Spatial patterns of plankton biomass and stable isotopes reflect the influence of the nitrogen-fixer Trichodesmium along the subtropical North Atlantic.

1 Journal of Plankton Research, 2013 , doi:10.1093/plankt/fbt011 http://plankt.oxfordjournals.org/content/early/2013/02/23/plankt.fbt011.abstract Spatial patterns of plankton biomass and stable isotopes reflect the influence of the 1 nitrogen-fixer Trichodesmium along the subtropical North Atlantic 2 Carmen Mompeán* 1 , Antonio Bode 1 , V.M. Benítez-Barrios 2 , J. Francisco Domínguez3 Yanes 2 , José Escánez 2 and Eugenio Fraile-Nuez 2 4 1 Instituto Español de Oceanografía. Centro Oceanográfico de A Coruña. E15080 A Coruña 5 (Spain) 6 2 Instituto Español de Oceanografía. Centro Oceanográfico de Canarias. Vía Espaldón, 7 Dársena Pesquera, Parcela 8, E38180 Santa Cruz de Tenerife (Spain) 8 *Corresponding author. Phone: +34-981205362, Fax: +34-981229077, E-mail: 9 [email protected] 10 Abstract 11 The spatial variability of biomass and stable isotopes in plankton size fractions in the upper 12 200 m was studied in a high spatial resolution transect along 24ºN from Canary Islands to 13 Florida to determine nitrogen and carbon sources. Vertical advection of waters predominated 14 in lateral zones while the central Atlantic (30-70º W) was characterised by a strong 15 stratification and oligotrophic surface waters. Plankton biomass was low in the central zone 16 and high in both eastern and western sides, with most of the variability due to either large 17 (>2000 µm) and small plankton (<500 µm). Carbon isotopes reflected mainly the advection 18 the deep water in lateral zones. Stable nitrogen isotopes showed a nearly symmetrical spatial 19 distribution in all fractions, with the lowest values (δ 15 N<1‰) in the central zone, and were 20 inversely correlated to carbon stable isotopes (δ 13 C) and to the abundance of the nitrogen21 fixer Trichodesmium. Diazotrophy was estimated to account for >50% of organic nitrogen in 22 the central zone, and even >30% in eastern and western zones. The impact of diazotrophy 23 increased with the size of the organisms, supporting the wide participation of all trophic levels 24 in the processing of recently fixed nitrogen. These results indicate that atmospheric sources of 25 2 carbon and nitrogen prevail over deep water sources in the subtropical North Atlantic and that 26 the zone influenced by diazotrophy is much larger than reported in previous studies. 27 Keywords: Stable isotopes, plankton, Subtropical North Atlantic, Trichodesmium 28 29 3 INTRODUCTION 30 Large regions of the ocean at subtropical latitudes are characterised by gyres of ocean currents 31 rotating clockwise in the Northern Hemisphere. Biological production in the central regions 32 of these gyres is generally low because of low nutrient inputs while production is enhanced at 33 their borders (e.g. Behrenfeld et al., 2006). For instance, the supply of nitrogen from deep 34 waters to the photic zone is lowest in the middle oceanic gyres, where a deep thermocline and 35 smooth nutrient gradients determine slow rates of nutrient supply by diffusion (Mouriño36 Carballido et al., 2011). Notwithstanding their low production, these gyres contribute a large 37 fraction of global biogenic carbon export into the deep ocean because of their size (Emerson 38 et al., 1997; Karl et al., 2008). 39 A variety of physical mechanisms are known to contribute to nitrogen inputs in oligotrophic 40 gyres, including mesoscale and submesoscale turbulence (Oschlies and Garçon, 1998), lateral 41 transport from other regions (Williams and Follows, 1998; Torres-Valdes et al., 2009), and 42 atmospheric deposition (Duce et al., 2008). However, biological fixation of atmospheric N 2 43 (diazotrophy) can be also a major input of nitrogen in the oligotrophic ocean (Gruber and 44 Sarmiento, 1997; Capone et al., 2005; Moore et al., 2009). In the North Atlantic, diazotrophy 45 contributed to a large fraction of new production, even exceeding the contributions by nitrate 46 diffusion across the pycnocline (Capone et al., 2005; Fernández et al. 2010; Mouriño47 Carballido et al., 2010). Nitrogen fixation is controlled by temperature (Breitbarth et al., 48 2007), CO 2 (Barcelos e Ramos et al., 2007) and the availability of other nutrients, notably 49 phosphorus and iron, the latter provided by atmospheric dust inputs (Moore et al., 2009; 50 Sohm et al., 2011). Nitrogen of diazotrophic origin is made available to the pelagic food web 51 through excretion and mortality of cyanobacteria (Glibert and Bronk, 1994) and further 52 processing by microbes and planktonic metazoa (Montoya et al., 2002). 