You are on page 1of 96

Author's Accepted Manuscript

Influence of the Nazar Canyon, central

Portuguese margin, on late winter coccolithophore assemblages
Catarina Guerreiro, Carolina S, Henko de
Stigter, Anabela Oliveira, Mrio Cacho, Llusa
Cros, Carlos Borges, Luis Quaresma, Ana I.
Santos, Jos-Manuel Fortuo, Aurora Rodrigues



To appear in:

Deep-Sea Research II

Cite this article as: Catarina Guerreiro, Carolina S, Henko de Stigter, Anabela
Oliveira, Mrio Cacho, Llusa Cros, Carlos Borges, Luis Quaresma, Ana I.
Santos, Jos-Manuel Fortuo, Aurora Rodrigues, Influence of the Nazar
Canyon, central Portuguese margin, on late winter coccolithophore
assemblages, Deep-Sea Research II,
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal

Influence of the Nazar Canyon, central Portuguese margin,

on late winter coccolithophore assemblages

Catarina Guerreiroa,b,c*, Carolina Sd, Henko de Stigterb, Anabela Oliveiraa, Mrio Cachoc,e,
Llusa Crosf, Carlos Borgesa, Luis Quaresmaa, Ana I. Santosa, Jos-Manuel Fortuof, Aurora

a Portuguese Hydrographic Institute (IH), Rua das Trinas 49, 1249-093 Lisbon, Portugal
b Royal Netherlands Institute for Sea Research (NIOZ), Marine Geology Dep., Texel, The
c Geology Centre of the University of Lisbon, 1749-016 Lisbon, Portugal
d Oceanography Centre Fac. Sciences of the University of Lisbon, 1749-016 Lisbon, Portugal
e Department of Geology, Fac. Sciences, University of Lisbon, 1749-016 Lisbon, Portugal
f Institut de Ciencies del Mar (CSIC), Passeig Martim de la Barceloneta, 37-49. E-08003
Barcelona, Spain
* Corresponding author Fax: +351 217500119, E-mail:

Key words: Living coccolithophores; Chl-a; ENACWst; Submarine canyon


This paper presents a first attempt to characterize coccolithophore assemblages occurring

in the context of an active submarine canyon. Coccolithophores from the upper-middle sections


of the Nazar Canyon (central Portuguese margin) - one of the largest canyons of the European
continental margin - were investigated during a late winter period (9 12 March 2010). Species
distributions were analyzed in a multiparameter environmental context (temperature, salinity,
turbidity, Chl-a and nutrient concentrations). Monthly averaged surface water Chl-a
concentrations between 2006 and 2011 assessed from satellite data are also presented, as a
framework for interpreting spatial and temporal distribution of phytoplanktonin the Nazar
Canyon. The Nazar Canyon was observed to act as a conduit for advection of relatively
nutrient-poor oceanic waters of ENACWst origin into nearshore areas of the continental shelf
(less than 10 km off the coast), whilst at the surface a nutrient-rich buoyant plume resulting from
intensive coastal runoff prior and during the beginning of the cruise was spreading in oceanward
direction. Two distinct coccolithophore assemblages appear representative for the coast to openocean gradient: (1) Emiliania huxleyi together with Gephyrocapsa ericsonii and Coronosphaera
mediterranea dominated the more productive assemblage present within coastal-neritic surface
waters; and (2) Syracosphaera spp. and Ophiaster spp. displayed a higher affinity with openocean conditions, and also generally a broader vertical distribution. Local hotspots of
coccolithophore and phytoplankton biomass potentially associated with perturbations of surface
water circulation by the canyon are discussed.

1. Introduction

Submarine canyons incising the continental margins are prominent topographic features that
modify the coastal circulation. By intensifying shelf-slope exchange of water and
organic/inorganic matter they play a key-role in global biogeochemical cycling (e.g. Durrieu de
Madron, 1994; Gardner, 1989; Hickey et al., 1986; Monaco et al., 1999; Puig et al., 2003).

Narrow canyons tend to have a stronger effect on low-frequency circulation, whereas wider
canyons mainly cause bottom flow adjustment along isobaths (Klinck, 1988). Stratification of the
water column reduces the canyons topographic effect on the coastal flow (Hickey, 1997; She
and Klinck, 2000).
In the upper water layers (above 100 m), the influence of the canyon is only gentle, with
the along-shelf flow turning slightly onshore upstream of the canyon and turning offshore
downstream. Closer to the canyon rims (100-200 m) the along-shelf flow is more strongly
deflected in onshore direction, turning back on the downstream side of the canyon, with
upwelling or downwelling occurring above the rims, depending on the wind direction (She and
Klinck, 2000). In the Northern Hemisphere right-bounded flows (i.e. coast to the right, looking
downstream) induce downwelling- conditions within the canyon, whereas left-bounded flows
favor the occurrence of upwelling (Klinck, 1996; She and Klinck, 2000). Upwelling occurs
mostly at the canyon head and downstream rim and adjacent shelf (Allen, 1996; Klinck, 1996;
Mendes et al., 2011; She and Klinck, 2000). Under downwelling conditions, the canyon acts as a
trap for converging shelf water (Skliris and Djenidi, 2006).
The intensification of both coast to ocean and vertical water transport within submarine
canyons is expected to affect the dynamics of plankton ecosystems in the vicinity of canyons
(see Bosley et al., 2004; Hickey, 1995; Kampf, 2006; Ryan et al., 2005; 2010; Skliris et al.,
2002; Skliris and Djenidi, 2006). Indeed a strong response of phytoplankton production to
canyon flows, and concentration of marine organisms by physical processes within and around
canyons were reported from several studies (e.g. Bosley et al. 2004; Macquart-Moulin and
Patriti, 1996; Skliris and Djenidi, 2006).
The Nazar Canyon, located at the central Portuguese margin and one of the largest


submarine canyons of Europe, has been relatively intensely explored with regards to its geology,
geomorphology, oceanography and benthic biology (e.g. Tyler et al., 2009). Little is known,
however, about the plankton communities thriving in this region, and about the canyons effect
on their ecology.
Guerreiro et al. (submitted) observed a relatively higher diversity of coccolith species,
including both oceanic and coastal-neritic taxa but with a relative dominance of the latter, in the
Nazar Canyon in comparison to the adjacent shelf/slope regions. This was interpreted as
reflecting the exchange of water masses between coastal and oceanic regions through the canyon,
as well as the dynamic and nutrient-rich conditions where the coastal species are better adapted
to survive. Locally enhanced productivity in the surroundings of the canyon may be related to
persistent physical phenomena associated with the canyon such as vertical mixing by solitary
internal waves (Quaresma et al. 2007), and/or upwelling in the canyon head (Guerreiro et al.,
2009). Evidence for local enhancement of phytoplankton productivity is also provided by
observations on phytoplankton pigments reported by Mendes et al. (2011), with maximum values
of Chl-a (indicative of phytoplankton in general) near the canyon head and maximum values of
19 hexanoyloxyfucoxantine pigment (indicative of coccolithophores) found in the area north of
the canyon.
Here we report the results obtained from a plankton survey on living coccolithophores from
the upper-middle Nazar Canyon, during late winter (9 12 March 2010) (Figure 1). On the
basis of a detailed characterization of the coccolithophore assemblages together with a general
characterization of environmental conditions prevailing during the sampling period, the impact
of this major submarine canyon on coccolithophores and phytoplankton biomass is discussed.

2. Regional setting

2.1. Oceanography

The central Portuguese continental margin is characterized by a relatively narrow shelf of

a few tens of km width, with a maximum of ~70 km where it projects oceanward in the
Estremadura promontory, but cut back to very close to shore where it is incised by the Nazar
and Lisbon-Setbal Canyons. The Douro and Tagus are the most important rivers debouching on
the shelf, with relatively minor contribution of continental runoff from other rivers. From the
shelf edge located at 160-200 m, a steep upper slope and more gently inclined lower slope
incised by numerous gullies and canyons, lead down to the Iberia and Tagus abyssal plains.
Surface water circulation along the Portuguese margin is directly dependent on two main current
systems transporting water eastwards across the North Atlantic: the North Atlantic Current
extending to the north of the Iberian Peninsula, and the Azores Current south of Iberia (Barton,
2001; Peliz et al., 2005; Pollard and Pu, 1985; Saunders, 1982). As the Azores Current extends
eastwards, branches of this current loop smoothly into the Portugal Current and further south into
the Canary Current. The Portugal Current slowly flows southwards, west of Portugal, carrying
about 2106 m3/s in the upper 200 m of the water column. It partially continues further south into
the Canary Current, while another part apparently enters the Mediterranean within a shallow
surface layer (Barton, 2001; Saunders, 1982).
The upper 500 m of water column off Portugal, including the surface mixed layer and the
first thermocline, is constituted by the Eastern North Atlantic Central Water (ENACW). This
water mass, representing the main source of the nutrient-rich upwelled waters on the Portuguese
coast, shows considerable variation in its hydrological features as it travels along the coast
(Fiza, 1984; McCave and Hall, 2002). The ENACW has two main components of different

origin that converge to this region: a lighter, relatively warm and salty subtropical branch
(ENACWst) formed along the Azores Front, which gradually loses its characteristics as it travels
further northwards along the Iberian margin; a less saline colder water mass of subpolar origin
(ENACWsp) slowly flowing southwards below the poleward subtropical branch, related with the
Subpolar Mode Water formed in the eastern North Atlantic by winter cooling and deep
convection (Fiza et al., 1998).
Beneath the near-surface equatorward flow of the Portugal and Canary currents, the
Iberian Poleward Current (IPC) can be recognized traveling poleward, counter to the general
circulation and closely bound to the continental slope, its core extending about 300-400 m
vertically. This current is mostly restricted to the subsurface layers along most of the eastern
subtropical gyre, but surfaces whenever the Trade Winds weaken or turn northward (Barton,
Circulation along the Portuguese shelf and upper slope is markedly seasonal, associated
to the annual cycle of two major atmospheric systems: the Azores high and Iceland low pressure
system, respectively (e.g. Barton, 2001; Haynes et al., 1993; Relvas et al., 2007). During
summer, the Azores high pressure system migrates towards the central Atlantic, typically
inducing Trade Winds to become northerly, inducing an equatorward circulation over the upper
150-200 m of the water column off Portugal. During winter, when the Azores high pressure
system is located further south and the Iceland low pressure system intensifies, the dominant
wind regime becomes southerly along the western Portuguese margin. This induces shoaling of
the IPC over the upper slope and shelf, where the poleward flow produces an onshore Ekman
transport, in turn resulting in downwelling conditions over the shelf (Fiza, 1983; Vitorino et al.,


