Shift in the species composition of the diatom ...

Progress in Oceanography 159 (2017) 31–44

Contents lists available at ScienceDirect

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

Shift in the species composition of the diatom community in the eutrophic Mauritanian coastal upwelling: Results from a multi-year sediment trap experiment (2003–2010)

MARK

⁎

Oscar E. Romeroa, , Gerhard Fischera,b a b

University of Bremen, Marum, Center for Marine and Environmental Sciences, Leobener Str. 13, 28359 Bremen, Germany University of Bremen, Geosciences Department, Klagenfurter Str., 28359 Bremen, Germany

A R T I C L E I N F O

A B S T R A C T

Keywords: Diatoms Fluxes Eastern Boundary Upwelling Ecosystems Mauritania Nepheloid layers Sediment traps

A multiannual, continuous sediment trap experiment was conducted at the mooring site CBeu (Cape Blanc eutrophic, ca. 20 °N, ca. 18 °W; trap depth = 1256–1296 m) in the high-productive Mauritanian coastal upwelling. Here we present fluxes and the species-specific composition of the diatom assemblage, and fluxes of biogenic silica (BSi, opal) and total organic carbon (TOC) for the time interval June 2003-Feb 2010. Flux ranges of studied parameters are (i) total diatoms = 1.2 ∗ 108–4.7 ∗ 104 valves m−2 d−1 (average = 5.9 × 106 valves ± (ii) BSi = 296–0.5 mg m−2 d−1 (average = 41.1 ± 53.5 mg m−2 d−1), and (iii) 1.4 × 107); TOC = 97–1 mg m−2 d−1 (average = 20.5 ± 17.8 mg m−2 d−1). Throughout the experiment, the overall good match of total diatom, BSi and TOC fluxes is reasonably consistent and reflects well the temporal occurrence of the main Mauritanian upwelling season. Spring and summer are the most favorable seasons for diatom production and sedimentation: out of the recorded 14 diatom maxima of different magnitude, six occurred in spring and four in summer. The diverse diatom community at site CBeu is composed of four main assemblages: benthic, coastal upwelling, coastal planktonic and open-ocean diatoms, reflecting different productivity conditions and water masses. A striking feature of the temporal variability of the diatom populations is the persistent pattern of seasonal groups’ contribution: benthic and coastal upwelling taxa dominated during the main upwelling season in spring, while open-ocean diatoms were more abundant in fall and winter, when the upper water column becomes stratified, upwelling relaxes and productivity decreases. The relative abundance of benthic diatoms strongly increased after 2006, yet their spring-summer contribution remained high until the end of the trap experiment. The occurrence of large populations of benthic diatoms at the hemipelagic CBeu site is interpreted to indicate transport from shallow waters via nepheloid layers. We argue that a significant amount of valves, BSi and TOC produced in waters overlying the Banc d’Arguin and the Mauritanian shelf is effectively transported to the CBeu trap in intermediate waters at the outer Mauritanian slope. The impact of the intermediate and bottom-near nepheloid layers-driven transport in the transfer of valves and bulk particulates and its potential contribution to the export of biogenic materials from the shelf and uppermost slope might play a significant role in hemipelagial fluxes off Mauritania.

1. Introduction Among present-day Eastern Boundary Upwelling Ecosystems (EBUEs: Benguela, Canary, Californian and Humboldt), the Canary Current (CC) system is perhaps the least understood due to the paucity of information and the region’s complex topography and circulation (Mittelstaedt, 1983; Arístegui et al., 2009; Cropper et al., 2014). Within the Canary EBUE, the Mauritanian coastal upwelling ecosystem is characterized by intense offshore Ekman transport and strong

⁎

mesoscale heterogeneity that facilitates the exchange of neritic and pelagic water masses (Chavez and Messié, 2009). The Ekman transport is augmented by the offshore channeling of waters through mesoscale instabilities of the coastal jet, like upwelling filaments, squirts, eddies and submarine canyons that dramatically alter the large-scale picture (Zenk et al., 1991; Van Camp et al., 1991; Krastel et al., 2006; Meunier et al., 2012). Regional factors, like nutrient trapping efficiency of the upwelling cells (Arístegui et al., 2009), dust deposition at the ocean surface (Fischer et al., 2016; Friese et al., 2016) or the shelf width

Corresponding author. E-mail address: [email protected] (O.E. Romero).

http://dx.doi.org/10.1016/j.pocean.2017.09.010 Received 27 January 2017; Received in revised form 5 September 2017; Accepted 13 September 2017 Available online 15 September 2017 0079-6611/ © 2017 Elsevier Ltd. All rights reserved.


O.E. Romero, G. Fischer

from depths of 100 to 200 m south off the Banc d’Arguin (Mittelstaedt, 1983). North of it, the SACW merges gradually into deeper layers (200–400 m) below the CC (Fig. 1a) (Mittelstaedt, 1983). During the upwelling season, the stratification of shelf waters weakens, as does the stratification offshore, usually within the upper 100 m (Mittelstaedt, 1991). The biological response is drastically accelerated in the upwelled waters when the nutrient-richer SACW feeds the onshore transport of intermediate layers to form mixed-water types over the shelf (Zenk et al., 1991). The Mauritanian Current (MC, surface coastal countercurrent, Fig. 1a) gradually flows northward along the coast to about 20°N (Mittelstaedt, 1991). It brings warmer surface water masses from the equatorial realm into the study area. Towards late autumn, the MC is gradually replaced again by a southward flow associated with upwelling water due to the increasing influence of trade winds south of 20°N (Zenk et al., 1991), and becomes a narrow strip of less than 100 km width in winter (Mittelstaedt, 1983). The MC advances onto the shelf during summer and is enhanced by the relatively strong Equatorial Countercurrent and the southerly monsoons (Mittelstaedt, 1983). The existence of the strong coastal currents during the upwelling season causes substantial horizontal shear within the surface layer, where currents tend to converge (Mittelstaedt, 1983). Below the surface waters between ∼150 and ∼600 m south of the CBeu site, the unstable Cape Verde Frontal Zone (CVFZ) occurs. This front builds where the NACW meets the SACW (Zenk et al., 1991). The boundary between both water masses is convoluted, variable in position and characterized by intense mixing and interleaving processes (Barton et al., 1998, and references therein; Fig. 1a). Year-to-year perturbations of the location of the CVFZ might influence the nutrient availability off Mauritania. The giant Cape Blanc filament extends more than 300 km offshore (Fig. 1a; Van Camp et al., 1991; Zenk et al., 1991; Hagen, 2001; Santos et al., 2008; Arístegui et al., 2009; Cropper et al., 2014). It carries SACW offshore through an intense jet-like flow (Meunier et al., 2012; Fig. 1a). The chlorophyll concentration in the filament lies normally above 1 mg m−3 year-round, hence representing eutrophic conditions. Convergence of southern and northern waters at the frontal zone produces offshore export of the relatively cold upwelling waters. The upwelling cell drives vertically nutrient-rich subsurface waters to the baroclinic zone. In this manner, the upwelling jet, if deep enough, may play a decisive role in the along-shore advection of the northern water masses (Santos et al., 2008). Climate over Northwest Africa is also influenced by the latitudinal migration of the continental Intertropical Convergence Zone (ITCZ; also named Intertropical Front, Nicholson, 2013). This low-pressure zone separates the warm and moist SW monsoon flow from the dry NE trade winds (TW), blowing from the northern Sahara (Fig. 1a). Northeasterly TW dominates in fall-winter and are restricted to a shallow layer (approximately 1.5 km, Chiapello et al., 1995). A further source of nutrients is the deposition of significant amounts of Saharan dust into the low-latitude northern Atlantic by the westward, mid-tropospheric Harmattan winds within the Saharan Air Layer (SAL, altitude = 5–6 km), originated in the central Sahara (Mittelstaedt, 1983; Swap et al., 1996). Highest dust load within the SAL occurs during summer (Prospero, 1990). Highest dust deposition in the study area, however, occurs predominantly in winter (Fischer et al., 2016) due to gravitational settling (Friese et al., 2016). Dust supply to the NE Atlantic Ocean depends not only on the strength of the transporting wind systems but also on the rainfall and dryness in the multiple source regions in northwestern Africa (Chiapello et al., 1995; Nicholson, 2013).