53 The colonial cyanobacteria of the genus Trichodesmium is the best known diazotroph, with a 54 widespread distribution across tropical and subtropical regions of the ocean (Capone et al., 55 1997; Luo et al., 2012) where surface water temperature exceeds 20°C (Breitbarth et al., 56 2007). In the North Atlantic Trichodesmium is more abundant between 20°N and 20°S 57 (Tyrrell et al., 2003; Davis and McGillicudy, 2006; Fernández et al., 2010, 2012) but most 58 studies on N 2 fixation have been focused in the subtropical and tropical regions where blooms 59 are frequent (Voss et al., 2004; Capone et al., 2005; Mulholland et al., 2006; Montoya et al., 60 2007). Only a few studies have measured concurrently Trichodesmium abundance and 61 4 nitrogen fixation over large spatial scales in the Atlantic, as reviewed by Luo et al. (Luo et al., 62 2012). However, further evidence of the impact of diazotrophy at regional scales was 63 provided by measurements of the natural abundance of stable nitrogen isotopes in seston and 64 plankton (Waser et al., 2000; Mino et al., 2002; Montoya et al., 2002; Reynolds et al., 2007; 65 Landrum et al., 2011). 66 Stable isotopes can trace N 2 inputs because atmospheric nitrogen is relatively depleted in 67 heavy ( 15 N) isotopes compared to marine nitrate (Owens, 1987). Assimilation of this light N 2 68 by diazotrophs produces organic matter with a characteristic isotopic signature that can be 69 traced along the food web. Because of the different turnover time of planktonic organisms 70 (hours to days in bacteria and phytoplankton, and up to several months in large zooplankton) 71 the isotopic signature of organic matter in various compartments provides an integrative, in 72 situ tracer of the movement and transformation of nitrogen in the water column beyond the 73 instantaneous effects reported during N 2 -uptake measurements. Nitrogen isotopes in seston 74 reflect the uptake of atmospheric N 2 by cyanobacteria (e.g. Montoya et al., 2002) while those 75 in zooplankton show the assimilation of organic matter initially produced by diazotrophs 76 (McClelland et al., 2003). This feature allows an estimation of the contribution of diazotrophy 77 to net nitrogen assimilation in different components of the food web (Mino et al., 2002; 78 Montoya et al., 2002; Reynolds et al., 2007; Landrum et al., 2011). Previous estimates using 79 measurements of natural abundance of nitrogen isotopes in seston and zooplankton revealed a 80 large contribution of diazotrophic nitrogen (up to 100%) in the north-eastern tropical and 81 subtropical Atlantic (Montoya et al., 2002). Landrum et al. (Landrum et al., 2011) reported 82 lower contributions in the central and eastern subtropical Atlantic compared to those in the 83 eastern region, however this study was made in waters near 30ºN, where Trichodesmium 84 abundances were lower than in southern waters (Tyrrell et al., 2003; Davis and McGillicuddy, 85 2006; Fernández et al., 2010). Direct measurements revealed significant N 2 fixation also in 86 the eastern subtropical Atlantic (Fernández et al., 2010; Wannicke et al., 2010; Benavides et 87 al., 2011; Fernández et al., 2012), although there are few measurements of either abundance 88 or N 2 fixation in the central region of the gyre (Luo et al., 2012). 89 The objective of this study is to characterize spatial patterns of plankton in the oligotrophic 90 subtropical North Atlantic by means of the analysis of size-fractionated plankton biomass and 91 natural abundance of stable carbon and nitrogen isotopes. The patterns are related to the 92 abundance of Trichodesmium and indicate a large influence of diazotrophy across plankton 93 size classes over most of the subtropical northern Atlantic. 94 5 95 MATERIAL AND METHODS 96 Samples and water column measurements were obtained during Leg 8 of Malaspina-2010 97 expedition (http://www.expedicionmalaspina.es) on R/V Sarmiento de Gamboa (January98 March 2011) in a transect mostly along 24 °N between Canary Islands and Florida (Fig. 1). 99 The transect was arbitrarily divided in eastern, central, and western zones to summarize its 100 oceanographic and plankton characteristics. 101 Plankton samples were collected by vertical tows of a microplankton net (40 µm mesh size) 102 and a mesoplankton net (200 µm mesh size) through the upper 200 m of the water column. 103 Sampling was made between 10:00 and 16:00 h GMT. Plankton was separated into five size 104 fractions (40-200, 200-500, 500-1000, 1000-2000 and >2000 µm) by gentle filtration of the 105 samples by a graded series of nylon sieves (2000, 1000, 500, 200 and 40 µm). Large 106 gelatinous organisms were removed before filtration. Aliquots for each size-fraction were 107 collected on pre-weighted glass-fibre filters, dried (60ºC, 48 h) and stored in a dessicator 108 before determination of biomass (dry weight), carbon and nitrogen content and natural 109 abundance of stable carbon and nitrogen isotopes ashore. 