ll!i/,

, / ' +

I I I l i . t , ?

Spatio-temporal distribution of phytoplankton pigments in Northumberland Strait: CZCS imagery and in situ data

C. ~uentes-~aco' , A.F. ~ézina ' , T. platt2, W.G. ~ a r r i s o n ~ , P.G. cormier3, L.E. waite3, and L. ~ev ine '

' Regional Science Branch Department of Fisheries and Oceans Maurice Lamontagne Institute, P.O. Box 1000 Mont-Joli, Québec, G5H 324

Regional Science Branch Department of Fisheries and Oceans Bedford Institute of Oceanography, P.O. Box 1006 Dartmouth, Nova Scotia, B2Y 4A2

Regional Science Branch Department of Fisheries and Oceans Gulf Fisheries Center, P.O. Box 5030 Moncton, New Brunswick, E1C 9B6

Canadian Technical Report of Hydrography and Ocean Sciences No. 195

Fisheries and Oceans

Canadian Technical Report of Fisheries and Aquatic Sciences

Technical reports contain scientific and technical information that contribute to existing knowledge but that are not normally appropriate for primary literature. Technical reports are directed primarily toward a worldwide audience and have an international distribution. No restriction is placed on subject matter, and the series reflects the broad interests and policies of the Department of Fisheries &d Oceans, namely, fisheries and aquatic sciences.

Technical reports may be cited as full publications. The correct citation appears above the abstract of each report. Each report is indexed in the data base Aquatic Sciences and Fisheries Abstracts.

Numbers 1-456 in this series were issued as Technical Reports of the Fisheries Research Board of Canada. Numbers 457-714 were issued as Department of the Environment, Fisheries and Marine Service, Research and Development Directorate Technical Reports. Numbers 715-924 were issued as Department of Fisheries and the Environment, Fisheries and Marine Service Technical Reports. The current series name was changed with report number 925.

Technical reports are produced regionally but are numbered nationally. Requests for individual reports will be filled by the issuing establishment listed on the front cover and title page. Out-of-stock reports will be supplied for a fee by commercial agents.

Rapport technique canadien des sciences halieutiques et aquatiques

Les rapports techniques contiennent des renseignements scientifiques et techniques qui constituent une contribution aux connaissances actuelles, mais qui ne sont pas nonnalement appropriés pour la publication dans un journal scientifique. Les rapports techniques sont destinés essentiellement à un public international et ils sont distribués a cet échelon. il n'y a aucune restriction quant au sujet; de fait, la série reflète la vaste gamme des intérêts et des politiques du ministère des Pêches et des Océans, c'est-a-dire les sciences halieutiques et aquatiques.

Les rapports techniques peuvent être cités comme des publications intégrales. Le titre exact paraît au- dessus du résumé de chaque rapport. Les rapports techniques sont indexés dans la base de données Aquatic Sciences and Fisheries Abstracts.

Les numéros 1 a 456 de cette série ont été publiés à titre de rapports techniques de l'Office des : recherches sur les pêcheries du Canada. Les numéros 457 à 714 sont parus à titre de rapports techniques de la Direction générale de recherche et du développement, Service des pêches et de la mer, ministère de l'Environnement. Les numéros 715 à 924 ont été publiés à titre de rapports techniques du Service des pêches et de la mer, ministère des Pêches et de l'Environnement. Le nom actuel de la série a été établi lors de la parution du numéro 925.

Les rapports techniques sont produits à l'échelon régional, mais numérotés à l'échelon national. Les demandes de rapports seront satisfaites par l'établissement d'origine dont le nom figure sur la couverture et la page du titre. Les rapports épuisés seront fournis contre rétribution par des agents commerciaux.

Canadian Technical Report of Hydrography and Ocean Sciences 195

SPATIO-TEMPORAL DISTRIBUTION OF PHYTOPLANKTON PIGMENTS IN NORTHUMBERLAND STRAIT: CZCS IMAGRY AND IN SITU DATA

César ~uentes-~aco ' , Alain F. ~éz ina ' , Trevor platt2, William G. ~arrison', Paryse G. cormier3, Lynda E. waite3, and Laure ~ e v i n e '

1 Regional Science Branch Department of Fisheries and Oceans

Maurice Lamontagne Institute, P.O. Box 1000 Mont-Joli, Québec, G5H 3Z4

2~egional Science Branch Department of Fisheries and Oceans

Bedford Institute of Oceanography, P.O. Box 1006 Dartmouth, Nova Scotia, B2Y 4A2

3~egional Science Branch Department of Fisheries and Oceans

Gulf Fisheries Centre, P.O. Box 5030 Moncton, New Brunswick, E1C 9B6

O Minister of Public Works and Government Services Canada 1998 Cat. No. Fs 97- 181195E ISSN 071 1-6764

