Production and respiration control the marine microbial metabolic balance in the eastern North Atlantic subtropical gyre

Deep-Sea Research I 58 (2011) 768–775

Contents lists available at ScienceDirect

Deep-Sea Research I journal homepage: www.elsevier.com/locate/dsri

Production and respiration control the marine microbial metabolic balance in the eastern North Atlantic subtropical gyre Marı´a Aranguren-Gassis a,n, Pablo Serret a, Emilio Ferna´ndez a, Juan L. Herrera b, Jose F. Domı´nguez c, Valesca Pe´rez d, Jose Escanez c a

Universidad de Vigo, Departamento de Ecologı´a y Biologı´a Animal, Carretera Colegio Universitario s/n, 36310, Vigo, Pontevedra, Spain. Universidad de Vigo, Departamento de Fı´sica Aplicada, Carretera Colegio Universitario s/n, 36310, Vigo, Pontevedra, Spain c ´fico de Canarias, Calle General Gutie´rrez, 4, 38003 Santa Cruz de Tenerife, Canarias, Spain Centro Oceanogra d ´n, de la Consellerı´a de Medio Ambiente e Desenvolvemento Sostible, Xunta de Galicia. Carretera de Marı´n km. ´n e Informacio ´n Ambiental - CINAM de Louriza Centro de Investigacio 4, Apdo. 127, 36080 Pontevedra, Spain b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 November 2010 Received in revised form 6 May 2011 Accepted 10 May 2011 Available online 24 May 2011

Two main contrasted hypotheses have arisen during the last decades about the factors controlling the planktonic net metabolic balance in oligotrophic waters: gross primary production controls net community production vs. variability of net community production is also influenced by changes in microbial respiration. This work discusses both hypotheses analyzing the variability of metabolic rates along a gradient from the margin to the centre of the North Atlantic oligotrophic gyre, i.e. from relatively productive to more oligotrophic conditions. Net community production (NCP) was close to zero (between  3.34 and  11.77 mmol O2 m  2 d  1) at the margin of the gyre and tended towards net heterotrophy ( 44.03 mmol O2 m  2 d  1) to the centre of the gyre as both gross primary production (GPP) and community respiration (CR) decreased. The strong relationships found between nutrient availability and both NCP and GPP suggest that factors controlling GPP are prevalent in determining NCP variability in this biogeographic region. However implementation of existing models to predict NCP from the measured GPP indicates that the precise estimation of NCP in different oligotrophic systems requires consideration of the magnitude and variability of microbial respiration rates. & 2011 Elsevier Ltd. All rights reserved.

Keywords: North Atlantic subtropical gyre Oligotrophic ocean Marine plankton Net community production Production Respiration

1. Introduction An intense debate about the trophic status of open ocean oligotrophic ecosystems has been addressed during the last decades. Since del Giorgio et al. (1997) found that bacterial respiration exceeds phytoplankton net production in unproductive marine ecosystems, many studies have focused on clarifying the magnitude and trends of net microbial plankton metabolism in the open ocean. Most investigations based on the in vitro change in dissolved oxygen after controlled incubations, have systematically reported net heterotrophy to prevail in low production systems (Duarte and Agustı´, 1998; Duarte et al., 2001; Gist et al., 2009; Gonza´lez et al., 2001; Mora´n et al., 2004; Robinson et al., 2002; Serret et al., 2001; Williams et al., 2004); although using similar in vitro methods, other studies found net autotrophic or balanced metabolism (Karl et al., 2003; Riser and Johnson, 2008; Serret et al., 2006; Williams and Purdie, 1991).

n

Corresponding author. Tel.: þ34 986814087; fax: þ 34 986812556. E-mail address: [email protected] (M. Aranguren-Gassis).

0967-0637/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr.2011.05.003

Regional to global DO2 gross primary production (GPP) and community respiration (CR) databases have been used to validate theoretical models (Lo´pez-Urrutia et al., 2006; Rivkin and Legendre, 2001 for example) and to derive generalizations and empirical models (del Giorgio and Duarte, 2002; Duarte and Regaudie-deGioux, 2009), whose predictive power has been empirically tested (Serret et al., 2009). However neither generalizations nor predictions derived from different databases, regions of the ocean, or by different authors, agree and as a result a new debate arose about what kind of information empirical GPP:CR relationships provide about factors controlling the functional relation of GPP with CR, and ultimately the net community metabolism of the microbial plankton communities in oligotrophic oceans. Several works have reported a relative constancy in respiration rates against more variable primary production rates (Duarte et al., 2001; Gonza´lez et al., 2001; Mora´n et al., 2004), what derives in strong relations between the GPP/CR ratio and GPP (Duarte et al., 2001; Gonza´lez et al., 2001, 2002) that are interpreted as indicative of a control of the microbial metabolic balance by factors controlling GPP (Arı´stegui and Harrison, 2002; Gonza´lez et al., 2002). According to this view, some authors interpret net heterotrophy observations in open ocean systems