110 After determination of dry weight, finely ground aliquots of each size fraction were packed in 111 tin capsules for elemental and stable isotope analysis by conversion into CO 2 and N 2 in an 112 elemental analyser (Carlo Erba CHNSO 1108) coupled to an isotope-ratio mass-spectrometer 113 (Finnigan Mat Delta Plus). Samples were not acidified to remove carbonates because other 114 studies showed that the acidification may not cause substantial modification in carbon isotope 115 results, but it may affect nitrogen determinations (Bunn et al., 1995; Bode at al., 2003). 116 Similarly no corrections were made for lipid content potentially affecting carbon isotope 117 composition (Symantec et al., 2007). In this case, the average (±se) C:N molar ratio of all 118 samples was 4.8±0.0 (n=218) and showed little variations among size fractions, suggesting 119 low influence of lipids. Carbon and nitrogen stable isotope abundance was expressed as δ 13 C 120 and δ 15 N relative to VPDB (Vienna PeeDee Belemnite carbonate) and atmospheric N 2 isotope 121 standards. Precision (± standard error) of replicate determinations of both C and N stable 122 isotopes was <0.03‰. 123 Water properties were estimated from CTD casts (SBE-911 Plus) in the upper 300 m. In 124 absence of more detailed observations, sea surface temperature (SST 0-10 m) was used as a 125 6 surrogate of nutrient supply to the surface by advection from deeper layers, and in vivo 126 fluorescence (SFluor) as an estimate of phytoplankton biomass. Total nitrate (NO 3- + NO 2- ) 127 and phosphate were analysed colorimetrically (Grashoff et al., 1983) on frozen samples 128 collected by Niskin bottles at standard depths. 129 Abundance of the diazotroph Trichodesmium sp. was estimated by counts of 50 ml aliquots of 130 the sample from the microplankton net preserved in glutaraldehyde (25% final concentration) 131 using a FlowCAM® system (Fluid Imaging Technologies). Prior to analysis the samples were 132 screened by a 100 µm nylon mesh to prevent clogging of the FlowCAM cell. Results are 133 reported as number of colonies (trichomes) per volume of seawater. Abundance of total 134 microzooplankton and phytoplankton (100 to 200 µm) was also determined in the same 135 samples. Total abundance of mesozooplankton was determined by counts of aliquots of the 136 200 µm net preserved in 4% formalin and observed under a binocular microscope. The 137 relative frequency of the main taxa was also recorded. 138 The contribution of nitrogen fixed by diazotrophs (diazotroph N) to plankton fractions was 139 estimated using the isotope mass balance approach of Montoya et al. (2002): 140 141 where δ 15 N m is the measured isotopic composition in the sample, δ 15 N ref is the isotopic 142 reference value for plankton not influenced by diazotroph N and δ 15 N d is the isotopic 143 composition for diazotrophs (-2‰, Montoya et al., 2002). Reference values δ 15 N ref were 3.7, 144 4.3, 5.1 and 5.8 ‰ for 200-500, 500-1000, 1000-2000 and >2000 µm size-classes, 145 respectively, corresponding to plankton in tropical equatorial regions (Landrum et al., 2011). 146 No estimations of diazotroph N contribution were made for the 40-200 µm class because of 147 potential bias caused by the presence of Trichodesmium filaments. 148 RESULTS 149 Temperature, salinity, fluorescence and nutrients 150 A large range in temperature (10 to 25 °C) was found in the upper 300 m along the transect 151 (Fig. 2). Isotherms raised in the eastern end tracing the influence of the Canary upwelling, and 152 also at other points along the transect indicating mesoscale features favouring upwelling (e.g. 153 near 50 and 70 °W). The highest surface temperature values were found in the western and 154 %diazotroph= 100 δ 15 N m - δ 15 N ref δ 15 N d - δ 15 N ref 7 central regions, and the lowest in the eastern region. Salinity showed a pattern similar to the 155 described for temperature, but in this case there was a core of high salinity (>37.4) between 156 25 and 48 °W in the upper 150 m. Strong salinity gradients characterised the eastern region 157 while the western region had in general low salinity values. 158 Low values of in vivo fluorescence prevailed along the transect and showed a characteristic 159 subsurface maximum in nearly all stations. This maximum was less developed in the eastern 160 region where fluorescence was more uniformly distributed in the upper 100 m but was sharper 161 and deeper in the central and western regions where it reached ca. 150 m deep. 162 Nitrate was almost depleted (<0.05 µM) in the upper 200 m for most of the transect but in the 163 central zone a layer of relatively high concentration was found between 75 and 100 m depth 164 (Fig. 2). Phosphate also showed low concentrations in most of the transect but in this case the 165 whole water column had higher concentrations (>0.05 µM) in the eastern than in the other 166 zones. Both nutrients showed higher concentrations in deep waters at the borders of the 167 transect. 