spatialvariabilityofphytoplanktoncompositionandbiomass ...queguiner/files... ·...
TRANSCRIPT
Continental Shelf Research 22 (2002) 1225–1247
Spatial variability of phytoplankton composition and biomasson the eastern continental shelf of the Bay of Biscay (north-eastAtlantic Ocean). Evidence for a bloom of Emiliania huxleyi
(Prymnesiophyceae) in spring 1998
L. Lamperta,b, B. Qu!eguinerc,*, T. Labasquea, A. Pichona, N. Lebretond
aCentre Militaire d’Oc!eanographie, EPSHOM, Brest, Franceb Institut Universitaire Europ!een de la Mer, Laboratoire des Sciences de l’Environnement Marin, UMR CNRS 6539 ,
Technop #ole BREST-IROISE, Place Nicolas Copernic - 29280 Plouzan!e, FrancecCentre Oc!eanologique de Marseille, Laboratoire d’Oc!eanographie et de Biog!eochimie, UMR CNRS 6535, FR CNRS 6106,
Universit!e de la M!editerran!ee, Parc Scientifique et Technologique de Luminy, Case 901, F-13288 Marseille cedex 09, FrancedBiotop, Penfeld braz - 29820 Bohars, France
Received 6 December 1999; received in revised form 19 June 2001; accepted 26 June 2001
Abstract
A coccolithophorid bloom, dominated by Emiliania huxleyi, was detected by sea viewing wide field of view sensor
(SeaWiFS) images on the French continental shelf break in April 1998. Concentrations of up to 3.2� 106
coccospheres l�1 and up to 8.6� 107 coccoliths l�1 were measured by microscope countings of samples taken during thefirst days of the bloom. Moderate chlorophyll a concentrations (range: 0.8–1.1 mg l�1) characterised the study area.Chlorophyll and carotenoid pigments, analysed by high performance liquid chromatography (HPLC), confirmed the
dominance of Pry-mnesiophytes in the bloom area. The bloom was not monospecific and diatoms, mainly belonging to
the genus Rhizosolenia, as well as silicoflagellates were observed in the phytoplankton. Outside of the bloom area,
‘‘green algae’’ and cryptophytes dominated the phytoplankton. Diatoms were a dominant group of the Vilaine plume
community and dinoflagellates were dominant in the southern part of the study area. The development of the
dissipative phase of coccolithophorid bloom and its persistence for at least 4 weeks is explained by the conjunction of
water mass preconditioning by river inputs on the continental shelf, increasing PAR during spring, and internal wave
formation at the shelf break during spring tides. Partial dissolving of coccoliths and lack of horizontal displacement of
the bloom, during the 4 weeks, are interpreted in terms of rapid settling of coccoliths due to packaging by grazers as well
as ongoing pro-duction maintained by nutrient injection via the action of internal waves.r 2002 Elsevier Science Ltd.
All rights reserved.
Keywords: Algal blooms; Emiliania huxleyi; Coccoliths; Hydrodynamics; Chemotaxonomy; HPLC; Prymnesiophytes; Remote
sensing; Riverine inputs; France; Bay of Biscay; 44–481N and 001–0051W
*Corresponding author. Tel.: +33-04-9182-9205; fax: +33-04-9182-1991.
E-mail address: [email protected]
(B. Qu!eguiner).
0278-4343/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 2 7 8 - 4 3 4 3 ( 0 1 ) 0 0 1 0 3 - 0
1. Introduction
Coccolithophorids are nanoplanktonic algaebelonging to the Prymnesiophyceae. Emiliania
huxleyi is the most representative and cosmopoli-tan species, capable of generating vast blooms inall oceans, including the polar oceans (Winteret al., 1999); such blooms have been clearlyidentified following the developments of electronmicroscopy in the 1950s (Braarud et al., 1952). Atthe global scale, blooms (i.e. cell concentration>106 cells l�1) cover 1.4� 106 km2 surface area eachyear, with temperate regions accounting for 71%of the total; in North Atlantic waters blooms cancover areas >105 km2 (Brown and Yoder, 1994).The recent bloom of E. huxleyi in the easternBearing Sea shelf occurred over an area >2� 105
km2 and was observed from early July untilNovember 1997 (Sukhanova and Flint, 1998).Blooms have significant environmental impacts,
via increased water albedo (reflectance), dimethyl-sulphide (DMS) production, large fluxes of cal-cium carbonate out of the surface waters andchanges in the oceanic uptake of CO2 (Westbroeket al., 1993). E. huxleyi cells achieve the process ofcalcification with the release of CO2 according tothe reaction: 2HCO�
3 þ Ca2þ-CaCO3 þ CO2 þH2O; hence, acting as a source of CO2 rather thana sink (Tyrrell and Taylor, 1995).Bloom dynamics have been related to the
physical structure of the surface water: E. huxleyi
blooms occur most often during stratification inthe North Atlantic Ocean (Nanninga and Tyrrell,1996). Lack of photoinhibition at light intensitiesof up to 1500 mEinsteinm�2 s�1 may contribute tothe dominance of E. huxleyi in surface waters inthe shallow mixed layers (Nanninga and Tyrrell,1996). The close relationship between blooms andstratification seems an essential, but not sufficient,requirement to bloom formation (Nanninga andTyrrell, 1996).Townsend et al. (1994) have coined the term
‘‘mature waters’’ for water masses with historiesresulting in depletion or alterations in macro- andmicro-nutrient levels that could provide the rightsets of conditions to facilitate the formation ofE. huxleyi blooms. In general, E. huxleyi bloomsfollow diatom blooms in waters that have been
recently depleted in inorganic nutrients and arebecoming more stable in terms of vertical mixing.Salinity does not seem to be a critical factor per se,except in association with nutrient distributions(Holligan et al., 1993a). Low orthosilicic acidlevels have been suggested as a possible explana-tion for the biogeographical distribution ofE. huxleyi (Brown and Yoder, 1994). The meso-cosm studies in Norwegian fjords have shown thatdiatoms always bloom (usually to the exclusion ofeverything else) when orthosilicic acid is present atconcentrations of X2 mM, but are less likely tobloom at concentrations o2 mM (Egge andAksnes, 1992). In the North Atlantic, the springshift from diatoms to Prymnesiophytes observedduring NABE/JGOFS experiment (Sieracki et al.,1993) was also attributed to the depletion oforthosilicic acid (Lochte et al., 1993).Other hypotheses include seeding effects related
to water mass advection (Townsend et al., 1994),advantage at low nitrate and ammonia concentra-tions (Eppley et al., 1969), and tolerance to lowiron concentrations (Brand, 1991). E. huxleyi isalso known to have a requirement for thiamine(vitamin B1), which is not present in water in theabsence of biological activity. Additionally, graz-ing can act as a regulating factor of E. huxleyi
biomass. The microzooplankton is capable ofresponding rapidly to changes in phytoplanktonbiomass and may control the biomass of smallalgae (Thingstad and Sakshaug, 1990; Riegmanet al., 1993). For an oceanic bloom, it was reportedthat microzooplankton removed approximately44% of the E. huxleyi stock per day (Holliganet al., 1993b).
