime flux of 1 h duration, and hlight and hday are the durations of the illuminated (14 h) and total length (24 h) of a diurnal cycle. To examine the photosynthetic response of seagrass meadows to their changing light environment, photosynthesis-irradiance relationships were fit with a hyperbolic tangent function (Jassby and 150 Platt, 1976) which was modified to account for respiration according to Rheuban et al., (2014a). The fit was calculated following Eq. (6): where is the maximum photosynthetic rate, is the light-saturation parameter, and is respiration. The light-saturation parameter marks an optimal irradiance above which the quantum yield of photosynthesis substantially declines and photoinhibition 155 may occur (Talling 1957;Falkowski and Raven 1997). The irradiance at which net oxygen production occurs is the irradiance compensation point ( ).

Additional measurements
Supporting measurements were made of nutrient concentrations and the carbonate chemistry in the water column at each of the study sites. Water samples, collected 1 m above the bed by divers, were filtered over 0.2 µM and then frozen until analysis. Nutrient 160 concentrations (NH4 + , NO3 -, PO4 + ) were determined with a QuAAttro nutrient analyzer according to the methods employed by Lichtschlag et al., (2015). Samples for both alkalinity and DIC were collected by divers at the beginning and end of each deployment. Alkalinity samples were kept cool and analyzed by titration on the day of their collection. Samples for DIC analysis were preserved with a saturated mercuric chloride solution (1% by volume), then quantified by flow injection (Hall and Aller, 1992). Time-varying CO2(aq) concentrations at the CO2 vent were calculated assuming a stable alkalinity and time-varying pH 165 supplied as inputs to the software package CO2SYS (Lewis et al., 1998). Diurnal alkalinity variation caused by seagrass photosynthesis and respiration had a negligible effect on these calculations.

Oxygen storage in the open-water meadow
Diel cycles of oxygen fluxes were measured over seagrass meadows (Figs. 1, 3, 4, S1). When plotting eddy covariance fluxes 170 versus light intensities in order to obtain photosynthesis-irradiance curves we observed a hysteresis, namely oxygen fluxes were lower in the morning than in the afternoon at the same light intensities (Fig. 2a). For example, O2 flux at 18:00 h was 175 mmol m -2 d -1 while at the same light intensity in the morning (08:00) it was close to zero. The correction for storage below the measurement volume (Eq. 2) was insufficient to fully reduce this hysteresis. Storage of O2 within P. oceanica gas-bearing tissues could also not explain this pattern. Our laboratory measurements showed that O2 production by the leaves ceased within a few 175 minutes of the cessation of light (data not shown). Instead, we found that the hysteresis in O2 flux at the open-water meadow was caused by O2 storage in the water column within the meadow. The concentration of dissolved O2 within the meadow, at 0.2 m above the bed, differed substantially from the concentration in the eddy covariance measuring volume at 0.3 m above the meadow 5 Biogeosciences Discuss., https://doi.org/10. 5194/bg-2018-199 Manuscript under review for journal Biogeosciences Discussion started: 27 April 2018 c Author(s) 2018. CC BY 4.0 License. (Fig. 1). Thus, there is a mass transfer resistance between the water in-and above the meadow that effectively separates the water in the meadow from the seawater. 180 To obtain metabolic rates from measured fluxes, an additional calculation had to be performed to correct for oxygen storage within the open-water meadow. We assumed a linear O2 concentration profile between the measurement volume and an O2 logger placed 0.2 m above the bed. Storage was calculated according Eq. 2, but now from the change over time in the estimated mean O2 concentration below the eddy covariance measurement volume (Fig. S1). Respiration and GPP, calculated from the thus corrected O2 fluxes (Eqns. 3 & 4), were 1.5 and 1.2 times the measured above-meadow fluxes, respectively. The photosynthesis 185 irradiance curves of corrected fluxes showed a smaller, reversed hysteresis (Fig. 2b). After the correction, the photosynthetic production in the early morning was greater than production in the evening at the same light levels, consistent with the diel photosynthetic production of a Zostera marina meadow (Rheuban et al., 2014b). This additional correction for the retention of water within the meadow was not required at the other seagrass meadows where dissolved O2 concentrations in overlying water matched concentrations within the meadow during the day and at night (data not shown). 190