M. Aranguren-Gassis et al. / Deep-Sea Research I 58 (2011) 768–775

as the result of universal thresholds of GPP that points out to the prevalence of heterotrophy whenever primary production is low (Duarte and Regaudie-de-Gioux, 2009). Other authors, while supporting the prevailing role of autotrophic processes, highlight the importance of the temporal decoupling of GPP and CR, so that pulses of GPP during autotrophic events would allow the maintenance of net heterotrophy (Williams et al., 2004). The first hypothesis would lead to the conclusion that low production habitats are systematically net heterotrophic, thus requiring an explanation to the source of allochthonous organic matter sustaining the prevalence of CR in open ocean oligotrophic provinces. In the later hypothesis, the excess CR is supported by previous local GPP, so that net autotrophy would prevail regionally in the long-term, even in oligotrophic provinces; this hypothesis in turn requires an explanation to the processes supporting such high episodic primary production rates in these unproductive areas. In any case, both hypotheses coincide in considering that the main controlling factor of NCP is GPP; consequently the variability of NCP in these provinces would be related to nutrient limitation due to stratification derived from the vertical thermohaline ˜ o´n and structure (Gist et al., 2009; Gonza´lez et al., 2002; Maran Holligan, 1999; Mc Andrew et al., 2007). Other works have found that the balance between GPP and CR does not only depend on the total amount of photosynthesis. The observation of both positive and negative NCP in similarly oligotrophic low production systems (Serret et al., 2002) leads to the hypothesis that respiration, and not only GPP variability, plays an important role on the emergence of geographic and seasonal patterns of the metabolic balance (Biddanda et al., 2001; del Giorgio and Williams, 2005; Serret et al., 2006). According to this view, no universal GPP:CR relation or thresholds may exist because the factors controlling GPP (nutrients limitation) and CR (organic matter availability) do not always co-vary, especially in oligotrophic regions, so GPP:CR relations become system-dependant (Serret et al., 2009). Net heterotrophy in low production conditions would be either a result of the delayed consumption of organic matter previously produced locally (e.g., Serret et al.,

769

1999), or the result of the consumption of allochthonous organic matter, mainly in dissolved form (DOM) (e.g., Robinson et al., 2002). In the first case the long-term community metabolism would be controlled by autotrophic processes within the system, while in the latter, trophic dynamics would be donor-controlled and geographic patterns of community metabolism would be associated to the spatial distribution of allochthonous DOM. With the aim to elucidate the relative importance of the factors controlling the microbial NCP in oligotrophic waters, a quasi-longitudinal transect from the margin of the North Atlantic Oligotrophic Gyre to the centre was sampled in autumn 2006. We expected to find a gradient towards more oligotrophic conditions, thus enabling to study the variation of the balance between GPP and CR related to nutrient availability. Moreover, moving away from the main source of dissolved organic matter into the region, the highly productive eastern coastal upwelling system (A´lvarezSalgado et al., 2007; Hansell et al., 2004), the zonal section would also contribute to assess the influence of changes in CR variation on the microbial metabolic balance. In addition to the spatial analyses of observations, NCP predictions from two contrasting published empirical models based on GPP (Duarte et al., 2001; Serret et al., 2009) have been compared with our data as a tool to ascertain the relevance of autotrophy in determining the net community metabolism. Summarizing, we aim to test two of the contrasted hypotheses described above: 1- Production control hypothesis: Primary production is more variable than respiration. Regional production thresholds determine the microbial community net metabolic balance (Arı´stegui and Harrison, 2002; Duarte et al., 2001, Duarte and Regaudie-de-Gioux, 2009) and, therefore, NCP can be reliably predicted from GPP based empirical models. If this hypothesis is correct, GPP and consequently, NCP should tend to decrease as primary production decreased. 2- Production and respiration control hypothesis: Respiration can be as variable as production and then the metabolic balance

Fig. 1. Map of monthly averaged surface chlorophyll concentration (mg m  3) in October 2006. Data were obtained from the ocean colour website (http://oceancolor.gsfc. nasa.gov/; Feldman and McClain, 2006) level-3 products (Modis-aqua mission) with 9 km resolution. White crosses signal the location of the stations sampled.

770

M. Aranguren-Gassis et al. / Deep-Sea Research I 58 (2011) 768–775

can be controlled by any of the two rates depending on the system under study (Serret et al., 2002, 2006). Consequently, GPP rates would not suffice to predict NCP. If this hypothesis is correct, in low production systems respiration should decrease as the availability of allochthonous DOM decreases and therefore, assuming the Eastern coastal upwelling system as the main source of dissolved organic matter (DOM) to the North Atlantic oligotrophic gyre, respiration should decrease as the distance to the allochthonous DOM source increases and, therefore, the microbial metabolic balance might change from net heterotrophy close to coast to balanced at the centre of the oligotrophic gyre.

2. Methods 2.1. Study area Sampling was conducted along the North Atlantic Subtropical Gyral Province East (NASTE in Longhurst, 1998) in October– November 2006 on R.V. Hespe´rides as part of the CARPOS cruise. Seven stations were occupied along a transect crossing the NorthEast bound of the region from 131W 341N to 341W 261N (Fig. 1). Stations were spaced at 360–420 km intervals. The NASTE region is bounded to the North by the Azores current, that flows to the southeast originating a thermal front at about 34–361N (Fasham et al., 1985; Martins et al., 2002). This front fades by the influence of eddies associated with the Canary current, on the eastern bound of the region. The Subtropical Convergence, located around 25–351N (Longhurst, 1998), limits the southern bound of the region. To the west, the region is separated from the North Atlantic Subtropical Gyral Province West (NASTW) by the mid-Atlantic Ridge (40–501W southern than 301N; Sy, 1988), although the 18 1C mode waters of Sargasso Sea (included on NASTW) can extend toward the east in winter. 2.2. Sampling and hydrographic characterization At each station, vertical profiles of temperature and conductivity were carried out with a Seabird 911plus CTD probe. Data were processed using software provided by the manufacturer following UNESCO recommendations (UNESCO technical papers, 1998), and salinity and density calculations were computed following Fofonoff and Millard Jr. (1983). The CTD probe was calibrated by the manufacturer just before the cruise. Vertical irradiance profiles were made at midday with a Satlantic ICP-100 FF radiometer. A statistically significant relationship was found between the depth of the 1% incident irradiance and the depth of the deep fluorescence maximum (linear parametric regression Type I: r2 ¼0.58 p¼0.046; slope not significantly different from 1). Water samples for chemical and biological analyses were drawn from 6–11 discrete depths between 0 and 300 m with 20 l Niskin metal-free bottles rosette fitted to the CTD. Sampling was always conducted before dawn. The mixed layer depth (MLD) was defined as the depth at which the density value differs more than 0.03 kg m  3 from the 10 m depth density value (de Boyer Monte´gut et al., 2004). 2.3. Nitrate and nitrite concentrations 15 ml water samples were collected in stoppered polypropylene conical centrifuge tubes. Samples for nanomolar analysis of NO3 and NO2 were fitted directly onto the AutoSampler of a six channels Technicon—Bran Luebbe AA II AutoAnalyzer