E. huxleyi blooms can be identified by satelliteimagery due to the strong reflectance signalproduced by the light-scattering coccoliths (Holli-gan and Groom, 1986). Coccoliths surround thecells and a few detach during the first phase of thebloom; detachment increases during the matureand dissipative phase of the bloom (Westbroeket al., 1993), characterised by moderate chloro-phyll a levels in nutrient-depleted waters (Holliganet al., 1993a). Satellite imagery has revealed thatE. huxleyi blooms occur between May and Augustin the North Atlantic with the greatest frequencyin June and July. Blooms in the Atlantic basin
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471226
have been documented using LANDSAT, coastalzone color scanner (CZCS), the visible band ofthe advanced very high resolution radiometer(AVHRR) and sea viewing wide field of viewsensor (SeaWiFS) (Le F"evre et al., 1983; Holliganet al., 1993b; Garcia-Soto et al., 1995; Holliganet al., 1983; Townsend et al., 1994; Sukhanova andFlint, 1998). In contrast to the abundant satellitedata, sea-truth data is scarce.During the Bio-Modycot 98 cruise, in the area
of the continental shelf of the Bay of Biscay, weobserved high concentrations of E. huxleyi in thevicinity of the continental slope. The aims of thispaper are (1) to document the spatial variability ofphytoplankton biomass, abundance, and composi-tion, in the study area at the beginning of spring;and (2) to derive the possible controlling factorsaffecting E. huxleyi development, with specialemphasis on the physical processes in the watermasses. We have followed a multi-parametricalapproach by using improved methods of pigmentanalysis (high performance liquid chromatogra-phy, HPLC) and satellite imagery (SeaWiFS), aswell as a more classical approach involvingphytoplankton examination by direct microscopy.
2. Materials and methods
Sampling was performed during the Bio-Modycot 98 cruise (Service Hydrographique etOc!eanographique de la Marine (SHOM) andIFREMER joint project) on board of the BH2
Lap!erouse research vessel (SHOM). A network of47 stations was sampled between 22 April and 27April 1998. Vertical profiles of temperature,conductivity and depth (CTD Sea Bird 911+)were recorded at every station (Lebreton andWolff, 1998). Three transects have been analysedin detail to define the physical properties (Fig. 1).To characterise the degree of water column
stability, a stratification index (SI) was computed(Bustillos-Guzman et al., 1995) as the difference ofdensity from surface to bottom, calculated foreach 5m layer (m�1), using the following equation:
SI ¼Pn
i¼1 Dst=Dz
nð1Þ
with i=number of layers of 5m in the watercolumn.For phytoplankton observations and pigment
analysis 5 l water samples were taken at 5m depth.Phytoplankton samples were taken at every stationwhereas pigment analysis was restricted to lowturbidity oceanic stations (Fig. 1).Water samples for phytoplankton identification
and counting were drawn in glass bottles. At eachstation, aliquots of 100ml seawater were preservedby adding 250 ml of Lugol’s iodine and 300ml werepreserved by adding 6ml of cacodylate-bufferedglutaraldehyde. Microphytoplankton (>20 mm)and nanophytoplankton (2–20 mm) were countedwithin 6 months by the Uterm .ohl (1931) methodusing a Nikon inverted phase-contrast microscope.For coccolithophorid identification, glutaralde-hyde-preserved samples were filtered onto 0.8 mmpolycarbonate membranes, which were then de-hydrated by keeping in an ethanol series andwere dried using a critical point apparatus. The
-200
-200
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
0˚
0˚
1˚E
1˚E
43˚N 43˚N
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
1
2
5
67
8
915
16
23
26 27
2829
30 31
3233
3435
36
39
40
41
42
43
44
46
47
3
4
10
11
12
13
14
17
18
19
20
21
22
24
25
37
38
45
LOIREVIL
AINE
GIRONDE
ADOURTra
nsec
t 1Tra
nsec
t 2
Transect 3
FRANCE
SPAIN
Micom transect
Fig. 1. Network of 47 stations sampled during Bio-Modycot 98
cruise (22–27 April 1998). The three transects 1, 2, and 3 are
also shown in the figure. Stars: CTD and HPLC sampling
network; Points: CTD-only sampling network. The dotted line
shows the location of the MICOM numerical model transect.
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1227
membranes were then affixed to stubs, coated with20nm gold, and then observed by scanning electronmicroscopy (SEM) using a HITACHI S-3200N.For pigment analysis, 1 l seawater samples were
prefiltered through 200 mm mesh nylon gauze andthen filtered onto 25mm GF/F fibre filters underlow-pressure vacuum (o0.5 bar, following DelAmo et al., 1997) for further HPLC analysis.Filters were stored immediately at �201C onboard for the duration of the cruise and were keptin the laboratory for a maximum of 7 months at�801C. Pigments were extracted and analysed bythe reverse-phase HPLC method slightly modifiedfrom Wright et al. (1991). The frozen GF/F filterswere ground and sonicated into 3ml of acetone–water (90/10, v/v). For each sample, 500 ml ofacetone–water extract were mixed with 165 ml ionpairing solution (tetrabutylammonium acetatebuffered with ammonium acetate) and 35 ml oftrans-canthaxanthin (internal standard), and100 ml were injected automatically by a refrigerated(41C) automatic sampler Thermo AS3000 in aODS2 C18 column (150mm� 4.6mm, with 3 mmsilica particles). The Thermo UV3000 detectorscanned the range spectrum between 400 and700 nm, and the effective detection was performedat 440 nm. The constametric pump used was aLDC Analytical Constametric 4100 with a flowrate of 1mlmn�1.Meteorological data (wind speed, irradiance)
were obtained from the AVISO database (M!et!eoFrance) on a 0.51 latitude� 0.51 longitude grid.The daily wind speed (knots) is the 10minaveraged speed at 46.51N and 31W, at 06h00UTC. Irradiance (Einsteinm�2 d�1) is the dailyintegrated value in the visible spectrum at the seasurface. Three surface global positioning system(GPS) buoys were deployed at Stations 7, 8 and 9to monitor surface water mass motion (positionrecording time-step: 1 h). River flow data wereobtained from the Service Hydrologique Centrali-sateur (Nantes, France). Customary values of thetide coefficient were obtained from the SHOM:they represent the geographically normalised tidalamplitude and range between 20 (neap tideminimum) and 120 (spring tide maximum).The SeaWiFS pictures were obtained from the
RSDAS research group (NERC, Plymouth La-
boratory); we used the composite colours RGB(stretched colour composite composed of the 555,510 and 443 nm wavebands) in parallel with thechlorophyll a product (in-water chlorophyll a
concentration calculated using the SeaBAM algo-rithm, McClain, 1997). These processing techni-ques are detailed at: http://www.npm.ac.uk/rsdas/doc/description.html. The comparison betweenHPCL chlorophyll and SeaWiFS chlorophyll a hasbeen presented elsewhere (Gohin et al., in press).We have used the MICOM numerical model
(code shallow water isopycnal) (Pichon, 1996) tosimulate internal wave amplitudes in the vicinity ofthe continental slope (Maz!e, 1987; Langlois et al.,1990) in April 1998. The 3-D multi-layer modelran over 10 layers with semidiurnal (M2) tidalforcing without thermodynamical coupling. Thedensity profile at Station 47, located at the shelfbreak, was used as initial reference (see Fig. 17b).
3. Results
3.1. Physical environment
The meteorological time series for March andApril showed inverse evolution trends betweenirradiance and wind speed (Fig. 2).During the study period two maxima in river
flow were observed (Fig. 3). In January 1998 heavyrainfalls resulted in increasing river flows of both
Fig. 2. Time-series of meteorological data (heavy line: daily
irradiance; light line: 10-mn averaged wind speed at 06h00
UTC, 461N–31W).