168 Spatial patterns of plankton and stable isotopes 169 Plankton biomass decreased towards the central zone of the transect and had high values at 170 both western and eastern zones (Fig. 3). This pattern was similar in all size classes although 171 mean values were significantly higher in the western zone for plankton <500 µm and lower in 172 the central zone for plankton >2000 µm (Table 1). There were significant differences in mean 173 values of total plankton biomass, with the lowest value central zone, and the highest in the 174 western zone. 175 Microzooplankton abundance was similar in all zones while phytoplankton was significantly 176 more abundant in the western zone (Table 2). Mean abundance of mesozooplankton followed 177 a similar pattern to total plankton biomass, with equivalent values in the eastern and western 178 zones and minimum values in the central zone (Table 2). Copepods were generally dominant 179 in all zones, but some genera were more frequent in the lateral zones (Oithona) than in the 180 central region (Calanus, Macrosetella). Ostracoda were also more frequent in lateral zones 181 while salps and appendicularia showed higher frequencies in central and eastern zones. 182 The spatial variability in nitrogen isotopes was similar to the pattern described for biomass 183 (Fig. 4), as significant, positive correlations (P<0.05) were found between δ 15 N and biomass 184 in all plankton size-fractions. In this case all fractions showed mean δ 15 N values in the central 185 8 zone (<2‰) significantly lower than values in either eastern or western zones (Table 1). 186 Isotopic enrichment was only noticeable between the smallest and largest size classes, while 187 plankton between 200 and 2000 µm showed similar mean δ 15 N values within zones. 188 In contrast, δ 13 C displayed an opposite pattern to the one described for δ 15 N and biomass, but 189 with more differences between size-fractions (Fig. 5). Mean values for 40-200, 200-500 and 190 1000-2000 µm classes were significantly lower in the eastern zone, while no significant 191 differences between zones were found for other classes (Table 1). Biomass was only 192 significantly correlated with δ 13 C for 200-500 and 500-1000 µm classes. 193 Relationships with surface temperature, salinity and in vivo fluorescence 194 Biomass was negatively correlated with surface temperature only for the largest size-class, 195 and also negatively with surface salinity for <500 µm classes, while non significant 196 correlations resulted between biomass and surface fluorescence (Fig. 6). However, 197 fluorescence was positively correlated with δ 15 N for all classes and with δ 13 C for <1000 µm 198 classes. Surface temperature was also correlated with δ 15 N (negatively) or δ 13 C (positively) 199 for <1000 µm (and in case of δ 13 C also for >1000 µm) classes. 200 Linearity between δ δδ δ 15 N and δ 13 C 201 Carbon and nitrogen isotope abundances showed a significant negative linear relationship 202 within size-classes, except for the >2000 µm class (Fig. 7). The slopes and intercepts of the 203 lines were equivalent for classes <1000 µm and also for classes >1000 µm, while within these 204 groups there were no significant differences (ANCOVA, P<0.05). 205 Trichodesmium abundance and δ δδ δ 15 N 206 With the exception of the two easternmost stations, Trichodesmium was recorded at all 207 stations of the transect (Fig. 8). Its abundance showed an abrupt increase at ca. 25ºW followed 208 by a general decrease to the west. Mean values were significantly higher in the eastern and 209 central zones (mean±se = 4.77±0.73 trichomes L -1 n=29) than in the western zone (2.04±0.57 210 trichomes L -1 n=14, ANOVA, P<0.05). 211 A negative linear relationship was found between δ 15 N and log-transformed Trichodesmium 212 abundance for all size-classes (Fig. 9). The slope of the line was similar for all classes (mean 213 slope = -1.42±0.11, n=84) while there were differences in the intercept between the 40-200 214 9 µm and the other classes and between >2000 µm and classes 200-1000 µm (ANCOVA, 215 P<0.05). 216 Contribution of N from diazotrophs 217 Diazotroph N contributed to all size fractions in almost all stations (Fig. 10). Only few 218 stations in the eastern zone without Trichodesmium showed zero contribution while maximum 219 values (>70%) occurred in general in the central zone. Mean contributions were ca. 50% in 220 the central zone but between 22 and 38% in the eastern and western zones (Table 3). The 221 contributions of diazotroph N increased for larger classes. On average there was an increase in 222 the contribution of diazotroph N between the 200-500 and the >2000 µm classes of 16, 10 and 223 4% for the eastern, central and western zones. 