Primary production and respiration at Elba
O2 production and consumption, inferred from storage-corrected eddy covariance O2 fluxes, followed similar diurnal patterns at both meadows in Elba (Fig. 3). Daytime O2 production exceeded nighttime O2 consumption at both meadows on both days. As a result, both meadows were net autotrophic (Table 2). Peak GPP was greater in the nearshore meadow, but nighttime O2 uptake was similar between meadows. The mean NEM of the nearshore meadow exceeded that of the open-water meadow (Table 2). 195 Downwelling PAR was similar in the meadows at Elba and slightly greater over nearshore sands (Fig. 3). The bare sands at Elba were also net autotrophic. Bare sand GPP exceeded R by 40%. However, GPP was 10-to 17-fold smaller than meadow GPP ( Table   2). As a result, the NEM of seagrass meadows was up to 20-fold greater than the NEM of sands.

Primary production and respiration at a CO2 vent
At the CO2 vent at the island of Panarea, gas bubbles emerged from sediments at both the seagrass meadow and at the bare sand 200 sites. The rate of gaseous CO2 release increased at low tide, and this increase coincided with a decrease in bottom water pH (Fig.   S2). These observations show that gas efflux was affected by tidally-driven expansion of gas reservoirs in the sediments. The pH fell from 8.05 at high tide to a low of 7.55 at low tide, corresponding to a 3.7-fold increase in CO2(aq) concentration (Lewis et al., 1998). Thus, CO2 was elevated by seepage. Oxygen production followed the same diurnal patterns as the Elba meadows, with daytime production and nighttime uptake (Fig. 4). Oxygen production by the meadow did not closely follow irradiance. 205 Surprisingly, the GPP of the CO2 vent meadow was less than half of the Elba meadows (Table 2). Mean R at the CO2 vent meadow, however, was similar to the lowest R at Elba. As a result, the meadow was autotrophic but with an NEM five-to ten-fold smaller than the NEM of the Elba meadows. Over neighboring bare sands, the pH of bottom waters dropped from 8.05 at high tide to 7.75 at low tide, corresponding to a doubling of the CO2(aq) concentration. The GPP of the CO2 vent sands was similar to the GPP of sands at Elba ( Table 2). Respiration of the CO2 vent sands exceeded R in bare sands at Elba. The sands were net autotrophic, but 210 with a small NEM.

Flux-irradiance curves
In all meadows, the diel oxygen flux was a non-linear function of irradiance (Fig. 5), resembling a saturation curve. In none of the meadows was reached. The light compensation point ( ) varied between 6 and 17% of the peak irradiance. The light- saturation parameter ( ) varied was one-third of peak irradiance in the Elba meadows and two-thirds of peak irradiance in the 215 CO2 vent meadow.
was greatest at the nearshore meadow, followed by the open-water and CO2 vent meadows.