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471228
the Loire and Gironde (Q > 2500m3 s�1). At theend of the study period a similar maximum wasobserved in early May (QE3000m3 s�1). In be-tween these periods, the Loire and Gironde riverflows remained quite high (between 500 and1000m3 s�1) during February and March. Riverflows in 1998 fell within the average range observedin the last decade (Lazure and J!egou, 1998). In thestudy area, the northward geostrophic spreading ofriver plumes is known to favour the offshoreadvection of coastal waters as far as the continentalslope during the spring transition (Lazure andJ!egou, 1998; Hermida et al., 1998).At the end of April 1998, the main river plumes
extended over the continental shelf as shown bythe surface salinity distribution (Fig. 4a). Thesurface seawater exhibited a strong vertical halinestratification characteristic of the spring situationin that area. The Gironde plume was restricted tothe coastal zone near the mouth of the estuary(Fig. 5c). To the north, the Loire and Vilaineplumes tended to spread towards the entire shelfsection, especially at the northernmost boundaryof the study area (Fig. 5a and b). The surfacetemperature distribution was characterised by anorth-south gradient in the study area reflectingthe beginning of spring warming (Fig. 4b). In frontof the Gironde estuary, the seasonal thermoclinewas starting to develop offshore even though the
Fig. 3. River flows of Gironde, Loire, and Vilaine between 01
January and 15 May 1998.
34
34.2
34.2
34.4
34.4
34.634.6
34.6
34.8
34.8
34.8
34.8
34.8
35
35
35
35
35
35
35.2
35.2
35.2
35.2
35.2
35.4
35.4
35.4
35.4
35.4
35.6
35.6
35.6
35.6
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(a)
11.811.8
11.8
1212
12
12
12.2
12.2
12.212.2
12.4
12.4
12.4
12.6
12.6
12.612.6
12.8
12.812.8
13
13
13.213.2
13.4
13.4
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(b)Fig. 4. Salinity (a) and temperature (b) distribution in surface
waters (5m depth).
"
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1229
temperature gradient still remained low (o0.51Cbetween surface and 40m) (Fig. 6c). In the coastalarea, the presence of the river plumes resulted in athermal stratification superimposed on the halinestratification (Fig. 6); this was also the mostimportant influence on the northern boundary ofthe study area (Fig. 6a). The density structurefollowed closely that of the salinity distribution,which demonstrated the major role of freshwaterinputs in determining the spring physical structureover the study area (Fig. 7). The offshore thermalstratification slightly influenced the density struc-ture, especially in front of the Gironde estuary(Fig. 7c).During the study period, the trajectories of the
GPS buoys gave information on the speed anddirection of wind-induced surface currents. Thebuoys travelled between 55–95 km at an average
speed of 1.7 cm s�1 in a south-western residualdirection (Fig. 8).
3.2. Phytoplankton distribution
Microphytoplankton and nanophytoplanktonpopulations showed different distribution patternsover the study area (Fig. 9).Microphytoplankton concentrations were fairly
uniform in surface waters (range: 7� 103–3� 104
cell l�1) over the main part of the study area;however, higher concentrations (up to 3� 105
cell l�1) were observed in the northern part, infront of the Vilaine estuary (Fig. 9a). In the areaof high cell numbers, diatoms dominated themicrophytoplankton, accounting for 89–93% oftotal cell numbers. Due to the relatively highand sustained river flow of Vilaine and the
Fig. 5. Vertical distribution of salinity on the three transects:
transect 1(a), transect 2(b), and transect 3(c).
Fig. 6. Vertical distribution of temperature on the three
transects: transect 1(a), transect 2(b), and transect 3(c).
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471230
wind-induced general surface water circulation,the high microphytoplankton cell numbers mayhave originated from the coastal Bay of Vilaineand then advected offshore. Chapelle et al. (1994)indicated that wind-induced circulation appears tobe the major process for water renewal in thatcoastal ecosystem. The diatom distribution insurface waters (Fig. 9b) also showed the occur-rence offshore of two maxima close to each otherat Stations 7 and 46 (3.5� 104 and 6� 104 cell l�1,respectively). The dominant diatom species areshown in Fig. 10. Cerataulina pelagica dominatedthe diatom population in the northern area,influenced by the plumes of Loire and Vilaine(Fig. 10a). Leptocylindrus spp. dominated a transi-tion area in the centre of the study area (Fig. 10b).Rhizosolenia spp. (mainly Rh. delicatula) appearedas typical of a diatom community located offshore
in front of the Gironde estuary (Fig. 10c). Skele-
tonema costatum dominated the inshore watersbetween the mouth of Gironde and Loire estuaries(Fig. 10d).In the southern part of the study area, the
contribution of diatoms was very low: this grouprepresented only a few percent (o10%) of themicrophytoplankton cell numbers, south of themouth of the Gironde estuary. The microphyto-plankton was mainly composed of a mixeddinoflagellate population characterised by highnumbers of Gymnodinium spp.Nanophytoplankton maximum concentrations
were located in offshore surface waters (values upto 2.6� 106 cell l�1) (Fig. 9c), with coccolithophor-ids restricted to a particular band extendingbetween the estuaries of Loire and Gironde(Fig. 9d). Optical microscopic observations re-vealed the existence of the greatest concentrationsof coccolithophorids and nanophytoplankton inthe same area, as shown in Fig. 9d; in the area ofhigh nanophytoplankton cell numbers, coccolitho-phorid concentrations (counted on glutaraldehydepreserved samples) reached up to 3.2� 106 cells l�1
and 8.6� 107 free coccoliths l�1 at Station 6. Atstations located in the band of maximal nanophy-toplankton concentration (1, 6, 8, 40, 42, 43, and47), coccolithophorid cell numbers were character-istic of bloom populations, i.e. >1� 106 cell l�1,according to Tyrrell and Taylor (1996). To thenorth, the coccolithophorid patch extended toStations 9, 14, and 15, but with lower values(p5� 105 cell l�1). On the north-western bound-ary of the study area, a discrete maximum ofnanophytoplankton (2.9� 106 cell l�1) was ob-served at Station 23, not related to coccolitho-phorid distribution. SEM observations (Fig. 11)enabled identification of the coccolithophorids asE. huxleyi. Coccolith shape and element numberper coccolith (ranging from 32 to 38), as shown inFig. 11a and f, were characteristic of E.huxleyi
type A (Van Emburg, 1989; Van Bleijswijk et al.,1991). An important observation concerned thestate of preservation of coccospheres: some wereintact and showed well-preserved coccoliths,but we observed a gradation towards cocco-spheres showing apparently dissolved coccoliths(Fig. 11a–d). Occasionally, empty coccospheres
Fig. 7. Vertical distribution of density on the three transects:
transect 1(a), transect 2(b), and transect 3(c).
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1231
(Fig. 11e) surrounded by intact coccoliths wereobserved, suggesting the escape of protoplastsfrom the coccolith envelope, a life cycle stagepreceding cell division (Braarud, 1963; Paasche,1964). It is noteworthy that the E. huxleyi bloomwas not monospecific but rather occurred in theoffshore area already characterised by the mixedpopulation of Rhizosolenia spp. (see above) wherewe also observed the highest Chrysophyte (silico-flagellates) cell numbers (range: 2� 102–1.2� 103
cell l�1). In the same area, colonies of Phaeocystis
pouchetii were also present.
3.3. Pigment distribution
In the area where HPLC samples weretaken, Chl a concentrations increased southward
(Fig. 12) from 0.6 mg Chl a l�1 at Station 15 to3.1 mg Chl a l�1 south of 451N at Station 27.As primary taxonomic markers, alloxanthin
(allo), 190-butanoyloxyfucoxanthin (19BF), fucox-anthin (fuco), peridinin (peri), prasinoxanthin(prasi), zeaxanthin (zea), 190-hexanoyloxyfucox-anthin (19HF) and chlorophyll b (Chl b) are,respectively, typical pigments of cryptophytes,chrysophytes (and pelagophytes), diatoms, photo-synthetic dinoflagellates, prasinophytes, coccoidcyanobacteria, prymnesiophytes and ‘‘green
50
50
50
200
200
6˚W
6˚W
4˚W
4˚W
2˚W
2˚W
46˚N 46˚N
29/04/98
25/05/98
04/05/98
25/05/98
24/04/98
25/05/98
*
**
Fig. 8. Trajectories of surface-free drifting GPS buoys with starting and end times of release.