Correct citation for this publication:

Fuentes-Yaco, C., A.F. Vézina, T. Platt, W.G. Harrison, P.G. Connier, L.E. Waite, and L. Devine. 1998. .Spatio-temporal distribution of phytoplankton pigments in Northumberland Strait: CZCS imagery and in situ data. Can. Tech. Rep. Hydrogr. Ocean Sci. 195:vii + 30 pp.

TABLE OF CONTENTS

................................................................................................................................ List of Tables iv

................................................................................................................................ List of Figures iv

.......................................................................................................................................... Abstract vi

. . .......................................................................................................................................... Résumé V l l

2 . Methodology ............................................................................................................................ 1

..................................................................................................................... 2.1. Study area 1 .................................................................................. 2.2. Remote sensing (CZCS) images 2

............................................................................................................. 2.3. In situ database 3

3 Results ...................................................................................................................................... 4 .

3.1. CZCS monthly composite images ................................................................................ 4 ............................................................................................................ 3.2. Regionalization 4

................................................................................................... 3.2.1. NAFO regions 4 ..................................................................................................... 3.2.2. Petrie regions 5

....................................................................................... 3.2.3. CZCS-derived regions 5

4 . Discussion ................................................................................................................................ 5

.............................................................................................................. 4.1. NAFO regions 6 4.2. Petrie regions ................................................................................................................ 6 4.3. Satellite-derived regions ............................................................................................... 7

5 . Conclusions and Recornrnendations ........................................................................................ 8

6 . Acknowledgements .................................................................................................................. 9

7 . References ................................................................................................................................ 9

LIST OF TABLES

Table 1. Dates and number of remote sensing images ........ ........... . .............. ........... . . ........ . . 12

Table 2. Research missions in the southem Gulf of St. Lawrence between May 1994 and October 1995 .......... ............................................................................... ........ ......... 13

Table 3. Local names and coordinates of the five regions in the study area identified using the CZCS images ................................................................................................... 14

LIST OF FIGURES

Figure 1. The southem Gulf of St. Lawrence. P.E.I.: Prince Edward Island; 1: Miscou Point; 2: Miramichi Bay; 3: North Cape; 4: Richibucto Cape; 5: Carey Point; 6: Cape Tormentine; 7: Tryon Head; 8: Rice Point; 9: Cape Bear; 10: Macquarrie Point; 1 1 : Arisaig Point; 12: St. Georges Bay; 13: Black Point; 14: East Point ............................................................................................................ 15

Figure 2. Residual surface circulation in Northumberland Strait inferred from drift bottles (from Lauzier, 1965) ....... . .. . ........ . ... . ....... . .............. . . . . . . . . . . . . . . . . . . . . . 16

Figure 3. Sampling locations for research missions between July 1994 and October 1995 ......... 17

Figure 4. CZCS composite images of phytoplankton pigments (mg m-3) from April to September (1979-198 1) in the southern Gulf of St. Lawrence ...................................... 18

Figure 5. Global average of CZCS composite images of phytoplankton pigments (mg 3 m- ) (1979-198 1) in the southern Gulf of St. Lawrence ................................................. 20

Figure 6. NAFO regions 4TL, 4TH, and 4TG ................................................................................ 22

Figure 7. Mean of CZCS (1979-1981) and in situ (1994-1995) pigments (mg m-'), standard enors, and number of data for NAFO regions 4TL, 4TH, and 4TG ............... 23

Figure 8. Monthly mean of CZCS (1979-198 1) and in situ (1994-1995) pigments (mg m-5, standard errors, and number of data for NAFO regions 4TL, 4TH, and

Figure 9. Petrie regions 1 1 (Shediac Valley), 14 (W Northumberland Strait), and 17 (E Northumberland Strait) .................................................................................................. 25

Figure 10. Mean of CZCS (1979-1981) and in situ (1994-1995) pigments (mg m-3), standard errors, and number of data for Petrie regions 11, 14, and 17 ........................... 26

Figure 1 1. Monthly mean of CZCS (1979-198 1) and in situ (1994-1995) pigments (mg m-3), standard errors, and number of data for Petrie regions 11, 14, and 17 .................. 27

........................................................ Figure 12. CZCS-derived subdivisions NW, W, C, E, and SE 28