(Hagen, 2001; Cropper et al., 2014), additionally affect the magnitude and intensity of primary production in and particle export from surface waters along the Mauritanian coast. The intense offshore transport is an important mechanism for the export of cool, nutrient-rich shelf and upper slope waters offshore Mauritania (Gabric et al., 1993). Based on satellite imagery and in situ data, it has been estimated that the giant Cape Blanc filament could export about 50% of the particulate coastal new production to the open ocean during intervals of most intense upwelling. Coastal phytoplankton might be transported at the surface as far as 400 km offshore (Gabric et al., 1993; Barton et al., 1998; Lange et al., 1998; Helmke et al., 2005). The effect of this transport could extend to even more distant regions in the deep ocean, since particles sink not only vertically but also laterally (Fischer and Karakaş, 2009). Observations off Mauritania during the past two decades prove that substantial year-to-year variations in fluxes and SST occur (Fischer et al., 1996, 2009, 2016; Bory et al., 2001; Müller and Fischer, 2001; Romero et al., 1999, 2002, 2003; Marcello et al., 2011; Skonieczny et al., 2013; Mollenhauer et al., 2015). Whether these variations are related to global- or Atlantic-scale climatic variations or to a natural level of basin-wide atmospheric and/or oceanic variability is still under discussion (Barton et al., 1998; Cropper et al., 2014; Fischer et al., 2016). Here, we present the first continuous record of diatom, biogenic silica and total organic carbon fluxes collected between June 2003 and February 2010 (ca. 2500 days of sampling) at the mooring site CBeu (Cape Blanc eutrophic), located within the giant Cape Blanc filament around 80 nm west of the Mauritanian coastline (Fig. 1). Seasonal and multi-year variability in fluxes and species-specific composition of the diatom assemblage, and biogenic silica and total organic carbon fluxes are presented and discussed in the context of variations of hydrographic and atmospheric conditions. 2. The study area This study was carried out in the Mauritanian upwelling area (Fig. 1a), one of the prominent EBUEs located at the eastern border of the North Atlantic Subtropical Gyre (Arístegui et al., 2009; Chavez and Messié, 2009; Cropper et al., 2014). The occurrence and intensity of the Mauritanian upwelling depends on the shelf width and seafloor topography along NW Africa (Mittelstaedt, 1983; Hagen, 2001). The Mauritanian shelf, which is wider than that to the north and to the south along the northwestern African margin, gently slopes from the coastline up to water depths of 200–300 m (Fig. 1b; Hagen, 2001). The shelf break zone with its steep continental slope extends over a distance of approximately 100 km (Hagen, 2001). According to Lathuilière et al. (2008), our study area is located within the Cape Blanc intergyre region (19–24°N), which is characterized by a weak seasonality. Following the definition by Cropper et al. (2014), our study area is situated on the southern rim of the strong and permanent coastal upwelling zone (21°–26°N). Details of hydrography, upwelling dynamics and main wind systems affecting the area off Mauritania are given below. 2.1. Hydrography, wind and upwelling dynamics The source of upwelling waters in the Cape Blanc area are either North Atlantic Central Water (NACW), north of about 23°N, or South Atlantic Central Water (SACW), south of 21°N, or both (Fig. 1a). The nutrient-poorer NACW has its source in the North Atlantic subducting zone and reaches subsurface waters at lower latitudes through the thermocline circulation (Sarmiento et al., 2004). The nutrient-richer, less saline SACW originates in the Subantarctic Zone of the Southern Ocean and travels via the South Atlantic thermocline equatorward into low-latitude regions (Sarmiento et al., 2004). The SACW occurs in layers between 100 and 400 m depth off Cape Blanc. The hydrographic properties of the upwelling waters on the shelf suggest that they ascend 32



(caption on next page)

33



Fig. 1. Map of the study area. (a) Schematic of the dynamics of the giant Cape Blanc filament (light green area), winds and water masses off Mauritania. Surface currents (Canary Current, CC, light blue arrows; North Equatorial Counter Current, NECC, light blue arrow; Mauritanian Current, MC, light orange arrows) and subsurface currents (South Atlantic Central Water, SACW, orange arrow) are depicted after Mittelstaedt (1983, 1991) and Zenk et al. (1991). The Cape Verde Frontal Zone (CVFZ) builds where the NACW meets the SACW (Zenk et al., 1991). Associated with the CVFZ and the giant filament, eddies (closed black circles) are seen (Meunier et al., 2012). The light yellow area close to the Mauritanian shoreline represents the warm Banc d’Arguin Water (BAW). The light-brown, diagonal lines represent the Banc d’Arguin. Dotted colors represent upwelled (orange) and non-upwelled water (blue), respectively (modified after Meunier et al., 2012). Trade winds (TW) and Saharan Air Layer (SAL) are represented by light yellow arrows (Nicholson, 2013). (b) Offshore transect along ca. 20.5°–21°N off Mauritania showing the general distribution of chlorophyll within the giant filament in the uppermost water column (green-shaded horizontal bar and arrows), and intermediate nepheloid (INL) and bottom nepheloid layers (BNL, yellow-shaded) (after Karakaș et al., 2006, 2009; Nowald et al., 2006, 2015). Encircled dots indicate poleward currents and encircled crosses equatorward currents (Mittelstaedt, 1983, 1991). Inverted red triangles represent deployed traps CBeu (large triangle, this study) and the outer CBmeso (small triangles; Romero et al., 2002, 2003; Fischer et al., 2016). Note that vertical scales in (a) and (b) are not identical. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3. Material and methods

Total organic carbon (TOC) and calcium carbonate were measured by combustion with a CHN-Analyser (HERAEUS, Geosciences Dept., University of Bremen). TOC was measured after removal of carbonate with 2N HCl. Overall analytical precision based on internal lab standards was better than 0.1% ( ± 1σ). Carbonate was determined by subtracting organic carbon from total carbon, the latter being measured by combustion without pre-treatment with 2N HCl. Organic matter was estimated by multiplying the content of total organic carbon by a factor of two as about 50–60% of marine organic matter is constituted by organic carbon (Hedges et al., 2002). BSi was determined with a sequential leaching technique with 1M NaOH at 85 °C (Müller and Schneider, 1993). The precision of the overall method based on replicate analyses is mostly between ± 0.2 and ± 0.4%, depending on the composition of the material analyzed. For standard deviations of BSi measurements we refer to Müller and Schneider (1993).

3.1. Moorings and sediment traps Sediment trap moorings were deployed at Site CBeu, located off Mauritania in the Canary EBUE (Fig. 1). Details on the trap depth, trap identification (GeoB No.), samples amount and resolution, and sampling intervals are presented in Table 1. Large-aperture time-series sediment traps of the Kiel and Honjo type with 20–40 cups –depending on ship-time availability– and 0.5 m2 openings, equipped with a honeycomb baffle (Kremling et al., 1996) were used. As the traps were moored in intermediate waters (trap depth = 1256–1296 m; water depth = 2693–2761 m, Table 1), uncertainties with the trapping efficiency due to strong currents (e.g. undersampling, Buesseler et al., 2007) and/or due to the migration and activity of zooplankton migrators (‘swimmer problem’) are assumed to be minimal in this depth range. Prior to each deployment, sampling cups were poisoned with 1 ml of concentrated HgCL2 per 100 ml of filtered seawater. Pure NaCl was used to increase the density in the sampling cups up to 40‰. Upon recovery, samples were stored at 4 °C and wet-split in the Marum sediment trap laboratory (University of Bremen) using a rotating McLANE wet splitter system. Larger swimmers –such as crustaceans– were handpicked with forceps and removed by carefully filtering through a 1-mm sieve. All flux data hereafter refer to the size fraction of < 1 mm. A total of 195 samples were processed for chemical analyses. Component fluxes for individual cups over the entire collection period are deposited at www.pangaea.de.

3.3. Diatom studies For this study 1/25 and 1/125 splits of the original samples were used. Samples were rinsed with distilled water and prepared for diatom studies following the method proposed by Schrader and Gersonde (1978). A total of 185 sediment trap samples were processed for diatom analysis. Each split was treated with potassium permanganate, hydrogen peroxide and concentrated hydrochloric acid following previously used methodology (Romero et al., 2002, 2009a,b, 2016). Qualitative and quantitative analyses were done at x1000 magnifications using a Zeiss®Axioscop with phase-contrast illumination (Marum, University of Bremen). Counts were carried out on permanent slides of acid cleaned material (Mountex® mounting medium). Depending on valve abundances in each sample, several traverses across each slide were examined. Total amount of counted valves per slide ranged between 300 and 800. At least two cover slips per sample were scanned in this way. Diatom counts of replicate slides indicate that the analytical error of concentration estimates is ≤10%. The counting procedure and definition of counting units for diatoms follows Schrader and Gersonde

3.2. Determination of the flux of total organic carbon and biogenic silica Analysis of the fraction < 1 mm, using 1/4 or 1/5 wet splits, was performed according to Fischer and Wefer (1991). Samples were freezedried and homogenized before being analyzed for bulk (total mass), organic carbon, carbonate and biogenic opal (BSi, opal). Total mass and carbonate fluxes will be published elsewhere (Fischer et al., in prep.).

