This article appeared in a journal published by Elsevier.

The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elseviers archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Author's personal copy

Progress in Oceanography 83 (2009) 217227

Contents lists available at ScienceDirect

Progress in Oceanography
journal homepage: www.elsevier.com/locate/pocean

Carbon uxes within the epipelagic zone of the Humboldt Current System off Chile: The signicance of euphausiids and diatoms as key functional groups for the biological pump
Humberto E. Gonzlez a,b,c,*, Giovanni Daneri b,c, Jos L. Iriarte c,d, Beatriz Yannicelli b,f, Eduardo Menschel a,b, Claudio Barra a, Silvio Pantoja b,e, Lorena Lizrraga b,c
a

Universidad Austral de Chile, Instituto de Biologa Marina, Casilla 567, Valdivia, Chile Centro de Investigacin Oceanogrca en el Pacco Sur-Oriental (COPAS), Universidad de Concepcin, Chile c Centro de Investigacin de Ecosistemas de la Patagonia (CIEP), Bilbao 466, Coyhaique, Chile d Universidad Austral de Chile, Instituto de Acuicultura, Casilla 1327, P. Montt, Chile e Universidad de Concepcin, Departamento de Oceanografa, Casilla 160-C, Concepcin, Chile f Centro de Estudios Avanzados en Zonas Aridas (CEAZA), Universidad Catlica del Norte, Departamento de Biologa Marina, Larrondo 1281, Casilla 117, Coquimbo, Chile
b

a r t i c l e

i n f o

a b s t r a c t
The information from 54 drifting sediment traps deployed between 1997 and 2006 along the Humboldt Current System off Chile (from 19.9S to 42.2S) was analyzed to contribute to unveiling the recurrent global-ocean issue of the lack of relationship between gross primary production (GPP) and particulate organic carbon (POC) export below 50 m depth. When the proportion of carbon that effectively sinks is relatively low compared to the carbon being xed through GPP, a signicant amount (average of 32%) of the sinking organic matter is composed of diatoms, regardless of GPP rates. Such a fraction seems to be affected by the physiological state of phytoplankton. In contrast, when the fraction of carbon sinking is high relative to GPP, most of sinking organic matter is composed of euphausid faecal strings. Such a situation occurs at relatively low values of GPP and chlorophyll-a. Most of these high sinking rates of pellets and low phytoplankton biomass occur during summer, when physical conditions favour the presence of phytoplankton blooms, and when the GPP/Biomass ratio indicates healthy phytoplankton physiological conditions. All this evidence supports the assessment of the relevance of euphausiids as key species in the Humboldt Current System pointing to (i) the topdown control that euphausiids are capable of exerting over primary producer biomass, and (ii) euphausiids paramount role on total organic carbon ux over the Concepcin continental shelf, regarding both POC export to the sediments and possibly the channelling of GPP directly to higher trophic levels. 2009 Elsevier Ltd. All rights reserved.

Article history: Received 18 July 2008 Received in revised form 25 February 2009 Accepted 16 July 2009 Available online 30 July 2009

1. Introduction The Humboldt Current System (HCS) off Chile is one of the most productive areas ever recorded (Daneri et al., 2000). The underlying factors encompassing such high primary production (PP) are diverse, but the continuous (northern area: 2030S) and seasonal (central-southern area: 3040S) fertilization of surface waters due to upwelling events seems to be partially the cause. The importance of the continental margins in the carbon export (sensu biological pump concept) is widely recognized (Walsh, 1991; Liu et al., 2000). However, in the vast area covered by the HCS, little

* Corresponding author. Address: Universidad Austral de Chile, Instituto de Biologa Marina, Casilla 567, Valdivia, Chile. Tel.: +56 63 221559; fax: +56 63 221455. E-mail address: hgonzale@uach.cl (H.E. Gonzlez). 0079-6611/$ - see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.pocean.2009.07.036

information on the functioning of the biological pump is available and it is restricted to specic upwelling centers in the central (Grnewald et al., 2002; Gonzlez et al., 2007) and northern (Gonzlez et al., 2004; Pantoja et al., 2004) HCS. Traditionally, a positive relationship between primary production (PP) and particulate organic carbon (POC) ux has been assumed in both biogeochemical cycling and global climate change models (Berelson, 2001). However, the usefulness of such models is strongly limited by the varying natural complexity of the processes involved (reviewed by Boyd and Trull, 2007; Buesseler et al., 2007; De La Rocha and Passow, 2007). High PP in the upper productive layer would enhance the carbon export due to diatom aggregation and sinking after the bloom and/or faecal pellet production and export due to zooplankton grazing. The organic carbon export mediated by these processes might be enhanced when refractory and dense mineral ballast provided by mineral skeletons

Author's personal copy

218

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227

of calcium carbonate or silicate is the main constituent of faecal material and marine snow (Klaas and Archer, 2002; Amstrong et al., 2002). However, although the ux decrease appears to be governed by a power function, the controls on its regional variability, and whether these controls derive primarily from particle properties that are imprinted in the surface ocean remain unclear (Boyd and Trull, 2007). Thus, it seems critical not only to focus on bulk parameters of the export-ux, but also to allocate more efforts analyzing the key components of sinking detritus such as faecal pellets and phytodetritus. Results using sediment traps and large-volume water samplers have concluded that faecal material makes up the bulk of the POC ux in different areas of the world oceans (Bishop et al., 1977; Pilskaln and Honjo, 1987; Wassmann et al., 2000; Turner, 2002) and in the HCS off Chile (Gonzlez et al., 2000; Gonzlez et al., 2007). In addition, episodic pulses of phytodetritus, especially in upwelling systems after the nutrient exhaustion, might result in the delivery of enormous amounts of POC towards the sediments (Billett et al., 1983; Smetacek, 1985; Turner, 2002). The HCS has been divided into coastal (rst 50 km from shore), transition (50200 km) and oceanic (>200 km) zones, where the annual cycle of surface chlorophyll-a showed maximum in summer, undened spring and autumn and minimum in winter months (Yuras et al., 2005). The data here reported have been collected mainly within the rst 50 km from shore and within the 200 m isobath. Thus, although most data on GPP can be assigned to highly productive (springsummer) or poorly productive (winter-autumn) seasons, strong within-season variability on GPP and POC export remains, probably because upwelling forcing in the HCS also varies at a synoptic scale (315 days) (Rutllant and Montecino, 2002; Sobarzo et al., 2007). We aimed to study the relationship between gross primary production (GPP) and POC ux along the HCS off Chile, considering two spatialtemporal scales (Fig. 1): (1) The central-northern HCS area stretching from 19 to 42S, covering one decade (19972006) of measurements in different sampling sites, collected during different periods of the year (largely during spring-summer) and (2) the time-series station of the upwelling area off Concepcin (36S), covering three years of monthly data at the same station located at 36S in the HCS. In addition, the main type of particle (phytodetritus or faecal material) and the key group involved in phytoplankton aggregation (diatoms) and faecal pellet production (euphausiids, appendicularians) were studied.