River runoff is an important feature of the winter circulation over the western Portuguese
margin, through which a significant discharge of low salinity water occurs into the coastal ocean.
This results in buoyant plumes that either develop into inshore currents (Relvas et al., 2007;
Otero et al., 2008) or expand further offshore, under the influence of, respectively, southerly or
northerly-winds over the shelf and slope (Otero et al., 2008). The Western Iberian Buoyant
Plume, characterized by low salinity (<35.8) and low temperature compared to normal shelf
waters, is mostly fed by outflow from rivers of northen Portugal (Mondego, Douro, Minho,
Lima, Vouga),. Besides these major rivers, other smaller rivers and lagoons contribute as well.
Interannual variability of circulation along the Portuguese shelf and slope is influenced by
the North Atlantic Oscillation (NAO), resulting from fluctuations in the difference of
atmospheric pressure between the Azores high and the Iceland low. NAO high index conditions
typically are associated with an increase of the trade winds that bring moist air into Europe,
resulting in cool summers and mild and wet winters in Europe and its Atlantic forefront. On the
contrary, NAO low index conditions leads to more extreme atmospheric temperatures, producing
heat-waves and deep freezing, and an increase of storm activity and rainfall in southern Europe
and North Africa.
Several studies have indicated a decreasing intensity and increasing frequency in
upwelling events, occurring even during the winter period (e.g. Alvarez et al., 2009; Barton,
2001; Ribeiro et al., 2005; Santos et al., 2004; Silva et al., 2008; Vitorino et al., 2002),
apparently linked with a trend towards the high index mode of the NAO observed over the last
decades, leading also to mild, wet winters over northern Europe and dry conditions over Portugal
(Barton, 2001; Wallace, 2002).
In addition to thermohaline and wind-driven circulation, tidal currents are also important in


influencing the hydrodynamics of the Portuguese margin. Particularly where the M2 semidiurnal tidal current, the dominant tidal constituent over the Portuguese margin, forces stratified
upper ocean water over the abrupt topography of the slope and shelf-break (e.g. Quaresma and
Pichon, 2011), tidal energy is transferred into baroclinic motions in the form of internal waves
and internal tides. These are very important in mixing the ocean water column, enhancing
vertical nutrient transport and thus phytoplankton productivity (e.g. Guerreiro et al., 2009), as
well as in increasing bottom turbulence over the continental shelf and slope, triggering bottom
sediment resuspension and transport (Huthnance et al., 2002). Hotspots of internal tide
generation on the Portuguese margin appear associated with submarine canyons cutting across
the shelf and slope (e.g. Portimo canyon, Bruno et al., 2006; Nazar Canyon, Quaresma et al.,
2007; Quaresma and Pichon, 2011) and with promontories of the continental shelf (e.g.
Estremadura spur, Quaresma and Pichon, 2011).

2.1. Nazar Canyon

The Nazar Canyon, the largest submarine canyon of the Portuguese margin, cuts completely
across the shelf and slope, from less than 1 km from the coastline off the village of Nazar at a
water depth of about 50 m to a distance of >210 km from the coast and a water depth of 5000 m.
An upper, middle and lower section can be distinguished on the basis of general morphology and
characteristics of the hydrodynamic and sedimentary environment (Vanney and Mougenot, 1990;
De Stigter et al., 2007; Lastras et al., 2009). The upper canyon section consists of a narrow and
distinctly V-shaped meandering valley that lies deeply entrenched in the shelf. Beyond the shelf
edge, it passes into the much broader and U-shaped middle section incised in the continental
slope. The lower canyon section consists of a broad and flat-floored valley at the base of the

slope, opening at 5000 m water depth into the Iberia Abyssal Plain.
The physical oceanography of the Nazar Canyon has been summarised by Tyler et al.
(2009), largely on the basis of CTD and current meter data collected by the Portuguese
Hydrographic Institute and Royal NIOZ (final reports of the EUROSTRATAFORM, HERMES
and HERMIONE European projects).
Inside the Nazar Canyon residual currents are generally aligned along the canyon axis as the
result of strong topographical control. The current alignment extends well above the canyon
edges (~150 m depth) implying substantial disturbance of the predominant north-south
circulation parallel to the general trend of the shelf and slope.
At depths shallower than 300 m, the residual currents inside the canyon show a distinct
coupling to the wind-driven current regime over the continental shelf. During winter, the
occurrence of downwelling conditions over the shelf results in a down-canyon residual flow near
or just above the canyon edge. Under strong upwelling conditions and southward flow across the
shelf, onshore (up-canyon) flow is observed in the upper canyon, with intensification of
upwelling near the canyon head. The enhancement of upwelling and associated bottom
resuspension can be expected to provide a nutrient source supporting enhanced phytoplankton
concentration south of the canyon (e.g. Hickey, 1995; Kampf, 2006). This seems to be confirmed
by observations by Mendes et al. (2011) regarding phytopigment distribution patterns in surficial
waters around the Nazar Canyon, with maximum concentrations of diatoms occurring south of
the canyon.
The interaction of the external (barotropic) tide off the Portuguese coast with the canyon
topography, in the presence of water stratification, leads to the generation of internal (baroclinic)
tides (i.e. internal waves of tidal period), which radiate from the generation point and propagate


the tidal energy vertically (Quaresma et al., 2007; Tyler et al., 2009). Strong semi-diurnal bottom
currents occur in all parts of the canyon, particularly in its upper and middle sections (commonly
exceeding 30 cm/s) which, along with the ample supply of fine-grained sediments from the shelf,
result in the permanent haze of suspended matter in the upper canyon (De Stigter et al., 2007).
The generation of non-linear internal waves (NIW) at the canyons northern shelf break and
their refraction towards NE was observed by Quaresma et al. (2007) mainly during summer,
when stratification of the shelf waters supports the waves. Observations indicate that the NIW
most likely result from the interaction of the semidiurnal M2 barotropic tide with the canyon rim,
displaying horizontal and vertical velocities strong enough to resuspend bottom sediments along
the wave propagation path from the middle to the inner shelf. The injection of nutrients from the
lower toward the upper levels of the water column forced by the shoreward propagation of these
NIW has been invoked to explain high concentrations of coccoliths found in the sedimentary
cover in this near shore position (Guerreiro et al., 2009). Although this mechanism occurs mainly
during spring and summer, it seems persistent enough to explain such anomaly in coccolith
In autumn and winter, violent westerly storms generating waves with significant height up to
9 m cause widespread sediment resuspension on the shelf and downwelling of turbid waters
towards the canyon. The location of the canyon head at less than 1 km from the shore makes it
particularly prone to trap particulate matter transported as bedload and in suspension along the
shelf (De Stigter et al., 2007; Oliveira et al., 2007).

3. Material and methods

3.1. Sample collection


Sampling was conducted between 9th and 12th of March 2010, on board of NRP Almirante
Gago Coutinho during the 2nd HERMIONE (Hotspot Ecosystem Research and Mans Impact
On European Seas) scientific cruise of the Portuguese Hydrographic Institute. Coccolithophore
communities were investigated in 97 water column samples collected at discrete water depth
levels between 5 and 110 m depth from 25 CTD (conductivity, temperature, depth) casts in and
around the Nazar Canyon (Figure 1, Table 1).
Physical oceanographic, biological and chemical data (i.e. temperature, salinity, turbidity,
fluorometry and nutrients) and water column samples were collected using a combined Neil
Brown MKIIIC CTD profiler equipped with an Aquatracka nephelometer, a Seapoint
fluorometer and a rosette sampler (12 Niskin bottles of 8 litres). A total of 192 suspended matter
samples were collected from surface, intermediate and bottom nepheloid layers in order to define
the particulate matter concentration (PMC) and to calibrate the nephelometer response
(turbidity). The PMC (g/m3) was compared to a laboratory calibration of the instrument with a
standard formazine solution (FTU). The turbidity calibration for March 2010, was FTU =
0.112*PMC with r = 0.88.

3.2. Meteorological and hydrological data

Hydrographic conditions during the cruise as determined from CTD profiles are represented
as contour plots using inverse distance to power gridding in Surfer Version 8 software. A WSWENE oriented transect covering the entire length of the upper-middle Nazar Canyon axis (23
CTD casts) was built to represent density, temperature, salinity and turbidity conditions during
the sampling period (casts indicated in Figure 1; CTD profiles in Figures 2a-d). For a more

detailed description of the data referring to wind, sea wave, and river discharge and sky
conditions, the reader is referred to Guerreiro et al. (2013).

3.3. Satellite data

Monthly averaged surface water chlorophyll a (Chl-a) concentration between 2006 and 2011
was assessed from satellite data as a framework for interpreting spatial and temporal distribution
of phytoplankton in the Nazar Canyon. Chl-a data acquired by the Moderate-resolution imaging
spectroradiometer (MODIS) on NASAs Aqua satellite and processed by The Ocean Biology










( After quality checking and masking, valid data were

interpolated from a grid of regular latitude-longitude inteval. For each image, with nominal
resolution of 1 km, data corresponding to three defined transects (one transect along the canyon,
two other crossing it) were extracted and averaged per month.

3.4. Laboratory and microscope analysis

3.4.1. Coccolithophores

For the study of coccolithophores, seawater samples of around 2l were filtered over
cellulose acetate filters (47 mm diameter and 0.45 m pore size) using a low pressure vacuum
system. The filters were then rinsed with tap water to remove salt and oven-dried at 40 C for 24
hours. A randomly chosen section (approx. 30 45) of each filter was cut and permanently


mounted on a glass slide. Coccospheres (cells) were identified and counted under polarized light
microscope (PLM) (Olympus BX-40) at 1250 magnification. The scanned area per filter varied
between 0.1 and 3.5 mm2, depending on the general cell density. The number of cells per liter of
seawater was estimated from the number of counted coccospheres multiplied with the ratio of
filled filter area to observed area and divided by the volume of filtered water (Cros, 2001).
For the study of the living assemblages (cells) only the water column between 5 and 110
m water depth was considered. To refine the taxonomic differentiation of Alisphaera spp.,
Algirosphaera robusta, Gephyrocapsa spp., Ophiaster spp., Syracolithus dalmaticus and
Syracosphaera spp., 13 samples were investigated using Scanning Electron Microscope (SEM
Hitachi S-3500N, at 5 kV). Samples were selected for containing relatively higher cell densities
and species diversity. A randomly chosen section of the selected filters was fixed with colloidal
Ag on a SEM stub and sputtered with an Au-Pd coating of maximum 20 nm thick; then, a
minimum number of 100 vision fields (VF) were observed and counted using magnifications
between 1000 (observation area of each VF: 126.52 94.84 m) and 2000 (observation area
of each VF: 63.26 47.42 m).
Identification of coccolithophore species followed Jordan et al. (2004) and Young et al.
(2003), whilst the new website on nannoplankton taxonomy (Young et al.,
2011) and specific literature on light microscopy (Frada et al. 2009), Mediterranean
coccolithophores (Cros and Fortuo 2002) and Syracosphaera genus (Kleijne and Cros, 2009)
provided useful additional guidance for classification.

3.4.2. Phytoplankton pigments (Chl-a) and nutrients


Chl-a concentrations were used as an indicator for phytoplankton biomass. Water samples
of 2l were filtered over Whatman GF/F filters (0.7 m pore size, 25 mm diameter), and the filters
were immediately deep-frozen and stored at 80 C. Phytoplankton pigments were extracted
with 2-3 ml of 95 % cold-buffered methanol (2 % ammonium acetate) and analysed with highperformance liquid chromatography (HPLC). Chromatographic separation was carried out
following Zapata et al. (2000). Chl-a concentrations obtained from 25 HPLC samples were then
used to calibrate fluorometry measurements obtained from CTD casts (r2 = 0.7, with p < 0.01).
Nutrient concentrations (nitrate, nitrite, ammonium, phosphate and silicate) were
determined using a Skalar SANplus Segmented Flow AutoAnalyzer specially developed for the
analysis of saline waters. NNOx and NNO2 were determined according to Strickland and
Parsons (1972), with NNO3 being estimated by the difference between the previous two; N
NH4 and SiSiO2 were determined according to Koroleff (1976); PPO4 was determined
according to Murphy and Riley (1962). All methods were adapted to the methodology of
segmented flow analysis and uncertainties were determined following Mendes et al. (2011).