for determination by Continuous Flow Analysis using the method described by Raimbault et al. (1990). Samples for the analysis of micromolar concentration of NO3 and NO2 were frozen and stored for the subsequent analysis at the laboratory following the methodology described by Tre´guer and Le Corre (1975). The depth of the nitracline was calculated by linear interpolation and taken as the depth where the value of 1 mmol m  3 NO3 concentration was reached (Robinson et al., 2006). 2.4. Chlorophyll-a Fluorescence was measured at all stations with a SeaPoint fluorometer fitted to the CTD and converted to chlorophyll-a units through the calibration with simultaneous chlorophyll-a measurements made on acetone extracts from filtered water samples (chlorophyll¼0.9046 fluorescenceþ0.0437; r2 ¼0.81 p o0.001). 250 ml water samples for chlorophyll-a determination were sequentially filtered through 2 and 0.2 mm pore size polycarbonate filters. Chlorophyll-a was extracted with 5 ml of 90% acetone during 12 h at 4 1C in darkness and measured with a Turner Designs 700 fluorimeter calibrated with pure chlorophyll-a standard (Welschmeyer, 1994). 2.5. Oxygen production and consumption Oxygen production and consumption rates were estimated by in vitro changes of dissolved oxygen after 24 h light and dark incubations. Water samples were transferred from the Niskin bottles to 100 ml nominal volume borosilicate bottles individually calibrated, overflowing 4200 ml. Incubations always started pre-dawn, and irradiance levels for incubations were determined assuming that the deep fluorescence maximum corresponded to 1% incident irradiance (Poulton et al., 2006; Robinson et al., 2006) in accordance with the high correlation coefficient obtained (see above). For each depth, four dark bottles samples were fixed immediately for initial oxygen concentration measurement and the rest of the samples (4 in light and 4 in dark bottles) were fixed after 24 h incubation, all of them following Grasshoff et al. (1999) recommendations. On-deck incubations were conducted with in situ simulated temperatures and light conditions. Irradiances between 0.3% and 87% of surface values were adjusted with neutral density meshes. Samples from the upper mixed layer were incubated with running surface seawater while samples collected from the subsurface layer were incubated with water cooled at the corresponding in situ temperature at the time of sampling (18–22 1C). Dark bottles were placed inside opaque fabric bags and then incubated together with light bottles. Dissolved oxygen was measured by precision Winkler titration performed with a Metrohm 716 DMS Titrino utilizing a potentiometric end point (Oudot et al., 1988; Serret et al., 1999). Community respiration (CR) was calculated from the difference between the averaged dissolved oxygen concentration in the incubated dark bottles and that in the initial samples. Net community production (NCP) was calculated from the difference between the averaged dissolved oxygen concentration in the incubated light bottles and that in the initial samples. Gross primary production (GPP) was calculated from the difference between the averaged dissolved oxygen concentration in the incubated light bottles and that in the incubated dark bottles. Median coefficient of variation of the dissolved oxygen replicates was 0.16% and means of the standard errors of CR, GPP and NCP were 0.29, 0.3 and 0.28 mmol O2 m  3 d  1, respectively. Photic zone integrated values were calculated by trapezoidal integration of the volumetric data from the surface to the depth of the 1% incident irradiance. Photic zone integrated metabolic rates were weighted by depth for the zonal trend evaluation to avoid

M. Aranguren-Gassis et al. / Deep-Sea Research I 58 (2011) 768–775

differences due to geographical variations in the depth of the photic zone. Results are presented with propagated standard deviations calculated following propagation procedures for independent measurements described by Miller and Miller (1988). 2.6. Statistics Normal distribution of the data was always checked before statistical analysis through the Kolmogorov–Smirnov and Shapiro–Wilk tests with a 0.05 significance level. In the case of Type II regression, the p value of the correspondent type I regression is presented. All statistical tests except type II regressions were made in SPSS 15.0 support. For type II regressions calculations the RMA software designed by the San Diego University was used.

3. Results 3.1. Thermohaline vertical distribution Temperature and salinity distributions (Fig. 2) show a gradual westward transition towards warmer and more saline waters. Surface temperature increased more than 2 1C from 23 1C at the easternmost station to 25.8 1C at station 6. Surface salinity also increased from 36.83 to 37.51 along the transect. Sinking of isolines at station 5 is coherent with the observation of a positive sea level anomaly, suggesting the presence of an anticyclonic mesoscale structure at this station (Gonza´lez-Taboada et al., 2010).

6

5

STATION NUMBER 2 4 3

3.2. Nitrate concentration and phytoplankton chlorophyll-a Nitrate concentration (Fig. 3a) was lower than 0.5 mmol m  3 in the upper 100 m in most of the transect and increased markedly below the photic zone. The nitracline significantly deepened westward (nitracline depth vs. distance covered linear parametric regression Type I: r2 ¼0.75 p ¼0.027). Surface chlorophyll-a concentration was lower than 0.2 mg m  3 at all the stations (Fig. 3b). A slight increase was found at surface waters of stations 1 and 5. A well developed deep chlorophyll-a maximum (DCM) was observed at the base of the photic layer along the transect (Fig. 3b). The depth of the DCM increased westward from 96 m at station 0 to 143 m at station 6 (DCM depth vs. distance covered linear parametric regression Type I: r2 ¼0.81 p ¼0.014). Chlorophyll-a concentration at the DCM decreased westward from 0.52 mg m  3 at station 0 to 0.34 mg m  3 at station 6 (DCM chlorophyll-a concentration vs. distance covered linear parametric regression Type I: r2 ¼0.77 p¼0.022). Both zonal trends were interrupted by a ca. 30 m shallowing of the DCM associated to 0.09 mg m  3 chlorophyll-a concentration increase at station 5. 3.3. Metabolic rates Measured values of GPP, CR and NCP shown in Fig. 4a, b and c, were within the range of values reported for the same region and season (Gist et al., 2009; Gonza´lez et al., 2002; Mora´n et al., 2004; Robinson et al., 2002; Serret et al., 2006; Williams, 1998).