224 225 DISCUSSION 226 Plankton biomass across the subtropical Atlantic 227 The measured plankton biomass reflected well the oligotrophy of most of the subtropical 228 North Atlantic. To our knowledge these results are the first obtained in this region of the deep 229 ocean at such small spatial resolution showing a gradual decrease from zones near the 230 continental shelves to the central basin. However, the studied transect included also 231 productive areas. Biomass of nearly all size classes was higher in the western than in the 232 eastern and central zones, with mean values equivalent to those previously reported for both 233 western (Madin et al., 2001) and eastern zones (Hernández-León et al., 2007). The high 234 productivity of lateral zones can be attributed to the influence of the Canary upwelling in the 235 east (indicated by the upward trend in isotherms and the shallow chlorophyll maxima in Fig. 236 2) and to mesoscale eddy pumping (McGillicudy et al., 2001) in the west (as indicated by 237 sharp changes in isotherms near 70°W). Zooplankton biomass has been shown to track 238 upwelling dynamics and extend the influence of the high productive waters near north239 western Africa well into the deep ocean (Hernández-León et al., 2007). Also, zooplankton 240 biomass peaks follow winter-spring blooms in the western zone (Madin et al., 2001). The 241 high plankton productivity is directly linked to large advective fluxes of nitrate from deep 242 waters in both eastern (Pelegrí et al., 2005) and western zones (Lipschulz, 2001) while low 243 production in the central zone of the basin is attributed to low nutrient advection (Marañón et 244 al., 2000). 245 16 Emerson, S., Quay, P., Karl, D., Winn, C., Tupas, L. and Landry, M. (1997) Experimental 435 determination of the organic carbon flux from open-ocean surface waters. Nature, 389, 436 951-954. 437 Fernández, A., Mouriño-Carballido, B., Bode, A., Varela, M. and Marañón, E. (2010) 438 Latitudinal distribution of Trichodesmium spp. and N 2 fixation in the Atlantic Ocean. 439 Biogeosci., 7, 3167-3176. 440 Fernández, A., Graña, R., Mouriño-Carballido, B., Bode, A., Varela, M., Domínguez, J. F., 441 Escánez, J., De Armas, D. and Marañón, E. (2012) Community N 2 fixation and 442 Trichodesmium spp. abundance along longitudinal gradients in the eastern subtropical 443 North Atlantic. ICES J. Mar. Sci., in press. 444 Fry, B. and Wainright, S. C. (1991) Diatom sources of 13 C-rich carbon in marine food webs. 445 Mar. Ecol. Prog. Ser., 76, 149-157. 446 Glibert, P. M. and Bronk, D. A. (1994) Release of dissolved organic nitrogen by marine 447 diazotrophic cyanobacteria, Trichodesmium spp. Appl. Environ. Microbiol., 60, 3996448 4000. 449 Goericke, R. (1998) Response of phytoplankton community structure and taxon-specific 450 growth rates to seasonally varying physical forcing in the Sargasso Sea off Bermuda. 451 Limnol. Oceanogr., 43, 921-935. 452 Grasshoff, K., Ehrhardt, M. and Kremling, K. (1983) Methods of seawater analysis. Verlag 453 Chemie, Weinheim. 454 Gruber, N. and Sarmiento, J. (1997) Global patterns of marine nitrogen fixation and 455 denitrification. Global Biogeochem. Cycles, 11, 235-266. 456 Gruber, N., Keeling, C. D., Bacastow, R. B., Guenther, P. R., Lueker, T. J., Wahlen, M., 457 Meijer, H. A. J., Mook, W. G. and Stocker, T. F. (1999) Spatiotemporal patterns of carbon458 13 in the global surface oceans and the oceanic Suess effect. Global Biogeochem. Cycles, 459 13, 307-335. 460 Hawser, S.P., O’Neil, J.M., Roman, M.R. and Codd, G.A. (1992) Toxicity of blooms of the 461 cyanobacterium Trichodesmium to zooplankton. J. Appl. Phycol.. 4, 79-86. 462 17 Hernández-León, S., Gómez, M. and Arístegui, J. (2007) Mesozooplankton in the Canary 463 Current system: The coastal–ocean transition zone. Prog. Oceanogr., 74, 397-421. 464 Karl, D. M., Bidigare, R. R., Church, M. J., Dore, J. E., Letelier, R. M., Mahaffey, C. and 465 Zehr, J. P. (2008) The nitrogen cycle in the North Pacific Trades biome: An evolving 466 paradigm. In Capone, D. G., Bronk, D. A., Mulholland. M. R. and Carpenter, E. J. (eds) 467 Nitrogen in the marine environment. Elsevier, pp. 705-769. 468 Knapp, A. N., Sigman, D. M., Lipschultz, F., Kustka, A. B. and Capone, D. G. (2011) 469 Interbasin isotopic correspondence between upper-ocean bulk don and subsurface nitrate 470 and its implications for marine nitrogen cycling. Global Biogeochem. Cycles, 25, 471 doi:10.1029/2010GB003878. 472 Landrum, J. P., Altabet, M. A. and Montoya, J. P. (2011) Basin-scale distributions of stable 473 nitrogen isotopes in the subtropical North Atlantic Ocean: Contribution of diazotroph 474 nitrogen to particulate organic matter and mesozooplankton. Deep Sea Res., 58, 615-625. 