3.5. Statistical analysis

A statistical multivariate analysis (r-mode Factor Analysis by Statistica 10) was performed
upon the data matrix with coccolithophore concentrations, nutrient concentrations (NOx, PO4,
SiO2), biomass (fluorometry calibrated with Chl-a concentrations measured by HPLC) and
physical parameters (temperature, salinity, turbidity) as columns (variables). Results from the
original data matrices were optimized through Varimax Raw rotation.

4. Results

4.1. Environmental conditions during the cruise

The plankton cruise took place under transient environmental conditions in late winter
2010, as discussed in detail in Guerreiro et al. (2013). Sampling was performed at the end of an
unusually cold winter in Europe (2009 2010) under an exceptionally negative phase of the
North Atlantic Oscillation (NAO) (Cattiaux et al., 2010; Troupin and Machin, 2012). Whilst a
northerly wind regime began to settle around the start of the cruise, the winter mixed layer was
still occupying most of the water column over the shelf and upper slope (uppermost ~150200 m
water depth), as normally the case during winter off Portugal (Oliveira et al., 2004). However,
intense river runoff that occurred prior to and continued during the cruise had produced a wellestablished colder and less saline surface layer extending from near the coast to more than 50 km
offshore, overlying the warmer and saltier winter mixed layer waters (Figures 2a, b and c). The
lowest TS values within this buoyant plume fed by runoff water were measured at the surface,
approximately between 16 and 30 km off the coast (stations 79 and 122, respectively).
The warmer and saltier winter mixed layer associated with the flow of the IPC along the
Portuguese margin was noticeable below the surface buoyant plume in the entire investigated
region, generally below 15-20 m water depth, appearing continuous in north-south direction
close to the shelf-break, at the upper-middle Nazar Canyon transition. Further offshore it was
mostly noticed along the southern flank of the middle canyon but weakening northwards where
significant mixing apparently occurred with colder water masses from north. The TS contrast
between the superficial BP and the winter mixed layer below was particularly pronounced in the
upper Nazar Canyon where the core of the IPC penetrated up-canyon to less than 10 km off the
coast. TS profiles along the canyon axis show evidence of strong vertical oscillation around

Belatina Valley (station 120) possibly driven by internal tides in this part of the canyon
(Quaresma et al., 2007) (Figures 2a, b and c).
Turbidity was generally low, with relatively higher FTU values noticed in the surficial
water layer, as well as at the bottom layer of the upper canyon (i.e. in the canyon head and close
to the intersection with Vitria tributary). Highest turbidity values recorded around 200 300 m
water depth appear to reflect bottom sediment resuspension caused by the canyons internal tide
(Figure 2d).
Highest nutrient concentrations were recorded in the relatively cool and low-saline
surface water of the BP, decreasing to lower concentrations in the winter mixed layer water
underneath. The vertical decreasing trend was particularly noticeable in the case of SiO2 (Figure
3). NOx/PO4 ratio was generally close to the 16:1 Redfield Ratio typical for marine waters
(Redfield et al., 1963). A slight deviation toward lower NOx concentrations relative to PO4 in
most samples suggests that NOx was the major limiting nutrient for phytoplankton growth at that
Phytoplankton biomass inferred from Chl-a concentrations (max. <0.7 g/l) was
generally low during the cruise, with the highest values reached at the uppermost part of the
water column (above 50 m depth) (Figure 4). Highest Chl-a and nutrient concentrations (NOx
and SiO2) and the lowest salinities were measured at the surface (5 m) near Belatina Valley
(stations 118 and 120) and north of the upper canyon (stations 112 and 111).

4.2. Coccolithophores

4.2.1. Species diversity, cell density and distribution


A total of 35 distinct taxa of coccolithophores (coccospheres) were recognized. Nineteen

species and 4 genera were identified using polarizing light microscopy (PLM) whereas additional
Scanning Electron Microscope (SEM) analysis revealed an additional 16 species belonging to the
genera Syracosphaera, Ophiaster, Alisphaera and Acanthoica, and one holococcolithophore,
Syracolithus dalmaticus (see Table 2). The list of observed species is presented in Appendix A.
Coccolithophore cells occurred within the BP and the upper layers of the winter mixed
layer as indicated in Figure 5. Total cell densities ranged between 4.0103 and 6.0105 cells/l
(Table 2). The highest cell densities along a transect covering the upper-middle Nazar Canyon
axis were noticed close to Belatina Valley, associated to minimum TS values within the BP
(stations 118 and 120) (Figure 6). High cell densities were also noticed closer to the coast, at the
canyons head (station 87), less than 2 km off the coast, and above the southern canyon rim
(station 89) (Fig. 7a). Further offshore toward the open ocean (station 132) lower cell densities
were observed, distributed more homogeneously over the water column (Fig. 7b-c).
Of the 35 identified taxa, only ten reached significant cell densities of more than 2000
coccospheres per litre: Emiliania huxleyi, Syracosphaera spp., Gephyrocapsa ericsonii, G.
oceanica, G. muellerae, Coronosphaera mediterranea, Ophiaster spp., Helicosphaera carteri,
Syracolithus dalmaticus and Algirosphaera robusta. E. huxleyi was the dominant species during
the cruise, particularly at the surface close to the shelf-coastal region (Figure 8, Table 2). Below
the surface and further offshore, other species gained in relative importance within the total
assemblage, generally displaying a broader vertical distribution (Figure 9; Guerreiro al., 2013).
G. ericsonii, A. robusta, Acanthoica spp., Syracosphaera pulchra, S. dalmaticus, Coccolithus
pelagicus, Michaelsarsia elegans and, to a lesser extent G. oceanica, displayed a downward


decreasing trend in cell density, similar to that of E. huxleyi. Other groups of species such as
Syracosphaera spp, Ophiaster spp. and Gephyrocapsa muellerae revealed a more uniform
vertical distribution. The remaining taxa did not reveal a specific vertical distribution pattern
(Figure 8).
A coast to open ocean ecological and hydrological dichotomy is well illustrated in Figure
9: E. huxleyi dominated at the surface in coastal waters (stations 87 and 120, Figures 9c,b),
whereas Syracosphaera spp. and Ophiaster spp. were dominant further offshore in more openocean conditions, and showing a broader vertical distribution (station 132, Figure 9a). Closer to
Belatina Valley, the three taxa co-existed, with E. huxleyi largely dominating at the surface, and
the latter taxa relatively increasing in the subsurface water mass (Figure 9b). G. muellerae, G.
oceanica and S. dalmaticus were also more abundant near the coast, whereas G. ericsonii and C.
mediterranea revealed a broader lateral distribution.

4.2.2. Multivariate analysis

Results from factor analysis revealed four distinct factor assemblages explaining 46 % of
the total variance in the data (Table 2, Figure 10). Factor 1 (F1) explains 22 % of total
variability, with NOx, SiO2, Acha and Eh recording the highest (positive) factor loadings, in
opposition to S, T (and Syraco and Ophi) (negative loadings). Factor 2 (F2) explains 10 % of
total variance, being represented by Ge, Biom and Cm (and Eh) (positive loadings). Factor 3 (F3)
explains 8 % of total variability and is represented by Dtub, Alisph (and PO4) (negative
loadings). Factor 4 (F4) explains 7 % of total variability and it is represented by the Go (and
Turb) (positive loadings) in opposition to Meleg (negative loadings).


Samples influenced by factor assemblage NOx, SiO2, Acanthoica spp. and E. huxleyi (F1
positive scores) were better represented at the surface, particularly close to Belatina Valley
(stations 118 and 120) but also around the uppermost reaches of Nazar Canyon (stations 112,
111, 102) and the canyon head (stations 85, 87) (Figure 10a). Below the surface this assemblage
was practically inexistent or weakly represented. Samples influenced by salinity, temperature,
Syracosphaera spp. and Ophiaster spp. (F1 negative scores) were preferentially represented
further offshore and showed a relatively broader lateral distribution, and at Belatina Valley
region below the surface (Figure 10a).
Samples influenced by G. ericsonii, phytoplankton biomass (Chl-a), C. mediterranea
(and to a lesser extent, E. huxleyi) (F2 positive scores) revealed a rather broad lateral distribution,
preferentially at the uppermost 25 m in the Nazar Canyon head and at Belatina Valley, whereas
further offshore a broader vertical distribution is noticed (Figure 10b).
Samples influenced by Discosphaera tubifera, Alisphaera spp. and PO4 (F3 negative
scores) recorded their strongest signal at 50 m water depth, at the Nazar Canyon head (station
87) (Figure 10c).
Samples influenced by G. oceanica and turbidity (F4 positive scores) were consistently
distributed in more neritic-coastal regions, particularly at the canyon head and surroundings
(stations 87, 93), at all water depths. In the intersection between the canyon axis and Vitria
tributary (station 80) and at Belatina Valley, this assemblage was well represented at the
uppermost 25 m (Figure 10d).
The relatively low percentage of variance explained by F1-F4 (< 50%) reflects the highly
transient meteorological and hydrological conditions during the cruise, where water masses (both
oceanic and continental) were still adjusting to the circulation imposed by the shifting wind


regime (see Guerreiro et al., 2013). Nevertheless, factor analysis helped to reveal and understand
the most important ecological signals during the cruise:
(a) coccolithophore cell density and diversity hotspot at the Nazar Canyon head, despite of
abundant detritic material (i.e. terrigenous particles, reworked coccoliths). Significant amounts of
perfectly preserved cells, particularly of E. huxleyi, together with several other species, testify of
the high diversity found in this part of the canyon (see Appendix B). Gephyrocapsa muellerae,
Syracolithus dalmaticus, Acanthoica spp. and Michaelsarsia elegans had their maxima in this
area (Table 2). Additional SEM observations confirmed the relative increase of Calcidiscus
leptoporus, Coccolithus pelagicus and Helicosphaera carteri in the canyon head, together with
the single occurrence of Syracosphaera amoena (formerly S. bannockii, see Dimiza et al., 2008),
Syracosphaera molischii, Palusphaera vandelii and Syracosphaera anthos. G. oceanica was also
systematically better represented in the canyon head at all water depths, associated to turbidity;
(b) Stations close to Belatina Valley seemed to represent a nutrient, Chl-a and coccolithophore
hotspot, with E. huxleyi, G. ericsonii, C. mediterranea (and G. oceanica) dominating the
assemblages at the uppermost 25 m water depth, and Syracosphaera spp. and Ophiaster spp.
dominating underneath; (c) E. huxleyi was clearly displaced towards the neritic/coastal zone, and
G. ericsonii and C. mediterranea more towards the neritic-oceanic zone. Syracosphaera spp. and
Ophiaster spp. were consistently better represented below the surface and further offshore.