Fig. 1. Map of the Humboldt Current System (HCS) off Chile showing the position of the 54 sites where drifting sediment traps were deployed and gross primary production was measured between 1997 and 2006 (different dots and names of the expeditions included in Table 1). The thin line close to the coast shows the position of the 200 m isobath. The time-series station is called Station 18 (St. 18) and is the one located off Concepcin.

(36460 S; 73030 W), four times a day (at 2 and 8 am and 2 and 8 pm). 2.1. Sample collection and laboratory analysis

2. Materials and methods A total of 54 deployments of surface tethered sediment traps were conducted from the northern (19.9S) to the southern (42.2S) HCS off Chile between 1997 and 2006 (Table 1, Fig. 1). Each deployment consisted of a drifting array of paired-sediment traps, deployed at 50 or 100 m depth for 1024 h, together with an array of white and dark bottles to estimate GPP (see below). We characterized the quantity and quality of POC exported from the euphotic layer along the HCS, and its relationship with GPP, chlorophyll-a concentration and diatom biomass. For such analyses, we considered data from 37 deployments conducted during 13 oceanographic cruises, plus data from a quasi-monthly visited time-series station off Concepcin (3630.80S; 7307.75W, Station 18 FONDAP COPAS Station www.copas.udec.cl/eng/research/timeseries.php), 17 times between September 2003 and May 2005 (Table 1, Fig. 1). Automatic wind data (wind speed and direction) were also recorded close to st. 18, at Carriel Sur airport

The vertical ux of POC was estimated by using surface-tethered, cylindrical, paired-sediment traps (122 cm2 area and 8.3 aspect ratio) or four cylindrical tubes (50 cm2 area and 7.3 aspect ratio), mounted in a cross frame to ensure that the tubes remain vertical. No poison was used because of the short deployment time. All estimates of particle uxes are subject to biases, resulting in possible under- or over-trapping (Gardner, 2000). We did not estimate trapping efciency, but specications of our sediment traps are similar to those calibrated with 234Th, in the Barents Sea (Coppola et al., 2002). We acknowledge that there is no clear way to correct for these biases (Boyd and Trull, 2007), in particular, when a high variability in the contribution of different size particles such as aggregates of phytodetritus and faecal pellets affect POC and 234 Th uxes differently (Burd et al., 2007). POC content was analyzed in an Europa Hydra 2020 continuous ow isotope ration mass spectrometer following combustion at 1000 C at the UC Davis Stable Isotope Facility Laboratory, using acetanilide as a standard (Bodungen et al., 1991).

Author's personal copy

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227

219

Table 1 Cruise, date, place, season and position of the stations where GPP experiment and drifting sediment trap deployment were conducted along the Humboldt Current System off Chile, during 19972006. Latitude and longitude are given in decimals of degree. No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 Cruise/program Dinamo Dinamo Pro-depoy (St. 2) Pro-depoy (St. 1) CHUPS (St.70) CHUPS (St.4) CHUPS (St.mooring) RECLAN oceanic RECLAN front RECLAN coast Geminis Geminis Geminis Geminis Geminis FLUMO oceanic FLUMO coast Sectorial Sectorial Sectorial Sectorial Sectorial Sectorial Sectorial Sectorial MiNOX MIRC oceanic MIRC oceanic MIRC Shelf MIRCShelf Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station Concepcin time serie station EXPLORA IAAW (St.40) EXPLORA IAAW (St.80) EXPLORA IAAW (St.mooring) Comau fjord Comau fjord Comau fjord Comau fjord Date (m/d/y) 03/01/04 03/01/04 07/01/04 07/01/04 03/01/03 03/01/03 03/01/03 01/27/98 01/26/98 01/01/98 10/01/01 10/01/01 08/01/01 08/01/01 02/01/01 04/01/01 04/01/01 01/01/97 07/01/97 07/01/97 07/01/97 01/01/97 01/01/97 07/01/97 01/01/97 03/01/00 10/01/98 07/01/99 07/01/99 10/01/98 07/06/04 04/07/04 09/24/03 09/08/04 03/22/05 08/18/04 05/26/04 12/30/03 01/11/04 02/26/05 12/01/04 04/12/05 04/29/04 10/05/04 01/27/04 11/03/04 05/31/05 11/01/03 11/01/03 11/01/03 11/04/05 11/04/05 01/26/06 01/26/06 Season Summer Summer Winter Winter Summer Summer Summer Summer Summer Summer Spring Spring Winter Winter Summer Autumn Autumn Summer Winter Winter Winter Summer Summer Winter Summer Summer Spring Winter Winter Spring Winter Autumn Spring Spring Autumn Winter Autumn Summer Summer Summer Summer Autumn Autumn Spring Summer Spring Autumn Spring Spring Spring Spring Spring Summer Summer Place Iquique Iquique Iquique Iquique Iquique Iquique Iquique Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Antofagasta Iquique Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcion Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Concepcin Chilo Chilo Chilo Chilo Latitude S 19.90 19.90 20.25 20.25 20.25 20.29 20.30 22.72 22.78 22.79 22.99 22.99 22.99 22.99 22.99 23.20 23.26 23.35 23.35 23.35 23.35 23.35 23.35 23.35 23.35 19.40 36.30 36.30 36.60 36.60 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.51 36.36 37.05 37.12 42.22 42.22 42.22 42.22 Longitude W 70.18 70.18 71.02 71.02 71.46 71.24 72.30 70.61 70.54 70.41 70.51 70.51 70.51 70.51 70.51 72.80 70.67 70.74 70.74 70.74 70.74 70.74 70.74 70.74 70.74 70.67 76.90 76.90 73.20 73.20 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 76.13 73.73 74.88 75.45 72.24 72.24 72.24 72.24 Depth (m) of deployment 100 100 50 50 100 70 70 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 50 100 100 100 100 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 100 100 100 50 50 50 50