Fig. 9. Spatial distribution of major phytoplankton groups
(103 cell l�1) in surface waters (5m): (a) total microphytoplank-
ton, (b) diatoms, (c) total nanophytoplankton, (d) coccolitho-
phorids. In the area without contour plots values are lower than
the nearest contour.
"
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471232
3030
30
30
30
30
30
30
30
30
30
30
30
30
30
30
303030
30
30
30
30
30
30
60
60
60
60 60
60
60
60
90
90
90
90
90
120
120
120
150 150180
210
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(a)
30
30
30
30
30
30
30
30
60
60
60
6060
60
90
9090
120
120120150
180
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
46˚N 47˚N
48˚N 48˚N
(b)
300
900
900900
900
900
900900
900
900
900
900
900900
900
900
900
1500
1500
1500
1500
1500
1500
1500
150015
00
1500
1500
1500
15001500
1500
1500 1500
1500
1500
2100
2100
2100
2100
2100
2100
2100
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(c)
500
500
500
500
500
500
500
500500
500
1000
1000
1000
1000
1000
1000
1000
1000
1000
1500
1500
1500
2000
2000 2500
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(d)
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1233
5
5
55
5
5
5
5
25
25
25
25
25
25
25
25
45
45
45
45
4545
45
45
4545
65
65
65
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(a)
10
10
10
10
10
10
10
10
10
10
10
20
20
20
20
20
2020
20
30
30
3030
30
30
40
40
40 50
50
60
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(b)
10
10
10
10
10
10
10
10
10
10
10
10
10
1010
10
10
10
30
3030
30
30
30
30
3030
30
30
50
50
50
50
50
5050
50
70
70
70
70
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(c)
5
5
5
5
5
55
5
5
5
55
5
5
5
5
5
25
25
25
25
25
25
25
25
2525
4545
45
45
65
65
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(d)
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471234
algae’’ (Jeffrey, 1997). However, all these markersare not specific to particular taxa and this pointwill be discussed latter.
We have calculated the contribution of eachtaxonomic group to the total Chl a stock by amultiple regression analysis with the followingassumptions: (1) the Chl a:taxonomic markerratios were constant during the sampling period,and (2) the pigments chosen are independent. Weare aware that both criteria were not necessarilymet. The use of multiple linear regression assumesexclusive pigment–algal class relationship, which
Fig. 10. Spatial distribution of dominant diatom species (%
over total diatoms) in surface waters (5m): (a) Cerataulina
pelagica, (b) Leptocylindrus spp., (c) Rhizosolenia spp., (d)
Skeletonema costatum. In the area without contour plots values
are lower than the nearest contour.
3
Fig. 11. Scanning electron micrographs of Emiliania huxleyi morphotype A: (a) entire well-preserved coccosphere, Station 1, (b) entire
coccosphere showing partial dissolution of coccoliths, Station 43, (c) entire ‘‘coccosphere’’ showing remains of coccoliths, Station 1, (d)
naked cell of size similar to E. huxleyi coccospheres, (e) group of coccospheres and coccoliths, Station 1: arrow indicates an empty
coccosphere, (f) detail of well-preserved free coccoliths, Station 1.
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1235
may lead to underestimation to the major classes.To address this problem, a two-way approach wasdeveloped using the regression analysis togetherwith cell counts and identification. The followingequation has been obtained:
½Chl a ðng l�1Þ ¼ 102:7þ 3:32½allo þ 1:35½Chl b
þ 1:32½fuco þ 1:38½19HF
þ 3:77½peri: ð2Þ
Our coefficient values are consistent with thosefound in the literature, except for the coefficientfor peridinin which was slightly higher (Gieskesand Kraay, 1983; Gieskes et al., 1988; Everitt et al.,
1990; Barlow et al., 1993; Letelier et al., 1993;Claustre et al., 1994; Andersen et al., 1996; Peeken,1997). Eq. (2) was the best fit of the multiple linearregression model describing the relationship be-tween Chl a and the 8 independent variableschosen (pigment taxonomic markers). The equa-tion obtained by the forward selection methodaccounted for 98.3% (R2 ¼ 0:983) of the Chl a
variance with 5 pigment concentrations and aconstant term; the latter was interpreted as theresidual Chl a concentration not explained by thetaxonomic marker considered (for example, Chry-sophyte and cyanobacteria markers are omitted inthe final model); that amount of Chl a accounted
0.8
0.8
0.8
0.8
0.8
0.8
0.8
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.4
1.41.4
1.41.4
1.7
1.7
1.7
1.71.
7
2
2
22
2.3
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
Fig. 12. Spatial distribution of chlorophyll a (mg l�1) in surface waters (5m).
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471236
for 3–20% of the total phytoplankton crop(Table 1).The distribution of 19HF (Fig. 13a) showed
elevated concentrations in the area affected by E.
huxleyi bloom (range: 114–522 ng l�1). The highestcontribution to the biomass was observed atStation 6 where Prymnesiophytes accounted for55% of the Chl a (Table 1) which was in agreementwith the cell number distribution. Within thepatch, the lowest contribution was 27% forstations 9 and 15, which were not characterisedby bloom concentrations. Prymnesiophyte contri-bution to the standing stock shown in Table 1indicated again the nonmonospecific signature ofthe bloom. The relationship (regression model)between the Prymnesiophyte contribution (percent
over total biomass obtained from HPLC) andE. huxleyi number (counted by optical microscopyand normalised by log-transformation) was highlysignificant (Po0:01; R2 ¼ 0:78; Fig. 14). Thisconfirmed the dominance of E. huxleyi in thePrymnesiophyte community of the bloom area. InFig. 14, the positive intercept was interpreted asthe contribution of the other 19HF-relatedalgal group in the absence of E. huxleyi (wehave mentioned previously the occurrence ofPh. pouchetii within the bloom area). At thesouthern boundary of the study area high con-centrations of 19HF were also observed. Thesewere not related to E. huxleyi distribution butcoincided with an area of high contribution ofdinoflagellates on the basis of cell numbers.
Table 1
Contribution of pigment-related taxonomic groups to total Chl a stocks (results of the multiple regression analysis)
Station
number
Allo
crypto. %
Chl b
‘‘green algae’’ %
Fuco
diatoms %
19HF
prymnes. %
Peri
dinoflag. %
Others
%
Sum
%
1 11 26 14 43 0 12 106
2 20 33 16 13 0 15 97
5 20 34 17 19 0 12 102
6 10 19 14 55 0 8 106
7 12 24 23 18 0 9 86
8 18 18 10 42 0 11 100
9 23 22 12 27 9 10 103
15 20 25 15 27 0 18 105
16 28 23 13 25 9 12 110
23 39 11 11 11 11 8 90
26 13 45 14 12 6 9 99
27 8 22 20 28 19 3 100
28 9 28 16 34 0 5 91
29 17 47 10 14 5 8 101
30 24 50 9 9 0 7 99
31 20 21 4 30 14 7 95
32 23 30 11 20 0 13 97
33 20 37 19 19 6 4 104
34 15 37 11 19 0 20 102
35 22 30 19 22 6 5 105
36 23 16 20 38 0 8 105
39 15 40 19 10 6 7 97
40 25 9 20 39 0 7 100
41 19 29 19 21 0 10 99
42 11 19 20 43 0 15 107
43 16 29 15 33 0 10 104
44 13 33 17 15 14 14 106
46 10 19 22 38 0 11 100
47 9 24 17 38 0 10 99
The sums of individual contributions are indicated in the last column. The stations belonging to E. huxleyi patch (as defined by cell
numbers, see text) are in bold italics.