4.3. Monthly averaged Chl-a from satellite imagery

Time-series of monthly averaged Chl-a between 2006 and 2011 calculated from satellite
data are shown for three transects: transect A, WSW ENE oriented, covering the whole upper-


middle canyon axis (Figure 11a); transect B, N S oriented, cutting across the canyon axis at
Belatina Valley (station 120) (Figure 11b); and transect C oriented at a low angle to the coastline
and cutting across the canyon head (station 87).
The along-canyon time series (Figure 11a) illustrates the recurrent maximum of Chl-a in
offshore waters occurring around March and April of all years, and a more persistent presence of
high Chl-a concentrations in the coastal zone during spring and summer months. There is no
evidence of persistent or particularly high Chl-a at Belatina Valley, although the transition zone
between Chl-a enriched waters extending from the coast and Chl-a poorer offshore waters is
often located approximately in this region. A map of average Chl-a concentration for March
2010 (Figure 12) illustrates the broad spatial spread of this Chl-a enrichment, occupying a
significant portion of the continental shelf and extending approximately up to the middle shelf
region, apparently coming from north. Slightly higher Chl-a concentrations are noticed along the
canyon axis in comparison to the shelf immediately north of it, particularly at Belatina Valley,
where the highest coccolithophore cell densities and Chl-a were recorded in situ during the
cruise. Similar offshore outbreaks of Chl-a enrichment were also observed in March 2006 and
2009, extending almost to the shelf-break in 2006, and even beyond in 2009 (data not shown).
The Chl-a time series for the transect across the canyon at Belatina Valley (Figure 11b)
shows higher concentrations in the canyon meander and adjacent northern and southern shelf in
March of 2006, 2009 and 2010, reflecting the seasonal offshore spread of Chl-a enrichment. Chla concentrations are consistently higher south of 39.4N where the transect is located in the more
productive near shore area, whereas concentration is much lower along the northern part of the
transect located in the less productive offshore area.
Persistently high Chl-a concentration is observed close to the coast, particularly in spring


and summer months (between March and October), reaching the highest concentrations in
August-October 2007, and June-September 2010. The time series for the NNE-SSW near-shore
transect cutting across the canyon head (Figure 11c) shows maximum Chl-a concentrations in the
canyon head and immediate vicinity in these time-intervals, exceeding concentrations on the
surrounding shelf. The timing of Chl-a peaks in the canyon head, between mid-August and midSeptember 2007; between mid-May and mid-June 2009; in March and between mid-June and
mid-August 2010, is conspicuously different from that of the widespread Chl-a enrichment in
early spring extending across the shelf.

5. Discussion

5.1. Late winter coccolithophore assemblages off Portugal

Moderately low coccolithophore cell densities (between 3.6103 and 6105 cells/l) and
low phytoplankton biomass (Chl-a) (max. <0.7 g/l) were observed during the sampling period,
which is in good agreement with observations by Moita (2001) and Silva et al. (2009) regarding
wintertime phytoplankton production off Portugal. The lack of a clear correlation between Chl-a
and nutrients (Figure 4) suggests that phytoplankton production was generally low and not
limited by nutrient availability within the BP. In comparison, previous studies reporting a much
stronger relationship with nutrient concentration attributed a decisive role to haline-stratification
in promoting phytoplankton blooms during late winter upwelling events off Portugal (Peliz and
Fiza, 1999; Ribeiro et al., 2005; Santos et al., 2004).The subdued phytoplankton growth here
observed is interpreted as resulting from important advective mixing promoted by the BP during


this period of intense runoff, sub-optimal light conditions due to cloud cover and initial relatively
high turbidity within the superficial BP (discussed in Guerreiro et al., 2013).
Four distinct ecological signatures (factors) were extracted from multivariate statistical
analysis applied to the present dataset, explaining 46 % of the variability within the data and
revealing the most important environmental and ecological signals during the cruise (Figure 10,
Table 3).
Factor 1 (positive loadings) is interpreted as representing the gradient between the runoffinfluenced coastal-neritic zone, where relatively high nutrient concentrations were retained in the
superficial BP, and the oceanic mixed water that characterizes the Portuguese margin during
winter, present below the BP closer to the coast and surfacing further offshore. Acanthoica spp.
and Emiliania huxleyi appear positively correlated with nutrients at the surface within more
coastal-neritic conditions. F1 was strongly expressed close to Belatina Valley and around the
Nazar Canyon head and surroundings, but nearly absent below the surface, highlighting the
strong vertical density gradient of the BP and the clear preference of these taxa to develop at the
sunlit nutrient-rich surface water layer (Figure 9b,c and 10a).
The large dominance of E. huxleyi at the less saline sunlit surface layer and its preference
for more coastal/neritic conditions is in good agreement with several authors describing this
species as having a highly cosmopolitan distribution independent of sea surface temperature, and
attaining high cell densities in both oligotrophic and eutrophic environments (Andruleit, 2007;
Baumann et al., 2000; Winter et al., 1994). This species was considered to be a possible indicator
for more stable regions regarding with nutrient availability (Andruleit and Rogalla, 2002), and is
often found associated with nutrient-rich and productive coastal regions (e.g. Andruleit, 2007;
Giraudeau and Bailey, 1995; Sprengel et al., 2002; Silva et al., 2008). From various locations it


has been reported as responsible for major blooms (e.g. Garcia et al., 2011; Knappertsbush and
Brummer, 1995).
E. huxleyi was also weakly positively correlated to Gephyrocapsa ericsonii,
Coronosphaera mediterranea and Chl-a in Factor 2 (positive loadings), particularly in what
revealed to be the most productive station monitored during the cruise, located around Belatina
Valley (Figure 10b). Further offshore where E. huxleyi was not dominating the coccolithophore
community, the two taxa were also important. G. ericsonii was the second most abundant species
during the cruise, next to E. huxleyi, which is in good agreement with several studies indicating
its preference for nutrient-enriched coastal-neritic regions (Giraudeau and Bailey, 1995; Silva et
al., 2009; Winter et al., 1979). C. mediterranea was also significantly present during the cruise,
supporting previous observations reporting high cell densities of this species off the Nazar
region (Moita et al., 2010; Silva et al., 2008), and fast response to nutrient availability in this area
during winter (Guerreiro et al., 2013).
On the contrary, Syracosphaera spp. and Ophiaster spp. (negative loadings of Factor 1)
showed a higher affinity for warmer and saltier open ocean waters, and these species appeared
more broadly distributed along the water column (Figures 9a and 10a). Closer to the coast, these
species were generally less frequent, although higher cell densities were observed below the
surface, where they were apparently able to compete with E. huxleyi. This suggests that the lower
light and nutrient level within neritic subsurface waters were less limiting for these taxa than for
E. huxleyi. Results are consistent with Andruleit (2007) in terms of the broad depth range of
Syracosphaera spp. but not concerning the affinity of this group for nutrient availability in
coastal regions as reported in earlier studies (e.g. Andruleit, 2007; Andruleit and Rogalla, 2002;
Giraudeau and Bailey, 1995). Whereas SEM observations indicated that Syracosphaera


marginoporata was the dominant species within the group (max.1.4105 cells/l), the low level of
taxonomical differentiation of Syracosphaera spp. from the above mentioned studies


explain the discrepancy between its reportedly association with relatively eutrophic conditions
and its preferential distribution in the relatively oligotrophic oceanic waters of the Nazar
Canyon region.
The same applies to Ophiaster spp. of which little is known yet in terms of both
ecological preferences and biogeographic distribution. Whereas our late winter observations
seem to indicate an association of this genus with oceanic-oligotrophic conditions, other studies
describe it as associated to nutrient-rich environments such as subtropical frontal zones and
upwelling areas (e.g. Boeckel and Baumann, 2008; Kleijne, 1993).
Discosphaera tubifera and Alisphaera spp. were not abundant during the cruise (<2000
cells/l), and found weakly correlated to PO4 (negative loadings of Factor 3), very close to the
coast at the Nazar Canyon head (station 87, 50 m water depth) (Figure 10c). In the recent
literature, D. tubifera is typically associated with subtropical gyres (Boeckel and Baumann,
2008) and oligotrophic waters, always outside the upwelling area (e.g. Andruleit et al., 2003;
Kleijne, 1992). Ecological preferences of Alisphaera spp. are still poorly known,whereas Kleijne
(1993) found it associated with G. oceanica in the warm waters of the Indian Ocean and
Sourthern Red Sea, both increasing towards the central upwelling zone. In addition to these taxa,
the sporadic occurrence of other species of subtropical affinity in the upper Nazar Canyon, i.e.
Rhabdosphaera clavigera, Palusphaera vandelii, Umbellosphaera irregularis, Scyphosphaera
apsteinii and Umbilicosphaera hulburtiana, and the local relative increase of species that
exhibited an oceanic affinity during the cruise, i.e. Calcidiscus leptoporus, Coccolithus
pelagicus, Syracosphaera amoena, and Syracosphaera molischii were also noticed. The


relatively oligotrophic signature of coccolithophore assemblages observed in subsurface waters

along the Nazar Canyon axis seems in favour of the hypothesis that the canyon acts as a
preferential pathway for advection of oceanic waters derived from ENACWst from offshore onto
more nearshore areas during winter (see section 5.2).
Gephyrocapsa oceanica appears to be related to turbidity, although the correlation is
somewhat weak (positive loadings of Factor 4). This species was consistently distributed closer
to the coast (< 10 km) at all water depths, particularly at the canyon head and adjacent shelf
(Figure 10d). The broad depth range of G. oceanica was also recognized by Andruleit (2007) and
Houghton and Guptha (1991), as well as its tolerance for lithogenic particles, which would be in
accordance with the occurrence of G. oceanica in the dynamic canyon head area; water samples
collected from this area displayed a highly content in terrigenous particles and reworked
coccoliths; see Appendix B). The coastal preference of G. oceanica is also in good agreement
with Silva et al. (2008) and Guerreiro et al. (2013; submitted), who described this species as a
typical coastal coccolithophore, well adapted to the nutrient-rich and productive environment off
Portugal. The species seems able to quickly respond to nutrient input (Andruleit and Rogalla,
2002; Andruleit et al., 2003; Broerse et al., 2000; Giraudeau and Bailey, 1995; Sprengel et al.,
2002; Winter et al., 1994). The relatively low cell densities of this species confirms the generally
low-productive conditions during the cruise.