Seagrasses versus sands
The seagrass meadows at Elba and Panarea were 10-20 times as productive as surrounding sands. This is consistent with the high productivity of P. oceanica individuals determined by laboratory-, chamber-, leaf tagging-, or biomass measurements (Zieman, 220 1974;Ott, 1980;Drew, 1979;Libes, 1986;Pergent et al., 1994), thus emphasizing the importance of seagrass for CO2 sequestration (Fourqurean et al., 2012). Implicit in these prior observations is the high productivity of seagrasses relative to surrounding sands, however, direct comparisons have rarely been made. Previous comparisons were made with benthic chambers, where sand oxygen fluxes were close to the detection limit, and seagrass oxygen fluxes were of the same order as our quantifications (Holmer et al., 2004;Barrón et al., 2006a). Using DIC mass balance, Gazeau et al., (2005) found smaller primary production than we observed in 225 seagrass and negligible primary production in bare sands.
Meadow productivity in this study was consistent with other open water measurements of P. oceanica seagrass meadow productivity. Respiration in P. oceanica meadows is much lower than primary production, thus the meadows are highly autotrophic (Champenois and Borges, 2012). Therefore, the decomposition of leaf litter within the ecosystem was substantially smaller than production. Prior studies utilizing the eddy covariance technique in summer identified greatrer similarity between respiration and 230 primary production in Zostera marina meadows in the mid-Atlantic Bight (Rheuban et al., 2014a) and in Thalassia testudinum meadows in Florida Bay, USA (Long et al., 2015). In those studies NEM was one-fifth of GPP. Thus, P. oceanica meadows may function differently in the highly oligotrophic Mediterranean than Zostera and Thalassia meadows in a more eutrophic sea. It is of interest why seagrass respiration in the Mediterranean is so much lower than in meadows from Florida or the mid-Atlantic Bight.
Nutrient limitation is an explanation. Nitrogen in pore water can limit the primary productivity of P. oceanica meadows in summer 235 (Alcoverro et al., 1997). However, heterotrophic activity is also limited by nutrient availability. The addition of inorganic nitrogen to P. oceanica meadow pore water enhances benthic respiration and diminishes the amount of stored organic carbon (Lopez et al., 1998). Because of this, the oligotrophic environment may contribute to the high carbon storage in P. oceanica meadows.
In addition to the high productivity of meadows, we found that sands from the Mediterranean are unusually unproductive.
Indeed, R in Elba sands was comparable to that of the continental shelf of the South Atlantic Bight (Jahnke et al., 2000), and among 240 the lowest reported for coastal systems (Huettel et al., 2014). Pelagic primary production is limited in the oligotrophic Mediterranean. Thus, there is little organic carbon to stimulate respiration in this environment. Organic matter concentrations in Mediterranean waters become so low in the summer that many benthic suspension feeders undergo dormancy to avoid starvation (Coma and Ribes, 2003).
We may understand the strong differences in activity between seagrasses and sands in oligotrophic waters in the 245 acquisition and retention of nutrients by seagrass meadows. P. oceanica take up nutrients through roots and leaves primarily in winter and early spring, when nutrients are more abundant (Lepoint et al., 2002). Nutrients are reallocated before leaves are shed (Alcoverro et al., 2000). In acquiring nutrients from surrounding waters, the filtering of particles by seagrasses or their epiphytes may be similar to coral reefs (e.g., Rasheed et al., 2002). Seagrasses collect particles passively, through flow-attenuation and particle deposition (Gacia et al., 1999;Gacia and Duarte, 2001). Seagrass also collect particles actively. Macro-and epifaunal 250 suspension feeders (e.g., hydroids; Fig. S3) enhance particle filtering by orders of magnitude over surrounding sands (Lemmens et al, 1996). Epifauna biomass on P. oceanica is one-third of total epiphytic biomass (Lepoint, 1999). Mineralization of captured 7 Biogeosciences Discuss., https://doi. org/10.5194/bg-2018-199 Manuscript under review for journal Biogeosciences Discussion started: 27 April 2018 c Author(s) 2018. CC BY 4.0 License. particles in seagrass sediment supplies nutrients for seagrass growth (Evrard et al., 2005;Barrón et al., 2006b). Their mineralization by epiphytes may be an additional source of nutrients for seagrass growth, as dissolved nutrients can be retained within the waters of seagrass meadows . 255