Table 1 Deployment data of traps at Site CBeu (Cape Blanc eutrophic, off Mauritania): coordinates, GeoB location and cruise, trap depth, sample amount, capture duration of each sample and sampling interval. Note that ten samples corresponding to CBeu 5 were not available for diatom analyses. This gap extends between 12/13/2007 and 03/17/2008. Mooring CBeu

Coordinates

GeoB-#/cruise

Trap depth (m)

Sample amount

Capture duration (sample/days)

Sampling interval

1

20°45′N 18°42′W

1296

20

1 = 10.5, 2–20 = 15.5

06/05/2003–04/05/2004

2

20°45′N 18°42′W

1296

20

1–20 = 22, 2–19 = 23

04/18/2004–07/20/2005

3

20°45.5′N 18°41.9′W

1277

20

21.5

07/25/05–09/28/2006

4

20°45.7′N 18°42.4′W

1256

20

1 = 3.5, 2–20 = 7.5

10/28/2006–03/23/2007

5

20°44.9′N 18°42.7′W

1263

38

1, 2 = 6.5, 3–38 = 9.5

03/28/2007–03/17/2008

6

20°45.1′N 18°41.9′W

1263

40

1, 2 = 3.5, 2–40 = 8.5

04/26/2008–03/22/2009

7

20°44.6′N 18°42.7′W

– POS 310 9630-2 M 65-2 11404-3 POS 344-1 11835-2 MSM 04b 12910-2 POS 365-2 13612-12 MSM 1114202-4 POS 396

1364

37

9

04/01/2009–02/28/2010

34



Fig. 2. Total fluxes at the trap site CBeu between June 2003 and February 2010: (a) total diatoms (valves m−2 d−1, red bars) (b) biogenic silica (mg m−2 d−1, blue bars) and (c) total organic carbon (mg m−2 d−1, dark gray bars). The light gray shadings highlight maxima of total diatom fluxes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

average = 41.1 ± 53.5 mg m−2 d−1, and TOC, 0.9–97.4 mg m−2 d−1, average = 20.5 ± 17.8 mg m−2 d−1; daily fluxes of total mass, calcium carbonate and lithogenics will be published elsewhere, Fischer et al., in prep.). The BSi flux mostly peaked in spring at the CBeu site (Fig. 2b). Four major mass flux maxima events (> 225 mg m−2 d−1) occurred: early spring 2004, early winter 2005, early spring 2007 and mid-spring 2008. The fluxes of TOC closely parallel that of BSi (R2=0.839). Overall spring fluxes BSi and TOC exhibit also the largest standard deviations, which points to significant interannual variability (Fig. 3a and b).

(1978). The resulting counts yielded abundance of individual diatom taxa as well as daily fluxes of valves per m−2 d−1 (DF), calculated according to Sancetta and Calvert (1988), as follows:

DF =

[N] x [A/a] x [V] x [Split] [days] x [D]

where, [N] number of valves in an known area [a], as a fraction of the total area of a petri dish [A] and the dilution volume [V] in ml. This value is multiplied by the sample split [Split], representing the fraction of total material in the trap, and then divided by the number of [days] of sample deployment and the sediment trap collection area [D].

4.2. Diatoms 4.2.1. Total diatom fluxes and species-specific composition of the diatom assemblage The siliceous fraction of the CBeu trap samples was mainly composed of marine diatoms. Silicoflagellates, radiolarians, the dinoflagellate Actiniscus pentasterias and land-derived freshwater diatoms and phytoliths (silica bodies of epidermic grass cells) were present in minor quantities. In term of number of individuals, marine diatoms dominate the BSi fraction throughout: their flux was always one order to four orders of magnitude higher than that of the above-mentioned siliceous organisms encountered (not shown here). The total diatom flux ranged 4.7 × 104–1.2 × 108 valves m−2 d−1 (average = 5.9 × 106 valves ± 1.4 × 107, Fig. 2). Spring was the most favorable season for diatom production at the CBeu site. Two highest diatom maxima events (> 8.0 × 107 valves m−2 d−1) occurred in late winter/early spring 2005 and mid spring 2008, and secondary maxima were recorded mostly in spring. Other minor peaks occurred during season transitions (winter-spring 2007, spring-summer 2006, 2008,

3.4. Calculation of seasonal fluxes Due to logistical reasons, the time resolution of the sediment trap collections differs (Table 1). This limits comparisons between specific intervals and years. Seasonal fluxes were calculated to allow comparison among seasons. Seasons were defined using the dates of opening and closure of the sampling cups closest to the start of the astronomical seasons (March 21st, June 21st, September 21st, December 21st). 4. Results 4.1. Variations of biogenic silica and organic carbon fluxes For the studied interval, main contributors to the mass flux were biogenic components (calcium carbonate, 1.9–922.3 mg m−2 d−1, average = 288.84 ± 242.7 mg m−2 d−1; BSi, 0.5–296.2 mg m−2 d−1, 35



Fig. 3. Seasonal mean of fluxes and the respective standard deviation (1 SD) of fluxes (ac) and relative contribution (%) of diatom groups (d–g) for the entire CBeu trap experiment (June 2003 – February 2010). (a) Biogenic silica (BSi, mg m−2 d−1, blue bars), (b) total organic carbon (TOC, mg m−2 d−1, gray bars), (c) total diatoms (valves m−2 d−1, red bars, (d) benthic (%, light green bars), (e) coastal upwelling (%, dark green bars), (f) coastal planktonic (%, black bars), and (g) open-ocean diatoms (%, orange bars). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2009; fall-winter 2008) (Fig. 2a). The diverse diatom community at the CBeu site was composed by 170 marine and 20 freshwater species. Out of 170 marine species and in order to better understand the temporal variations of species, the 70 most abundant diatom taxa (average relative contribution > 0.75% for the entire studied interval) were distributed in four groups, according to the main ecological conditions they represent: (1) benthic, (2) coastal upwelling, (3) coastal planktonic, and (4) open-ocean diatoms (see Table 2 for the species-composition of the diatom groups): (1) The benthic group is dominated by Delphineis surirella, with very minor contributions by Cocconeis spp., Grammatophora sp. and Tabullaria spp. Delphineis surirella is a benthic marine diatom widespread in the euphotic zone of sandy shores (as part of the episammic community), shelf and uppermost slope waters along temperate to cool seas (Andrews, 1981), forming either short or long chains (Round et al., 1990). (2) The coastal upwelling group is composed by several species of Chaetoceros resting spores (RS). Vegetative cells of numerous Chaetoceros species (mainly section Hyalochaete) rapidly respond to the decay of upwelling intensity and nutrient depletion by forming endogenous resting spores. Their occurrence in high numbers is usually interpreted to represent the end of the seasonal upwelling (Hasle and Syvertsen, 1996). Spores of Chaetoceros are common components of the coastal and hemipelagial upwelling assemblages in EBUEs (Abrantes et al., 2002; Nave et al., 2001; Romero et al., 2002; Romero and Armand, 2010). (3) Coastal planktonic diatoms thrive in neritic, oligo-to-mesotrophic waters with high to moderate DSi levels, becoming more abundant during intervals of weak turbulence and respond mainly to moderate upwelling (Crosta et al., 2012; Romero et al., 2009a,b, 2016; Romero and Armand, 2010). Main components of this group at the CBeu site were well-silicified Thalassionema nitzschioides var. nitzschioides and Thalassiosira oestrupii var. venrickae, with Actinocyclus spp., Cyclotella litoralis and Coscinodiscus radiatus as accompanying taxa. (4) Open-ocean diatoms thrive in pelagic, oligo-to-mesotrophic and warm to temperate surface waters where siliceous productivity is low to moderate due to moderate/low DSi levels and weak mixing (Nave et al., 2001; Romero et al., 2005; Crosta et al., 2012). At site CBeu, this group was dominated by Fragilariopsis doliolus, several species of Pseudo-nitzschia, Nitzschia bicapitata, Nitzschia interruptestriata, Ditylum brightwellii, Roperia tesselata and Planktoniella sol, with lesser contributions by Thalassiosira ferelineata, Rhizosolenia bergonii, Asteromphalus flabellatus, Thalassionema bacillare and Azpeitia tabularis. 4.2.2. Temporal variations of groups of marine diatoms The average relative contribution of each group was as follows: (1) benthic = 42.43 ± 20.98%, (2) coastal upwelling = 18.95 ± 11.74%, (3) coastal planktonic = 12.90 ± 8.64% and (4) openocean = 16.57 ± 12.32% (sum of the averages = 90.9%). The four diatom groups show a clear seasonal pattern (Fig. 4b). Benthic diatoms had higher relative contributions during spring and summer (Figs. 3d and 4b). Coastal upwelling mainly contributed between late spring and early fall, while open-ocean diatoms tended to be more abundant between fall and early spring. The coastal planktonic group 36



Table 2 (continued)

Table 2 Species composition of the diatom groups at Site CBeu and main ecological conditions.

Group of species Group of species (1)

(2)

(3)

(4)

R. imbricatae R. setigera Roperia tessellata Stellarima stellaris Thalassionema bacillare T. frauenfeldii T. nitzschioides var. capitulata T. nitzschioides var. inflata T. nitzschioides var. parva Thalassiosira eccentrica T. endoseriata T. ferelineata T. lineata T. nanolineata T. oestrupii var. oestrupii T. sacketii var. sacketii T. sacketii var. plana T. subtilis

Benthic Actinoptychus spp. Amphora spp. Cocconeis spp. Cymatosira belgica Delphineis surirella Grammatophora marina Licmophora sp. Odontella mobiliensis Psammodyction panduriformis Tabullaria spp. Coastal upwelling C. affinis C. cinctus C. compresus C. constrictus C. coronatus C. debilis C. diadema C. radicans

Coastal planktonic Actinocyclus curvatulus A. octonarius A. subtilis Chaetoceros concavicornis (vegetative cell, VC) C. lorenzianus (VC) C. pseudobrevis (VC) Coscinosdiscus argus C. decrescens C. radiatus Cyclotella litoralis Skeletonema costatum Thalassionema nitzschioides var. capitulata Thalassiosira angulata T. conferta T. oestrupii var. venrickae T. poro-irregulata Open-ocean Asteromphalus flabellatus A. sarcophagus Azpetia neocrenulata A. nodulifera A. tabularis Detonula pumila Dytilum brightwellii Fragilariopsis doliolus Hemiaulus hauckii Hemidiscus membranaceus Leptocyclindrus mediterraneus Neodelphineis denticula Nitzschia bicapitata N. capuluspalae N. interruptestriata N. sicula Planktoniella sol Pseudo-nitzschia inflata var. capitata P. pungens P. subfraudulenta Pseudosolenia calcar-avis Pseudotriceratium punctatum Rhizosolenia acuminata R. bergonii