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1237
Several dinoflagellates are known to containpigments characteristic of their endosymbionts,e.g. Chrysophytes, green algae or Prymnesiophytes(Bj�rnland and Liaaen-Jensen, 1989), and thiscould be the reason of the southern 19HFmaximum.The fucoxanthin distribution could indicate the
importance of diatoms. In the coccolithophoridbloom area, the diatom contribution to total Chl a
ranged between 10% and 20% (Table 1) and wehave mentioned previously that diatoms wereobserved in phytoplankton samples of this area.However, although fucoxanthin is often consid-ered as a diatom marker, some species of otheralgal groups, especially belonging to prymnesio-phytes, chrysophytes and dinoflagellates, maycontain significant amounts of fucoxanthin (Bumaet al., 1991). So, it was difficult to conclude on thereal contribution of diatom to Chl a biomass inour study where diatoms and coccolithophoridsoccurred together. Outside the bloom area, thephytoplankton pigments were indicative of thepresence of a mixed nanophytoplankton commu-nity represented by ‘‘green algae’’ and crypto-phytes (Table 1).
140
140
140140
220
220
220
220
220
220
220
220
220
220
220
220
300
300
300
300
300
300
300
300
300
300
300
380
380
380
460
460
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(a)
20
20
20
20
20
20
30
30
30
30
30
30
30
30
30
30
40
4040
40
40
40
50
5˚W
5˚W
4˚W
4˚W
3˚W
3˚W
2˚W
2˚W
1˚W
1˚W
44˚N 44˚N
45˚N 45˚N
46˚N 46˚N
47˚N 47˚N
48˚N 48˚N
(b)
Fig. 14. Relationship between coccosphere concentrations
(Log10 cell l�1) and chlorophyll a contributions of prymnesio-
phytes (% total chlorophyll a) as derived from the multiple
regression analysis.
Fig. 13. Spatial distribution of (a) 190-hexanoyloxyfucoxanthin
(ng l�1), and (b) chlorophyll a contribution of prymnesiophytes
(% total chlorophyll a) as derived from the multiple regression
analysis, in surface waters (5m).
3
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471238
3.4. Remote sensing
The temporal evolution of water-leaving radi-ance (Fig. 15) showed an area of high radiancecoinciding with the area of high E. huxleyi cellnumbers and of elevated 19HF concentrations. Atits maximal spatial extent, the bloom area coveredabout 15,000 km2 on 07 May 1998 (Fig. 15b).SeaWiFS images prior to 18 April 1998 did notshow any evidence of high water-leaving radianceover the study area. The heavy cloud coveragebetween 20 and 25 April 1998 did not allow thedetermination of the exact starting of the bloom,but the bloom was clearly seen from 26 April 1998.We can then argue that the structure appearedwithin a week or less. The high water-leavingradiance area persisted for more than 4 weeks and
the structure faded from 24 May 1998 and almostdisappeared on 28 May 1998 (Fig. 15d). Duringthe period of high water-leaving radiance, nomajor lateral advection of the patch was detectedon the SeaWiFS images, contrary to the observa-tions made by the surface drifting buoys.Outside of the study area, in the northern part
of the Bay of Biscay, two areas of high water-leaving radiance were observed on the internalArmorican continental shelf break and in theCeltic Sea. These structures appeared later, ascompared to the bloom we document, butpersisted for a longer period until 14 July 1998.They most certainly also reflected the occurrenceof coccolithophorid blooms which have beendocumented previously in the northern part ofthe Bay of Biscay (Holligan et al., 1983; Le F"evre
Fig. 15. Selected SeaWiFS images showing the temporal evolution of the reflective stage of E. huxleyi bloom on the eastern continental
shelf of the Bay of Biscay as evidenced by sea-surface reflectance: (a) 28 April 1998, (b) 07 May 1998, (c) 18 May 1998, (d) 28 May
1998. Arrows indicate the occurrence of other coccolithophore blooms in the Celtic Sea.
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1239
et al., 1983; Viollier et al., 1987). After that date,no RGB signature could be attributed to cocco-lithophorid bloom in the Bay of Biscay.
4. Discussion
4.1. Spring phytoplankton communities
Diatoms did not dominate the microphyto-plankton communities of the eastern continentalshelf of the Bay of Biscay in spring 1998. Diatomswere abundant near shore and especially in theplume of the Vilaine river. Diatom assemblageswere similar to that which was found in nearbywater masses like the Ushant thermal front(Birrien et al., 1985). In the outer shelf area,dinoflagellates dominated the microphytoplanktoncommunities beyond the 50m isobath and at everystation south of 451N. Nanophytoplankton dis-tribution resembled closely that of the dinoflagel-late distributions. HPLC measurements showedthat nanophytoplankton dominated the autotrophstanding stock over the entire study area (see Table1). This agreed well with the observations made byHerbland et al. (1998), who found up to 70%phytoplankton biomass in theo3 mm size-fractionin offshore waters in front of the Gironde in May1995. Also, this was the first observation ofPhaeocystis pouchetii at such low latitudes on theEuropean continental shelf.The present study confirmed the importance of
coccolithophorid developments at the shelf breakof the Bay of Biscay. Such developments have beenmentioned already in the Celtic Sea by Holliganet al. (1983) and on the Armorican continentalshelf by Le F"evre et al. (1983) and Viollier et al.(1987). However, coccolithophorid blooms werenever sampled in shelf waters in front of theGironde estuary until now, even though Beaufortand Heussner (1999) mentioned that they were animportant component of the mass flux in sedimenttraps deployed over the continental slope, to thesouth of our study area. The morphotype A,observed in this study, is the most widespread inthe North Atlantic. The larger morphotype B hasbeen only mentioned rarely, such as in the waterseast of Scotland (Van Bleijswijk, 1991) and may be
a possible old stage of morphotype A, at 50mdepth west of Brittany (GREPMA, 1988).
4.2. Dynamics of the spring coccolithophorid
bloom onset
Surface salinity values measured during the 1998cruise fell within the range of SHOM averagemonthly values compiled from 22,000 recordsfrom the beginning of the century (Buleon andHassani, 1995); this precluded the occurrence of anatypical year as related to water mass evolution,contrary to what was observed for the 1997 BeringSea coccolithophorid bloom (Sukhanova and Flint1998).Lazure and J!egou (1998) and Hermida et al.