6

1

771

5

STATION NUMBER 2 4 3

1

0

0

0 50

DEPTH (m)

DEPTH (m)

50 100 150

100 150 200

200 250

250

NO (mmol m )

300 0

TEMPERATURE (ºC)

300 0

50

DEPTH (m)

DEPTH (m)

50 100 150

100 150 200

200 250

250 300

2200 2000 1800 1600 1400 1200 1000 800 600 400 200

CHLOROPHYLL-A (mg m )

300

SALINITY

2200 2000 1800 1600 1400 1200 1000 800 600 400 200

0

Km Fig. 2. Spatial distribution of (a) temperature (1C) and (b) salinity along the transect. Numbers on the top indicate the location of the seven stations sampled. The dashed line signals the MLD.

0

Km Fig. 3. Vertical distribution of (a) NO3 concentration (mmol m  3) and (b) chlorophylla concentration (mg m  3) along the transect. Numbers on the top indicate the location of the seven stations sampled and black crosses show sampling depths. The dashed line signals the MLD. The dash-dotted line shows the depth of the nitracline.

772

M. Aranguren-Gassis et al. / Deep-Sea Research I 58 (2011) 768–775

0

6

5

STATION NUMBER 2 4 3

2

1

mmol O2 m-3 d-1

DEPTH (m)

100 150

1

0

-1

-2 2400 GPP (mmol O m d )

300 0

0

4

6

NCP CR GPP

200 250

2100

1800

1500

1200

900

600

300

0

Km Fig. 5. Photic layer integrated values of the metabolic rates weighted by depth (mmol O2 m  3 d  1): GPP (dashed dark-grey line), CR (dashed light-grey line) and NCP (solid black line) at the seven stations of the transect. Error bars represent propagated standard deviation.

50

DEPTH (m)

2

3

1

50

100 150 200 250 CR (mmol O m d )

300 0 50

DEPTH (m)

5

0

100 150 200 250 300

NCP (mmol O m d )

2200 2000 1800 1600 1400 1200 1000 800 600 400 200

0

Km Fig. 4. Distribution of metabolic rates (mmol O2 m  3 d  1) along transect: (a) GPP, (b) CR and (c) NCP. Numbers on the top indicate the locations of the seven stations sampled and black crosses show sampling depths. The dashed line signals the depth of the DCM and the dash-dotted line shows the depth of the nitracline.

GPP and CR varied between 0.02 and 2.6 mmol O2 m  3 d  1. The highest rates were measured at the subsurface of station 5 (3–4 times higher than the average for all other stations). The complete volumetric data set, including the standard error of every measurement, is available at the global respiration data base: http://www.uea.ac.uk/env/people/facstaff/plankton (data compiled and maintained by Carol Robinson initially for Robinson, 2008). GPP and CR rates (Fig. 4a and b) presented similar trends along the transect (RMA Type II regression: r2 ¼0.49; p o0.001) according to the equation: log CR¼0.8 log GPP 0.03. The slope of this regression indicates that CR tended to be slightly higher than GPP. Consequently, NCP was generally heterotrophic and varied between  1 and 0.8 mmol O2 m  3 d  1 (Fig. 4c). Autotrophy

was estimated at stations 2 and 5, the last one affected by the mesoscale structure (Figs. 2 and 3) where maxima values of GPP and CR were measured. At station 2 estimates of autotrophic metabolism correspond with enhanced nutrient concentration at the DCM (Fig. 3) consistent with the proximity of the Canary Current. A net metabolic O2 balance close to 0 mmol O2 m  3 d  1 was found at the eastern stations (stations 0 and 1), where the nitracline and DCM were shallower and chlorophyll-a concentration were higher compared to the western stations. Photic layer integrated GPP rates (Fig. 5) decreased slightly to the west, except at station 5 (GPP vs. distance covered linear parametric regression Type I: r2 ¼0.58 p¼0.077). Photic layer integrated CR rates presented a similar, although less marked, westward decreasing trend when station 5 was excluded (CR vs. distance covered linear parametric regression Type I: r2 ¼0.54 p¼0.095). Maxima rates of both CR and GPP were estimated at station 5. The variability of GPP and CR measurements was analyzed by calculating the coefficient of variation (C.V.) of both photic layer integrated and volumetric rates. The C.V of photic layer integrated GPP along the transect (72%) was higher than that of the photic layer integrated CR rate (41%). This difference increased when the mesoscale affected station 5 was excluded (GPP C.V.¼60% vs. CR C.V.¼19%). Similarly, the averaged C.V. of volumetric GPP rates for each station (83%) was higher than that of volumetric CR (61%). Photic layer integrated NCP rates were close to zero at the easternmost stations (stations 0, 1 and 2) and tended towards net heterotrophy to the west (NCP vs. distance covered linear parametric regression Type I: r2 ¼ 0.29 p ¼0.266), with the exception of station 5 where the high depth integrated GPP and CR rates were balanced. The corresponding photic layer integrated NCP not weighted by depth varied between  58.42 715.38 and 19.01718.59 mmol O2 m  2 d  1 (NCP with standard error). To explore the relationship between O2 metabolic rates and nutrient availability a nutrient limitation index was calculated as the difference between the mixed layer depth and the nitracline depth (Behrendfeld et al., 2002; Gonza´lez et al., 2002; Gist et al., 2009). Station 5, which represented a marked zonal discontinuity in all the physical, chemical and biological variables studied, was excluded from this analysis as GPP and CR rates measured at this station were out of the 95% confidence interval of the complete data set. A negative and statistically significant relationship was found between the nutrient availability index and both GPP and NCP (Fig. 6).