Aliquots of the collected material were used for diatoms, microzooplankton and faecal material counting and sizing with standard microscopy methods (Utermhl, 1958). For Chl-a, 250 mL samples of water were ltered (GF/F glass ber lters), in triplicate, and immediately frozen (20 C) until later analysis by uorometry (Turner Design AU-10) according to standard procedures (Parsons et al., 1984). Plankton carbon content was estimated after measuring the cell volume (CV in lm3) of the identied taxa, using a factor of pgC = CV0.811 0.288 and pgC = CV0.939 0.216 for diatom and microzooplankton, respectively (Menden-Deuer and Lessard, 2000). Faeces collected were identied based on published descriptions (e.g., Gonzlez, 1994) and classied as copepod, euphausiid, appendicularian, or undetermined faecal matter. Cylindrical faeces

were assumed to be from copepods and euphausiids, and ellipsoid pellets to be from appendicularians. The total volume of faecal pellets collected was calculated after counting (50300 pellets per sample) and sizing (length, width) all intact and broken faecal pellets. Faecal pellet carbon from sediment-trap samples was estimated after calculating the total volume and using an average volume-to-carbon ratio of 0.076 mg C mm3 (Gonzlez et al., 2000). Gross primary production (GPP) incubations were performed with water samples obtained from within the euphotic zone; four water depths (i.e. surface, 10, 20, and 35 m) were usually selected to collect and incubate water samples. The euphotic zone was determined as the depth where the scalar photosynthetically active radiation (PAR) fell to 1% of the surface values. GPP measurements were estimated from the changes in dissolved oxygen

Author's personal copy

220

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227

concentrations observed after incubating in situ light and dark bottles (Strickland, 1960). Water samples were collected at dawn using Niskin bottles. Samples were transferred to 125 mL borosilicate bottles (gravimetrically calibrated) using a silicone tube to ll ve initial bottles, ve light bottles, and ve dark bottles for each incubation depth. The light and dark bottles were incubated in a surface tethered mooring system for 12 h, the initial bottles were xed at the beginning of each experiment. Dissolved oxygen concentration was measured using a semi-automatic version of the Winkler method (Williams and Jenkinson, 1982; Knap et al., 1993) based on an end-point photometric detector, a Dosimat 665 (Metrohom), and a chart recorder.

3. Results GPP and POC ux values along the HCS ranged between 270 21180 and 321045 mg C m2 d1, respectively. The export ratio (e-ratio) uctuated between 1% and 63% (Fig. 2), with an average of $12% (32 out of 57 pairs of data were 610%), which do not depart signicantly from e-ratios reported for more oligotrophic areas. However, the HCS shows a POC ux one order of magnitude higher than those oligotrophic areas (see Table 3 of Kawakami and Honda, 2007). When all GPP values along the HCS are plotted against the POC ux (as % GPP), a negative exponential curve appeared and three zones could be dened (a) Zone 1, where a relatively narrow range of low production estimates resulted in a wide range of export productions as % PP (1063%), (b) Zone 2, where a relatively narrow range of low production estimates resulted in a narrow range of export productions as % PP (110%), and (c) Zone 3, where a wide range of GPP estimations (321 g C m2 d1), resulted in a relatively narrow range (111%) of POC ux as % PP (Fig. 3, bottom). A detailed analysis showed that the dominant type of particle contributing to the POC ux was clearly faecal material from zooplankton. However, from Zone 1 to Zone 3, the increase in GPP is paralleled by an increase in the relative contribution of diatom cells in the sediment traps (Fig. 3, upper panel). Vertical uxes of zooplankton faecal pellets and diatom were similarly important during the productive period (late-spring and early summer), while during the rest of the year, faecal material became more important in the removal of organic matter from the photic zone. Similar analysis, but for only the samples collected at the timeseries station on the shelf of Concepcion (Station 18), showed the same pattern, where only two zones could be identied (a) Zone 1, where a relatively narrow range of low production estimates (0.92.2 g C m2 d1) resulted in a wide range of export productions as% PP (963%) and (b) Zone 2, where a wide range of GPP estimations (421 g C m2 d1), resulted in a relatively narrow range (2.59.6%) of POC ux as % PP (Fig. 4, bottom). The dominant type of particle in the sediment traps showed that in Zone 1 the faecal material contributed almost 90% of the collected particles. Within Zone 2, faecal material shared dominance with diatoms, contributing 61 and 33% of the carbon ux, respectively (Fig. 4, upper panel). At Station 18, where several biological (i.e. Chl-a and faecal material), physical (i.e. u and v components of wind stress) and chemical (i.e. nutrients) variables were simultaneously measured

2.2. Data analysis 2.2.1. Concepcin time-series station (Station 18) For each sampling occasion, the percent of the daily GPP that sank as diatoms (carbon units) was estimated. The natural logarithm of this index was correlated with the biomass specic primary production (GPP/Chl-a). The biomass specic primary production has been used as an indicator of diatom physiological condition in a given situation, since a dominance of senescent cells within the community or the occurrence of environments with poor nutrient (and/or light) conditions would yield low GPP/Chl-a values. Although GPP/Chl-a is also dependent on community structure, previous analyses of St. 18 data indicate that diatoms contribute the bulk of phytoplankton biomass and production all year round (Montero et al., 2007). Therefore, it was likely that environmental conditions played a larger role on GPP/Chl-a ratio in St. 18 than community composition. It was also expected that the metabolic condition of diatoms was reected in the percentage of sinking particles. In addition, a linear model was tted using Statistica 7.0 to analyze the ln(diatom ux) dependency on ln(Chl-a) and GPP/Chl-a ratio. The last model was applied in order to analyze whether the actual raw values of diatom ux depended on diatom biomass and biomass-specic gross primary production. We expected that a situation with high diatom biomass but low production to biomass ratio would reect decaying blooms and therefore it would yield higher diatom uxes. Since GPP has been proven to be highly related to diatom biomass (as Chl-a) in the Concepcin time-series station (Montero et al. 2007), we also analyzed the relationship between euphausiid faecal pellet ux, with Chl-a integrated in the upper 15 m of the water column and.wind conditions. First, the relationship between the across-shore wind component (EW, u), taken daily at 2 pm and averaged over the ve days previous to sampling (from now on u-index) and diatom as Chl-a was explored for the Concepcion time-series station. Since most of the chlorophyll biomass variance could be explained by such a simple index (see results) but a few exceptional departures from the correlation existed, we split the data with a criterion of high/low euphausiid pellet carbon ux (with 100 mg C m2 d1 as cutting point). Then, we re-analyzed Chl-a relationship with the u index for low ux dates. For high ux dates, a generalized lineal model, based on normal distribution and a logarithmic link was adjusted for Chl-a (dependent variable) and u index and fecal pellet ux (independent variables). For the later analysis, McCullagh and Nelder (1999) were followed, and the Statistica 7.0 package was used. Since our goal is to complement the knowledge about the relationship (or lack of relationship) between GPP and carbon ux, we emphasize the relationships and ndings from the dates with high pellet ux in the later analysis. A deeper description of diatom biomass dependency on physical forcing in the studied area is beyond our goals in the present paper, and it will be presented elsewhere (B. Yannicelli, unpublished data).