(1998) have described the seasonal evolution ofmajor hydrological features over the easterncontinental shelf of the Bay of Biscay: (1) Duringwinter, coastal shelf waters are separated fromoffshore shelf waters by a thermal front parallel tothe coastline; near shore, the water columnremains homogeneous until the setting of halinestratification, related to river flow, increases. Thehaline stratification we observed, in early 1998,could then have been induced by increased riverflows at the end of January as shown in Fig. 3. (2)In spring, the seasonal thermocline, isolating warmsurface waters from cold bottom waters, starts todevelop in the oceanic area before spreading overthe continental shelf. (3) The shelf thermalstratification favours advection of estuarine inputsin the surface over the entire area and evensometimes beyond the shelf break limit.By using SeaWiFS imagery we were able to
determine the period of appearance of E. huxleyi.The period corresponded to an increase in averagedaily irradiance with values >25 Einsteinm�2 d�1
(see Fig. 2), which was in agreement with thethreshold value of 23 Einsteinm�2 d�1 obtained byEgge and Heimdal (1994) in a mesocosm experi-ment. Critical depths have been calculated for thebeginning of the study period by using the formulaof Sverdrup (1953) modified by Nelson and Smith(1991). Using in-water diffuse attenuation coeffi-cient K490 (calculated in SeaWiFS processingversion 2 from an empirical algorithm that usesthe 443 and 555 nm bands) and the daily irra-
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471240
diance, we estimated critical depths much greaterthan the mixed layer depths over the entire studyarea except, in near-shore river-influenced waters.This means that light conditions, which enablephytoplankton spring bloom development, wereprobably encountered over the main part of thecontinental shelf before the beginning of the cruise.Due to the spring advection of estuarine inputs
in the surface water, outer shelf surface waterswere influenced by the freshwater chemical proper-ties. In the shelf waters influenced by the Girondeplume, Herbland et al. (1998) have observed verylow phosphate levels (0.035–0.09 mM in early May1995) with residual nitrate concentrations (averagevalue: 4.63–0.17 mM) and, conversely, high nitra-te:phosphate molar ratios (mean value: 161) on atransect between the mouth of Gironde and thecontinental slope; they related the chemical prop-erties of shelf waters to the specificity of Girondewaters characterised by a large excess of nitraterelative to phosphate together with a balance ofnitrate relative to orthosilicic acid. Herbland et al.(1998) suggested that early phosphorus depletioncould limit the spring phytoplankton bloom. Sucha potential limitation of phytoplankton growth byphosphorus availability could explain the outburstof E. huxleyi in the outer shelf area influenced bythe Gironde plume; Tyrrell and Taylor (1996)reported very low half-saturation constants forphosphate uptake for this species (0.005 mM) ascompared to small flagellates (0.05 mM). Also,Herbland et al. (1998) measured decreased silicicacid concentrations (1.41–0.78 mM: average valueswithin the 0–25m layer) which could also be acompetitive advantage for E. huxleyi againstdiatoms: Egge and Aksnes (1992) indicated thatE. huxleyi overtook diatoms at orthosilicic acidconcentrationso2 mM in a mesocosm experiment.It is notable that the advection of Gironde
plume waters (impoverished in phosphate relativeto nitrate and silicate), as far as the shelf break, isnot a regular annual feature of the area. Theintensity of advection is controlled by river flows(Lazure and J!egou, 1998) and we hypothesise thathigh river flows and biological activity duringwinter could precondition the shelf water beforethe thermocline was established. After high winterriver discharge, initial nutritional conditions at the
onset of the spring bloom would favour thenonsiliceous low-phosphate adapted E. huxleyi
against diatoms and small flagellates.
4.3. Fate of the spring coccolithophorid bloom
The SeaWiFS image sequence in April–May1998 (Fig. 15) showed that the coccolithophoridbloom observed lasted at least 4 weeks. As shownin Fig. 16, the onset of the dissipative phase ofcoccolithophorid bloom, as well as those occurringon the Armorican shelf and in the Celtic Sea,coincided with the higher tide coefficients. Theonset of high scattering is due to the presence offree coccoliths and coccolithophorids. Westbroeket al. (1993) presented mesocosm data showingthat free coccoliths appear after 7 days into abloom and lag behind peak cell numbers. Further-more, the high water-leaving radiance patchescould have started a few weeks before appearing ashigh radiance. However, we suggest a controlmechanism by mixing, induced by internal wavesat the shelf break, in the period of thermohalinestratification as has already been demonstrated byMaz!e (1987). The intensity of the semidiurnalvertical mixing induced by diffusion and shearingat the pycnocline interface was thus strongerduring spring tides which could have resulted ina semidiurnal injection of deep-water originatednutrients in to the surface layer.Results of the MICOM numerical model runs
for an average tide coefficient in April 1998(Fig. 17) confirmed a vertical displacement of
Fig. 16. Temporal evolution of tide coefficients: the beginning
of blooms in the Armorican shelf and Celtic Sea (according to
the RGB SeaWiFS pictures) are indicated.
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1241
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471242
10m at the pycnocline depth (20m) and 30m forthe bottom layer (130m) related to internal wavepropagation. Internal wave propagation thenappears as a probable mechanism responsible forthe long duration of the dissipative phase ofE. huxleyi bloom observed on SeaWiFS images.
The spatial distribution of the stratificationindex (Fig. 18) evidenced an area of very lowvalues at the shelf break suggesting the intensifica-tion of vertical mixing at that location. Holliganand Groom (1986) already emphasised tidalcurrents and internal waves as physical mechan-isms favouring the phytoplankton activity andgrowth in the Celtic Sea. Owing to low sedimenta-tion rates in the order of a few cmd�1 (Honjo,1976), it was expected that the coccoliths wouldhave drifted with the surface waters during thestudy period. However, the SeaWiFS images didnot show any indication of bloom drifting, in spiteof the observed SW surface currents (1.7 cm s�1 onaverage) shown by the trajectories of surface GPS
Fig. 17. Results of the 3-D multi-layer Micom numerical model
(isopycnal shallow water code): (a) simulation of internal wave
amplitudes in the vicinity of the continental slope (see Fig. 1) in
April 1998. Connected line: model layers (left scale); dotted line:
bathymetry profile showing the shelf break location (right
scale). (b) Density profiles: connected line: st profile used in the
Micom model; dotted line: st profile obtained from the CTD
sensors.
3
Fig. 18. Spatial distribution of the stratification index (104m�1) during the study period.
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1243
buoys (see Fig. 8). This could reflect two proper-ties: (1) a low residence time of coccoliths withinthe surface layer, which has been indicated alreadyby sediment trap records south of our study area(Beaufort and Heussner, 1999), allied to (2) acontinuous production of coccoliths in the surfacewaters in that area. The latter could be explainedby quasi-continuous input of nutrients related tointernal waves; the former could be related toaggregation processes as has already been postu-lated (Holligan et al., 1983, 1993a; Cad!ee, 1985;Beaufort and Heussner, 1999). Coccolith aggrega-tion and rapid settling could have arisen throughmucus excretion by living cells (Cad!ee, 1985) orthey could result from packaging in grazer faecalpellets. We had no opportunity to sample aggre-gates or faecal pellets but the SEM observations ofpartially dissolved coccoliths (see Fig. 11) togetherwith well-preserved coccospheres (which dismissedsample conservation biases) was in favour of thegrazing mechanism.Few studies have focused on the dynamics of
coccolithophorid blooms in this study area. How-ever, Beaufort and Heussner (1999) have providedclear evidence of the occurrence of phytoplanktonpopulations dominated by E. huxleyi, showingseasonal and interseasonal variations at the south-eastern shelf break of the Bay of Biscay. The fate ofthe population after the bloom is still a matter ofdebate and we have no definitive explanation of theempty coccospheres we observed. Those questionsdeserve further investigations in the study area.
5. Conclusion
E. huxleyi was responsible for the coccolitho-phorid bloom detected by SeaWiFS on the Frenchcontinental shelf break in April 1998. From ourobservations we have proposed the followingscenario taking into account of the water masspreconditioning described by Townsend et al.(1994) and of a competitive advantage in nutrientutilisation (Tyrrell and Taylor, 1996; Brown andYoder, 1994; Egge and Aksnes, 1992; Sierackiet al., 1993; Lochte et al., 1993):
(1) The waters of the continental shelf areconditioned by river inputs and the biological
activities (especially from the Gironde estuary)in winter and early spring. The extension ofthe area influenced by river inputs is related tothe intensity of winter–spring rainfall, domi-nant winds, and current trajectories over thecontinental shelf.
(2) River inputs together with increasing radia-tion in spring result in the establishment of apycnocline over the continental shelf. Thiscreates the conditions favouring both internalwave formation at the shelf break and exten-sion of the area influenced by the physical andchemical characteristics of coastal waters.
(3) Daily PAR reaching values >23Einsteinm�2 d�1 together with large amplitude internalwaves, favouring mixing at the pycnoclineinterface during spring tides, are finallyresponsible for the onset of the coccolitho-phorid bloom. When phosphate and orthosi-licic acid concentrations are low at the shelfbreak, the coccolithophorid ability to grow atphosphate concentrations down to 10% ofthose required for the other phytoplanktongroups gives them the competitive advantageenabling the occurrence of the bloom.