Figure 13. Mean of CZCS (1979-1981) and in situ (1994-1995) pigments (mg m-'), standard errors, and number of data for CZCS-derived regions NW, W, C, E, and SE .............................................................................................................................. 29

Figure 14. Monthly mean of CZCS (1979-1981) and in situ (1994-1995) pigments (mg m"), standard errors, and number of data for CZCS-derived regions NW, W,

..................................................................................................................... C, E, and SE 30

ABSTRACT

Fuentes-Yaco C., A.F. Vézina, T. Platt, W.G. Harrison, P.G. Cormier, L.E. Waite, and L. Devine. 1998. Spatio-temporal distribution of phytoplankton pigments in Northumberland Strait: CZCS imagery and in situ data. Can. Tech. Rep. Hydrogr. Ocean Sci. 195: vii + 30 pp.

This report describes the spatial and seasonal variations of phytoplankton pigments in the southern Gulf of St. Lawrence using Coastal Zone Color Scanner (CZCS) images and in situ data collected during different periods. To compare the patterns obtained with both methods, we grouped the CZCS and in situ data into regional subdivisions and computed averages. Because there are no climatological standards for biological and chemical data to compare with this information, we used subdivisions developed for commercial fisheries (International Commission for Northwest Atlantic Fisheries [ICNAF/NAFO]) and physical oceanographic features (Petrie et al., 1996) as first attempts. Finally, we used new subareas based on visual analyses of satellite phytoplankton pigment imagery. Both the satellite and in situ data sets show similar synoptic patterns in the spatial distribution of phytoplankton pigments; however, the CZCS estimate gives a more detailed view, suggesting five subareas. The pigment concentrations evaluated in these subareas seemed to Vary temporally in agreement with in situ measurements, showing a bi-modal cycle with high and comparable pigment levels during both spring and late surnrner-early fall. In conclusion, we recommend: i) the study of possible links between in situ biological (phytoplankton biomass), physical, and nutrient data within the subareas identified using CZCS- derived phytoplankton pigment levels; ii) the confirmation or modification of the proposed regionalization using remote sensing data from new satellites like the SeaWIFS and with sensors of higher spatial resolution (CASI); and iii) that the grid of stations in future research missions be modified to adequately sarnple the different subareas identified using CZCS images.

Fuentes-Yaco C., A.F. Vézina, T. Platt, W.G. Harrison, P.G. Cormier, L.E. Waite. and L. Devine. 1998. Spatio-temporal distribution of phytoplankton pigments in Northumberland Strait: CZCS imagery and in situ data. Can. Tech. Rep. Hydrogr. Ocean Sci. 195: vii + 30 pp.

Ce rapport décrit des variations spatio-temporelles des pigments phytoplanctoniques dans le sud du golfe du Saint-Laurent en utilisant des images-duapeiir satellitaire « Coastal Zone Color Scanner >> (CZCS) et des données in situ échantillonnées à différentes périodes. Pour comparer les patrons obtenus avec ces deux méthodes, nous avons couplé les données CZCS et in situ dans des subdivisions régionales et pour lesquelles nous avons calculé des moyennes. Considérant qu'il n'y a pas de standards climatologiques pour les caractéristiques biologiques ou chimiques pour comparer avec nos données dans cette partie du golfe, nous avons suivi régions délimitées pour la gestion des pêcheries commerciales (« International Commission for Northwest Atlantic Fisheries » [ICNAF/NAFO]) de même que des subdivisions basées sur les caractéristiques physiques océanographiques (Petrie et al., 1996). Finalement, nous avons utilisé une nouvelle régionalisation, prenant en considération les analyses visuelles des images satellitaires. Les deux méthodes de travail, soit les images satellitaires ainsi que les données prises en mer, montrent de fortes variations synoptiques de la répartition spatiale des pigments. Néanmoins, les valeurs estimées avec le satellite donnent une appréciation plus détaillée. Les analyses visuelles des images du CZCS suggèrent l'identification de cinq régions. Les concentrations en pigments évaluées dans ces régions semblent varier temporellement en accord avec les données in situ suivant un cycle bi-modal. Des concentrations élevées en chlorophylle ont été observées au printemps et à la fin de l'été et au début de l'automne. En conclusion, nous suggérons : i) l'étude des liens possibles entre les données in situ de la biomasse phytoplanctonique et les caractéristiques physiques et chimiques du milieu pour chacune des régions identifiées par les images du capteur satellitaire CZCS ; ii) la confirmation ou la modification de la régionalisation proposée en utilisant les nouveaux capteurs tels que le SeaWIFS, ainsi qu'à l'aide des senseurs à plus haute résolution spatiale (CASI), et finalement; iii) que la grille de stations d'échantillonnage dans les futures missions de recherche soit modifiée afin de mieux échantillonner les différentes sous-régions identifiées à partir des images CZCS.