5.2. Influence of the submarine canyon and hydrological conditions

Although phytoplankton production apparently had not yet responded to higher nutrient
availability provided by runoff, as revealed by generally low coccolithophore cell and Chl-a


concentrations (see section 5.1, Guerreiro et al., 2013), local abundance and diversity hotspots
were noticed in the upper Nazar Canyon axis close to Belatina Valley (stations 120 and 118)
and in the canyon head (station 87).
Of particular interest for the canyon head dynamics are the sporadic occurrences of
typical subtropical-oligotrophic species, such as Discosphaera tubifera and Palusphaera
vandelii. These species were only observed in this proximal part of the canyon and may be
interpreted as tracers for the preferential onflow of ENACWst through the upper canyon during
winter, as revealed by TS profiles along the Nazar Canyon axis (Figures 2a,b). Shoreward
deflection of circulation in the upper water column is expected to be stronger when the water
column above the shelf and upper slope is relatively unstratified (see Allen, 1996; She and
Klinck, 2000), as typically the case off Portugal during this time of the year (Oliveira et al.,
Along with these subtropical species, a diverse assemblage dominated by the productive
Emiliania huxleyi, Gephyrocapsa ericsonii and Coronosphaera mediterranea was observed in
the canyon head. Maxima of other species, both neritic-coastal (i.e. Gephyrocapsa oceanica,
Acanthoica spp.) and neritic-oceanic (i.e. Gephyrocapsa muellerae, Syracolithus dalmaticus and
Alisphaera spp.) were also observed in this area. Whereas during the low productive winter
season the shoreward advection of oceanic waters through the canyon can be traced by relatively
diverse coccolithophore assemblages with oligotrophic subtropical affinity, satellite data clearly
show a maximum in Chl-a concentration at the canyon head between March and October
suggesting that upwelling of oceanic waters in the canyon head enhances phytoplankton
productivity making the canyon head the most persistently productive part of the inner shelf zone
(Figures 11a,c). Also during the years of lower productivity a relative increase of Chl-a is


noticed at the canyon head. Previous observations from Mendes et al. (2011) had already
indicated that the highest Chl-a concentrations during an upwelling event occurred at the canyon
head (> 4 g/l).
Enhanced Chl-a concentrations observed in satellite imagery south of the Nazar Canyon
and close to Cape Carvoeiro supports previous observations of Mendes et al. (2011) of
persistently high concentrations of diatoms in this area, interpreted as reflecting the occurrence
of intensified upwelling along the southern canyon rim extending its influence over the southern
shelf, and persisting even during relaxation of upwelling-favourable winds. Enhancement of
upwelling in the canyon head and nearby shelf is in accordance with studiesof Bosley et al.
(2004), Hickey (1995) and Skliris and Djenidi (2006). However, the recurrent generation of
upwelling filaments off Cape Carvoeiro during spring-summer should also be considered when
explaining high production of phytoplankton in this area (e.g. Fiza, 1983; Haynes et al., 1993;
Peliz et al., 2002).
The region close to Belatina Valley, where the upper canyon axis makes a tight turn
(stations 120 and 118), also stood out for particular hydrological and ecological characteristics.
The highest phytoplankton biomass and coccolithophore cell densities during the cruise were
observed in this area, with E. huxleyi, G. ericsonii and C. mediterranea dominating the
assemblage in the uppermost 25 m of the water column. The lowest TS values and the highest
nutrient concentrations within the superficial BP were also measured here, whereas indications
of enhanced vertical baroclinic oscillation were noticed underneath the BP (Figures 2a,b),
interpreted as resulting from the interaction of internal waves with the canyon topography. The
conversion of barotropic to baroclinic tidal motion occurs in the presence of water stratification
and leads to the generation of internal (baroclinic) tides (i.e. internal waves of tidal period),


which radiate from the generation point and propagate the tidal energy vertically (Quaresma et
al., 2007; Tyler et al., 2009). The vertical density gradient existing between the BP (above) and
the ENACWst (below) within the confined topography of the upper canyon will likely promote
the baroclinic oscillation of the water masses involved. In addition, the presence of a meander in
this part of the canyon axis appears to block the flow of the internal wave, leading to local
amplification of the vertical oscillation. One could speculate that this represents a typical
hydrological feature of the canyon during wintertime, given that it is during this time of the year
that the IPC usually surfaces and reaches particularly nearshore areas within the canyon. During
summer, when the IPC retreats down to slope water depths, baroclinic activity near the surface is
mainly associated to water column thermo-haline stratification typical of this season (e.g.
Quaresma et al., 2007).
The highest cell and Chl-a concentrations measured in situ close to Belatina Valley may
be interpreted to merely represent a local expression of shelf-wide high phytoplankton
production recurrently occurring during the month of March (Figures 11a and 12b). Although
slightly higher monthly Chl-a concentrations appear to be roughly aligned with the canyon axis
in comparison to the northern shelf, in particular close to the meander (Figure 12b), it is very
hard to decipher whether this reflects a recurrent physical phenomenon related to the canyon or
merely an artefact produced by the satellite acquisition. Given that the regional Chl-a outbreak
observed in satellite imagery consistently occurs in late winter/early spring of most years,
occasionally also in early autumn, but never in full winter or full summer time, it is likely
representing the early spring and autumn phytoplankton bloom, controlled primarily by the
increase in light availability in spring and replenishment of nutrients in autumn (see Figure 12a).
The more intense offshore blooms recorded in March of 2006, 2009 and 2010 may result from


late-winter and early-spring runoff in combination with short-term northerly winds over the
shelf, a condition described by several authors (i.e. Guerreiro et al., 2013; Peliz and Fiza., 1999;
Ribeiro et al., 2005; Santos et al., 2004).
In situ measurements indicating enhanced productivity in the surroundings of Belatina
Valley should, therefore, primarily reveal the presence of a front generated between the BP and
the shelf-slope waters during a hydrologically and meteorologically highly transient period in
this region (Guerreiro et al., 2013). Nevertheless, in view of the particular location, other
phenomena could be invoked to explain the local phytoplankton increase, which may be too
short-lived and localized to be identified within monthly averaged Chl-a distribution maps.
As suggested above, the abrupt seabed topography of the Nazar Canyon is likely to
induce perturbations in the flow of water masses on the shelf. Fronts between relatively
productive coastal water masses and less-productive open ocean water masses will tend to
meander across the canyon, adding complexity to spatial distribution of particulate matter in the
surface water. Since the sampling cruise took place during the winter-spring transition when the
water masses were still adjusting, the canyon topography might be expected to have a noticable
influence on the circulation. In contrast, during typical summer-conditions, wind forcing will
play a more prominent role in determining surface water circulation.
Belatina Valley appears to be a region of significant topographic effect on the front of the
low salinity BP, as indicated by the occurrence of the strongest vertical density gradients in this
area. Quaresma (2012) reported on the existence of a barotropic water mass flux of convergentdivergent periodic motion between the interior of the canyon and the shelf close to Belatina
Valley, driven by the barotropic onshore-offshore water flow. According to this author, the
canyon axis acts as drain for shelf water at this location during every low tide. This water


exchange may result in the concentration of nutrients within the surface water layer, whose timeintegrated effect would result in a local nutrient-enrichment favorable for phytoplankton growth.
Several modeling studies revealed the importance of ocean currents interacting with submarine
canyons, enhancing productivity and influencing phytoplankton distribution by funneling and
trapping plankton within the canyons. These studies highlight the predominant effect of local
primary production on the canyon food web, in comparison to other potential sources (Bosley et
al., 2004; Macquart-Moulin and Patriti, 1996; Skliris and Djenidi, 2006).
In addition, one could speculate that internal tidal pumping driven by intensified vertical
baroclinic oscillation around Belatina Valley could contribute to phytoplankton growth in this
area, similar to what has recently been described from Monterey Canyon (California, USA) by
Ryan et al. ( 2005; 2010). These authors described the upsurge of a wedge-shaped tongue of
cold, dense water from the canyon, flowing up onto the continental shelf. The intruding water
mass was observed to entrain a plume of nutrient-rich turbid water from the seafloor up to the
surface, above which high concentrations of phytoplankton were observed. In the Nazar
Canyon, during a cruise performed in November 2002, a vertical turbid plume was observed at
the Belatina Valley area, extending upward from a level of intense intermediate nepheloid layers
at 800-900 m water depths to about 300 m. This plume was interpreted as reflecting resuspension
by the canyons internal tide, enhanced by the strong density gradient between the ENACW
(above) and the denser Mediterranean Outflow Water (MOW) (below) (Oliveira et al., 2007).
Such baroclinic vertical oscillation, amplified in the canyon meander, may be responsible for
bringing nutrients from below the canyon rim during winter, promoting phytoplankton growth in
the upper part of the water column. However, our water column turbidity profiles and vertical
distribution of coccolithophores and nutrients show no evidence of the occurrence of this process


during the investigated late winter period, where enhanced nutrient concentrations appeared
predominantly associated with the BP (see Guerreiro et al., 2013). Baroclinic activity is more
likely to gain in importance during stratified summer conditions.

5.3. Satellite data versus in situ measurements

Our observations, both long-term Chl-a concentrations obtained from satellite data and in
situ quantification, suggest that the Nazar Canyon may locally favor, at least indirectly, the
development of phytoplankton, including coccolithophores. This is the case for the canyon head,
which appears to be the stage of recurrent higher productivity in comparison to the adjacent
shelf. However, in the case of Belatina Valley, where in situ observations revealed local Chl-a
and coccolithophore cell enhancement, monthly averaged productivity obtained from satellite
suggest that enhanced Chl-a production was not confined to that specific area but occurred over a
much wider area including most of the shelf, (transect B, Figure 12b).
On the one hand, lacking in situ observations from outside the canyon, we cannot
ascertain whether higher Chl-a and cell densities obtained from this area are actually confined to
the canyon axis, or are part of a larger pattern not necessarily related to the canyon. On the other
hand, it cannot be expected that monthly Chl-a averages obtained from satellite data will match
exactly the coccolithophore and Chl-a peaks measured in situ and only representing one instant
of the annual productivity. Different spatial and temporal scales are involved: whereas the
satellite data reveal patterns of phytoplankton distribution at the surface at relatively high
resolution, insight of phytoplankton productivity at deeper levels in the water column can only
be obtained from in situ measurements. Differences between the two will expectedly be largest


under the highly transient meteorological and oceanographic conditions characteristic of the late
winter period, as prevailing during the cruise.
Intensified baroclinic activity at Belatina Valley might well promote biological
production events that are too deep and short-lived to be detected by monthly Chl-a averages
obtained by satellite. The rapid response of certain species of coccolithophores to regional
meteorological and hydrological variations off central Portugal was recently demonstrated by
Guerreiro et al. (2013). Satellite imagery has a tremendous potential to describe larger-scale
phenomena prevailing on the Portuguese margin, but it may not be the best approach to
investigate smaller-scale processes, for which higher temporal and spatial resolution are probably
Validation of hypotheses presented here requires additional sampling surveys integrated
with meteorological and hydrological monitoring in order to address the seasonal and interannual
variability of phytoplankton (in general) and coccolithophores (in particular), in relation with
physical processes in the Nazar Canyon.

6. Conclusions

This study is the first attempt to characterize coccolithophore assemblages occurring in

the context of an active submarine canyon. A late winter low-productive period was investigated
in the Nazar Canyon (off central Portugal) during which warm and saline waters fed by the IPC
were still strongly influencing the hydrology of the shelf and slope, and the winter mixed layer
occupied the entire water column of the shelf-upper slope. The canyon was clearly acting as a
conduit for the onshore advection of relatively nutrient-poor oceanic waters to very nearshore


areas (less than 10 km off the coast).