Acknowledgements
We are grateful to J. Young, T. Tyrrell, R.Maz!e, B. Le Cann, G. Langlois and C. Ratsiva-laka for their help during the preparation of themanuscript. We also wish to thank the captain,officers, crew and hydrographs of BH2 Lap!erouse
for their support at sea during the cruise, as well asX. Lenhardt and F. Orvain for kind help in watersampling. Thanks are also due to G. Sinquin forhelp with the SEM study. We also acknowledgetwo anonymous referees for their very usefulcomments, which greatly helped to improve theformer manuscript.
References
Andersen, R., Bidigare, R., Keller, M.D., Latasa, M., 1996. A
comparison of HPLC pigment signatures and electron
microscopic observations for oligotrophic waters of the
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471244
North Atlantic and Pacific Oceans. Deep-Sea Research II
43, 517–537.
Barlow, R.G., Mantoura, R.F.C., Gough, M.A., Fileman,
T.W., 1993. Pigment signatures of the phytoplankton
composition in the northeastern Atlantic during the 1990
spring bloom. Deep-Sea Research 40, 459–477.
Beaufort, L., Heussner, S., 1999. Coccolithophorids on the
continental slope of the Bay of BiscayFproduction,
transport and contribution to mass fluxes. Deep-Sea
Research II 46, 2147–2174.
Birrien, J.L., Le Corre, P., Videau, C., 1985. Development of
Gyrodinium aureolum [Hulburt] in the Bay of Douarnenez
and the Iroise Sea in summer 1983. Actes du Colloque
Franco-Japonais d’Oceanographie, Vol. 2, Marseille 16–21
September, pp. 51–64.
Bj�rnland, T., Liaaen-Jensen, S., 1989. Distribution patterns ofcarotenoids in relation to chromophyte phylogeny and
systematics. In: Green, J.C., Leadbeater, B.S.C., Diver,
W.L. (Eds.), The Chromophyte Algae: Problems and
Perspectives. Clarendon Press, Oxford, pp. 37–61.
Braarud, T., 1963. Reproduction in the marine coccolithophor-
id coccolithus huxleyi in culture. Pubblicazioni della
Stazione Zoologica di Napoli 33, 110–116.
Braarud, T., Gaarder, K.R., Markali, J., Nordli, E., 1952.
Coccolithophorids studied in the electron microscope.
Observations on Coccolithus huxleyi and Syracosphaera
carterae. Nytt Magazine f .or Botanie 1, 129–134.
Brand, L.E., 1991. Minimum iron requirements of marine
phytoplankton and the implications for the biogeochemical
control of new production. Limnology and Oceanography
36, 1756–1771.
Brown, C.W., Yoder, J.A., 1994. Coccolithophorid blooms in
the global ocean. Journal of Geophysical Research 99,
7467–7482.
Buleon, L., Hassani, A., 1995. Salinit!e de surface dans le Golfe
de Gascogne. Rapport No. 106 EPSHOM/CMO/CM/NP
du 19 septembre 1995, EPSHOM, Brest, France 31 pp,
unpublished.
Buma, A.G.J., Bano, N., Veldhuis, M.J.W., Kraay, G.W.,
1991. Comparison of the pigmentation of two strains of the
prymnesiophyte phaeocystis sp. Netherlands Journal of Sea
Research 27, 173–182.
Bustillos-Guzman, J., Claustre, H., Marty, J.-C., 1995. Specific
phytoplankton signatures and their relationship to hydro-
graphic conditions in the coastal northwestern Mediterra-
nean Sea. Marine Ecology Progress Series 124, 247–258.
Cad!ee, G.C., 1985. Macroaggregates of emiliania huxleyi in
sediment traps. Marine Ecology 24, 193–196.
Claustre, H., Kerherv!e, P., Marty, J.-C., Prieur, L., Videau, C.,
Hecq, J.-H., 1994. Phytoplankton dynamics associated with
a geostrophic front: ecological and biogeochemical implica-
tions. Journal of Marine Research 52, 711–742.
Chapelle, A., Lazure, P., Menesguen, A., 1994. Modelling
eutrophication events in a coastal ecosystem. Sensitivity
analysis. Estuarine Coastal and Shelf Science 39, 529–548.
Del Amo, Y., Qu!eguiner, B., Tr!eguer, P., Breton, H., Lampert,
L., 1997. Impacts of high-nitrate freshwater inputs on
macrotidal ecosystems. II. Specific role of the silicic acid
pump in the year-round dominance of diatoms in the Bay of
Brest (France). Marine Ecology Progress Series 161,
225–237.
Egge, J.K., Aksnes, D.L., 1992. Silicate as regulating nutrient in
phytoplankton competition. Marine Ecology Progress
Series 83, 281–289.
Egge, J.K., Heimdal, B.R., 1994. Blooms of phytoplankton
including Emiliania huxleyi (Haptophyta). Effects of nu-
trient supply in different N:P ratios. Sarsia 79, 333–348.
Eppley, R.W., Rogers, J.N., McCarthy, J.J., 1969. Half-
saturation constants for uptake of nitrate and ammonium
by marine phytoplankton. Limnology and Oceanography
14, 912–920.
Everitt, D.A., Wright, S.W., Volkman, J.K., Thomas, D.P.,
Lindstrom, E.J., 1990. Phytoplankton community composi-
tions in the western equatorial pacific determined from
chlorophyll and carotenoid pigment distributions. Deep-Sea
Research 37, 975–997.
Garcia-Soto, C., Fernandez, E., Pingree, R.D., Harbour, D.S.,
1995. Evolution and structure of a shelf coccolithophore
bloom in the Western English Channel. Journal of Plankton
Research 17, 2011–2036.
Gieskes, W.W.C., Kraay, G.W., 1983. Dominance of crypto-
phyceae during the phytoplankton spring bloom in the
central North Sea detected by HPLC analysis of pigments.
Marine Biology 75, 179–185.
Gieskes, W.W.C., Kraay, G.W., Nontji, A., Setiapermana, D.,
Sutomo, 1988. Monsoonal alternation of a mixed and a
layered structure in the phytoplankton of the euphotic zone
of the Banda Sea (Indonesia): A mathematical analysis of
algal pigment fingerprints. Netherland Journal of Sea
Research 22, 123–137.
Gohin, F., Druon, J.-N., Lampert, L. A five channel
chlorophyll concentration algorithm applied to SeaWiFS
data processed by SeaDAS in coastal waters. International
Journal of Remote Sensing, in press.
GREPMAGroup, 1988. Satellite (AVHRR/NOAA-9) and ship
studies of a coccolithophorid bloom in the western English
Channel. Marine Nature 1, 1–14.
Herbland, A., Delmas, D., Laborde, P., Sautour, B., Artigas,
F., 1998. Phytoplankton spring bloom of the Gironde plume
waters in the Bay of Biscay: early phosphorus limitation and
food-web consequences. Oceanologica Acta 21, 279–291.
Hermida, J., Lazure, P., Froidefond, J.M., J!egou, A.-M.,
Castaing, P., 1998. La dispersion des apports de la Gironde
sur le plateau continental. Donn!ees in situ, satellitales et
num!eriques. Oceanologica Acta 21, 209–221.
Holligan, P.M., Groom, S.B., 1986. Phytoplankton distribu-
tions along the shelf break. Proceedings of the Royal
Society of Edinbourgh 88B, 239–263.
Holligan, P.M., Viollier, M., Harbour, D.S., Camus, P.,
Champagne-Philippe, M., 1983. Satellite and ship studies
of coccolithophore production along a continental shelf
edge. Nature 304, 339–342.