M. Aranguren-Gassis et al. / Deep-Sea Research I 58 (2011) 768–775

180

r 2=0.859 p=0.023

GPP (mmol O2 m-2 d-1)

160 140 120 100 80 60 40 20 0 40

r 2=0.920 p=0.010

NCP (mmol O2 m-2 d-1)

20 0 -20 -40 -60 -80 50

60

70

80

90

100

110

Nutrient limitation index (m) Fig. 6. Relation between the nutrient limitation index (m) and photic layer integrated values of GPP (a) or NCP (b) (mmol O2 m  2 d  1). Open circles represent station 5 and solid lines represent the Type II regression fit according to the equations: GPP ¼ 199.05þ 1.73(Nutrient limitation index) and NCP ¼ 118.53þ 1.75 (Nutrient limitation index).

4. Discussion As expected, a westward sinking of the nitracline and the DCM, together with an associated decreasing trend of chlorophyll-a concentrations were observed in this study, illustrating a zonal gradient towards more oligotrophic conditions. This spatial trend in microbial oxygen metabolism allows us to test the two main hypotheses on the control of net microbial plankton metabolism in the open ocean (see Section 1): (1) GPP is more variable than CR and NCP variability is controlled and could be predicted from GPP rates and (2) CR could be as variable as GPP and therefore knowledge of GPP is not enough to predict trends of NCP. In accordance with previous studies indicating that availability of dissolved inorganic nutrients is the main factor limiting GPP in ˜ o´n and Holligan, 1999), lower GPP this oligotrophic region (Maran rates were measured in the more nutrient-limited western stations. By contrast, only a very slight zonal decreasing trend in CR rates was detected, coherent with a potential reduction in the availability of allochthonous DOM from coastal origin. Furthermore, CR rates showed a consistently lower variability than GPP, because of the higher decrease of primary production at the stations with lowest nutrient availability as shown by the relation between GPP and the nutrient limitation index (Fig. 6). Similarly, lower variability of CR than GPP has been reported by previous studies (Arı´stegui and Harrison, 2002; Duarte et al., 2001; Gonza´lez et al., 2001; Mora´n et al., 2004) and it has been mainly attributed to the temporal decoupling between GPP and CR (Arı´stegui and Harrison, 2002). As a consequence, a westward decreasing trend in NCP rates was observed and thus, net heterotrophy increased as primary production decreased. The significant relationship found between NCP and the nutrient limitation index (Fig. 6b) further suggests that the variability of

773

NCP is related to factors controlling GPP. These observations agree with the hypothesis that GPP is more variable than CR, and NCP variability is controlled and could be predicted from GPP rates (see Section 1). The westward decreasing trend of NCP was significantly altered by the high GPP and CR rates measured at station 5, affected by mesoscale activity, where a balanced microbial metabolism was estimated. According to Gonza´lez-Taboada et al., 2010, this mesoscale eddy structure was, by the time of sampling, propagating westward with velocities around 100 km per month from the Canary Archipelago and was under the influence of a northward current coming from the southern boundary of the subtropical gyre one month before the date of sampling. A bloom of Trichodesmium spp. was detected at this station related to the retention of southern populations into the eddy structure (Gonza´lez-Taboada et al., 2010). It has been proposed that the intermittent nature of GPP in contrast with more constant CR causes temporal decoupling between autotrophic and heterotrophic processes allowing the maintenance of net heterotrophy in oligotrophic oceans (Williams et al., 2004). In this context, eddy structures have been suggested as one of the possible episodic sources of organic carbon to ˜ o et al., 2002). oligotrophic oceans (Gonza´lez et al., 2001; Mourin Thus, higher GPP than CR rates would be expected at the eddy affected station. However, the subsurface increase of GPP measured at station 5 was associated with a CR enhancement of similar magnitude. In addition, NO2 concentration below the photic zone at station 5 (0.313 mmol m  3) was one order of magnitude higher than the average values of the transect (0.030 mmol m  3; s.e.¼0.004; n¼58) suggesting a high and recent remineralisation activity usually associated to Trichodesmium dominated microbial communities (Sellner, 1992; Mullholand, 2007) probably resulting from a close coupling between autotrophic and heterotrophic processes by the time of sampling. When only stations not affected by mesoscale eddy dynamics are considered, the apparent relationship found between nutrient supply and NCP is consistent with the production control hypothesis, and consequently NCP rates should be reliably estimated from predictive empirical models dependant on GPP rates (del Giorgio and Duarte, 2002; Serret et al., 2002). To avoid the possible bias associated to the use of volumetric discrete data for the study of geographical patterns (Williams, 1998) we used areal integrated values. In addition, taking into account the possible system-dependence of the models (e.g. Serret et al., 2009) and given that factors limiting primary production seem to control the microbial metabolic balance in the present study, we considered that the best NCP predictive model should be that derived from mostly oligotrophic stations and a range of primary production similar to that measured at CARPOS cruise. Such conditions restrict the number of suitable models to be tested to the equation published by Duarte et al. (2001) with a data set collected at the NASTE region from 9 cruises (since March 1991 to March 2000), and that derived by Serret et al. (2009) with data from one Atlantic latitudinal cruise (AMT11, September–October 2000). As a testing exercise, we estimated NCP rates for the zonal transect from GPP and the Duarte et al. (2001) and Serret et al. (2009) models, and compared the results with in situ NCP measurements (Fig. 7a and b). We used a photosynthetic quotient (PQ) 1.15 for the application of the Serret et al. (2009) model. That PQ value is the average of the PQ range cited in Serret et al. (2009). Type II regressions equations describing predicted vs. measured NCP relations were calculated (Table 1) to validate the predictions. The two relationships were significantly fitted to a linear regression and slopes were not different from 1, pointing that both models were capable to simulate the spatial variation of

774

M. Aranguren-Gassis et al. / Deep-Sea Research I 58 (2011) 768–775

NCP rates measured during the CARPOS transect. However, only the Y-intercept obtained with the Duarte et al. (2001) model did not differ from 0 (Table 1), so only this model seems to yield reliable estimations of the magnitude, not only the variability, of the NCP rates measured along the CARPOS transect.