Meadow productivity at the CO2 vent
Despite the seemingly favorable dissolved CO2 availability and the elevated inorganic nitrogen in the water column at Panarea, seagrasses at the vents did not seem to prosper. So, our hypothesis that seagrass growth would be stimulated at the CO2 vent was incorrect. These results contrast with the increased productivity of seagrass at elevated CO2 and reduced pH (Beer and Koch, 1996;Invers et al., 2001;Palacios and Zimmerman, 2007). Nevertheless, P. oceanica is also not stimulated under comparable, 260 intermediate, ocean acidification scenarios at CO2 vents (Hall-Spencer et al., 2008;Cox et al., 2015;Cox et al., 2016). Thalassia testudinum is also not stimulated by intermediate CO2 enrichment in mesocosms (Campbell and Fourqurean, 2013). However, seagrass production at the CO2 vent in this study was substantially lower than the productivity of reference meadows. GPP was 40% and R was 70% of that at Elba, and NEM was negligible. The small meadow height (half that of Elba) may contribute to the lower GPP and R, but the negligible NEM suggests that this meadow was not storing organic carbon. A concern for seagrass 265 meadows exposed to high CO2 levels is the loss of calcifiers due to low pH (Martin and Gattuso, 2009;Donnarumma et al., 2014).
Calcifiers may be a significant component of meadow productivity. Calcification offsets photosynthetic CO2 uptake in seagrass meadows by half (Barrón et al., 2006a). However, the community of epibionts at the CO2 vent at Basiluzzo was intact and diverse (Guilini et al., 2017). An alternate explanation is that vent fluids alter the chemistry of the site in a way that limits seagrass productivity. Seagrass growing near the vent express more stress-related genes than seagrass at reference sites (Lauritano et al., 270 2015). The vent fluids and sediments at the vents are highly enriched in iron (Price et al., 2015). Iron (III) binds to phosphate and makes it less available in bottom waters (Slomp et al., 1996). Thus, phosphorus may limit seagrass productivity. Vent fluids may also be enriched in potentially harmful trace elements (Vizzini et al., 2013). Generally, the use of natural CO2 vents for studying ocean acidification may require caution. The effects of elevated CO2 may not always be separated from other factors that are induced by seepage. 275 In sands, respiration was enhanced at the CO2 vent relative to Elba but GPP was similar. Enhanced CO2 concentrations may enhance the productivity of microalgae and cyanobacteria by alleviating the energetic requirements of their CO2 concentrating mechanisms (Beardall and Giordano, 2002). Benthic diatoms dominated the benthic community at Basiluzzo (Molari et al., 2018).
Many benthic diatoms are insensitive to pH changes (Hinga, 2002). However, Johnson et al., (2013) found changes in benthic diatom assemblages and increases in chlorophyll a concentrations at a CO2 vent. Nevertheless, the negligible enhancement of 280 primary production at the CO2 vent is consistent with the small net effects of acidification on benthic diatoms and benthic primary production in other studies (Alsterberg et al., 2013;Fink et al., 2017). Nutrient limitation in oligotrophic environments may limit the response of phytoplankton to elevated CO2 (Gazeau et al., 2017). These results indicate that reduced seawater pH has little net effect on photosynthesis by microphytobenthos.

Hydrodynamic exchange 285
Reduced nighttime hydrodynamic exchange has ecological benefits for seagrass. The depletion of oxygen at the open-water meadow during the night (Fig. 1) is evidence that hydrodynamic exchange with surrounding waters is limited. Ecologically, the reduced exchange would benefit seagrass if limiting nutrients that were produced during mineralization at night were retained for primary production during the day. To estimate the potential benefit of nutrient retention we assumed Redfield ratios. A 20 µmol L -1 O2-deficit between the meadow and ambient waters corresponds to a 2.3 µmol L -1 increase in NO3 -. For a 0.6 m-high canopy 290 8 Biogeosciences Discuss., https://doi. org/10.5194/bg-2018-199 Manuscript under review for journal Biogeosciences Discussion started: 27 April 2018 c Author(s) 2018. CC BY 4.0 License. this amounts to 1.4 mmol m -2 of extra NO3 -, and 12 mmol m -2 d -1 of GPP. This is between 7 and 9% of GPP of the meadow. Using Redfield ratios for respiration and production, the same increase in GPP applies to phosphorus. These amounts of enrichment are approximately double the observed inorganic nitrogen and phosphorus enrichment near the bed in P. oceanica meadows . Seagrass are depleted in nitrogen and phosphorus relative to marine seston (Duarte et al., 1990), so the mineralization of marine seston may stimulate greater GPP. Mineralization would also be a source of CO2 for carbon fixation. Assuming a 295 respiratory quotient of 1, the 20 µmol L -1 DIC enrichment corresponds to a small, 0.04 pH decrease (Lewis et al., 1998). This would enhance CO2(aq) by 10%. The potential enrichment of seagrass primary production by nutrient and CO2(aq) retention would be greater for greater excursions in O2 and pH.
Posidonia oceanica may be adapted to diel oscillations in water velocity and waves caused by the solar-driven sea breeze.
For a seagrass meadow to take advantage of nutrient retention, the nutrients would be retained during the night and into the 300 morning. Late in the morning, after nutrients had been depleted, it would be an advantage for hydrodynamic exchange to be enhanced. This would prevent accumulation of O2 which can cause photorespiration (Falkowski and Raven, 1997). This diel pattern in hydrodynamic exchange would be caused by low nighttime water velocities and higher daytime water velocities. Interestingly, this is exactly the diel pattern in water velocities that occurs at the meadow (Fig. 1). The pattern is the result of the solar-driven sea breeze. This velocity pattern is dominant in the Mediterranean (Azorin-Molina et al., 2011), where tides are almost absent, and 305 thus may be exploited by P. oceanica.