Runoff prior and during the cruise was an important source of nutrients into the system,
as indicated by high nutrient concentrations that were measured in the relatively low saline
buoyant plume overlying the winter mixed layer in the coastal zone. Nevertheless, the weak
correlation of nutrients with biomass suggests that phytoplankton production had not yet
responded to higher nutrient availability, probably resulting from important advective mixing
promoted by the BP during this period of intense runoff, sub-optimal light conditions due to
cloud cover and initial relatively high suspended sediment load within the surface water layer
(discussed in Guerreiro et al., submitted).
Two main coccolithophore assemblages were distinguished, representing the gradient
between the runoff-influenced coastal-neritic zone and the oceanic mixed water conditions that
characterize the Portuguese margin during winter: (1) Emiliania huxleyi was the dominant taxon
at the surface within more coastal-neritic conditions and, together with Gephyrocapsa ericsonii
and Coronosphaera mediterranea, represent the more productive assemblage during the
sampling period. (2) Syracosphaera spp. and Ophiaster spp. showed a clearly higher affinity for
open-ocean conditions, displaying a generally broader vertical distribution. Closer to the coast,
these taxa were able to compete well with E. huxleyi in the subsurface layer, suggesting that
lower light and nutrient level within more neritic conditions were less limiting for Syracosphaera
spp. and Ophiaster spp. as it was for E. huxleyi.
Chl-a time series obtained from satellite data suggest that the Nazar Canyon head is
often the stage of high productivity between March and October, which makes the canyon head
the most persistently productive part of the upper-middle canyon. In situ observations also
revealed a coccolithophore diversity hotspot in this area, including both oligotrophic-oceanic


and opportunistic-coastal taxa. The single occurrence of typically subtropical-oligotrophic

species (i.e. Discosphaera tubifera, Palusphaera vandelii, Calcidiscus leptoporus) is interpreted
as indicative for the shoreward flow of ENACWst intensified along the upper canyon during
winter. In addition to these species, a diversified assemblage dominated by the productive E.
huxleyi, G. ericsonii and C. mediterranea, together with other species which have their
maximum occurrence in the canyon head area including both neritic/coastal (i.e. Gephyrocapsa
oceanica, Acanthoica spp.) and neritic/oceanic species (i.e. Gephyrocapsa muellerae,
Syracolithus dalmaticus, Alisphaera spp. and Michaelsarsia elegans) may also reflect exchange
of water masses between neritic-coastal and oceanic regions through the canyon during winter.
Local enhancement of nutrient concentration and coccolithophore cell concentration was
observed near the Belatina Valley, with E. huxleyi, G. ericsonii and C. mediterranea dominating
the assemblage at the uppermost 25 m of the water column. In addition, monthly averaged
satellite data reveal slightly higher Chl-a concentrations apparently roughly aligned with the
canyon axis, close to Belatina Valley. We hypothesize that this imprint may be tracing the timeintegrated effect of barotropic water mass flux into Belatina Valley and the meandering of the
low-salinity front into this location. Based on our in situ observations and on recent studies
identifying this narrower part of the canyon axis as an area of intensified vertical water
movement, we suggest that Belatina Valley may potentially be a favourable region for
phytoplankton local enhancement.
Results presented here provide some valuable indications with regards to the important
and persistet influence of the Nazar Canyon on the ecology and distribution of coccolithophores
and phytoplankton biomass at the central Portuguese margin. The results highlight the need of
long-term multi-proxy investigations in order to address the seasonal and interannual variability


in phytoplankton in relation with the seasonal- and/or topographically driven physical

phenomena associated with the Nazar Canyon.


This research was supported by the HERMIONE project (EC contract 226354) funded by
the European Commission and the Cd Tox-CoN project (FCT-PTDC/MAR/102800/2008)
funded by the Portuguese Science Foundation FCT. The first author benefits from an FCT PhD
grant (FRH/BD/41330/2007). Filter samples were collected during the 2nd HERMIONE cruise of
the Portuguese Hydrographic Institute (IH) on board of NRP Almirante Gago Coutinho. The
authors are grateful to all the crew of the NRP Almirante Gago Coutinho and several researchers
participating in the cruise for their valuable help during the collection of samples. A special
thanks to the OC-IH CTD data acquisition team, Joo Vitorino, Manuel Marreiros, Ins Martins,
Vnia Carvalho and Nuno Zacarias, and to Manuela Valena (QP-IH) for performing the
compilation of nutrient data. All the samples were prepared and analyzed at NANOLAB,
Geology Centre of Lisbon University (CEGUL). SEM observations were performed at the
Institut de Cincies del Mar (ICM CSIC, Barcelona, Spain). Constructive criticism and helpful
suggestions from two anonymous reviewers are most gratefully acknowledged by the authors.


Allen, S. E., 1996. Topographically generated, subinertial flows within a finite length canyon.
Journal of Physical Oceanography 26, 1608-1632.


Alvarez, I., Gomez-Gesteira, M., de Castro, M., Lorenzo, M. N., Crespo, A. J. C., Dias, J.M.,
2011. Comparative analysis of upwelling influence between the western and northern coast of
the Iberian. Continental Shelf Research 31(5), 388-399.
Andruleit, H., 2007. Status of the Java upwelling area (Indian Ocean) during the oligothrophic
northern hemisphere winter monsoon season as revealed by coccolithophores. Marine
Micropaleontology 64, 36-51.
Andruleit, H., Rogalla, U., 2002. Coccolithophores in surface sediments of the Arabian Sea in
relation to environmental gradients in surface waters. Marine Geology 186, 505-526.
Andruleit, H., Stager, S., Rogalla, U., Cepek, P., 2003. Living coccolithophores in the northern
Arabian Sea: ecological tolerances and environmental control. Marine Micropaleontology 49,
157- 181.
Barton, E., 2001. Canary and Portugal currents. In: Steele, J., Turekian, K., Thorpe, S. (Eds.), In:









Baumann, K.-H., Andruleit, H. A., Samtleben, C., 2000. Coccolithophores in the Nordic Seas:
comparison of living communities with surface sediment assemblages. Deep-Sea Research II
47, 1743-1772.
Boeckel, B., Baumann, K.-H., 2008. Vertical and lateral variations in coccolithophore
community structure across the subtropical frontal zone in the South Atlantic Ocean. Marine
Micropaleontology 76, 255-273.
Bosley, K. L., Lavelle, J. W., Brodeur, R. D., Wakefield, W. W., Emmett, R. L., Baker, E. T.,
Rehmke, K. M., 2004. Biological and physical processes in and around Astoria submarine
Canyon, Oregon, USA. Journal of Marine Systems 50, 21-37.


Broerse, A. T. C., Ziveri, P., van Hinte, J. E., Honjo, S., 2000. Coccolithophore export
production, species composition and coccolith-CaCO3 fluxes in the NE Atlantic (34N 21W
and 48N 21W). Deep-Sea Research II 47, 1877-1906.
Bruno, M., Vzquez, A., Gmez-Enria, J., Vargas, J. M., Lafuente, J. G., Ruiz-Caavate, A.,
Mariscal, L., Vidal, J., 2006. Observations of internal waves and associated mixing
phenomena in the Portimao Canyon area. Deep Sea Research Part II: Topical Studies in
Oceanography 53 (11-13), 1219-1240.
Cattiaux, J., Masson-Delmotte,V., Vautard, R., Yiou, P., Cassou, C., Codron, F., 2010. Winter
2010 in Europe: A cold extreme in a warming climate. Geophysical Research Letters
Cros, L., 2001. Planktonic coccolithophores of the NW Mediterranean. PhD Thesis, Universitat
de Barcelona, 181pp.
Cros, L., Fortuo, J.-M. 2002. Atlas of Northwestern Mediterranean coccolithophores. Scientia
Marina 66 (1), 1-186.
De Stigter, H., Boer, W., Mendes, P., Jesus, C., Thomsen, L., van den Bergh, G., van Weering,
T., 2007. Recent sediment transport and deposition in the Nazar Canyon, Portuguese
continental margin. Marine Geology 256, 144-164.
Dimiza, M. D., Triantaphyllou, M. V., Dermitzakis, M. D., 2008. Seasonality and ecology of
living coccolithophores in Eastern Mediterranean coastal environments (Andros Island,
Middle Aegean Sea). Micropaleontology 54 (2), 159-175.
Durrieu de Madron, X., 1994. Hydrology and nepheloid structure in the Grand-Rhne canyon.
Continental Shelf Research 14, 457-477.


Fiza, A., 1983. Upwelling patterns off Portugal. In: E. Suess and J. Thie (Eds). Coastal
Upwelling: Its sediment Record. Plenum, New York, 85-98.
Fiza, A. F. G., 1984. Hidrologia e dinmica das guas costeiras de Portugal (Hydrology and
dynamics of the Portuguese coastal waters). PhD. dissertation, Universidade de Lisboa, 294
Fiza, A. F. G., Hamann, M., Ambar, I., Del Ro, G., Gonzlez, N., Cabanas, J., 1998. Water
masses and their circulation off western Iberia during May 1993. Deep Sea Research Part I:
Oceanographic Research Papers 45 (7), 1127-116.
Frada, M., Young, J., Cacho, M., Lino, S., Martins, A., Probert, I., de Vargas, C. 2010. A guide
to extant coccolithophores (Calcihaptophycidae, Haptophyta) using light microscopy. 2010.
Journal of Nannoplankton Research 31 (2), 58-112.
Garcia, C., Garcia, V., Dogliotti, A., Ferreira, A., Romero, S., Mannino, Souza, M. S., Mata, M.
M., 2011. Environmental conditions and bio-optical signature of a coccolithophore bloom in
the Patagonian shelf. Journal of Geophysical Research, doi:10.1029/2010JC006595.
Gardner, W. D., 1989. Baltimore Canyon as a modern conduit of sediment to the deep sea. DeepSea Research 36, 323-358.
Giraudeau, J., Bayley, G. W., 1995. Spatial dynamics of coccolithophore communities during an
upwelling event in the Southern Benguela system. Continental Shelf Research 15, 1825-1852.
Guerreiro, C., de Stigter, H., Cacho, M., Oliveira, A., Rodrigues, A., submitted. Coccoliths
from recent sediments of the Central Portuguese Margin: taphonomical and ecological
inferences. Marine Micropaleontology.


Guerreiro, C., Oliveira, A., Cacho, M., de Stigter, H., S, C., Borges, C., Cros, C., Santos, A.,
Rodrigues, A., 2013. Late winter coccolithophore bloom off central Portugal in response to
river discharge and upwelling. Continental Shelf Research 59, 65-83.
Guerreiro, C., Rosa, F., Oliveira, A., Cacho, M., Fatela, F., Rodrigues, A., 2009. Calcareous
nannoplankton and benthic foraminiferal assemblages from the Nazar Canyon (Portuguese
continental margin): preliminary results. IOP Conference Series: Earth and Environmental
Science 5012004, 11pp.
Haynes, R., Barton, E. D., Pilling, I., 1993. Development, persistence and variability of
upwelling filaments off the Atlantic coast of the Iberian Peninsula. Journal of Geophysical
Research 98, 22681-22692.
Hickey, B. M., 1995. Coastal Submarine Canyons. In: Proceedings Aha Huliko'a Hawaiian
Winter Workshop 1995, Topographic Effects in the Ocean, 95-110.
Hickey, B. M., 1997. The response of a steep-sided, narrow canyon to time-variable wind
forcing. Journal Physical Oceanography 27, 697-726.
Hickey, B.M., Baker, E., Kachel, N., 1986. Suspended particle movement in and around
Quinault Submarine Canyon. Marine Geology 71, 35-83.
Houghton, S. D., Guptha, M. V. S., 1991. Monsoonal and fertility controls on Recent marginal
sea and continental shelf coccolith assemblages from the western Pacific and northern Indian
oceans. Marine Geology 97, 251-259.
Huthnance, J. M., van Aken, H., White, M., Barton, E. D., Le Cann, B., Coelho, E. F., Fanjul, E.
A., Miller, P., Vitorino, J., 2002. Ocean margin exchangewater flux estimates. Journal of
Marine Systems 32, 107-137.