1200 1000 POC flux (mg m d )

-1

10%

800 600 400 200 0 0 5000 10000 15000 20000

-2 -1

-2

2%

25000

Gross primary production (mgC m d )

Fig. 2. Gross primary production (GPP) data plotted against POC ux (mg C m2 d1). Dashed lines represented situations, where 2% or 10% of the GPP were collected in sediment traps.

Author's personal copy

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227

221

14000

GPP (mgCm-2 d-1)

10000 8000

400 300

6000 4000 2000 0 1 2 3 200 100 0

Zone

70 60
ZONE 1

POC fiux (as %PP)

50 40 30 20 10 0 0

ZONE 2

ZONE 3

5000

10000

15000

20000

POC flux (mgCm-2d-1)

12000

GPP POC flux Pellet flux (%) Diatoms flux (%) Dinoflagellate flux (%)

600 500

25000

Gross Primary Production (mgC m-2 d-1)

Fig. 3. The bottom panel shows a power t of GPP data plotted against POC ux (as % PP) (y = 2219.5x07394 r2 = 0.48) for the HCS (open circles and black diamonds are winter and summer samples, respectively). The data set was split into three zones based on the magnitude and variability of the data (see text for explanation). Upper panel: GPP and POC ux rates (black and grey bars) and the relative (%) contribution of faecal pellets, diatoms, and microzooplankton to the POC ux are depicted in the pie graphs.

along with GPP and POC uxes, we explored relationships between these uxes and those of euphausiid, appendicularian faecal material, diatom frustules as well as Chl-a and the GPP/Chl-a ratio in the water column from August 2002 to May 2007. Peaks in faecal material coincide with low abundances of diatoms in the water column and very low uxes of diatoms in sediment traps. In addition, high uxes of diatoms (arrows in Fig. 5) coincide with high diatom abundances in the water column and low GPP/Chl-a ratio (Fig. 5). The largest uxes of carbon as euphausiid faecal pellets were found at low Chl-a concentrations. On the contrary, when high Chl-a concentrations were found in the water column, faecal pellet ux was scant (Figs. 5 and 6 upper panel). Roughly, high uxes of faecal pellets (above 100 mg C m2 d1) occurred at low Chl-a concentration. Therefore, we took into account those sampling dates when faecal pellet ux exceeded 100 mg C m2 d1, and found that the relationship between the standing stock of Chl-a and acrossshore wind fell below the expected from the upwelling-favourable wind pattern (B. Yannicelli, unpublished data). In fact, low Chl-a concentrations were found at environmental conditions favourable for bloom development. In addition, those same dates coincided with a high production to biomass ratio indicating also healthy diatom condition. Only when we combined physical forcing (u index), and faecal pellet ux (as proxy for in situ predation) as explanatory variables during those dates, over 50% of Chl-a variance could be explained (Fig. 6, Table 3), denoting the topdown effect of primary consumers (largely krill) on diatom biomass.

Diatom ux (mg C m2 d1) at Station 18 usually represented less than 1% of daily PP, and it never exceeded 7%. Over 50% of the ln(diatom ux) variance was signicantly explained by Chl-a concentration in the euphotic layer, and the physiological condition of the bloom (production to biomass ratio) (Fig. 7 and Table 2). Higher diatom sedimentation is favoured by large quantities of phytoplankton and poor physiological condition, typical of senescent blooms after nutrient exhaustion (see arrows in Fig. 5). Additionally, the physiological condition of the diatom biomass was able to explain 49% of the variability of the fraction of GPP that sediments as whole diatoms (Fig. 7 upper panel). 4. Discussion 4.1. Conceptual ux model The HCS off Chile is one of the Eastern Boundary Current Systems (EBCS), seasonally centered in the spring-summer period and characterized by high GPP, opal-dominated coastal zones, and a high f-ratio. Overall, a lack of coincidence between peaks of GPP and POC uxes was found. Although net primary production (NPP) would be a more suitable measure to relate to vertical uxes of particulates than GPP, we note that usually neither measure correlated with estimates of vertical POC uxes. This is mainly because of the remarkable variability in the time lag between (1) the production of suspended POC, (2) the physical, chemical and biological processes that enhance the biological aggregation/

Author's personal copy

222

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227

16000

GPP (mgCm-2 d-1)

12000 10000 8000 6000 4000 2000 0 1 2

500 400 300 200 100 0

Zone

70 60

POC flux (as%PP)

50 40 30 20 10 0 0

ZONE 1 ZONE 2

5000

10000

15000

20000
-2 -1

25000

Gross Prim ary Production (m gC m d )

Fig. 4. Power t (y = 7120.3x
0.834

r = 0.59) as in Fig. 3, but including only the data collected at the time-series station (Station 18) located on the shelf area of Concepcin.

sinking of particles and (3) the biological processes that recycle/ oxidise the particles in the ocean. During recent years, it has been suggested that these processes operate at different temporal and spatial scales and are mediated by complex physical and biological interactions that are not well understood (Boyd and Newton 1999; Lutz et al., 2002). Thus, the time lag between the maximum in GPP or NPP and that of POC uxes remains poorly understood (Buesseler et al., 1998; Kawakami and Honda, 2007). Based on these arguments, we did not expect a close correlation between our GPP estimates and the POC uxes, as one might for new production versus POC export in large areas and integrated for the whole year. In addition, in coastal waters, the input of allochthonous organic matter from rivers contributes to mask the relationship between photosynthetic production and vertical ux. In the Concepcin shelf ecosystem the combined annual ow of the Itata and Bobo rivers, the two major rivers in the area, is estimated at ca. 63 km3 (Direccin General de Aguas, Chile). If we use a conservative estimate for DOC of 8200 mg C m3 (Pantoja et al., 2009), the input of DOC would be about 5 1011 g C per year which is roughly equivalent to the biomass of landed sh in the area (Cubillos et al., 1999).