475 Lipschultz, F. (2001) A time-series assessment of the nitrogen cycle at BATS. Deep Sea Res. 476 II, 48, 1897-1924. 477 Luo, Y.-W., Doney, S. C., Anderson, L. A., Benavides, M., Bode, A., Bonnet, S., Boström, K. 478 H., Böttjer, D., Capone, D. G., Carpenter, E. J., Chen, Y. L., Church, M. J., Dore, J. E., 479 Falcón, L. I., Fernández, A., Foster, R. A., Furuya, K., Gómez, F., Gundersen, K., Hynes, 480 A. M., Karl, D. M., Kitajima, S., Langlois, R. J., Laroche, J., Letelier, R. M., Marañón, E., 481 McGillicuddy Jr., D. J., Moisander, P. H., Moore, C. M., Mouriño-Carballido, B., 482 Mulholland, M. R., Needoba, J. A., Orcutt, K. M., Poulton, A. J., Raimbault, P., Rees, A. 483 P., Riemann, L., Shiozaki, T., Subramaniam, A., Tyrrell, T., Turk-Kubo, K. A., Varela, M., 484 Villareal, T. A., Webb, E. A., White, A. E., Wu, J. and Zehr, J. P. (2012) Database of 485 diazotrophs in global ocean: Abundances, biomass and nitrogen fixation rates. Earth Syst. 486 Sci. Data Discuss., 5, 47-106. 487 Madin, L. P., Horgan, E. F. and Steinberg, D. K. (2001) Zooplankton at the Bermuda Atlantic 488 Time-Series study (BATS) station: Diel, seasonal and interannual variation in biomass, 489 1994-1998. Deep Sea Res. II, 48, 2063-2082. 490 18 Marañón, E., Holligan, P. M., Varela, M., Mouriño, B. and Bale, A. J. (2000) Basin-scale 491 variability of phytoplankton biomass, production and growth in the Atlantic Ocean. Deep 492 Sea Res., 47, 825-857. 493 Margalef, R. (1978) Phytoplankton communities in upwelling areas. The example of NW 494 Africa. Oecologia Aquatica, 3, 97-132. 495 McClelland, J. W., Holl, C. M. and Montoya, J.-P. (2003) Relating low δ 15 N values of 496 zooplankton to N 2 -fixation in the Tropical North Atlantic: Insights provided by stable 497 isotope ratios of amino acids. Deep Sea Res., 50, 849-861. 498 McGillicuddy Jr., D. J., Kosnyrev, V. K., Ryan, J. P. and Yoder, J. A. (2001) Covariation of 499 mesoscale ocean color and sea-surface temperature patterns in the Sargasso Sea. Deep Sea 500 Res. II, 48, 1823-1836. 501 Mino, Y., Saino, T., Suzuki, K. and Marañón, E. (2002) Isotopic composition of suspended 502 particulate nitrogen (δ 15 N sus ) in surface waters of the Atlantic Ocean from 50ºN to 50ºS. 503 Global Biogeochemical Cycles, 16, doi:10.1029/2001GB001635. 504 Moisander, P. H., Beinart, R. A., Hewson, I., White, A. E., Johnson, K. S., Carlson, C. A., 505 Montoya, J. P. and Zehr, J. P. (2010) Unicellular cyanobacterial distributions broaden the 506 oceanic N 2 fixation domain. Science, 327, 1512-1514. 507 Montoya, J. P., Carpenter, E. J. and Capone, D. G. (2002) Nitrogen fixation and nitrogen 508 isotope abundances in zooplankton of the oligotrophic North Atlantic. Limnol. Oceanogr., 509 47, 1617-1628. 510 Montoya, J. P., Voss, M. and Capone, D. G. (2007) Spatial variation in N 2 -fixation rate and 511 diazotroph activity in the Tropical Atlantic. Biogeosci., 4, 369-376. 512 Moore, K., Doney, S. C., Lindsay, K., Mahowald, N. and Michaels, A. F. (2006) Nitrogen 513 fixation amplifies the ocean biogeochemical response to decadal timescale variations in 514 mineral dust deposition. Tellus, 58B, 560-572. 515 Moore, C. M., Mills, M. M., Achterberg, E. P., Geider, R. J., Laroche, J., Lucas, M. I., 516 Mcdonagh, E. L., Pan, X., Poulton, A. J., Rijkenberg, M. J. A., Suggett, D. J., Ussher, S. J. 517 and Woodward, E. M. S. (2009) Large-scale distribution of Atlantic nitrogen fixation 518 controlled by iron availability. Nature Geoscience, 2, 867-871. 519 19 Mouriño-Carballido, B., Graña, R., Fernández, A., Bode, A., Varela, M., Domínguez, J. F., 520 Escánez, J., De Armas, D. and Marañón, E. (2011) Importance of N 2 fixation vs. Nitrate 521 eddy diffusion along a latitudinal transect in the atlantic ocean. Limnol. Oceanogr., 56, 522 999-1007. 523 Mulholland, M. R. (2007) The fate of nitrogen fixed by diazotrophs in the ocean. Biogeosci., 524 4, 37-51. 525 Mulholland, M. R., Bernhardt, P. W., Heil, C. A., Bronk, D. A. and O’neil, J. M. (2006) 526 Nitrogen fixation and release of fixed nitrogen by Trichodesmium spp. in the Gulf of 527 Mexico. Limnol. Oceanogr., 51, 1762-1776. 528 O’Neil, J.M., Metzler, P.M. and Glibert, P.M. (1996) Ingestion of 15 N 2 -labelled 529 Trichodesmium spp. and ammonium regeneration by the harpacticoid copepod 530 Macrosetella gracilis. Mar. Biol. , 125, 89-96. 531 Oschlies, A. and Garçon, V. (1998) Eddy-induced enhancement of primary production in a 532 model of the North Atlantic Ocean. Nature, 394, 266-269. 533 Owens, N. J. P. (1987) Natural variations in 15 N in the marine environment. Adv. Mar. Biol., 534 24, 389-451. 535 Pelegrí, J. L., Arístegui, J., Cana, L., González-Dávila, M., Hernández-Guerra, A., 536 Hernández-León, S., Marrero-Díaz, A., Montero, M. F., Sangrá, P. and Santana-Casiano, 537 M. (2005) Coupling between the open ocean and the coastal upwelling region off 538 Northwest Africa: Water recirculation and offshore pumping of organic matter. J. Mar. 539 Systems, 54, 3-37. 