1. INTRODUCTION

The Fisheries Resources Conservation Council of Canada has suggested an ecosystem approach to fisheries management; in consequence, the Fisheries Oceanography Cornmittee of the Department of Fisheries and Oceans (DFO) discussed and proposed studies of environmental variability and its effect on fisheries resources (Harrison and Sameoto, 1996). Harrison and Platt (1995) identified a component of the variability in exploitable fish stocks that is due to changes in the marine ecosystem mediated by environmental change; they proposed that the only practical way to monitor these changes is by satellite remote sensing. Harrison and Sameoto (1996) consider that this satellite information will be critical for the development of plankton productivity indices.

Harrison and Sameoto (1996) noted that within DFO there currently exists a lirnited number of databases from sea-going operations that may provide oceanographic information, but that the nature, quality, and extent of this data has not yet been fully evaluated. For several years, a monitoring program in the southern Gulf of St. Lawrence (GSL) was conducted by DFO's Maritime Region. Their results showed high phytoplankton concentrations through Northumberland Strait, between Miscou Point and St. Georges Bay (Waite et al., 1997). Satellite images of the water color also suggest that the pigment concentrations are high in this region (Fuentes-Yaco et al., 1997a). However, the relationships between satellite and in situ data are not easy to evaluate because of the presence of suspended sediments that contaminate the ocean color signal.

The main objective of this report is to describe the spatial and seasonal variations of phytoplankton pigments using Coastal Zone Color Scanner (CZCS) archives and in situ data collected during different periods. To compare the patterns obtained with these two methods, we grouped the CZCS and in situ data into regional boxes and computed averages. Because there are no climatological standards for biological and chernical data to compare with this information, we used previous regionalizations developed for other purposes. Subdivisions based on commercial fisheries (International Commission for Northwest Atlantic Fisheries [ICNAF/NAFO]) and physical oceanographic features (Petrie et al., 1996) were selected as first attempts. Finally, we propose a new classification of the regions based on visual analyses of satellite phytoplankton pigment imagey.

2. METHODOLOGY

2.1 S tudy area

The study area was mainly Northumberland Strait, extending from Miscou Point in the West to St. Georges Bay in the east (Figure 1). The strait, located between the mainland and Prince Edward Island, is shallow (5 - 40 m) and is an important commercial fishery area for lobster and herring.

The surface circulation of Northumberland Strait was described by Lauzier (1965) using drift bottles (Figure 2). The main pattern is a non-tidal drift of the surface waters through the strait

from the northwest and West to the southeast and east at various points. The results suggest a predorninant drift from the New Brunswick coast toward the Prince Edward Island coast that is stronger during sumrner than during autumn. Tidal currents are complex in this region due to the presence of the MZ arnphidromic point somewhere near the western end of the strait (Godin, 1979). Tidally-induced residual flows can be expected around several capes. Ln addition to local wind forcing, the strait circulation probably responds to non-local forcing from the Gulf, such as setup and relaxation during large-scale cyclonic wind events (Koutitonsky and Bugden, 1991).

De Lafontaine et al. (1991) divided the GSL into four regions based on general bathyrnetry and physical oceanographic regimes and-tried to explain the biological dynamics using historical data. Our study area is within their region III (Magdalen Shallows). It is a large, shallow (average depth 50 m) area where wind and bottom stresses become relatively important mixing agents for the water colurnn. The surface circulation is generally cyclonic and characterized by moving eddies, internal waves, and transient frontogenesis (Côté et al., 1986; Koutitonsky and Bugden, 199 1). This region has been a traditional fishing ground, yielding about 50% of the total fish landings in the Gulf (de Lafontaine et al., 1991). However, these authors note that nutrient and phytoplankton dynamics over the Magdalen Shallows and other coastal zones in the GSL are not well understood.

2.2 Remote sensing (CZCS) images

The remote sensing database we used is made up of 59 satellite images of phytoplankton pigments (chlorophyll a plus phaeophytin) recorded by the CZCS between April and September from 1979 to 1981 (Fuentes-Yaco et al., 1995) (Table 1). Satellite images are not available between October and March due to excessive noise in many of the images and interference by ice cover.