Holligan, P.M., Groom, S.B., Harbour, D.S., 1993a. What
controls the distribution of the coccolithophore Emiliania
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1245
huxleyi, in the North Sea? Fisheries Oceanography 2,
175–183.
Holligan, P.M., Fernandez, E., Aiken, J., Balch, W.M., Boyd,
P., Burkill, P.H., Finch, M., Groom, S.B., Malin, G.,
Muller, K., Purdie, D.A., Robinson, C., Trees, C.C.,
Turner, S.M., van der Wal, P., 1993b. A biogeochemical
study of the coccolithophore, Emiliania huxleyi, in the
North Atlantic. Global Biogeochemical Cycles 7 (4),
879–900.
Honjo, S., 1976. Coccoliths: production, transportation and
sedimentation. Marine Micropaleontology 1, 65–79.
Jeffrey, S.W., 1997. Application of pigment methods to
oceanography. In: Jeffrey, S.W., Mantoura, R.F.C., Wright,
S.W. (Eds.), Phytoplankton Pigments in Oceanography:
Guidelines to Modern Methods. SCOR-UNESCO, Paris,
pp. 127–166.
Langlois, G., Gohin, F., Serpette, A., 1990. Refroidissements
locaux aux abords du talus continental Armoricain.
Oceanologica Acta 13, 159–169.
Lazure, P., J!egou, A.-M., 1998. 3D modelling of seasonal
evolution of Loire and Gironde plumes on Biscay Bay
continental shelf. Oceanologica Acta 21, 165–177.
Lebreton, N., Wolff, J., 1998. Modycot 98.3FBio-Modycot
98.3 Les donn!ees hydrologiques. Rapport No. 111/EP-
SHOM/CMO/CM/NP du 23 juin 1998, 80 pp.
Le F"evre, J., Viollier, M., Le Corre, P., Dupouy, C., Grall, J.R.,
1983. Remote sensing observations of biological material by
Landsat along a tidal thermal front and their relevancy to
the available field date. Estuarine Coastal and Shelf Science
16, 37–50.
Letelier, R.M., Bidigare, R.R., Hebel, D.V., Ondrusek, M.,
Winn, C.D., Karl, D.M., 1993. Temporal variability of
phytoplankton community structure based on pigment
analysis. Limnology and Oceanography 38, 1420–1437.
Lochte, K., Ducklow, H.W., Fasham, M.J.R., Stienen, C.,
1993. Plankton succession and carbon cycling at 471N 201W
during the JGOFS North Atlantic bloom experiment. Deep-
Sea Research II 40, 91–114.
Maz!e, R., 1987. Generation and propagation of non-linear
internal waves induced by the tide over a continental slope.
Continental Shelf Research 7, 1079–1104.
McClain, C.R., 1997. SeaWiFS Bio-optical Mini-workshop
(SeaBAM) Overview.
Nelson, D.M., Smith, W.O., 1991. Sverdrup revisited: critical
depths, maximum chlorophyll levels, and the control of
southern ocean productivity by the irradiance-mixing
regime. Limnology and Oceanography 36, 1650–1661.
Nanninga, H.J., Tyrrell, T., 1996. Importance of light for the
formation of algal blooms by Emiliania huxleyi. Marine
Ecology Progress Series 136, 195–203.
Paasche, E., 1964. A tracer study of the inorganic carbon
uptake during coccolith formation and photosynthesis in
the coccolithophore Coccolithus huxleyi. In: Lund (Ed.),
Physiologia Plantarum Supplementum III, University of
London, 82 pp.
Peeken, I., 1997. Photosynthetic pigment fingerprints as
indicators of phytoplankton biomass and development in
different water masses of the Southern Ocean during austral
spring. Deep-Sea Research II 44, 261–282.
Pichon, A., 1996. Introduction a la mod!elisation des ondes
internes de mar!ee "a l’aide du mod"ele MICOM. Rapport
d’Etude N1003/96 EPSHOM/CMO/RE, EPSHOM, Brest,
France, 111 pp, unpublished.
Riegman, R., Kuipers, B.R., Noordeloos, A.A.N., Witte, H.J.,
1993. Size-differential control of phytoplankton and the
structure of plankton communities. Netherland Journal of
Sea Research 31, 255–265.
Sieracki, M.E., Verity, P.G., Stoecker, D.K., 1993. Plankton
community response to sequential silicate and nitrate
depletion during the 1989 North Atlantic spring bloom.
Deep-Sea Research II 40, 213–225.
Sukhanova, I.N., Flint, M.V., 1998. Anomalous blooming of
coccolithophorids over the Eastern Bering Sea shelf.
Oceanology 38, 502–505.
Sverdrup, H.U., 1953. On conditions for the vernal blooming of
phytoplankton. Journal du Conseil international pour
l’Exploration de la Mer 18, 287–295.
Thingstad, T.F., Sakshaug, E., 1990. Control of phytoplankton
growth in nutrient recycling ecosystems. Theory
and terminology. Marine Ecology Progress Series 63,
261–272.
Townsend, D.W., Keller, M.D., Holligan, P.M., Ackleson,
S.G., Balch, W., 1994. Blooms of the coccolithophore
Emiliania huxleyi with respect to hydrography in the Gulf of
Maine. Continental Shelf Research 14, 979–1000.
Tyrrell, T., Taylor, A.H., 1995. Latitudinal and seasonal
variations in carbon dioxide and oxygen in the northeast
Atlantic and the effects on Emiliania huxleyi and
other phytoplankton. Global Biogeochemical Cycles 9,
585–604.
Tyrrell, T., Taylor, A.H., 1996. A modelling study of Emiliania
huxleyi in the NE Atlantic. Journal of Marine Systems 9,
83–112.
Uterm .ohl, M., 1931. .Uber das umgekehehrte mikroskop.
Archiv f .ur Hydrobiologie Beihefte Ergebnisse Plankton
22, 643–645.
Van Bleijswijk, J.D.L., Van der Wal, P., Kempers, R., Veldhuis,
M., Young, J.R., Muyzer, G., de Vrind-de Jong, E.,
Westbroek, P., 1991. Distribution of two types of emiliania
huxleyi (prymnesiophyceae) in the northeast Atlantic region
as determined by inmunofluorescence and coccolith mor-
phology. Journal of Phycology 27, 566–570.
Van Emburg, P.E., 1989. Coccolith formation in Emiliania
huxleyi. Ph. D. Thesis, University of Leiden, The Nether-
lands, 145 pp, unpublished.
Viollier, M., Sournia, A., Birrien, J.L., Morin, P., 1987.
Observations satellitaires du phytoplancton dans les zones
de discontinuit!e hydrologique au large de la Bretagne.
Oceanologica Acta Num!ero Sp!ecial, Spatial Oceanography
Symposium, Brest, 19–20 November 1985, Proceedings, pp.
51–56.
Winter, A., Elbr.achter, M., Krause, G., 1999. Subtropical
coccolithophores in the Weddell Sea. Deep-Sea Research 46,
439–449.
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–12471246
Westbroek, P., Brown, C.W., Van Bleijswijk, J.D.L., Brownlee,
C., Brummer, G.J., Conte, M., Egge, J., Fernandez, E.,
Jordan, R., Knappertsbusch, M., Stefels, J., Veldhuis,
M.J.W., Van der Wal, P., Young, J., 1993. A model system
approach to biological climate forcing: the example of
Emiliania huxleyi. Global and Planetary Change 8, 27–46.
Wright, S.W., Jeffrey, S.W., Mantoura, R.F.C., Llewellyn,
C.A., Bj�rnland, T., Repeta, D., Welschmeyer, N., 1991.Improved HPLC method for the analysis of chlorophylls
and carotenoids from marine phytoplankton. Marine
Ecology Progress Series 77, 183–196.
L. Lampert et al. / Continental Shelf Research 22 (2002) 1225–1247 1247