Main ecological conditions

Main ecological conditions

Widespread in the euphotic zones of shallow marine and uppermost slope waters, and sandy shores (as part of the episammic community) along the coasts of temperate to cool seas (Round et al., 1990)

Vegetative cells of several Chaetoceros species (section Hyalochaete) rapidly respond to the decay of upwelling intensity and nutrient depletion by forming endogenous resting spores. High numbers of RS indicate intense upwelling (Hasle and Syvertsen, 1996). Spores of Chaetoceros are common components of the coastal and hemipelagial upwelling assemblages in EBUEs (Abrantes et al., 2002; Nave et al., 2001; Romero et al., 2002; Romero and Armand, 2010)

T. symmetrica

was more abundant during fall and winter (Fig. 3f), although its seasonal pattern remains less clear (Fig. 4b). A major shift in the relative contribution of the diatom occurred in spring-summer 2006 (Fig. 4b). The percentage of benthic diatoms strongly increased from 2006 onward, compared to 2003–2005. The spring-summer contribution remained high until the end of the trap experiment in February 2010. In spite of the increased relative contribution of benthic diatoms after 2005, the seasonal pattern of high spring-summer values remained unaltered.

Thrive in neritic, oligo-tomesotrophic waters with high to moderate DSi levels, becoming more abundant during intervals of weak turbulence and respond only to moderate upwelling (Crosta et al., 2012; Romero et al., 2009, 2012, 2016; Romero and Armand, 2010)

4.2.3. Temporal variations of freshwater diatom fluxes The concentration of freshwater diatoms was at least three orders of magnitude lower than total diatom flux and ranged 0–1.6 ∗ 105 valves m−2 d−1 (average = 2.4 ∗ 104 ± 3.1 ∗ 104 valves m−2 d−1). Highest flux of freshwater diatoms occurred mainly between late summer fall and late winter (Fig. 4c). Two maxima of freshwater diatoms were recorded in spring (2006 and 2008). Phytoliths abundance was lower than that of freshwater diatoms (range = 0–3.6 ∗ 104 bodies m−2 d−1, average = 2.3 ∗ 103–0 ± 6.0 ∗ 103 bodies m−2 d−1). Around 20 species corresponding to the following genera were found: Aulacoseira spp., Cyclotella spp., Navicula spp., Nitzschia spp. and Pinnularia spp. Freshwater diatoms thrive in African rivers and lakes south of the Sahara and in dried Holocene lake basins, beyond the marine influence (Romero et al., 2003; Crosta et al., 2012). The presentday occurrence of freshwater diatoms in marine waters off Mauritania is mostly assigned to wind transport (Romero et al., 2003), as no major riverine input occurs (Krastel et al., 2006).

Thrive in pelagic, oligo-tomesotrophic warm to temperate surface oceanic waters where siliceous productivity is low to moderate due to low DSi levels and weak mixing (Romero et al., 2005; Crosta et al., 2012)

5. Discussion 5.1. Seasonal variability of the total diatom flux at the CBeu site A common feature of the seasonal dynamics of the total diatom flux throughout the CBeu trap experiment was its increases predominantly between mid spring and late summer (Fig. 2a). Out of 14 maxima of different magnitude recorded between June 2003 and February 2010, six occurred in spring and four in summer: none of the summer maxima exceeded the highest spring peaks (Table 3). Considering that our time series record extended continuously six and a half years (except for the lack of diatom data in winter 2008), we argue this temporal pattern of variation to be reasonably consistent. In addition to the distinct seasonal distribution of the four diatom groups (Fig. 4; see discussion below in 5.2), the overall good match of the temporal evolution of total 37



Fig. 4. (a) Total diatom flux (valves m−2 d−1; red bars), (b) cumulative relative contribution of diatom groups: benthic (%, light green bars), coastal upwelling (%, dark green bars), coastal planktonic (%, black bars), and open-ocean diatoms (%, orange bars), and (c) freshwater diatom flux (valves m−2 d−1; dessert blue bars) at the trap site CBeu between June 2003 and February 2010. The light gray shadings highlight maxima of total diatom fluxes. Note that freshwater diatoms constitute only a small portion of total diatom fluxes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

diatom, BSi and TOC fluxes (Fig. 2) is interpreted to be the response of the CBeu diatom community to the seasonal cycle of water mass properties over the Mauritanian shelf and upper slope, and the upwelling dynamics (Mittelstaedt, 1991; Van Camp et al., 1991; Meunier et al., 2012; Cropper et al., 2014).

The temporal dynamics of the Mauritanian upwelling is mainly linked to the meridional component of upwelled water masses and wind direction and intensity. Although upwelling occurs almost year-round north of 20°N off the Mauritanian coast (Mittelstaedt, 1983; Lathuilière et al., 2008; Cropper et al., 2014), it becomes most intense between

Table 3 Seasonal occurrence of diatom peaks at Site CBeu provided for each year between June 2003 and February 2010 (see Table 1 for sampling intervals). For each peak, diatom flux (valves m−2 d−1), duration of peak (days) and percentage (%) of the total annual diatom flux are given. Boreal season Winter

Spring

Summer

Fall

a

Parameters Flux value Duration % Flux value Duration % Flux value Duration % Flux value Duration %

2003

2004 ----

Begin of Trap experiment Late spring 1.3 * 107 08/16–09/01 30.34 ----

2.0 * 107 03/20–04/05 28.05 ----

3.5 * 107 03–26/12 48.61

2005

2006 8

1.2 * 10 02/10–03/05 68.50 1.2 * 107 04/20–05/04 6.63 ----

----

----

4.5 * 107 04/09–30 22.27 4.2 * 107 06/12–07/04 20.91 ----

7

1.5 * 10 08–15/03 25.96 1.1 * 107 05/18–27 50.02 ----

----

Note that the 2008 (calendar year) sampling is not complete due to a temporal malfunction of the sediment trap.

38

2008a

2007

No data

9.1 * 107 04/29–05/3 33.34 2.4 * 107 06/23–07/01 8.94 ----

2009

2010 7

1.1 * 10 12/18–01/04 7.90 2.5 * 107 04/28–05/07 17.76 1.8 * 107 21–30/06 25.28 ----

----

End of trap Experiment Early spring No data

No data



Fig. 5. Seasonal means of (a) biogenic silica (BSi, wt%, blue bars), (b) total diatom flux (valves m−2 d−1, red bars), and (c) the benthic:pelagic diatoms ratio (light green bars) recorded at the trap site CBeu between summer 2003 and winter 2010. The y-axis scale for each season is identical. Note that the benthic:pelagic diatoms ratio clearly changed after 2005. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

diatom flux was intercepted in intervals shorter than three weeks (Table 3; see also Table 1 for the temporal resolution of the trap sampling). As evidenced by the deepening of mixed layer depth (indicative of reduced stratification) and the subsequent periods of high nitrate values in spring and summer (Fig. 6b and c), diatoms quickly responded to rapid nutrient injections by reproducing fast during these shortlasting events. However, some mismatches in the absolute magnitude of total diatom and BSi fluxes are seen (Fig. 2a and b). The two highest diatom maxima (early winter 2005 and spring 2008) were between 1 and 0.7 order of magnitude higher than secondary diatom peaks (> 4.5 ∗ 107 valves m−2 d−1), while differences in the magnitude of corresponding BSi peaks were smaller. This mismatch has several possible explanations. A main reason might be the method of quantification of the diatom flux (census of valves of different sizes, flux expressed as valves m−2 d−1), while fluxes of BSi and TOC are quantified as weight (mg m−2 d−1). Additional factors explaining this mismatch might be the degree of valve silicification (mainly affecting the BSi content, Romero and Armand, 2010), the varying size of valves (affects both BSi and TOC fluxes, Hamm et al., 2003), and the species-specific contribution to the diatom flux (Lončarić et al., 2007; Romero et al., 2002, 2009a,b, 2016). Highest fluxes of freshwater diatoms –indicative of stronger eolian input off Mauritania (Romero et al., 2003)– were mainly recorded in fall and winter (Fig. 4c). Since northeastern TW are more intense in fall and winter, we interpret the dominant seasonal pattern of freshwater diatom flux to reflect the transport within TW layers (ca. 1.5 km high), with minor influence of SAL wind trajectories (predominantly involved in late spring-summer) (Chiapello et al., 1995; Nicholson, 2013). This eolian-induced temporal pattern of land-derived microorganisms was also reflected by the flux of lithogenics particles at site CBeu (Fischer et al., in prep.) and reflects the dry seasonal deposition.