Serret et al (2009) AMT11 in CARPOS

NCP (mmol O2 m-2 d-1)

250

0

-250

Duarte et al (2001) in CARPOS

NCP (mmol O2 m-2 d-1)

250

0

-250

2500

2000

1500

1000

500

0

Km

Duarte et al (2001) in AMT11

NCP (mmol O2 m-2 d-1)

250

0

-250

-45 N

-30

-15

0 LATITUDE 2

15

30 S

1

Fig. 7. Photic layer integrated NCP (mmol O2 m d ) predicted (grey circles) for the CARPOS transect derived from the Serret et al. (2009) AMT11 model (a) and the Duarte et al. (2001) model (b), and for the AMT11 cruise derived from the Duarte et al. (2001) model (c). The Atlantic Meridional Transect (AMT)-11 cruise was conducted in September–October 2000 between Grimsby, UK (53.5N–0.1W) and Montevideo, Uruguay (34.8S–56.2W; Robinson et al., 2006). Grey bands show NCP predicted standard errors. Black circles are the in situ measured photic layer integrated NCP (mmol O2 m  2 d  1) and error bars represent the propagated standard errors.

NCP could be reliably predicted from NCP-GPP models provided that the metabolic balance of the microbial communities were exclusively controlled by primary production rates. However respiration rates reported by Serret et al. (2002) during the AMT11 cruise were lower than respiration rates determined during the CARPOS cruise for similar production rates. These differences in respiration rates explain the overestimations of NCP rates derived from the Serret et al. (2009) model (Fig. 7a). By contrast, the Duarte et al. (2001) model was constructed with data obtained from the same region studied in this investigation and a good correlation was obtained between our NCP data set and predicted values, suggesting that the data base used to construct the Duarte et al. (2001) model appears to be adequate to characterize NCP spatial variability and so that, to predict the microbial metabolic balance at the NASTE region. Nevertheless the wide range of the error associated to predicted NCP values (grey bands in Fig. 7) reflects the low precision of the prediction. The error associated to the prediction of NCP by the Duarte et al. (2001) model is ca. 100 mmol O2 m  2 d  1(Fig. 7). This uncertainty is broader than the range of measured NCP along the CARPOS transect (77 mmol O2 m  2 d  1). Such low precision prevents from accurately predicting small changes in the magnitude of the NCP. To further test the generality of the hypothesis that GPP suffice to predict NCP, we used the Duarte et al. (2001) model to predict the NCP rates measured at the AMT11 cruise, with the aim to assess the universality of this empirical model for NCP prediction (Fig. 7c). The range of measured NCP along the AMT11 latitudinal section (185 mmol O2 m  2 d  1) was larger than the uncertainty associated to the Duarte et al. (2001) model prediction (100 mmol O2 m  2 d  1). Consequently, these results should allow us to test the predictive power of the Duarte et al. (2001) model. The relationships between predicted and measured NCP rates (Table 1) were significantly fitted to a linear regression with a slope not different from 1 and a Y-intercept different from 0. The predicted NCP agreed well with AMT11 in situ measured NCP rates at the equatorial region where autotrophic metabolism was determined. By contrast, modelled NCP values subestimated in situ measured NCP at mid latitudes. The mismatch between predicted and measured NCP rates at the oceanic region covered by the Duarte et al. (2001) data base was related to the low respiration rates measured at this region during the AMT11 cruise, values which were lower than any of the respiration rates included in the Duarte et al. (2001) model data set. Consequently, it appears that neither the range of primary production values nor the geographical location are sufficient conditions as to assure universal predictive ability for empirical models aiming to predict microbial NCP rates from GPP. In conclusion, the results presented in this paper show that, in accordance with the production control hypothesis, factors limiting primary production appear to control the microbial metabolic balance at local scales. However, in agreement with the second hypothesis, variability in microbial respiration rates over spatial and/or temporal scales prevents reliable universal predictions to be made from models exclusively based on primary production rates.

Table 1 Equation parameters from the Type II linear regression for measured vs. predicted NCP rates. C.I. is the 95% confidence interval (Sokal and Rohlf, 1995). MEASURED NCP

MODEL FOR PREDICTED NCP

r2

p

SLOPE (C.I. 95%)

Y-INTERCEPT (C.I. 95%)

CARPOS CARPOS AMT11

Serret et al. (2009) AMT11 Duarte et al. (2001) Duarte et al. (2001)

0.74 0.62 0.70

n

1.45 (0.60;2.30) 1.99 (0.58;3.41) 1.35 (0.97;1.73)