The above described scenario give insights to understand why, despite a very low annual average GPP/CR ratio (GPP/ CR = 1.1 0.6, n = 18; CR = community respiration) (Montero et al., 2007), which includes negative values during winter; we recorded a relatively high vertical ux of POC at the time-series station off Concepcin (Fig. 4, upper panel). Thus, we treated the GPP values as a reference, and tackle this problem by focusing on the functional groups of plankton (diatoms, euphausiid faecal pellets) that mainly contribute to vertical ux of POC in coastal waters off the HCS. In this study, we observed that sinking uxes dominated by diatoms corresponded to highly productive areas and dates, even though diatom were less efcient at transferring POC to deeper layers of the water column than faecal pellet-dominated uxes (Figs. 3 and 4). At the HCS, the absolute vertical carbon uxes were relatively high, but the transfer efciency relatively low, and highly variable (Figs. 2 and 7). Variability in export efciency could also arise from the composition of the exported organic matter, i.e., the occurrence of biodegradable loose aggregates (Francois et al., 2002) or faecal material (Gonzlez et al., 2007). A large proportion of the daily phytoplankton production is ungrazed in the mixed layer, and is

POC flux (mgCm-2 d-1)

14000

GPP POC flux Pellet flux (%) Diatoms flux (%) Dinoflagellate flux (%)

700 600

Author's personal copy

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227

223

mg C m-2d-1

2000 1000 0 60

Euphausiid faecal strings

Sediment trap

mg C m-2d-1

45 30 15 0 200

Appendicularian pellets

mg C m-2d-1

150 100 50 0 800 600 400

Diatoms

mg m-2

Chl-a total

Water column

200 0 200

GPP/Chl-a

150 100 50 0

GPP/Chl-a (30)

A S O N D J F M A MJ J

A S O N D J A M J A S O N D J F M A M J A S N D J F M A M J J A S ON D J F M A M

2002

2003

2004

2005

2006

2007

Date (day, month, year)

Fig. 5. From top to bottom: Fluxes of euphausiid and appendicularian faecal material, diatom frustules collected in sediment traps, Chl-a, and the GPP/Chl-a ratio in phytoplankton collected in the upper 30 m of the water column at the time-series station (Station 18) on the shelf area of Concepcin from August 2002 to May 2007.

advected offshore from the production site, both in the California(Olivieri and Chavez, 2000) and the HCS-EBCSs (Morales et al., 2007). In contrast, in oceanic areas off the HCS that are carbonate-dominated (low f-ratio and more oligotrophic), the transfer efciency might be high because most of the highly biodegradable organic matter is usually consumed by a complex food web, and has undergone a highly efcient degradation, while a relatively small fraction is packaged in fast-sinking faecal pellets (Gonzlez et al., 2007). The structure and composition of the pelagic food-web were determinant in controlling the export of biogenic particles, a situation reported earlier (Legendre and LeFevre, 1989) even for diverse environments (Boyd and Newton, 1999). In the HCS, the overall fraction of PP exported below 100 m depth ranged from 160%, similar to the usually described range of 250% for different areas of the world oceans (Buesseler, 1998; Laws et al., 2000). This high variability seems to be of biological origin, stemming from differences in the quantity and quality of the food-web structure and functioning at various times and places (Legendre and Rivkin, 2002). The difculties of unravelling the mechanisms behind carbon ux control might be in part the disconnection between the proxies to estimate biogeochemical processes and the in situ biogeochemical processes itself. This disconnection might occur because uxes of bioelements cross through multiple vertical strata experiencing diverse processes, and where the origin of any bioel-

ement (i.e., carbon, calcite, opal) might be from different functional groups or key species (Verity and Smetacek, 1996). When the fraction (percentage) of GPP that effectively sinks is relatively low, a signicant part of the sinking organic matter is composed of diatoms, regardless of total GPP (Fig. 3, zone 3). Such a fraction appears to be affected by the physiological state of diatom biomass. On the contrary, when the fraction (percentage) of GPP that sinks is high, most of sinking organic matter is composed of euphausid pellets and occurs at relatively low GPP and Chl-a (Fig. 3, zone 1). Most of these high sinking rates of pellets and low diatom biomasses occur during summer months, when physical conditions favour the presence of phytoplankton blooms, and when the high P/B ratio indicates healthy physiological conditions. All this evidence points to (i) the top-down control of euphausiids over primary producer biomass, and (ii) euphausiids paramount role on total organic carbon ux over the Concepcin continental shelf. Overall, on the shelf, the faecal material ux was dominated by krill faecal strings loaded with diatoms in spring-summer and dinoagellates/diatoms in autumnwinter, while in oceanic deployments (>50 km from shore) the ux of appendularian faecal pellets loaded with coccolithophorids was also important. Along the whole HCS, the pivotal role of euphusiid faecal material in the export ux (pelagic-benthic coupling) has been reported for the upwelling systems off Concepcin (36S, Gonzlez et al., 2007), Coquimbo (30S, Gonzlez et al., 2004) and Antofagasta

Author's personal copy

224

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227

Chl-a integrated to 15m (mg m -2)

400 350 300 250 200 150 100 50 0 -1 300

R2 = 0.4566

0
R = 0.5558
2

250 200 150 100 50 0 -1 0 1 2 3 4

5 day averaged U (m s-1)

Sedimented faecal pellets< 100 mgCm d -1
300
-2

Sedimented faecal pellets> 100 mgCm-2 d-1

300 200 200 150 100 50

Chl-a integrated to 15m (mg m-2 )

R = 0.9454

250 200 150 100 50 0 -1 0 1 2 3 4

5 day averaged U (m

s-1)

Chl-a integrated to 15m (mg m-2 )

-1 300 250 200 150 100 50 0 0 300 250 200 150 100 50 0 0
R2= 0.7091

5 day averaged U (ms -1)

200

400

600

800

1000 1200

Sedimented faecal pellets (mgCm -2d -1)

50

100

150

200

Modeled chl-a (mg m -2 )

Fig. 6. (a) Chlorophyll-a integrated from surface to 15 m depth, as determined for time series station 18 quasi-monthly from September 2003 to May 2005, as a function of wind (u-component averaged over the 5 days previous to biological sampling: u index); (b) same as (a) but only for sampling occasions when sediment trap data is available; (c) same as (b) but selecting dates when euphausiid faecal pellets ux was below 100 mg m2 d1; (d) same as (b) but selecting dates when euphausiid faecal pellets ux was above 100 mg m2 d1; (e) Chl-a vs euphausiid faecal pellet ux for dates of high ux; (f) Chl-a vs linearly modeled Chl-a as a function of u index and faecal pellet ux.