spring and early fall, following the increased mixing of the uppermost water column (Mittelstaedt, 1983; Santos et al., 2008; Arístegui et al., 2009; Marcello et al., 2011). The overall pattern of seasonal average fluxes at site CBeu supports our interpretation of the good match between the temporal occurrence of upwelling and the main interval of diatom production and flux. The highest seasonal average values of BSi and TOC were recorded in spring (79.3 and 30.3 mg m−2 d−1, respectively) and were between 1.5 and 2.6 times higher than those recorded for other seasons (Fig. 3a and b). Similarly, the spring average of the total diatom flux was between 1.5 and 4 times higher than that recorded in other seasons (Fig. 3c). The standard deviation of both BSi and total diatom fluxes was, however, the second largest in spring (after winter), which suggests a wide range of interannual variability in the magnitude of diatom production during this season. In spite of differences in the absolute magnitude between total diatoms and BSi fluxes, the overall seasonal averages show a very similar temporal trend: BSi fluxes were low whenever diatom flux decreased (Fig. 5a and b). A further feature evident from Fig. 5 is the strong interannual variability of the calculated seasonal values and the absence of a clear trend –either decreasing or increasing– during the entire record. Although the three highest total BSi fluxes (2004, 2006 and 2008) were recorded in spring (main upwelling season), seasonal values as high of those of spring were also recorded in winter 2005 (Fig. 5a, Table 3). This reflects the complex dynamics of the upwelling scenario off Mauritania between 20° and 22°N with pronounced yearto-year variations of (i) hydrographic and atmospheric conditions (Mittelstaedt, 1983), (ii) nutrient availability (Van Camp et al., 1991; Nykjaer and Van Camp, 1994), and (iii) the amplitudes of mixed layer depth variations off Mauritania (Fig. 6b and c; Cropper et al., 2014). Diatom peaks recorded at site CBeu represented short-lived productivity and export events. Between 25 and 70% of the total annual 39



Fig. 6. Comparison of time-series of total diatom flux and modeled data. (a) Total diatom flux (valves m−2 d−1; red bar) at trap site CBeu between June 2003 and February 2010, (b) mixed layer depth (m, gray line), (c) nitrate content (μm L−1, green line), (d) diatom-averaged chlorophyll (mg m−3, red line) and (e) the benthic:pelagic diatom groups ratio (light green bars). Monthly data for b-d were downloaded from https://modis.gsfc.nasa.gov for the area between 19°–18°W and 20°–21°N. The gray-shaded area in (b) represents values above the total average for mixed layer depth (m). The green-shaded area in (c) and (d) represents values above the average. The vertical C-stripped line indicates the sharp shift in values of the modeled data (b-d) and the benthic:pelagic diatom ratio (e) between 2005 and 2006. The light gray shadings highlight maxima of total diatom fluxes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

–compared to that of the benthic group (Fig. 3d, e and 4b)– suggests moderate influence of hemipelagial productivity upon the CBeu site. At the outer trap location CBmeso (Fig. 1), Chaetoceros RS dominated the diatom assemblage during intervals of strong mixing and the largest offshore migration of the giant chlorophyll filament (Romero et al., 2002; Romero and Armand, 2010; Fischer et al., 2016). The third important group is the open-ocean group (Figs. 3f and 4b). Its represents the onshore transport of pelagic, oligo-to-mesotrophic waters from the North Atlantic Subtropical Gyre upon the Mauritanian hemipelagial and suggests the relaxation of upwelling, lower productivity and reduced influence of the giant chlorophyll filament upon site CBeu in fall and winter. Coastal planktonic diatoms paralleled the overall seasonal distribution of open-ocean diatom (Fig. 3f), therefore reflecting the relaxation of upwelling. The seasonal contribution of diatom groups was also well mirrored in the species composition of the peaks. Out of 14 peaks, eight were dominated by benthic diatoms (Fig. 7). Maxima of total diatoms were more diverse before 2006: Shannon-Weaver index values were always higher than 2.1, while they remained always below 1.7 after early 2006 (Fig. 7a; only exception is the 2009 mid spring peak). The highly diverse assemblage during the 2005 late winter peak was mostly composed by coastal planktonic (50.4%) and open-ocean species (32.9%) (Fig. 7c). Main species of both groups possess large (> 20 μm length or diameter) and dissolution-resistant valves (Romero et al., 2005; Lončarić et al., 2007). Due to their low area to volume ratio, it is speculated that large-sized diatoms might achieve positive buoyancy

5.2. Temporal trends in the species-specific composition of the diatom assemblage and productivity dynamics A striking feature of the species-specific composition of the diatom assemblage is the persistent seasonal pattern of the groups’ contribution. The higher relative contribution of the benthic group in spring and summer (Figs. 3–5d, 4b, 5c) matches well the interval of most intense upwelling and the predominant water masses off Mauritania (see 2. The study area). The quick response of benthic diatoms (mainly D. surirella) to high-turbulence conditions was reflected by the good correlation between the increased values of nitrate content in surface waters, the mixing of uppermost water column and the highest values of the benthic:pelagic diatoms ratio in spring and summer (Fig. 6b, c, e). The seasonal dominance of D. surirella might be related to the advantage of its small-sized valves (< 16 μm in our samples) in their efficiency of nutrient uptake compared to larger valves (> 20 μm; Smetacek, 1985; Armbrust, 2009). Delphineis surirella is a neritic diatom (Andrews, 1981; Round et al., 1990; Pokras, 1991) and the occurrence of large populations in the hemipelagic CBeu trap might not only be related to the seasonal occurrence of upwelling over the shelf. Lateral transport from the shallow shelf via surface and subsurface currents might have also played a role (see thorough discussion below in 5.3.). Complementary to the spring-summer dominance of the benthic D. surirella, coastal upwelling diatoms (RS of Chaetoceros) occurred when less intense upwelling prevailed. Although present throughout the year at site CBeu, the overall lower contribution of Chaetoceros spores 40



Fig. 7. (a) Total diatom flux (valves m−2 d−1, red bars), (b) diversity of the diatom assemblage (open circles, upper panel) and (c) relative contribution of main diatom groups (cumulative, %, lower panel) for all diatom maxima recorded at the trap site CBeu between June 2003 and February 2010. Maxima represented are those highlighted with light gray bars in Figs. 2 and 4. Horizontal stripped lines in the upper panel highlight periods of high (before mid spring 2006) or low (after mid spring 2005) diatom diversity. The vertical L-stripped arrow indicates an abrupt shift in the relative composition of the diatom assemblage between 2005 and 2006. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

in spring-summer 2006 (Figs. 4b, 6c, 7e) might have been possibly determined by the intensification of lateral advection upon the intermediate-waters deployed CBeu trap. Observational and model experiments show that the transport from the shelf and the upper slope via nepheloid layers contributes more to the deposition of particulates on the lower slope and beyond than the direct vertical settling of particles from the surface layer (Fischer et al., 2009). Based on the vigorous mixing due to the confluence of different water masses and strong winds off Mauritania (Fig. 1a; Mittelstaedt, 1983, 1991; Zenk et al., 1991), Karakaş et al. (2006) concluded that the offshore transport from shallow into deeper waters is most intense between 20.5°N and 23.5°N along the northwestern African margin. Nowald et al. (2006) observed a good correlation between modeled distributions of particles eroded from the Mauritanian shelf and particle clouds recorded with offshore in-situ camera observations. Previous trap experiments off Cape Blanc already discussed the possible role of nepheloid layers on the lateral transport and the temporal dynamics of particle fluxes observed at the outer CBmeso site. Fischer et al. (2009) estimated that the lateral contribution to the TOC collected with the deep CBmeso trap (deployment CB-13) was ca. 15% on the annual basis and reached up to 65% in spring.

and even migrate vertically in weakly mixed waters (Smetacek, 1985; Kemp et al., 2000), allowing the access to nutrient pools stored below the mixed layer (Armbrust, 2009). In spite of the stronger mixing and the nutrient depletion in fall and winter (Fig. 6b and c; Mittelstaedt, 1983; Cropper et al., 2014), the combination of the above-mentioned conditions might have contributed to the dominance of open-ocean and coastal planktonic species during the 2005 late winter peak. On their 24-year record of deep ocean mass fluxes at the outer trap site CBmeso (Fig. 1), Fischer et al. (2016) observed an extraordinary increase of BSi flux in winter 2005. In general, higher (lower) BSi fluxes are associated with a larger (smaller) filament area. However, a smaller filament area and lower chlorophyll standing stock were recorded in the Mauritanian EBUE during the 2005 winter than in previous and later winters (Fischer et al., 2016). This points to additional regulators for the export of BSi to the deep sea, such as the availability of ballast minerals (Fischer et al., 2016), microorganisms’ distribution and nutrient (Si) supply into surface waters. 5.3. Intensification of lateral transport off Mauritania? The remarkable increase in the contribution of the benthic diatoms 41