116.14 (86;146.28) 12.11 (  37.89;62.11)  56.8 ( 80.98;  32.62)

n nn

M. Aranguren-Gassis et al. / Deep-Sea Research I 58 (2011) 768–775

Acknowledgment We thank the scientific CARPOS team, especially the cruise principal scientist R. Varela, and the captain and crew on board R.V. Hespe´rides, especially U.T.M. staff. E. Teira and E. Aguiar measured size ˜a, fractionated chlorophyll-a concentration while E. Gonza´lez, R. Gran ˜o provided radiometer and CTD data. We also T. Martı´n and B. Mourin thank D.C. Lo´pez for her assistant with satellite image. M.A.G. was founded by a FPI-MICINN fellowship. This research was supported by project CARPOS (MEC, REN2003-09532-C03-03). References A´lvarez-Salgado, X.A., Aristegui, J., Barton, E.D., Hansell, D.A., 2007. Contribution of upwelling filaments to offshore carbon export in the subtropical Northeast Atlantic Ocean. Limnol. Oceanogr. 52, 1287–1292. Arı´stegui, J., Harrison, W.G., 2002. Decoupling of primary production and community respiration in the ocean: implications for regional carbon studies. Aquat. Microb. Ecol. 29, 199–209. ˜ on, E., Siegel, D.A., Hooker, S.B., 2002. Photoacclimatation Behrendfeld, M.J., Maran and nutrient-based model of light-saturated photosynthesis for quantifying oceanic primary production. Mar. Ecol. Prog. Ser. 228, 103–117. Biddanda, B., Ogdahl, M., Cotner, J., 2001. Dominance of bacterial metabolism in oligotrophic relative to eutrophic waters. Limnol. Oceanogr. 46 (3), 730–739. de Boyer Monte´gut, C., Madec, G., Fischer, A.S., Lazar, A., Iudicone, D., 2004. Mixed layer depth over the global ocean: An examination of profile data and a profile-based climatology. J. Geophys. Res. 109, C12003. doi:10.1029/ 2004JC002378. del Giorgio, P.A., Duarte, C.M., 2002. Respiration in the open ocean. Nature 420, 379–384. del Giorgio, P.A., Cole, J.J., Cimbleris, A., 1997. Respiration rates in bacteria exceed phytoplankton production in unproductive aquatic systems. Nature 385, 148–151. del Giorgio, P.A., Williams, P.J.le B., 2005. Respiration in aquatic ecosystem, 1st ed. Oxford University. Duarte, C.M., Regaudie-de-Gioux, A., 2009. Thresholds of gross primary production for the metabolic balance of marine planktonic communities. Limnol. Oceanogr. 54 (3), 1015–1022. Duarte, C.M., Agustı´, S., Aristegui, J., Gonza´lez, N., Anado´n, R., 2001. Evidence for a heterotrophic subtropical northeast Atlantic. Limnol. Oceanogr. 46 (2), 425–428. Duarte, C.M., Agustı´, S., 1998. The CO2 balance of unproductive aquatic ecosystems. Science 281, 234–236. Fasham, M.J.R., Platt, T., Irwin, B., Jones, K., 1985. Factors affecting the spatial pattern of the deep chlorophyll maximum in the region of the Azores front. Prog. Oceanogr. 14, 129–165. Feldman, G.C., McClain, C.R., 2006. In: Kuring, N., Bailey, S.W. (Eds.), Ocean Color Web, Modis-aqua. NASA Goddard Space Flight Center. Fofonoff, N.P., Millard Jr., R.C., 1983. Algorithms for Computation of fundamental Properties of Seawater. Endorsed by UNESCO/SCOR/ICES/IAPSO Joint Panel on Oceanographic Tables and Standards and SCOR Working Group 51. UNESCO technical papers in marine science. Paris (44): 53. Gist, N., Serret, P., Woodward, E.M.S., Chamberlain, K., Robinson, C., 2009. Seasonal and spatial variability in plankton production and respiration in the Subtropical Gyres of the Atlantic Ocean. Deep Sea Res. Part II 56, 931–940. ˜ o´n, E, 2002. Large-scale variability of planktonic Gonza´lez, N., Anado´n, R., Maran net community metabolism in the Atlantic Ocean: importance of temporal changes in oligotrophic subtropical waters. Mar. Ecol. Prog. Ser. 233, 21–30. ˜o, B., Ferna´ndez, E., Sinha, B., Esca´nez, J., de Armas, D., Gonza´lez, N., Anado´n, R., Mourin 2001. The metabolic balance of the planktonic community in the North Atlantic Subtropical Gyre: the role of mesoscale instabilities. Limnol. Oceanogr. 46 (4), 946–952. ¨ Gonza´lez-Taboada, F., Gonza´lez-Gil, R., Hofer, J., Gonza´lez, S., Anado´n, R., 2010. Trichodesmium spp. population structure in the eastern North Atlantic subtropical gyre. Deep Sea Res. Part I 57, 65–77. Grasshoff, K., Kremling, K., Ehrhardt, M., 1999. Determination of oxygen. In: Grasshoff, K. (Ed.), Methods of Seawater Analysis. Wiley-VCH, Winheim (Chapter 4). Hansell, D.A., Ducklow, H.W., Macdonald, A.M., Baringer, M.O., 2004. Metabolic poise in the North Atlantic Ocean diagnosed from organic matter transport. Limnol. Oceanogr. 49, 1084–1094. Karl, D.M., Laws, E.A., Morris, P.J., Williams, P.J.leB., Emerson, S., 2003. Metabolic balance of open sea. Nature 426, 32. Longhurst, A.R., 1998. Ecologycal Geography of the Sea. Academic Press. Lo´pez-Urrutia, A., San Martin, E., Harris, R.P., Irigoien, X., 2006. Scaling the metabolic balance of the oceans. Proc. Natl. Acad. Sci. USA 103, 8739–8744. ˜ o´n, E., Holligan, P.M, 1999. Photosynthetic parameters of phytoplankton Maran from 501N to 501S in the Atlantic Ocean. Mar. Ecol. Prog. Ser. 176, 191–203.