(23S, Gonzlez et al., 2000). Faecal pellet carbon is usually a significant, albeit highly variable, fraction of the POC ux in several, not only in coastal, productive environments such as the Norwegian

and Greenland Sea (790%, Bathmann et al., 1987, 242%, JuulPedersen et al., 2006), and Southern California Bight ($70%, Landry et al., 1994), but also in oceanic, oligotrophic regions such as the

Author's personal copy

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227

225

4 2 0 -2 -4 -6 -8

R 2 = 0.49

50 100 150 -1 GPP/Chl-a (mgC d / mgChl-a -1)

200

5 2 4 Multiple R = 0.50 3 2 1 0 -1 -2 -3 -6 -4 -2

ln(Diatomflux)

summer (Gonzlez et al., 2007). Diatoms and faecal pellet sinks are episodic events, where the latter seems to be a more continuous process that never reaches values close to zero. Diatom sinking might have been overlooked because several diatom mass-sedimentation events probably by-passed the traps due to deployments for short periods (1024 h), as previously shown (Gonzlez et al., 2007). Diatoms reported as key species in the downward ux of POC in the HCS are the genus Thalassiosira, Chaetoceros and Skeletonema (Gonzlez et al., 2007). The transfer of phytoplankton carbon was mainly due to diatoms, probably because of the ability of this group to produce more mucopolysaccarides and TEP than any other phytoplanktonic group. Species of the long-spined diatom genus Chaetoceros appear to have played a major role in aggregate formation and sinking events in highly diverse environments ranging from the highly productive coastal upwelling off Chile (Gonzlez et al., 1987; Gonzlez et al., 2007) to high nutrient low chlorophyll areas in the Southern Ocean (Smetacek et al., 2006). 4.2. Concepcion time-series station (Station 18)

ln(Diatomflux/GPP*100)

Modeled ln(diatom flux (mgCm -2d-1)) -2.08-0.02*(GPP/Chl-a)+0.95*ln(Chl-a)

Fig. 7. Upper panel: ln(diatom ux/PP 100) plotted against the GPP/Chl-a ratio (mgC d1 mg Chl-a1). Bottom panel: modeled versus measured diatom ux at the time-series station (Station 18) off Concepcin. The diatom ux was related positively to Chl-a and negatively to biomass-specic gross primary production.

Table 2 General linear model results for the dependency of ln(diatom ux) on GPP/Chl-a ratio and ln(mg Chl-a m3). SS Whole model Error 71.77 70.67 df 2 26 1 1 26 MS 35.88 2.71 17.78 38.77 2.71 F 13.2 P 0.0001 Multiple R2 0.50

Univariate results GPP/Chl-a 17.78 ln(Chl-a) 38.77 Error 70.67

6.54 14.26

0.016 0.0008

Table 3 General linear model results for the dependency of Chl-a (mg Chl-a m2, integrated down to 15 m depth) on cross-shore wind u (m s1) and euphausiids pellet ux Euph. ux (mg C m2 d1). SS Whole model Error Univariate results Euph ux u Error 437052 28025 28023 33449 28025 df 2 7 1 1 7 MS 21852 4003 28023 33449 4003 F 5.45 P 0.03 Multiple R2 0.60

6.99 8.35

0.03 0.02

ALOHA and K2 time-series stations, where the ratio algal:faecal export ux was in the range of 0.10.2 (Boyd et al., 2008). On an annual basis, fast-sinking faecal material and algal cells make a major and minor contribution, respectively, to the downward POC ux at 50 and 100 m depths in several environments such as the subarctic Pacic (St. Papa) (Thibault et al., 1999), Disko Bay, Greenland (Juul-Pedersen et al., 2006) and Monterey Bay, California (Olivieri and Chavez, 2000). For the time-series station off Concepcion, zooplankton faecal pellets were the most prominent type of particle contributing to the vertical ux of carbon during autumn, winter and early-spring, averaging 79.6% of the POC ux, while phytoplankton made a signicant contribution (17.3%) in

Off Concepcion, GPP positively correlates with both Chl-a and diatom abundance in the water column (Montero et al., 2007). The GPP/Chl-a ratio (as a proxy of the physiological state of diatom biomass) was dependent on daily maximum irradiance and mixed layer nitrate concentration (B. Yannicelli, unpublished data). Therefore, diatom sedimentation at Station 18 would strongly depend on the seasonal variability of irradiance and the synoptic surface nutrient variability that results from its injection in the mixed layer during upwelling events and exhaustion by the pulsing development of fast growing diatom blooms during spring-summer. In spite of the multiple factors that affect sedimentation rates (reviewed by Buesseler et al., 2007) and the general lack of direct relationship that is usually found between surface production and sinking organic matter (Boyd and Trull, 2007), variability in sinking rates seems to be largely explained by three main variables: GPP, Chl-a and zooplankton predation. These factors affect sinking uxes due to: (i) the rather shallow deployment of sediment traps, (ii) the strength of synoptic/annual bio-physical signals and (iii) the dominance of key zooplanktonic species in the HCS. Finally, euphausiids have been recognized as the main link between primary producers and the main sh species caught in the HCS (Antezana, 2007). Highly dense euphausiid swarms are able to remove most of the diatom biomass, constituting the key element channelling PP to higher trophic levels and into sinking organic carbon. During certain periods of the year (e.g. February 2004, May 2005, Fig. 5), gelatinous zooplankton (such as appendicularians) might overtake euphausiids role in exporting carbon to deep waters. The ability of euphausiids to remove large quantities of diatom biomass exerting a topdown control on GPP might also affect the development of the microbial loop which could be favoured during the absence or low occurrence of this large omnivorous zooplankton. The presence of euphausiids could preclude momentarily the increase of the microbial community by competing for the main organic carbon producers, and by directly consuming microzooplankton. The presence or absence of euphausiids could determine shifts from autotrophic to heterotrophic periods in the coastal HCS at the latitude of Concepcin. Any model addressing carbon uxes in the Concepcin area should include the tremendous impact of euphausiids, as evidenced by sediment trap data. Acknowledgements We thank the ofcers and crew of the research vessels Abate Molina, AGOR Vidal Gormaz, LC Kay Kay and LC Purihaalar and many