(Fig. 6d) increased one- to threefold, amplitudes of the seasonallyvarying mixed layer depth became larger and nitrate availability increased between 0.5 and 1 time (Fig. 6b and c). Altogether, these data point to increased surface productivity in Mauritanian coastal waters. Fluxes of total diatom, BSI and TOC at site CBeu, however, did not increase after early 2006 (Figs. 2 and 6a, b). Although the mismatch between absolute magnitudes of CBeu diatom fluxes and modeled data might be partly determined by differences in temporal resolution (trap sampling resolution between 3.5 and 23 days, Table 1, versus monthly averages of modeled parameters), the increased values of benthic:pelagic ratio after early 2006 supports the scenario of a significant shift in the composition of the diatom assemblage (Figs. 5c and 6e). Hydrographic changes that induce shifts in the community composition can occur relatively rapidly (within a year), yet the biological response to these changes can have different spatial and temporal patterns depending upon the species involved (Andersen et al., 2008). Most shifts within phytoplankton populations are inferred from abrupt changes over time; however, time itself is never the actual underlying driver and the identification of a change-point in time is therefore the natural first step toward identifying a potential driver (Andersen et al., 2008). At this stage, it cannot be fully disregarded that the shift in the community composition at the CBeu site might be also due to the natural long-term variability of the diatom populations due to external forcings such as the North Atlantic Oscillation. Recently recovered sediment traps at the CBeu site (2015–2016) confirm the dominance of D. surirella after 2010 and as well its increased contribution to the community at the outer site CBmeso (Romero, unpublished observations). Bottom and intermediate nepheloid layers are important phenomena concerning the lateral, long distance transport of particles from the shelf to slope and deep sea environments. Our long-term trap results from the Mauritanian EBUE emphasize the significance of lateral transport as an important mechanism, effectively transferring carbon from the atmosphere into the deep sea. The applicability of vertical particle flux models is not limited to the Mauritanian EBUE and is comparable to similar ecological and oceanographic settings (e.g., Benguela EBUE). Strong remobilization and lateral export of organic matter from shelf areas toward the slope and the deep sea during periods of rapid sea-level change have been suggested as major factors causing the formation of deep-water organic carbon depocenters in the geological record (Fütterer, 1983; Inthorn et al., 2006; Gallego-Torres et al., 2014). Results from our trap experiment stress the significance of long-term (multiannual) trap experiments on evaluating the impact of changing environmental conditions. The effect of climate change on biodiversity and ecosystems are currently assumed to be smooth, involving a continuous increase in impacts and extinctions as global temperature rises (IPCC, 2007). In this regard, time-series trap experiments conducted over many years –as those currently conducted in the Canary EBUE (Fischer et al., 2016; Romero et al., 2002, 2016)– deliver a broad observational basis on the occurrence of long-lasting variations of productivity parameters in response to key forcings, such as varying nutrient inputs, ecosystem fragmentation (Andersen et al., 2008) or climate change. Our multiannual trap experiment off Mauritania provides an unique opportunity to characterize temporal patterns of variability that can be extrapolated to other EBUEs, which might have not yet experienced regime changes.

The very dynamic activity in Mauritanian coastal waters plays a dominant role in determining the pathways and fluxes of particles across the shelf (Karakaș et al., 2006; Meunier et al., 2012). This activity also serves as a jet for cross-shore particle transfer and it produces sporadic particle clouds moving to the pelagial of the low-latitude NE Atlantic (Fischer and Karakaş, 2009; Nowald et al., 2015). The transport of particles and microorganisms from their source in shallow coastal waters into the hemipelagic realm probably occurs within weeks (Karakaș et al., 2006, 2009). In-situ observations show that erosional processes along the Mauritanian shelf and uppermost slope significantly contribute to the subsurface peaks of particle abundance observed in optical measurements. Nowald et al. (2015) observed that particles produced along the Mauritanian shelf are advected several hundreds kilometers offshore within intermediate and bottom-near nepheloid layers (see also Fischer et al., 2009). In turn, the subsurface layer (200–300 m water depth) might be the place of mixing processes of older, laterally-advected materials from the shelf by giant filament activity, with relatively fresh material derived from the open ocean surface (Fischer et al., 2009). At this stage, however, we are not able to provide additional evidence confirming that the shelf and slope undercurrents, which erode and transport fine-grained particles and microorganisms −such as benthic diatoms− to the hemipelagic CBeu trap, possibly intensified its transport after 2005 due to general circulation changes. Several studies indicate that lateral transport in shelf and slope-depth nepheloid layers contribute more significantly to the deposition of particulate organic matter on the lower slope than the direct vertical settling of particles from the surface layer (Inthorn et al., 2006; McPhee-Shaw et al., 2004). Accordingly, the area of final burial of organic matter can be effectively displaced from the area of production (Inthorn et al., 2006). Carbon depocenters generally occur at the continental slopes between 500 and 2000 m (the depths range of CBeu). The relevance of advective processes within nepheloid layers from the shelf towards slope and deep sea environments has been proposed for similar settings (McCave and Hall, 2002; McPhee-Shaw et al., 2004; Puig and Palanques, 1998; Inthorn et al., 2006). An important bathymetric feature influencing the dynamics of upwelling and lateral transport is the remarkably great number of narrow and deep canyons cutting into the Mauritanian slope off the Banc d’Arguin between Cape Blanc and Cape Timiris (Fig. 1a; Krastel et al., 2006). Within these canyons, the subsurface onshore flow accelerates shoreward. Through this drainage system, fine-grained sediments derived from wind-borne dust clouds and high productivity over the shelf are effectively collected and transported downslope (Krastel et al., 2006), either by semi-permanent suspension flows or by episodic or periodic turbidite activity (Zühlsdorff et al., 2008). Offshore waters penetrate into the Banc d’Arguin at its northwestern border and leave at its southwestern border after a time of residence of two to five weeks depending on the wind force (Mittelstaedt, 1983, 1991). Because of its high density, the shallow bank water (depth < 20 m) submerges beneath the cool upwelling water off the Banc d’Arguin, and it sinks and mixes along the continental slope down about to 400 m depth (Zenk et al., 1991). In addition to the nepheloid layer-mediated transport, this dynamics of water masses related to the existence of the canyon system off Mauritania might have contributed to the enhancement of transport from shallow water upon the trap site CBeu. 5.4. Further implications of long-term flux trends of the diatom production in the Mauritanian EBUE

6. Conclusions

The seasonal dynamics of the CBeu fluxes and modeled data reveal a remarkable match. Recorded maxima of total diatom flues at site CBeu closely followed the seasonal dynamics of hydrographic changes and the input of nutrients into the uppermost water column off Mauritania (Fig. 6). Within our 2003–2010 study period, a significant shift in the magnitude of modeled parameters in the 1 × 1 degree CBeu box occurred after 2005: the averaged area of diatom-produced chlorophyll

The 2003–2010 CBeu flux record provides insights into (i) seasonal and multiyear trends and (ii) the major mechanisms driving diatom production and sedimentation to the deep ocean in a highly productive EBUE. Considering the limited present-day availability of continuous records of diatom fluxes longer than two years (Romero and Armand, 2010; Romero et al., 2016), observations from our multiyear trap experiment are useful for understanding the main processes (drivers) 42