540 Reynolds, S. E., Mather, R. L., Wolff, G. A., Williams, R. G., Landolfi, A., Sanders, R. and 541 Woodward, E. M. S. (2007) How widespread and important is N 2 fixation in the North 542 Atlantic Ocean? Global Biogeochem Cycles, 21, doi:10.1029/2006GB002886. 543 Sañudo-Wilhelmy, S. A., Kustka, A. B., Gobler, C. J., Hutchins, D. A., Yang, M., Lwiza, K., 544 Burns, J., Capone, D. G., Raven, J. A. and Carpenter, E. J. (2001) Phosphorus limitation of 545 nitrogen fixation by Trichodesmium in the central Atlantic Ocean. Nature, 411, 66-69. 546 Sohm, J. A., Webb, E. A. and Capone, D. G. (2011) Emerging patterns of marine nitrogen 547 fixation. Nat. Rev. Microbiol., 9, 499-508. 548 20 Torres-Valdes, S., Roussenov, V. M., Sanders, R., Reynolds, S., Pan, X., Mather, R., 549 Landolfi, A., Wolff, G. A., Achterberg, E. P. and Williams, R. G. (2009) Distribution of 550 dissolved organic nutrients and their effect on export production over the Atlantic Ocean. 551 Glob. Biogeochem. Cycles, 23, doi:10.1029/2008GB003389 552 Tortell, P. D. and Morel, F. M. M. (2002) Sources of inorganic carbon for phytoplankton in 553 the eastern subtropical and equatorial Pacific Ocean. Limnol. Oceanogr., 47, 1012-1022. 554 Tyrrell, T., Marañón, E., Poulton, A. J., Bowie, A. R., Harbour, D. S. and Woodward, E. M. 555 S. (2003) Large-scale latitudinal distribution of Trichodesmium sp. in the Atlantic Ocean. 556 J. Plankton Res., 25, 405-416. 557 Voss, M., Croot, P., Lochte, K., Mills, M. and Peeken, I. (2004) Patterns of nitrogen fixation 558 along 10º N in the Tropical Atlantic. Geophys. Res. Lett., 31, doi:10.1029/2004GL020127. 559 Wannicke, N., Liskow, I. and Voss, M. (2010) Impact of diazotrophy on N stable isotope 560 signatures of nitrate and particulate organic nitrogen: Case studies in the north-eastern 561 tropical Atlantic Ocean. Isotopes Environ. Health Stud., 46, 337-354. 562 Waser, N. A. D., Harrison, W. G., Head, E. J. H., Nielsen, B., Lutz, V. A. and Calvert, S. E. 563 (2000) Geographic variations in the nitrogen isotope composition of surface particulate 564 nitrogen and new production across the North Atlantic Ocean. Deep-Sea Res., 47, 1207565 1226. 566 Williams, R. G. and Follows, M. J. (1998) The Ekman transfer of nutrients and maintenance 567 of new production over the North Atlantic. Deep-Sea Res., 45, 461-489. 568 569 21 Figure legends 570 Figure 1. CTD and plankton sampling stations during cruise Leg 8 of Malaspina-2010 571 expedition along 24 ⁰N (dashed line). The vertical lines indicate the limits of the eastern (E), 572 central (C) and western (W) regions described in the text. 573 Figure 2. Temperature, salinity and in vivo fluorescence in the upper 300 m along 24 ⁰N. 574 CTD stations are indicated in the upper panel. The dashed lines indicate the limits between 575 eastern (E), central (C) and western (W) zones described in the text. 576 Figure 3. Accumulated biomass (mg dry weight m -3 ) of size fractionated plankton along 24 577 ⁰N. The dashed lines indicate the limits between eastern (E), central (C) and western (W) 578 zones described in the text. 579 Figure 4. Natural abundance of stable nitrogen isotopes (δ 15 N, ‰) of size fractionated 580 plankton along 24 ⁰N. The dashed lines indicate the limits between eastern (E), central (C) 581 and western (W) zones described in the text. 582 Figure 5. Natural abundance of stable carbon isotopes (δ 13 C, ‰) of size fractionated plankton 583 along 24 ⁰N. The dashed lines indicate the limits between eastern (E), central (C) and western 584 (W) zones described in the text. 585 Figure 6. Correlation coefficients (Pearson r) between size fractionated plankton: a) biomass 586 (mg DW m-3), b) δ15N or c) δ13C of and sea surface temperature (SST, ⁰C), salinity (SSS) or 587 in vivo fluorescence (Sfluor) along 24 ⁰N. The dashed lines indicate the significance value 588 (P<0.05). 589 Figure 7. Relationships between δ 15 N and δ 13 C for size fractionated plankton (µm). All 590 regression lines are significant with P<0.01 except for the >2000 µm fraction (dashed line, 591 P<0.05). 592 Figure 8. Abundance of Trichodesmium (trichomes L -1 ) along 24 ⁰N. The dashed lines 593 indicate the limits between eastern (E), central (C) and western (W) zones described in the 594 text. 595 22 Figure 9. Relationships between δ 15 N and Trichodesmium abundance (log 10 (trichomes L -1 )). 596 All regression lines are significant with P<0.001.The dashed line indicates the regression line 597 for the >2000 µm fraction. 