Author's personal copy

226

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227 Coquimbo (30S), Chile, during 19931995. Deep-Sea Research II 51, 2457 2474. Gonzlez, H.E., Menschel, E., Aparicio, C., Barra, C., 2007. Spatial and temporal variability of microplankton and detritus, and their export to the shelf sediments in the upwelling area off Concepcin, Chile (36S), during 2002 2005. Progress in Oceanography 75, 435451. Grnewald, A., Morales, C., Gonzlez, H.E., Sylvester, C., Castro, L., 2002. Grazing impact of copepod assemblages and gravitational ux in coastal and oceanic waters off central Chile during two contrasting seasons. Journal of Plankton Research 24, 5568. Juul-Pedersen, T., Nielsen, T.G., Michel, C., Mller, E.F., Tiselius, P., Thor, P., Olesen, M., Selander, E., Gooding, S., 2006. Sedimentation following the spring bloom in Disko Bay, West Greenland, with special emphasis on the role of copepods. Marine Ecology Progress Series 314, 239255. Kawakami, H., Honda, M.C., 2007. Time-series observations of POC uxes estimated from 234Th in the northwestern North Pacic. Deep-Sea Research I 54, 1070 1090. Klaas, C., Archer, D.E., 2002. Association of sinking organic matter with various types of mineral ballast in the deep sea. Implications for the rain ratio. Global Biogeochemical Cycles 16. doi:10.1029/2001GB00176. Knap, A.H., Michaels, A.F., Dow, R.L., Johnson, R.J., Gundersen, K., Sorensen, J.C., Clos, A.R., Howse, F.A., Hammer, M., Bates, N., Doyle, A., Waterhouse, T., 1993. US Joint Global Flux Study, Bermuda Atlantic Time-series Study. BATS Methods Manual, Version 3, Woods Hole, MA. Landry, M.R., Lorenzen, C.J., Peterson, W.K., 1994. Mesozooplankton grazing in the Southern California Bight. II. Grazing impact and particulate ux. Marine Ecology Progress Series 115, 7385. Laws, E.A., Falkowski, P.G., Smith, W.O.J., Ducklow, H., McCarthy, J.J., 2000. Temperature effects on export production in the open ocean. Global Biogeochemical Cycles 14, 12311246. Legendre, L., LeFevre, J., 1989. Hydrodynamical singularities as controls of recycled vs. export production in the ocean. In: Berger, W.H., Smetacek, V.S., Wefer, G. (Eds.) Productivity of the Oceans: Present and Past. Dahlem Workshop Life Science Report. J. Wiley & Sons, pp. 155173. Legendre, L., Rivkin, R., 2002. Fluxes of carbon in the upper ocean: regulation by food-web control nodes. Marine Ecology Progress Series 242, 95109. Liu, K.K., Atkinson, L., Chen, C.T., Gao, S., Hall, J., McDonald, R.W., McManus, L.T., Quiones, R., 2000. Exporting continental margin carbon uxes on a global scale. Eos American Geophysical Union Transactions 185, 641644. Lutz, M., Dunbar, R., Caldeira, K., 2002. Regional variability in the vertical ux of particulate organic carbon in the ocean interior. Global Biogeochemical Cycles 16(3), doi: 10.1029/2000GB001383. McCullagh, P., Nelder, J.A., 1999. Generalized Linear Models, second ed. Monographs on Statistics and Applied Probability, vol. 37. Chapman & Hall. 51 pp. Menden-Deuer, S., Lessard, E.J., 2000. Carbon to volume relationships for dinoagellates, diatoms, and other protist plankton. Limnology and Oceanography 45, 569579. Montero, P., Daneri, G., Cuevas, L.A., Gonzlez, H.E., Jacob, B., Lizrraga, L., Menschel, E., 2007. Productivity cycles in the coastal upwelling area off Concepcion: The importance of diatoms and bacterioplankton in the organic carbon ux. Progress in Oceanography 75, 518530. Morales, C., Gonzlez, H.E., Hormazabal, S., Yuras, G., Letelier, J., Castro, L., 2007. The distribution of chlorophyll-a and dominant planktonic components in the coastal transition zone off Concepcin, central Chile, during different oceanographic conditions. Progress in Oceanography 75, 452469. Olivieri, R.A., Chavez, F.P., 2000. A model of plankton dynamics for the coastal upwelling system of Monterrey Bay, California. Deep-Sea Research II 47, 1077 1106. Pantoja, S., Seplveda, J., Gonzlez, H.E., 2004. Decomposition of sinking proteinaceous material during fall in the oxygen minimum zone off northern Chile. Deep-Sea Research I 51, 5570. Pantoja, S., Iriarte, J.L., Gutirrez, M., Calvete, C., 2009. The southern chile continental margin. In: Liu, K.-K., Atkinson, L., Quiones, R., Talaue-McManus, L. (Eds.), Carbon and Nutrients Fluxes in Continental Margins: A Global Syntheis. Global Change. The IGBP Series. Springer Verlag, New York, pp. 265 273. Parsons, T., Maita, Y., Lalli, C., 1984. A Manual of Chemical and Biological Methods for Seawater Analysis. Pergamon Press, Oxford. 173 pp.. Pilskaln, C.H., Honjo, S., 1987. The fecal pellet fraction of biogeochemical particles uxes to the deep sea. Global Biogeochemical Cycles 1, 3148. Rutllant, J., Montecino, V., 2002. Multiscale upwelling forcing cycles and biological response off northcentral Chile. Revista Chilena de Historia Natural 75, 217 231. Smetacek, V., 1985. Role of sinking in diatom life-history cycles: ecological, evolutionary, and geological signicance. Marine Biology 84, 239251. Smetacek, V., Assmy, P., Cisewski, B., Henjes, J., Klaas, C., Strass, V., Montresor, M., 2006. Aggregation and mass sinking of diatom species populations from an iron-fertilized bloom in the Southern Ocean, International Workshop on Marine Aggregates (IWOMA), 11+12 December 2006, MPI Bremen, Germany. Sobarzo, M., Bravo, L., Donoso, D., Garcs-Vargas, J., Schneider, W., 2007. Coastal upwelling and seasonal cycles that inuence the water column over the continental shelf off central Chile. Progress in Oceanography 75, 363382. Strickland, J.D.H., 1960. Measuring the production of marine phytoplankton. Bulletin of Fisheries Research Board Canada 122, 172.