A.R., Brink, K. (Eds.), The Global Coastal Ocean, vol. 11. John Wiley and Sons, New York, pp. 29–67. Bory, A., Jeandel, C., Leblond, N., Vangriesheim, A., Khripounoff, A., Beaufort, L., Rabouille, C., Nicolas, E., Tachikawa, F., Etcheber, H., Buat-Ménard, P., 2001. Downward particle fluxes within different productivity regimes off the Mauritanian upwelling zone (EUMELI programm). Deep-Sea Res. I 48, 2251–2282. Buesseler, K.O., Antia, A.A., Chen, M., Fowler, S.W., Gardner, W.D., Gustafsson, O., Harada, K., Michaels, A.F., Rutgers van der Loeff, M., Sarin, M., Steinberg, D.K., Trull, T., 2007. An assessment of the use of sediment traps for estimating upper ocean particle fluxes. J. Mar. Res. 65, 345–416. Chavez, F.P., Messié, M., 2009. A comparison of eastern boundary upwelling ecosystems. Prog. Oceanogr. 83, 80–96. Chiapello, I., Bergametti, G., Gomes, L., Chatenet, B., Dulac, F., Pimenta, J., Santos Suares, E., 1995. An additional low layer transport of Sahelian and Saharan dust over the North-Eastern Tropical Atlantic. Geophys. Res. Lett. 3, 191–194. Cropper, T.E., Hanna, E., Bigg, G.R., 2014. Spatial and temporal seasonal trends in coastal upwelling off Northwest Africa, 1981–2012. Deep Sea Res. I 86, 94–111. Crosta, X., Romero, O.E., Ther, O., Schneider, R.R., 2012. Climatically-controlled siliceous productivity in the eastern Gulf of Guinea during the last 40,000 yr. Climate of the Past 8, 415–431. Fischer, G., Wefer, G., 1991. Sampling, preparation and analysis of marine particulate matter. In: In: Hurd, D.C., Spencer, D.W. (Eds.), The Analysis and Characterization of Marine Particles, vol. 63. Geophysical Monograph Series, pp. 391–397. Fischer, G., Donner, B., Ratmeyer, V., Davenport, R., Wefer, G., 1996. Distinct year-toyear particle flux variations off Cape Blanc during 1988–1991: Relation to δ18O-deduced sea-surface temperatures and trade winds. J. Mar. Res. 54, 73–98. Fischer, G., Reuter, C., Karakaş, G., Nowald, N., Wefer, G., 2009. Offshore advection of particles within the Cape Blanc filament, Mauritania: Results from observational and modelling studies. Prog. Oceanogr. 83, 322–330. Fischer, G., Karakaş, G., 2009. Sinking rates and ballast composition of particles in the Atlantic Ocean: implications for the organic carbon fluxes to the deep ocean. Biogeosciences 6, 85–102. Fischer, G., Romero, O.E., Merkel, U., Donner, B., Iversen, M., Nowald, N., Ratmeyer, V., Ruhland, G., Klann, M., Wefer, G., 2016. Deep ocean mass fluxes in the coastal upwelling off Mauritania from 1988 to 2012: variability on seasonal to decadal timescales. Biogeosciences 13, 3071–3090. Friese, C.A., van der Does, M., Merkel, U., Iversen, M.H., Fischer, G., Stutt, J.-B., 2016. Environmental factors controlling the seasonal variability in particle size distribution of modern Saharan dust deposited off Cape Blanc. Aeol. Res. 22, 165–179. Fütterer, D.K., 1983. The modern upwelling record off Northwest Africa. In: Thiede, J., Suess, E. (Eds.), Coastal Upwelling, Its Sediment Record, Part B, Sedimentary Records of Ancient Coastal Upwelling, NATO Conference Series IV, Marine Science, pp. 105–121. Gabric, A.J., Garcia, L., Van Camp, L., Nykjaer, L., Eifler, W., Schrimpf, W., 1993. Offshore export of shelf production in the cape Blanc (Mauritania) giant filament as derived from coastal zone color scanner imagery. J. Geophys. Res. 98 (C3), 4697–4712. Gallego-Torres, D., Romero, O.E., Martínez-Ruiz, F., Kim, J.-H., Donner, B., OrtegaHuertas, M., 2014. Rapid bottom-water circulation changes during the last glacial cycle in the coastal low-latitude NE Atlantic. Quatern. Res. 81, 330–338. Hagen, E., 2001. Northwest African upwelling scenario. Oceanol. Acta 24 (Supplement), S113–S128. Hamm, C.E., Merkel, R., Springer, O., Jurkojc, P., Maier, C., Prechtel, K., Smetacek, V., 2003. Architecture and material properties of diatom shells provide effective mechanichal protection. Nature 421, 841–843. Hasle, G.A., Syvertsen, E.E., 1996. Marine diatoms. In: Thomas, C.R. (Ed.), Identifying Marine Diatoms and Dinoflagellates. Academic Press, Inc., San Diego, CA, pp. 1–385. Hedges, J.I., Baldock, J.A., Gélinas, Y., Lee, C., Peterson, M.L., Wakeham, S.G., 2002. The biochemical and elemental compositions of marine plankton: A NMR perspective. Mar. Chem. 78, 47–63. Helmke, P., Romero, O.E., Fischer, G., 2005. Northwest African upwelling and its effect on off-shore organic carbon export to the deep sea. Global Biogeochem. Cycle 19, GB4015. http://dx.doi.org/10.1029/2004GB002265. Inthorn, M., Wagner, T., Scheeder, G., Zabel, M., 2006. Lateral transport controls distribution, quality, and burial of organic matter along continental slopes in highproductivity areas. Geology 34, 205–208 doi: 210.1130/G22153.22151. IPCC, 2007: Climate Change 2007: Synthesis Report. In: Core Writing Team, Pachauri, R. K., Reisinger, A., (Eds.). Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland, pp. 104. Karakaș, G., Nowald, N., Blaas, M., Marchesiello, P., Frickenhaus, S., Schlitzer, R., 2006. High-resolution modeling of sediment erosion and particle transport across the northwest African shelf. J. Geophys. Res. 111, C06025. http://dx.doi.org/10.1029/ 2005JC003296. Karakaș, G., Nowald, N., Schäfer-Neth, C., Iversen, M.H., Barkmann, W., Fischer, G., Marchesiello, P., Schlitzer, R., 2009. Impact of particle aggregation on vertical fluxes of organic matter. Prog. Oceanogr. 83, 331–341. Kemp, A.E.S., Pike, J., Pearce, R.B., Lange, C.B., 2000. The “Fall dump” - a new perspective on the role of a “shade flora” in the annual cycle of diatom production and exportation. Deep-Sea Res. II 47, 2129–2154. Krastel, S., Wynn, R.B., Hanebuth, T.J.J., Henrich, R., Holz, C., Meggers, H., Kuhlmann, H., Georgiopoulou, A., Schulz, H.D., 2006. Mapping of seabed morphology and shallow sediment structure of the Mauritania continental margin, Northwest Africa: some implications for geohazard potential. Norw. J. Geol. 86, 163–176. Kremling, K., Lentz, U., Zeitzschell, B., Schulz-Bull, D.E., Duinker, J.C., 1996. New type of time-series sediment trap for the reliable collection of inorganic and organic trace

behind multi-year shifts in particle and microorganism dynamics in other EBUEs. The study of interannual time-series of diatom flux dynamics provides a data framework that can be utilized to test the ability of coupled biophysical ocean models and may therefore help to improve our understanding of environmental controls on plankton community structure over interannual time scales. Nepheloid layer-mediated transport of particles and microorganisms from the shelf and the upper slope into the deep ocean might act as a major contributor to their offshore transfer and dispersal off Mauritania. A significant amount of valves, BSi and TOC produced in waters overlying the Banc d’Arguin and the Mauritanian shelf is effectively transported from their original production area until their catchment in the CBeu trap in intermediate waters (1000–1500 m water depth) at the outer Mauritanian slope. While a certain portion of valves is produced in hemipelagial surface waters off Mauritania, the nepheloid layer transport seems crucial in establishing the magnitude and temporal pattern of diatom fluxes at the hemipelagial CBeu site. The shift observed in the species-specific composition of diatom populations between 2005 and 2006 points to a certain degree of perturbation of the coastal ecosystem along the Mauritanian coast. At this stage, however, it remains unclear whether the observed shift in the diatom composition is part of a natural multi-year cycle forced by regular climate variability or was due to some perturbation of the coastal ecosystem driven by other external forcings. The applicability of the flux dynamics scenario discussed here is not limited to the Mauritanian upwelling system, but it is comparable to similar EBUEs. We propose that the impact of nepheloid layer-mediated transport of valves, BSi and TOC and its potential contribution to the export of biogenic materials into the deep ocean needs to be more accurately considered in future biogeochemical models. Acknowledgements We are greatly indebted to the masters and crews of several RV Poseidon, RV Meteor and RV MS Merian expeditions (Table 1). Help by Mauritanian and German authorities, and the RV Poseidon headquarters (Geomar, Kiel, Germany) during the planning phases of the research expeditions is greatly appreciated. Special thanks to G. Ruhland, M. Klann, N. Nowald and G. Meinecke (all at MARUM, University of Bremen, Bremen, Germany), who were responsible for the preparation, deployment and recovery of sediment traps and home-laboratory work on collected samples. The comments and criticisms of two anonymous and Editor Dr. Nate Mantua reviewers greatly helped to improve this paper. This work was possible due to the funding by the German Research Foundation (DFG) through the Research Center Ocean Margins and the MARUM Excellence Cluster “The Ocean in the Earth System” (University of Bremen, Bremen, Germany). References Abrantes, F., Meggers, H., Nave, S., Bollman, J., Palma, S., Sprengel, C., Henderiks, J., Spies, A., Salgueiro, E., Moita, T., Neuer, S., 2002. Fluxes of micro-organisms along a productivity gradient in the Canary Islands region (29°N): implications for paleoreconstrucions. Deep-Sea Res. II 49, 3599–3629. Andersen, T., Carstensen, J., Hernández-García, E., Duarte, C.M., 2008. Ecological thresholds and regime shifts: approaches to identification. Trends Ecol. Evol. 24, 49–57. Andrews, G.W., 1981. Revision of the diatom genus Delphineis and morphology of Delphineis surirella (Ehrenberg) G. W. Andrews, n. comb. In: Ross, R. (Ed.), Proceedings of the Sixth Diatom Symposium on Recent and Fossil Diatoms. Otto Koeltz, Koenigstein, pp. 81–92. Arístegui, J., Barton, E.C., Álvarez-Salgado, X.A., Santos, A.M.P., Figueiras, F.G., Kifani, S., Hernández-León, S., Mason, E., Machú, E., Demarcq, H., 2009. Sub-regional ecosystem variability in the canary current upwelling. Prog. Oceanogr. 83, 33–48. Armbrust, E.V., 2009. The life of diatoms in the world's ocean. Nature 459, 185–192. Barton, E.D., Arístegui, J., Tett, P., Cantón, M., García-Braun, J., Hernández-León, S., Nykjaer, L., Almeida, C., Almunia, J., Ballesteros, S., Basterretxea, G., Escánez, J., García-Weill, L., Hernández-Guerra, A., López-Laatzen, F., Molina, P., Montero, M.F., Navarro-Pérez, E., Rodriguez, J.M., van Lenning, K., Vélez, H., Wild, K., 1998. Eastern boundary of the North Atlantic: northwest Africa and Iberia. In: In: Robinson,