Flux-irradiance curves
Our results showed that seagrass meadows are not always light-limited during a typical diel cycle (Fig. 5). Light limitation means that higher light levels will lead to higher photosynthesis rates. Essentially, self-shading from the thick meadow is thought to effectuate light limitation (Binzer et al., 2006;Sand-Jensen, 2007). Light saturation ( ) at the Elba meadows occurred at less than 310 half of peak irradiance. These results are surprisingly similar to light saturation of photosynthesis in P. oceanica fragments (Drew, 1979;Pirc, 1986;Figueroa et al., 2002). Unlike in incubations, saturation occurred for a complete meadow under low in situ light conditions. It contrasts with light-limitation of mature Z. marina and T. testudinum meadows under in situ irradiance (Rheuban et al. 2014;Long et al., 2015), and suggests that a factor other than light may limit primary production at peak irradiance. Nutrient limitation may contribute. Chlorophyll content and the maximum photosynthetic rate of seagrasses increase with nutrient 315 availability (Agawin et al., 1996;Alcoverro et al., 1998;Lee and Dunton 1999). Further measurements would improve our understanding of this response.

Summary
Open water techniques, including eddy covariance, allow measurements of productivity that integrate all ecosystem components under in situ environmental conditions. For seagrass, accurate metabolism measurements may require correction for diel changes 320 in dissolved O2 within meadows. Ecologically, the behavior of P. oceanica meadows differed in one fundamental way from expectationat a CO2 vent productivity was diminished. This may result from the effect of enhanced CO2 on seagrass ecosystem productivity, but it also may result from adverse effects of other vent fluid constituents. In other respects, P. oceanica ecosystem GPP, R, and NEM under in situ conditions behaved similarly to measurements of P. oceanica productivity alone in prior studies.
These meadows had a high productivity, close to half of GPP, and were far more productive than surrounding sands. This is in 325 agreement with prior studies and supports the assertion that P. oceanica is a significant location for biotic carbon storage (Fourqurean et al., 2012). Unlike in other seagrass meadows, photosynthesis did not increase linearly with irradiance during 9 Biogeosciences Discuss., https://doi. org/10.5194/bg-2018-199 Manuscript under review for journal Biogeosciences Discussion started: 27 April 2018 c Author(s) 2018. CC BY 4.0 License.
summer, but instead approached a saturating irradiance. This feature, the low productivity of surrounding sands, and low respiration in seagrass meadows, are all consistent with nutrient limitation in this oligotrophic system. Finally, diel cycles of dissolved oxygen within a meadow suggest a functional adaptation to nutrient limitation in this environment. Daytime peaks and nighttime lulls in 330 wind speed are characteristic of the region in summer and align well with an ideal scenario for a seagrass meadow. Nutrients remineralized during the night can be retained into the morning hours before hydrodynamic exchange is renewed.    at the open-water seagrass meadow. Eddy covariance O2 fluxes measured above the meadow as a function of irradiance (a). O2 fluxes corrected to account for O2 concentration change (i.e. storage) in the water between the eddy covariance measurement volume and the sediment, including the meadow (b). Labels represent the hour of the day.