post- and pre-graduate students who participated in our cruises during the last decade. We also would like to thank M.J. Caldern, C. Valenzuela, T. Cuevas, P. Montero, for their help during the laboratory analysis. We also thank Dr. C. Lange and P. Bernal for their valuable suggestions that substantially improved an earlier version of the manuscript. Financial support for this study was provided by the FONDAP-COPAS Grant 15010007 and Fondecyt Grants Nos. 5960002, 1000419, 1000366, and 1050487. References
Amstrong, R.A., Lee, C., Hedges, J.I., Honjo, S., Wakeham, S.G., 2002. A new, mechanistic model for organic carbon uxes in the ocean based on the quantitative association of POC with ballast minerals. Deep-Sea Research I 49, 219236. Antezana, T., 2007. Euphausia mucronata a central link in the Humboldt Current food web. In: Abstract Book of the 4th International Zooplankton Production Symposium: Human and Climate Forcing of Zooplankton Populations. Hiroshima, Japan, p. 82. Bathmann, U.V., Noji, T.T., Voss, M., Peinert, R., 1987. Copepod fecal pellets: abundance, sedimentation and content at a permanent station in the Norwegian Sea in May/June 1986. Marine Ecology Progress Series 38, 4551. Berelson, W.M., 2001. The ux of particulate organic carbon into the ocean interior: a comparison of four U.S. JGOFS regional studies. Oceanography 14, 5967. Billett, D.S.M., Lampitt, R.S., Rice, A.L., Mantoura, R.F.C., 1983. Seasonal sedimentation of phytoplankton to the deep sea benthos. Nature 302, 520522. Bishop, J.K.B., Edmond, J.M., Ketten, D.R., Bacon, M.P., Silker, W.B., 1977. The chemistry, biology, and vertical ux of particulate matter from the upper 400 m of the equatorial Atlantic Ocean. Deep-Sea Research 24, 511548. Bodungen, B.v., Wunsch, M., Frderer, H., 1991. Sampling and analysis of suspended and sinking particles in the Northern North Atlantic. In: Marine Particles: Analysis and Characterization. Geophysical Monograph, vol. 63. American Geophysical Union, pp. 4756. Boyd, P.W., Newton, P., 1999. Does planktonic community structure determine downward particulate organic carbon ux in different oceanic provinces. DeepSea Research I 46, 6391. Boyd, P.W., Trull, T.W., 2007. Understanding the export of biogenic particles in oceanic waters: is there consensus? Progress in Oceanography 72, 276312. Boyd, P.W., Gall, M.P., Silver, M.W., Coale, S.L., Bidigare, R.R., Bishop, J.L.K., 2008. Quantifying the surfacesubsurface biogeochemical coupling during the VERTIGO ALOHA and K2 studies. Deep-Sea Research II 55, 15781593. Buesseler, K.O., 1998. The decoupling of production and particulate export in the surface ocean. Global Biogeochemical Cycles 12, 297310. Buesseler, K.O., Ball, L., Andrews, J., Benitez-Nelson, C., Belastock, R., Chai, F., Chao, Y., 1998. Upper ocean export of particulate organic carbon in the Arabian Sea derived from thorium-234. Deep-Sea Research II 45, 24612487. Buesseler, K.O., Antia, A.N., Chen, M., Fowler, S.W., Gardner, W.D., Gustafsson, O., Harada, K., Michaels, A.F., Rutgers van der Loeff, M., Sarin, M., Steinberg, D., Trull, T., 2007. An assessment of the use of sediment traps for estimating upper ocean particle uxes. Journal of Marine Research 65, 345416. Burd, A.B., Jackson, G.A., Moran, S.B., 2007. The role of the particle size spectrum in estimating POC uxes from 234Th /238U disequilibrium. Deep-Sea Research I 54, 897918. Coppola, L., Roy-Barman, M., Wassmann, P., Mulsow, S., Jeandel, C., 2002. Calibration of sediment traps and particulate organic carbon export using 234 Th in the Barents Sea. Marine Chemistry 80, 1126. Cubillos, L., Seplveda, A., Glvez, M., Arcos, D., 1999. Situacin actual de los principales recursos pesqueros de la zona centro-sur de Chile. Instituto de Investigacin Pesquera, Talcahuano (Chile), Technical Report, 23 pp. Daneri, G., Dellarossa, V., Quiones, R., Jacob, B., Montero, P., Ulloa, O., 2000. Primary production and community respiration in the Humboldt Current System off Chile and associated oceanic areas. Marine Ecology Progress Series 197, 4149. De La Rocha, C., Passow, U., 2007. Factors inuencing the sinking of POC and the efciency of the biological carbon pump. Deep-Sea Research II 54, 639658. Francois, R., Honjo, S., Krisheld, R., Manganini, S., 2002. Factors controlling the ux of organic carbon to the bathypelagic zone of the ocean. Global Biogeochemical Cycles 16. doi:10.1029/2001GB00172. Gardner, W.D., 2000. Sediment trap sampling in surface waters. In: Hanson, R.B., Ducklow, H.W., Field, J.G. (Eds.), The Changing Ocean Carbon Cycle: a Midterm Synthesis of the Joint Global Ocean Flux Study, vol. 5. Cambridge University Press, Cambrige, pp. 240281. Gonzlez, H.E., 1994. Zooplankton faecal pellets in high-latitude marine ecosystems. PhD thesis, Biology/Chemical Department, University of Bremen, Germany, 223 pp. Gonzlez, H.E., Bernal, P., Ahumada, R., 1987. Desarrollo de dominancia local en la taxocenosis de toplancton de Baha de Concepcin, Chile, durante un evento de surgencia. Revista Chilena de Historia Natural 60, 1935. Gonzlez, H.E., Ortiz, V., Sobarzo, M., 2000. The role of faecal material in the particulate organic carbon ux in the northern Humboldt Current, Chile (23S), before and during the 19971998 El Nio. Journal of Plankton Research 22, 499529. Gonzlez, H.E., Hebbeln, D., Iriarte, J.L., Marchant, M., 2004. Downward uxes of faecal material and microplankton at 2300 m depth in the oceanic area off

Author's personal copy

H.E. Gonzlez et al. / Progress in Oceanography 83 (2009) 217227 Thibault, D., Roy, S., Wong, C.S., Bishop, J.K., 1999. The downward ux of biogenic material in the NE subarctic Pacic: importance of algal sinking and mesozooplankton herbivory. Deep-Sea Research II 46, 26692697. Turner, J.T., 2002. Zooplankton fecal pellets, marine snow and sinking phytoplankton blooms. Aquatic Microbial Ecology 27, 57102. Utermhl, H., 1958. Vervollkommnung der quantitativen Phytoplankton-Methodik. Mitteilungen Internationale Vereinigung Theorie Angewandte Limnologie vol. 9, 138. Verity, P., Smetacek, V., 1996. Organism life cycle, predation, and the structure of marine pelagic ecosystems. Marine Ecology Progress Series 130, 277 293.

227

Wassmann, P., Ypma, J.E., Tselepides, A., 2000. Vertical ux of faecal pellets and microplankton on the shelf of the oligotrophic Cretan Sea (NE Mediterranean Sea). Progress in Oceanography 46, 241258. Walsh, J.J., 1991. Importance of continental margins in the marine biogeochemical cycling of carbon and nitrogen. Nature 350, 5355. Williams, P.J.LeB., Jenkinson, N.W., 1982. A transportable micro-processorcontrolled precise Winkler titration suitable for eld station and shipboard use. Limnology and Oceanography 27, 576584. Yuras, G., Ulloa, O., Hormazabal, S., 2005. On the annual cycle of coastal and open ocean satellite chlorophyll off Chile (1840S). Geophysical Research Letters 32. doi:10.1029/2005GL023946.