43



Romero, O.E., Lange, C.B., Fischer, G., Treppke, U.F., Wefer, G., 1999. Variability in export production documented by downward fluxes and species composition of marine planktonic diatoms: Observations from the tropical and equatorial Atlantic. In: Fischer, G., Wefer, G. (Eds.), Use of Proxies in Paleoceanography, Examples from the South Atlantic. Springer Verlag, Berlin, pp. 365–392. Romero, O.E., Lange, C.B., Wefer, G., 2002. Interannual variability (1988–1991) of siliceous phytoplankton fluxes off northwest Africa. J. Plankton Res. 24, 1035–1046. Romero, O.E., Dupont, L., Wyputta, U., Jahns, S., Wefer, G., 2003. Temporal variability of fluxes of eolian-transported freshwater diatoms, phytoliths, and pollen grains off Cape Blanc as reflection of land-atmosphere-ocean interactions in northwest Africa. J. Geophys. Res. 108 (C5). http://dx.doi.org/10.1029/2000JC000375,002003. Romero, O.E., Armand, L.K., Crosta, X., Pichon, J.-J., 2005. The biogeography of major diatom taxa in Southern Ocean surface sediments: 3. Tropical/Subtropical species. Palaeogeogr. Palaeoclimatol. Palaeoecol. 223, 49–65. Romero, O.E., Thunell, R.C., Astor, Y., Varela, R., 2009a. Seasonal and interannual dynamics in diatom production in the Cariaco Basin, Venezuela. Deep-Sea Res. I 56, 571–581. Romero, O.E., Rixen, T., Herunadi, B., 2009b. Effects of hydrographic and climatic forcing on diatom production and export in the tropical southeastern Indian Ocean. Mar. Ecol. Prog. Ser. 384, 69–82. Romero, O.E., Armand, L.K., 2010. Marine diatoms as indicators of modern changes in oceanographic conditions. In: Smol, J.P., Stoermer, E.F. (Eds.), The Diatoms, Applications for the Environmental and Earth Sciences, second ed. Cambridge University Press, Cambridge, pp. 373–400. Romero, O.E., Fischer, G., Karstensen, J., Cermeño, P., 2016. Eddies as trigger for diatom productivity in the open-ocean Northeast Atlantic. Prog. Oceanogr. 147, 38–48. Round, F.E., Crawford, R.M., Mann, D.G., 1990. The Diatoms. Cambridge University Press, Cambridge, pp. 747. Sancetta, C., Calvert, S.E., 1988. The annual cycle of sedimentation in Saanich Inlet, British Columbia: implications for the interpretation of diatom fossil assemblages. Deep-Sea Res. 35 (1), 71–90. Santos, F., deCastro, M., Gómez-Gesteira, M., Alvarez, I., 2008. Differences in coastal and oceanic SST warming rates along the Canary upwelling ecosystem from 1982 to 2010. Cont. Shelf Res. 46, 1–6. Sarmiento, J.L., Gruber, N., Brzezinski, M.A., Dunne, J.P., 2004. High-latitude controls of thermocline nutrients and low latitude biological productivity. Nature 427, 56–60. Schrader, H.-J., Gersonde, R., 1978. Diatoms and silicoflagellates. In: In: Zachariasse, W.J., Riedel, W.R., Sanfilippo, A., Schmidt, R.R., Brolsma, M.J., Schrader, H., Gersonde, R., Drooger, M.M., Broekman, J.A. (Eds.), Micropaleontological counting methods and techniques - an exercise on an eight meter section of the Lower Pliocene of Capo Rosello, Sicily, vol. 17. Utrecht Micropaleontological Bulletin, Utrecht, pp. 129–176. Skonieczny, C., Bory, A., Bout-Roumazeilles, V., Abouchami, W., Galer, S.J.G., Crosta, X., Diallo, A., Ndiaye, T., 2013. A three-year time series of mineral dust deposits on the West African margin: Sedimentological and geochemical signatures and implications for interpretation of marine paleo-dust records. Earth Planet. Sci. Lett. 364, 145–156. Smetacek, V., 1985. Role of sinking in diatom life-history cycles: ecological, evolutionary and geological significance. Mar. Biol. 84, 239–251. Swap, R., Ulanski, S., Cobbett, M., Garstang, M., 1996. Temporal and spatial characteristics of Saharan dust outbreaks. J. Geophys. Res. 101 (D2), 4205–4220. Van Camp, L., Nykjær, L., Mittelstaedt, E., Schlittenhardt, P., 1991. Upwelling and boundary circulation off Northwest Africa as depicted by infrared and visible satellite observations. Prog. Oceanogr. 26, 357–402. Zenk, W., Klein, B., Schröder, M., 1991. Cape verde frontal zone. Deep-Sea Res. 38 (Suppl. 1), S505–S530. Zühlsdorff, C., Hanebuth, T.J.J., Heinrich, R., 2008. Persistent quasi-periodic turbidite activity off Saharan Africa and its comparability to orbital and climate cyclicities. Geo-Mar. Lett. 28, 87–95.

chemical substances. Rev. Sci. Instrum. 67, 4360–4363. Lange, C.B., Romero, O.E., Wefer, G., Gabric, A.J., 1998. Offshore influence of coastal upwelling off Mauritania, NW Africa, as recorded by diatoms in sediment traps at 2195 m water depth. Deep-Sea Res. I 45, 985–1013. Lathuilière, C., Echevin, V., Levy, M., 2008. Seasonal and intraseasonal surface chlorophyll-a variability along the northwest African coast. J. Geophys. Res. 13 (C5), C05007. http://dx.doi.org/10.1029/2007/JC004433. Lončarić, N., van Iperen, J., Kroon, D., Brummer, G.-J.A., 2007. Seasonal export and sediment preservation of diatomaceous, foraminiferal and organic matter mass fluxes in a trophic gradient across the SE Atlantic. Prog. Oceanogr. 73, 27–59. Marcello, J., Hernandez-Guerra, A., Eugenio, F., Fonte, A., 2011. Seasonal and temporal study of the northwest African upwelling system. Int. J. Remote Sens. 32, 1843–1859. McCave, I.N., Hall, I.R., 2002. Turbidity of waters over the Northwest Iberian continental margin. Prog. Oceanogr. 52, 299–313. McPhee-Shaw, E.E., Sternberg, R.W., Mullenbach, B., Ogston, A.S., 2004. Observations of intermediate nepheloid layers on the northern California continental margin. Cont. Shelf Res. 24, 693–720. Meunier, T., Barton, E.D., Barreiro, B., Torres, R., 2012. Upwelling filaments off Cape Blanc: interaction of the NW African upwelling current and the Cape Verde frontal zone eddy field? J. Geophys. Res. 117 (C8), C08031. http://dx.doi.org/10.1029/ 2012JC007905. Mittelstaedt, E., 1983. The upwelling area off Northwest Africa - a description of phenomena related to coastal upwelling. Prog. Oceanogr. 12, 307–331. Mittelstaedt, E., 1991. The ocean boundary along the northwest African coast: Circulation and oceanographic properties at the sea surface. Prog. Oceanogr. 26, 307–355. Mollenhauer, G., Basse, A., Kim, J.-H., Sinninghe Damsté, J.S., Fischer, G., 2015. A fouryear record of UK37 - and TEX86-derived sea surface temperature estimates from sinking particles in the filamentous upwelling region off Cape Blanc, Mauritania. Deep-Sea Res. I 97, 67–79. Müller, P., Fischer, G., 2001. A 4-year sediment trap record of alkenones from the filamentous region off Cape Blanc, NW Africa, and a comparison with distributions in underlying sediments. Deep-Sea Res. I 48, 1877–1903. Müller, P.J., Schneider, R.R., 1993. An automated leaching method for the determination of opal in sediments and particulate matter. Deep Sea Res. I 40, 425–444. Nave, S., Freitas, P., Abrantes, F., 2001. Coastal upwelling in the Canary Island region: spatial variability reflected by the surface sediment diatom record. Mar. Micropaleontol. 42, 1–23. Nicholson, S.E., 2013. The West African Sahel. A review of recent studies on the rainfall regime and its interannual variability. ISRN Meteorol. 2013, 453521. http://dx.doi. org/10.1155/2013/453521. Nowald, N., Karakaș, G., Ratmeyer, V., Fischer, G., Schlitzer, G., Davenport, R.A., Wefer, G., 2006. Distribution and transport processes of marine particulate matter off Cape Blanc (NW-Africa): results from vertical camera profiles. Ocean Sci. Discuss. 3, 903–938. 2006 www.ocean-sci-discuss.net/3/903/2006/. Nowald, N., Iversen, M.H., Fischer, G., Ratmeyer, V., Wefer, G., 2015. Time series of insitu particle properties and sediment trap fluxes in the coastal upwelling filament off Cape Blanc, Mauritania. Prog. Oceanogr. 137, 1–11. Nykjaer, L., Van Camp, L., 1994. Seasonal and interannual variability of coastal upwelling along northwest Africa and Portugal from 1981 to 1991. J. Geophys. Res. 99, 197–207. Pokras, E.M., 1991. A displaced neritic diatom (Delphineis karstenii) in pelagic sediments of the southeast Atlantic. Mar. Micropaleontol. 17, 311–317. Prospero, J.M., 1990. Mineral-aerosol transport to the North Atlantic and North Pacific: The impact of African and Asian sources. In: Knap, A.H. (Ed.), The Long-Range Atmospheric Transport of Natural and Contaminant Substances. Kluwer Academic Publishers, AA Dordrecht, pp. 59–86. Puig, P., Palanques, A., 1998. Nepheloid structure and hydrographic control on the Barcelona continental margin, northwestern Mediterranean. Mar. Geol. 149, 39–54.

44