598 Figure 10. Diazotrophic N contribution (%) to plankton size-classes >200 µm estimated from 599 δ 15 N along 24⁰ N. 600 601 23 Table 1. Mean (±se) biomass (mg DW m -3 ), δ 15 N and δ 13 C by size-fractions in the western (W), central (C) and eastern (E) zones along 24°N. 602 Total biomass (Total) is the sum of biomass for all size-fractions. n: number of data. Shaded values and letters indicate significant differences 603 between means (ANOVA and C-Dunnett a posteriori test, P<0.05). 604 DW δ 15 N δ 13 C Size fraction (µm) W C E W C E W C E 40-200 4.65±0.24 b 3.38±0.11 a 3.46±0.18 a 1.7±0.2 b 0.6±0.1 a 2.1±0.2 b -19.6±0.2 b -19.7±0.1 b -20.3±0.1 a 200-500 3.00±0.24 b 1.89±0.08 a 2.04±0.12 a 2.0±0.2 b 1.2±0.1 a 2.5±0.3 b -19.7±0.1 b -19.6±0.3 b -20.4±0.2 a 500-1000 2.69±0.24 a 1.95±0.09 a 2.48±0.22 a 2.3±0.2 b 1.2±0.1 a 2.5±0.3 b -20.0±0.2 a -20.1±0.3 a -20.6±0.2 a 1000-2000 2.31±0.16 a 1.74±0.20 a 2.08±0.11 a 2.4±0.2 b 1.5±0.2 a 2.6±0.3 b -19.4±0.4 b -18.7±0.4 b -20.9±0.7 a >2000 2.39±0.17 b 1.65±0.11 a 2.71±0.23 b 3.2±0.3 b 1.6±0.3 a 2.9±0.3 b -19.5±0.5 a -19.3±0.5 a -20.9±0.6 a Total 15.04±0.93 c 10.60±0.38 a 12.78±0.66 b n 14 12 17 14 12 17 14 12 17 24 Table 2. Mean (±se) abundance of microplankton (n L -1 ), and dominant taxa (% frequency) and mean (±sd) total abundance (n m -3 ) of mesozooplankton in the western (W), central (C) and eastern (E) zones along 24°N. Shaded values and letters indicate significant differences between means (ANOVA and C-Dunnett a posteriori test, P<0.05). zone Group Taxa W C E Total abundance (n L - 1 ) Microplankton (40-200 µm) Phytoplankton Mean ± se 8.1±1.0 b 4.5±0.5 a 3.7±0.4 a Zooplankton Mean ± se 11.0±1.0 a 7.3±0.7 a 7.5±1.4 a Frequency (%) Mesozooplankton Calanus 4.7 9.9 10.0 (>200 µm) Corycaeus 8.1 8.3 5.6 Macrosetella 4.0 7.4 4.4 Oithona 8.7 7.4 10.0 Appendicularia 2.0 2.5 6.1 Chaetognatha 6.7 7.4 7.8 Ostracoda 8.7 5.0 7.2 Polychaeta 7.4 3.3 5.6 Salps 2.7 5.8 5.0 Total abundance (n m - 3 ) Mean ± se 186.9±29.1 a 124.3±16.6 a 178.0±32.4 a n 14 12 17 25 Table 3. Mean (±se) contribution of N from diazotrophs (%) to plankton size-fractions in the western (W), central (C) and eastern (E) zones along 24°N. N: number of data. Shaded values and letters indicate significant differences between means (ANOVA and C-Dunnett a posteriori test, P<0.05). Size fraction (µm) W C E 200-500 29.4±3.9 a 43.1±1.4 b 22.5±3.9 a 500-1000 31.9±2.7 a 49.2±1.9 b 29.0±4.4 a 1000-2000 38.3±3.2 a 50.4±3.2 b 35.5±4.1 a >2000 33.2±4.3 a 53.5±4.1 b 38.7±4.7 a n 14 12 17 -26 -22 -18 -14 δ 13 C > 2000 ECW -14 - 22 -18 -14 δ 13 C 500 - 1000 -26 -22 -18 -14 δ 13 C 1000-2000 Figure 5. Natural abundance of stable carbon isotopes (δ13C, ‰) of size fractionated plankton along 24 ⁰N. The dashed lines indicate the limits between eastern (E), central (C) and western (W) zones described in the text. -26 -22 -18 -14 020406080100 δ 13 C 40-200 -26 -22 -18 δ13C 200-500 -26 - 22 500 - 1000 Longitude -1.0 -0.5 0.0 0.5 1.0 40-200 200-500 500-1000 1000-2000 > 2000 correlation coefficient size class (µm) SST SSS SFluor Biomass (DW) -0.5 0.0 0.5 1.0 correlation coefficient δ 15 N Figure 6. Correlation coefficients (Pearson r) between size fractionated plankton biomass (mg DW m-3)δ15N or δ13C of and sea surface temperature (SST, ⁰C), salinity (SSS) or in vivo fluorescence (Sfluor) along 24 ⁰N. The dashed lines indicate the significance value (P<0.05). -1.0 40-200 200-500 500-1000 1000-2000 > 2000 size class (µm) -1.0 -0.5 0.0 0.5 1.0 40-200 200-500 500-1000 1000-2000 > 2000 correlation coefficient size class (µm) δ 13 C -1 0 1 2 3 4 5 6 -26 -22 -18 -14 δ 15 N δ 13 C 40-200 -1 0 1 2 3 4 5 6 -26 -22 -18 -14 δ 15 N δ 13 C 200-500 3 4 5 6 15 N 3 4 5 6 15 N -1 0 1 2 -26 -22 -18 -14 δ 15 δ 13 C 500-1000 -1 0 1 2 -26 -22 -18 -14 δ 15 δ 13 C 1000-2000 >2000 Figure 7. Relationships between δ15N and δ13C for size fractionated plankton (µm). All regression lines are significantwith P<0.01 except for the >2000 µm fraction (dashed line, P<0.05). 10 15 20 Trichodesmium (trichomes L -1 ) WCE 0 5 10 020406080100 Trichodesmium Longitude Figure 8. Abundance of Trichodesmium (trichomes L-1) along 24 ⁰N. The dashed lines indicate the limits between eastern (E), central (C) and western (W) zones described in the text. -1 0 1 2 3 4 5 6 -2 -1 0 1 2 δ 15 N log 10 (Trichodesmium) 40-200 -1 0 1 2 3 4 5 6 -2 -1 0 1 2 δ 15 N log 10 (Trichodesmium) 200-500 5 6 5 6 -1 0 1 2 3 4 5 -2 -1 0 1 2 δ 15 N log 10 (Trichodesmium) 500-1000 -1 0 1 2 3 4 5 -2 -1 0 1 2 δ 15 N log 10 (Trichodesmium) 1000-2000 > 2000 Figure 9. Relationships between δ15N and Trichodesmium abundance (log10(trichomes L-1)). All regression lines are significant with P<0.001.The dashed line indicates the regression line for the >2000 µm fraction. 80 20 40 60 80 % N diazotrophs 1000-2000 0 20 40 60 80 % N diazotrophs >2000 ECW Figure 10. Diazotrophic N contribution (%) to plankton size-classes >200 µm estimated from δ15N along 24⁰ N. 0 20 40 60 80 020406080100 % N diazotrophs Longitude W 200-500 0 20 40 60 % N diazotrophs 500-1000 0