Jordan, R., Cros, L., Young, J., 2004. A revised classification scheme for living haptophytes.
Micropaleontology 50 (1), 5579.
Kampf, J., 2006. Transient wind-driven upwelling in a submarine canyon: A process-oriented
modelling study. Journal of Geophysical Research 111 (C11011) doi:10.1029/2006JC003497
Kleijne, A., 1992. Extant Rhabdosphaeraceae (coccolithophorids, class Prymnesiophyceae) from
the Indian Ocean, Red Sea, Mediterranean Sea and North Atlantic Ocean. Scripta Geologica
100, 1-63.
Kleijne, A., 1993. Morphology, Taxonomy and Distribution of extant coccolithophorids
(calcareous nannoplankton). Ph.D. Thesis, Vrije Universiteit, Amsterdam.
Kleijne, A., Cros, L. 2009. Ten new extant species of the coccolithophore Syracosphaera and a
revised classification scheme for the genus. Micropaleontology 55 (5), 425-462.
Klinck, J. M., 1988. The influence of a narrow transverse canyon on initially geostrophic flow.
Journal of Geophysical Research 93, 509-515.
Klinck, J. M., 1996. Circulation near submarine canyons: a modelling study. Journal of
Geophysical Research 101, 1211-1223.
Knappertsbusch, M., Brummer, G.-J. A., 1995. A sediment trap investigation of sinking
coccolithophores in the North Atlantic. Deep-Sea Research Part I 42 (7), 1083-1109.
Koroleff, F., 1976. Determination of ammonia. In: Grasshoff, K. (Ed.), Methods of Seawater
Analysis. Verlag Chemie, New York, 126158.
Lastras, G., Arzola, R., Masson, D., Wynn, R., Huvenne, V., Huhnerbach, V., Canals, M., 2009.
Geomorphology and sedimentary features in the Central Portuguese submarine canyons,
Western Iberian margin. Geomorphology 103, 310-329.


Macquart-Moulin, C., Patriti, G., 1996. Accumulation of migratory micronekton crustaceans

over the upper slope and submarine canyons of the northwestern Mediterranean. Deep-Sea
Research 43, 579-601.
McCave, I., Hall, I., 2002. Turbidity of waters over the Northwest Iberian continental margin.
Progress in Oceanography 52, 299-313.
Mendes, R., S, C., Vitorino, J., Borges, C., Garcia, V., Brotas, V., 2011. Distribution of
phytoplankton assemblages in the Nazar submarine canyon region (Portugal): HPLCCHEMTAX








Moita, M. T., 2001. Estrutura, Variabilidade e Dinmica do Fitoplncton na Costa de Portugal
Continental. PhD Thesis, Faculdade de Cincias da Univ. de Lisboa, 272 pp.
Moita, M. T., Silva, A., Palma, S., Vilarinho, M. G., 2010. The coccolithophore summerautumn
assemblage in the upwelling waters of Portugal: Patterns of mesoscale distribution (1985
2005). Estuarine, Coastal and Shelf Science 87, 411-419.
Monaco, A., Durrieu de Madron, X., Rodacovitch, O., Huessner, S., Carbone, J., 1999. Origin
and variability of downward biogeochemical fluxes on the Rhne continental margin (NW
Mediterranean). Deep Sea Research I 46, 1483-1511.
Murphy, J., Riley, J. P., 1962. A modified single solution method for the determination of
phosphate in natural waters. Analytica Chimica Acta 27, 3136.
Oliveira, A., Santos, A. I., Rodrigues, A., Vitorino, J., 2007. Sedimentary particle distribution
and dynamics on the Nazar canyon system and adjacent shelf (Portugal). Marine Geology
246, 105-122.


Oliveira, P. B., Peliz, A., Dubert, J., Rosa, R. T., Santos, A. M. P., 2004. Winter geostrophic
currents and eddies in the western Iberia coastal transition zone. Deep-Sea Research I 51, 36381.
Otero, P., Ruiz-Villarreal, M., Peliz, A., 2008. Variability of river plumes off Northwest Iberia in
response to wind events. Journal of Marine Systems 72, 238-255.
Peliz, A., Dubert, J., Santos, A. M. P., Oliveira, P. B., Le Cann, B., 2005. Winter upper ocean
circulation in the Western Iberian Basin Fronts, Eddies and Poleward Flows: an overview.
Deep Sea Research I 52, 621-646.
Peliz, A. J., Fiza, A. F. G., 1999. Temporal and spatial variability of CZCS-derived
phytoplankton pigment concentrations off the western Iberian Peninsula. International Journal
Remote Sensing 20 (7), 1363-1403.
Peliz, A., Rosa, T., Santos, A. M. P., Pissarra, J., 2002. Fronts, jets, and counter-flows in the
Western Iberia Upwelling System. Journal of Marine Systems 35, 6177.
Pollard, R., Pu, S., 1985. Structure and circulation of the upper Atlantic Ocean northeast of the
Azores. Progress in Oceanography 14, 443-462.
Puig, P., Ogston, A. S., Mullenbach, B. L., Nittrouer, C. A., Sternberg, R. W., 2003. Shelf-tocanyon sediment-transport processes on the Eel continental margin (northern California).
Marine Geology 193, 129-149.
Quaresma, L. S., 2012. Super-Inertial Tides Over Irregular Narrow Shelves, Ph.D thesis,
Universit de Bretagne Occidentale, Brest, France, 172 pp.
Quaresma, L. S., Pichon, A., 2011. Modelling the barotropic tide along the West-Iberian margin.
Journal of Marine Systems,


Quaresma, L. S., Vitorino, J., Oliveira, A., Silva, J., 2007. Evidence of sediment resuspension by
nonlinear internal waves on the western Portuguese mid shelf. Marine Geology 246, 123-143.
Relvas, P., Barton, E. D., Dubert, J., Oliveira, P. B., Peliz, A., da Silva, J. C. B., Santos, A. M.
P., 2007. Physical oceanography of the western Iberia ecosystem: Latest views and
challenges. Progress in Oceanography 74, 149-173.
Ribeiro, A., Peliz, A., Santos, M. P., 2005. A study of the response of chlorophyll-a biomass to a
winter upwelling event off Western Iberia using SeaWiFS and in situ data. Journal of Marine
Systems 53, 87-107.
Ryan, J. P., Dierssen, H. M., Kudela, R. M., Scholin, C. A., Johnson, K. S., Sullivan, J. M.,
Fischer, A. M., Rienecker, E. V., McEnaney, P. R., Chavez, F. P., 2005. Coastal ocean
physics and red tides: an example from Monterey Bay, California. Oceanography 18, 246255.
Ryan, J. P., McManus, M., Sullivan, J. M., 2010. Interacting physical, chemical and biological
forcing of phytoplankton thin-layer variability in Monterey Bay, California. Continental Shelf
Research 30, 7-16.
Santos, A. M. P., Peliz, A., Dubert, J., Oliveira, P. B., Anglico, M. M., R, P., 2004. Impact of a
winter upwelling event on the distribution and transport of sardine (Sardina pilchardus) eggs
and larvae off western Iberia: a retention mechanism. Continental Shelf Research 24, 149165.
Saunders, P., 1982. Circulation in the eastern North Atlantic. Journal of Marine Research 40,
She, J., Klinck, J. M., 2000. Flow near submarine canyons driven by constant winds. Journal
Geophysical Research 105, 28671-28694.


Silva, A., Palma, S. , Moita, M. T., 2008. Coccolithophores in the upwelling waters of Portugal:
Four years of weekly distribution in Lisbon bay. Continental Shelf Research 28, 2601-2613.
Silva, A., Palma, S., Oliveira, P. B., Moita, M.T., 2009. Composition and interannual variability
of phytoplankton in a coastal upwelling region (Lisbon Bay, Portugal). Journal Sea Research
62, 238-249.
Skliris, N., Djenidi, S., 2006. Plankton dynamics controlled by hydrodynamic processes near a
submarine canyon off NW corsican coast: A numerical modelling study. Continental Shelf
Research 26, 1336-1358.
Skliris, N., Hecq, J. H., Djenidi, S., 2002. Water fluxes at an ocean margin in the presence of a
submarine canyon. Journal of Marine Systems 32, 239-251.
Skliris, N., Lacroix, G., Djenidi, S., 2004. Effects of extreme meteorological conditions on
coastal dynamics near a submarine canyon. Continental Shelf Research 24, 1033-1045.
Sprengel, C., Baumann, K.-H., Henderiks, J., Henrich, R., Neuer, S., 2002. Modern
coccolithophore and carbonate sedimentation along a productivity gradient in the Canary
Islands region: seasonal export production and surface accumulation rates. Deep-Sea
Research II 49, 3577-3598.
Strickland, J. D. H., Parsons, T. R., 1972. A Practical Handbook of Seawater Analysis. Fisheries
Research Board of Canada, Ottawa.
Troupin, C., Machn, F., 2012. Negative wind anomalies generated a diminution of productivity









Tyler, P., Amaro, T., Arzola, R., Cunha, M., de Stigter, H., Gooday, A., Huvenne, V., Ingels, J.,
Kiriakoulakis, K., Lastras, G., Masson, D., Oliveira, A., Pattenden, A., Vanreusel, A., van


Weering, Vitorino, J., Witte, U., Wolf, G., 2009. Europes Grand Canyon: Nazar submarine
canyon. Oceanography 22 (1), 46-57.
Vanney, J., Mougenot, D., 1990. Un canyon sous-marin du type gouf: le Canho da Nazar
(Portugal). Oceanologica Acta 13, 13-14.
Vitorino, J., Oliveira, A., Jouanneau, J. M., Drago, T., 2002. Winter dynamics on the northern
Portuguese shelf. Part 1: physical processes. Progress in Oceanography 52, 129-153.
Wallace, J. M., 2002. Two faces of the North Atlantic Oscillation (NAO). Luso-American
Foundation, Lisbon, Portugal, 41pp.
Winter, A., Jordan, R., Roth, P., 1994. Biogeography of living coccolithophores in ocean waters.
In: A. Winter and W. Siesser (Eds). Coccolithophores. Cambridge University Press,
Cambridge, 161-177.
Winter, A., Reiss, Z., Luz, B., 1979. Distribution of living coccolithophore assemblages in the
Gulf of Elat (Aqaba). Marine Micropaleontology 4, 197-223.
Young, J., 1994. Functions of coccoliths. In: A. Winter and W. Siesser (Eds). Coccolithophores.
Cambridge University Press, Cambridge, 63-82.
Young, J. R., Bown P. R., Lees J. A. (eds) Nannotax website. International Nannoplankton
Association. 21 Sept 2011. URL:
Young, J., Geisen, M., Cros, L., Kleijne, A., Sprengel, C., Probert, I., stergaard, J.B., 2003. A
guide to extant calcareous nannoplankton taxonomy. Journal of Nannoplankton Research,
Special Issue 1, 1-125.
Zapata, M., Rodriguez, F., Garrido, J. L., 2000. Separation of chlorophylls and carotenoids from
marine phytoplankton: a new HPLC method using a reversed phase C8 column and pyridinecontaining mobile phases. Marine Ecology Progress Series 195, 29-45.


Appendix A
The taxonomic list includes, in alfabetical order, all taxa identified during the present study.