775

Martins, C.S., Hamann, M., Fiu´za, A.F.G., 2002. Surface circulation in the eastern North Atlantic, from drifters and altimetry. J. Geophys. Res. 107, C1. doi:10.1029/2000JC000345. ¨ Mc Andrew, P.M., Bjorkman, K.M., Church, M.J., Morris, P.J., Jachowski, N., Williams, P.J le B., Karl, D.M., 2007. Metabolic response of oligotrophic plankton communities to deep water nutrient enrichment. Mar. Ecol. Prog. Ser. 332, 63–75. Miller, J.N., Miller, J.C., 1988. Statistics for Analytical Chemistry, 2nd ed. Ellis Horwood Limited. Mora´n, X.A.G., Ferna´ndez, E., Pe´rez, V., 2004. Size-fractionated primary production, bacterial production and net community production in subtropical and tropical domains of the oligotrophic NE Atlantic in autumn. Mar. Ecol. Prog. Ser. 274, 17–29. ˜ o, B., Ferna´ndez, E., Esca´nez, J., de Armas, D., Giraud, S., Sinha, B., Pingree, R., Mourin 2002. A subtropical oceanic ring of magnitude (STORM) in the Eastern North Atlantic: physical, chemical and biological properties. Deep Sea Res. Part II 49, 4003–4021. Mullholand, M.R., 2007. The fate of nitrogen fixed by diatrophs in the ocean. Biogeosciences 4, 37–51. Oudot, G., Gerard, R., Morin, P., 1988. Precise shipboard determination of dissolved oxygen (Winkler procedure) for productivity studies with a commercial system. Limnol. Oceanogr. 33 (1), 146–150. Poulton, A.J., Holligan, P.M., Hickman, A., Kim, Y.-N., Adey, T.R., Stinchcombe, M.C., Holeton, C., Root, S., Woodward, E.M., 2006. Phytoplancton carbon fixation, clorophyll-biomass and diagnostic pigments in the Atlantic Ocean. Deep Sea Res. Part II 53, 1593–1610. Raimbault, P., Slawyk, G., Coste, B., Fry, J., 1990. Feasibility of using an automated colorimetric procedure for the determination of seawater nitrate in the 0 to 100 ZM range: examples from field and culture. Mar. Biol. 104, 347–351. Riser, S.C., Johnson, K.S., 2008. Net production of oxygen in the subtropical ocean. Nature 451, 323–326. Rivkin, R.B., Legendre, L., 2001. Biogenic carbon cycling in the upper ocean: effects of microbial respiration. Science 291, 2398–2400. Robinson, C., Poulton, A.J., Holligan, P.M., Baker, A.R., Forster, G., Gist, N., Jickells, T.D., Malin, G., Upstill-Goddard, R., Williams, R.G., Woodward, E.M.S., Zubkov, M.V., 2006. The Atlantic Meridional Transect (AMT) Programme: a contextual view 1995–2005. Deep Sea Res. Part II 53, 1485–1515. Robinson, C., Serret, P., Tilstone, G., Teira, E., Zubkov, M.V., Rees, A.P., Woodward, E.M., 2002. Plankton respiration in the Eastern Atlantic Ocean. Deep Sea Res. Part I 49, 787–813. Robinson, C., 2008. Heterotrophic bacterial respiration. In: Kirchman, D.L. (Ed.), 2008 Microbial Ecology of the Oceans 2nd ed. John Wiley & Sons, Inc. (Chapter 9). Sellner, K.G., 1992. Trophodynamics of marine cyanobacterial blooms. In: Carpenter, E.J., Capone, D.G., Rueter, J.G. (Eds.), Marine pelagic cyanobacteria: Trichodesmium and other diatrophs. Kluwer Academic Publisher Group (Chapter 6). Serret, P., Robinson, C., Ferna´ndez, E., Teira, E., Tilstone, G., Pe´rez, V., 2009. Predicting plankton net community production in the Atlantic Ocean. Deep Sea Res. Part II 56, 941–953. Serret, P., Ferna´ndez, E., Robinson, C., Woodward, E.M., Pe´rez, V., 2006. Local productivity does not control the balance between plankton photosynthesis and respiration in the open Atlantic Ocean. Deep Sea Res. Part II 53, 1611–1628. Serret, P., Ferna´ndez, E., Robinson, C., 2002. Biogeographic differences in the net ecosystem metabolism of the open ocean. Ecology 83, 3225–3234. Serret, P., Robinson, C., Ferna´ndez, E., Teira, E., Tilstone, G., 2001. Latitudinal variation of the balance between plankton photosynthesis and respiration in the eastern Atlantic Ocean. Limnol. Oceanogr. 46 (7), 1642–1652. Serret, P., Ferna´ndez, E., Sostres, J.A., Anado´n, R., 1999. Seasonal compensation of microbial production in a temperate sea. Mar. Ecol. Prog. Ser. 187, 43–57. Sokal, R.R., Rohlf, F.J., 1995. Biometry: The Principles and Practice of Statistics in Biological Research, 3rd ed. W.H. Freeman and Company, cop. Sy, A., 1988. Investigation of large-scale circulation patterns in the central North Atlantic: the North Atlantic Current, the Azores Current, and the Mediterranean Water plume in the area of the Mid-Atlantic Ridge. Deep Sea Res. 35, 383–413. Tre´guer, P., Le Corre, P., 1975. Manuel d’analyse des sels nutritifs dans l’eau de mer. Utilisation de l’autoanalyseur Technicon II, 2nd ed. Loc. Univ. Bretagne Occidentale, Brest. UNESCO, 1998. The UNESCO Technical Papers: The Acquisition, Calibration, and Analysis of CTD Data. UNESCO Technical Papers in Marine Sciences, 51. Welschmeyer, N.A., 1994. Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and pheopigments. Limnol. Oceanogr. 39 (8), 1985–1992. Williams, P.J.le B., Morris, P.J., Karl, D.M., 2004. Net community production and metabolic balance at the oligotrophic ocean site, station ALOHA. Deep Sea Res. Part I 51, 1563–1578. Williams, P.J.le B., 1998. The balance of plankton respiration and photosynthesis in the open oceans. Nature 394, 55–57. Williams, P.J.le B., Purdie, D.A., 1991. In vitro and in situ derived rates of gross production, net community production and respiration of oxygen in the oligotrophic subtropical gyre of the North Pacific Ocean. Deep Sea Res. 38, 891–910.