Acanthoica quattrospina Lohmann 1903

Acanthoica spp. Lohmann 1903; emend. Schiller 1913, Kleijne 1992
Algirosphaera robusta (Lohmann 1902) Norris 1984
Alisphaera extenta Kleijne et al. 2002
Alisphaera ordinata (Kamptner 1941) Heimdal 1973
Alisphaera pinnigera Kleijne et al. 2002
Braarudosphaera bigelowii (Gran et Braarud 1935) Deflandre 1947
Calcidiscus leptoporus (Murray et Blackman, 1898) Loeblich et Tappan, 1978
Coccolithus pelagicus subsp. braarudii (Gaarder, 1962) Geisen et al., 2002
Coronosphaera mediterranea. (Lohmann 1902) Gaarder in Gaarder and Heimdal 1977
Discosphaera tubifera (Murray et Blackman 1898) Ostenfeld 1900
Emiliania huxleyi (Lohmann, 1902) Hay et Mohler in Hay et al.,1967
Gephyrocapsa ericsonii McIntyre et B, 1967
Gephyrocapsa muellerae Brhret, 1978
Gephyrocapsa oceanica Kamptner, 1943
Helicosphaera carteri (Wallich, 1877) Kamptner, 1954
Michaelsarsia elegans Gran 1912; emend. Manton et al. 1984
Ophiaster formosus Gran 1912 emend. Manton et Oates 1983
Ophiaster hydroideus (Lohmann 1903) Lohmann 1913; emend. Manton et Oates 1983


Ophiaster cf. reductus Manton et Oates 1983

Palusphaera vandelii Lecal 1966a; emend. Norris 1984
Rhabdosphaera clavigera Murray et Blackman 1898
Scyphosphaera apsteinii Lohmann 1902
Syracolithus dalmaticus (Kamptner 1927) Loeblich Jr. et Tappan 1966
Syracosphaera amoena (Kamptner 1937) Dimiza et Triantaphyllou 2008
Syracosphaera anthos (Lohmann 1912) Janin 1987
Syracosphaera hirsuta Kleijne et Cros 2009
Syracosphaera marginaporata Knappertsbusch 1993
Syracosphaera molischii Schiller 1925
Syracosphaera nodosa Kamptner 1941
Syracosphaera ossa (Lecal 1966) Loeblich Jr. et Tappan 1968
Syracosphaera pulchra, Murray et Blackman 1898
Umbellosphaera irregularis Paasche in Markali and Paasche 1955
Umbilicosphaera hulburtiana Gaarder 1970
Umbilicosphaera sibogae (Weber-van Bosse, 1901) Gaarder, 1970

Figure captions

Figure 1 Geographical location of the study area and investigated CTD casts. Number labeled
stations indicate locations where samples for coccolithophore analysis were collected.


Figure 2 Density (a), salinity (b), temperature (c) and turbidity (d) sections obtained from CTD
casts along a WSW-ENE transect representing the hydrological conditions during the cruise
along the upper-middle Nazar Canyon axis. Labels refer to stations where plankton samples
were collected for coccolithophore studies.

Figure 3 - Relationship between nutrient concentration and salinity during the cruise. Water
depths were not differentiated in this analysis (i.e. nutrient data between 5 110 m water depths
were plotted all together).

Figure 4 Vertical distribution of phytoplankton biomass (Chl-a), salinity and nutrients along
the uppermost 110 m water depth, during the cruise. Grey squares refer to biomass and black
squares represent salinity and nutrient concentrations.

Figure 5 - Coccolithophore cell densities (cells/l) observed during the cruise plotted over a TS
diagram. Solid lines refer to CTD profiles from selected stations to illustrate the surface mixed
layer and the ENACWst as defined by Fiza (1984) and Fiza et al. (1998): stations 87, 118 and
132 located at 225 m, 854 m and 3478 m water depths, respectively.

Figure 6 - Coccolithophore total densities (cells/l) and isopycnals (kg/m3) recorded in the
uppermost 110 m water depth along a WSW ENE oriented transect covering the upper-middle
Nazar Canyon axis. Black arrows indicate the highest cell densities at the surface: stations 120
and 118 close to Belatina Valley, and station 87, at the canyon head.


Figure 7 Coccolithophore total densities (cells/l) and isopycnals (kg/m3) recorded along four
transects cutting across the Nazar Canyon upper reaches, approximately SSW NNE oriented:
T1 (a) is located at the canyon head and T4 (b) represents the more distal section.

Figure 8 - Vertical distribution of the most common coccolithophore taxa along the uppermost
110 m water depth.

Figure 9 CTD/turbidity and fluorometry casts and coccolithophore assemblages living in the
uppermost 100 m of three selected stations monitored during the cruise: (a) station 132 located at
the middle Nazar Canyon (3478 m), (b) station 118 located near Belatina Valley (854 m), and
(c) station 87 located at the Nazar Canyon head (224 m). Fluorometry was calibrated with in
situ Chl-a measurements.

Figure 10 - Water column density section (kg/m3) for a transect along the upper-middle Nazar
Canyon, with spatial distribution of scores from Factor 1 (a), Factor 2 (b), Factor 3 (c) and Factor
4 (d) obtained from the coccolithophore data. In each association species/variables are aligned
according to their factor loadings (in brackets those with less importance) and coded as an
equation: numerator = positive loadings; denominator = negative loadings. For taxonomic
complete references see Table 3.

Figure 11 Monthly averaged Chl-a production in the Nazar Canyon region during 2006-2011,
obtained from satellite data: (a) along-canyon oriented Transect A, between 39.59N, -9.1W and
39.51N, -9.9W; (b) north-south oriented Transect B crossing the canyon at station 120, between


39.85N, -9.41W and 39.2N, -9.41W; and (c) NNE-SSW oriented Transect C crossing the canyon
at station 87, between 39.85N, -9.1W and 39.4N, -9.41W. Dashed black lines indicate the
location of stations 87, 118 and 120.

Figure 12 Regional maps of monthly averaged Chl-a concentration in the Nazar Canyon
obtained from satellite data. (a) February, (b) March and (c) April of 2010.

Table captions

Table 1 Water column samples collected for the coccolithophore study (96 samples collected
from 28 stations), with station position, depth and sampling date. The bottom nepheloid layer
(BNL) was distinguished for stations at depths  110 m. Crosses indicate stations that provided
samples for HPLC (Chl-a) and/or nutrient (NOx, PO4 and SiO2) analyses.

Table 2 Maximum cell densities (cells/l) of the coccolithophore species observed under PLM
during the sampling period. The respective water depth, sampling station and location are
indicated (BNL = bottom nepheloid layer  110 m water depth). Minimum and maximum values
of total cell densities, temperature, salinity, turbidity, fluorometry, Chl-a measured by HPLC,
phytoplankton biomass and nutrients (NOx, PO4 and SiO2,) are indicated below. NC = Nazar

Table 3 - Results from factor analysis: eigenvalues and explained variance obtained for the
samples collected during the sampling period. The more significant variables were: temperature


(T), salinity (S), turbidity (Turb), Chl-a, nutrients (NOx, PO4 and SiO2), Acanthoica spp. (Acan),
Alisphaera spp. (Alisph), C. mediterranea (Cm), D. tubifera (Dtub), E. huxleyi (Eh), G. ericsonii
(Ge), G. oceanica (Go), M. elegans (Meleg), Ophiaster spp. (Ophi), and Syracosphaera spp.

Appendix B Figure captions

Appendix B, Figure B.1 - Typical aspect of loose coccoliths from samples collected at the

head (samples 87-25 m and 85-50 m, at 225 m and 306 m depths, respectively). [a-h] -

reworked and poorly preserved coccoliths (often belonging to larger species): (a, c) Coccolithus
pelagicus, (b) Emiliania huxleyi, coccosphere partially dissolved and collapsed, (d)
Umbilicosphaera sibogae, (e) Helicosphaera carteri, (f) Gephyrocapsa oceanica without bridge,
(g) Calcidiscus leptoporus, (h) Coronosphaera mediterranea. [ip] - well preserved coccoliths:
(i) C. pelagicus, (j, k, n) Syracosphaera pulchra, (l) E. huxleyi, (m) H. carteri, (n) C.
mediterranea, (o) Ophiaster sp., (p) U. sibogae. Scale bars: (o) =1 m; (a-n), (p) = 2 m.

Appendix B, Figure B.2 Acanthoica quatrospina (station 132-5 m, at 3478 m), Figure B.3
Alisphaera pinnigera (station 103-Bottom Nepheloid Layer (BL), at 109 m), Figure B.4
Alisphaera ordinata (station 132-5 m), Figure B.5 Alisphaera extenta (station 132-5 m). Scale
bars = 2 m.

Appendix B, Figure B.6 Algirosphaera robusta (station 131-25 m, at 3097 m), Figure B.7
Braarudosphaera bigelowii (station 103 - Bottom Nepheloid Layer (BNL), at 109 m), Figure B.8


Calcidiscus leptoporus (station 89-5 m, at 40 m), Figure B.9 Coronosphaera mediterranea

(station 89-5 m). Scale bars = 2 m.

Appendix B, Figure B.10 Coccolithus pelagicus subsp. braarudii (station 87-25 m, at 225 m),
Figure B.11 Emiliania huxleyi type A overcalcified (left side) and type B (right side) (station
87-25 m), Figure B.12 Gephyrocapsa ericsonii (station 89-5 m, at 40 m), Figure B.13
Gephyrocapsa muellerae (station 87-25 m). Scale bars: B.10, B.13 = 2 m; B.11 = 5 m; B.12 =

Appendix B, Figure B.14 Cluster of Gephyrocapsa oceanica (station 85-50 m, at 306 m),
Figure B.15 Helicosphaera carteri (station 87-25 m, at 225m), Figure B.16 Ophiaster
formosus (station 115-50 m, at 224 m), Figure B.17 Palusphaera vandelii (station 98-25 m, at
361 m). Scale bars = 2 m.

Appendix B, Figure B.18 Syracosphaera anthos (station 87-25 m, at 225 m), Figure B.19
Syracosphaera amoena (station 87-25 m), Figure B.20 Syracosphaera marginoporata (station
131-25 m, at 3097 m), Figure B.21 Syracosphaera molischii (station 96-Bottom Nepheloid
Layer (BNL), at 56 m). Scale bars: B.18, B.19, B.21 = 2 m; B.20 = 1m.

Appendix B, Figure B.22 Syracosphaera nodosa (station 103-Bottom Nepheloid Layer (BNL),
at 109 m depth), Figure B.23 Syracosphaera ossa (station 103-BNL), Figure B.24
Syracosphaera hirsuta and Emiliania huxleyi type A overcalcified (station 115-50m, at 224m


depth), Figure B.25 Syracolithus dalmaticus (station 85-50m, at 306m depth). Scale bars = 2


Table 1























































































































Water levels
40, 50, 100




16.4, 25



5, 25, 50




5, 25, 50,
5, 25, 50,
5, 25, BNL



5, 25, 50

5, 15, BNL



5, 15, BNL


5, 15, BNL

5, 25, BNL


5, 25, 50,
5, 25, 50,
5, 25, BNL














5, 25, 50,
5, 25, 50
5, 25, 50,
5, 25, 50,
5, 15, 50,
5, 15, 25,
5, 25, 50,
5, 25, 50


5, 25, 50




5, 25, 50,
5, 25, 50,
25, 50, 100


25, 50, 100


5, 25, 50



Table 3

Total variance (%)

Factor 1

Factor 2

Factor 3

Factor 4