The methodology for phytoplankton pigment calculations and validations from CZCS images has been fully discussed in Fuentes-Yaco et al. (1997a). Briefly, we used raw digital CZCS images and performed corrections for radiometry, clouds, land, overflow, and molecular scattering, adapting standard methodologies for CZCS imagery. We first tried the clear water pixel method (Gordon et al., 1983) to extract the aerosol signal from the images, but with limited success. We then tried an interactive aerosol removal technique (Arnone and Laviolette, 1984) that relies on the fact that aerosol radiances at different wavelengths are inter-correlated over large areas while water and aerosol spatial structures are not. The procedure allows the operator to interactively select the optimum Angstrom coefficient for each of the first three CZCS channels. In the present study, we selected the best channel and Angstrom coefficient to identify the aerosol pattern in the GSL area and used it for the other two channels of each image. After the atmospheric corrections were completed, the phytoplankton pigments were calculated using the algorithm proposed by Gordon et al. (1983).

Babin et al. (1993) classified the GSL outside of the estuary as Case 1 waters, where optical properties are determined largely by phytoplankton (More1 and Prieur, 1977). However, they did not sarnple the southern GSL, in particular the Northumberland Strait region, where constituents other than phytoplankton are likely to affect the ocean color signal, especially during high

freshwater discharge or strong tidal rnixing. To detect waters contaminated with suspended matter, the radiances of atmospherically corrected CZCS channels (443 nm, 520 nm, and 550 nm) were transformed into reflectances (Holligan et al., 1989). These images were then converted into Intensity/Hue/Saturation (IHS) (Kruse and Raines, 1984) values and used to constnict the mean spectral signature of areas having suspended matter. A supervised maximum likelihood classification (parallelepiped method) was then used to detect pixels contaminated by suspended matter that were subsequently flagged and dropped from the image. We also estimated pigment values without the reflectance-IHS correction; however, their persistently higher pigment levels compared to those with correction were outside the range of in situ measurements.

Geometric corrections were done in two phases: first with georeferenced data included in the raw CZCS data and second with a fine-tuned correction on a stereographic projection using the World Data Bank Feature Codes. Monthly composites were calculated by overlapping pixels free from clouds, overflow, and suspended matter from each geometrically corrected image over the three years.

2.3 In situ database

The in situ physical, chemical, and biological data include measurements from 19 research missions conducted between May 1994 and October 1995 in the southern GSL (Table 2 and Figure 3) (Waite et al., 1997). Within this database, 221 records contain information on phytoplankton pigments (chlorophyll a plus phaeophytin a). Water samples were collected at anchor (when weather permitted) at regular stations. To obtain higher resolution sampling for physical, chemical, and biological water properties, two types of stations were established and tenned "water" or "CTD" stations.

Discrete water samples were collected at the surface and at 4 m; other sample depths varied and were determined following the CTD cast. Data and samples collected included profiles of conductivity, pressure, temperature, and fluorescence using a Sea-Bird Electronics SEACAT SBE 19 CTD equipped with a Chelsea Instruments Mk III Aquatracka fluorometer calibrated for chlorophyll a measurements. Fluorometric calibration was based on acidified standards. The precision and accuracy of the CTD temperatures and conductivity are better than O.Ol°C and 0.01 Slm respectively. From the surface samples, the salinity was measured using a hand held salt refractometer with a precision of I 0.2 parts per thousand (p.p.t., Cole Palmer), and the temperature measured using a Barnant 115 thermocouple thermometer with an accuracy of k1.0 O

C. Al1 water samples were stored in clean plastic containers in the dark until they were processed. Light attenuation was measured with Secchi disk.

For chlorophyll a and phaeophytin a analysis, we followed the methods suggested in the literature (Parsons et al., 1984; Yentsch et al., 1963). Pigment analysis was performed using a Perkin Elmer LS3 spectrofluorometer; readings were taken at excitation wavelengths of 408 nm and 430 nm for phaeophytin a and chlorophyll a, respectively.

To make these pigment measurements comparable with satellite evaluations, in situ data were averaged over the maximum Secchi disk penetration depth. Phytoplankton pigment data from

both CZCS and in situ measurements were grouped and averaged following the NAFO and Petrie (Petrie et al., 1996) subdivisions. Finally, we used new subdivisions based on visual analyses of monthly and interannual composite CZCS images.

3.1 CZCS monthly composite images

CZCS monthly composite images are shown in Figure-4. Northumberland Strait (NTS) was ice covered during April, and pigment concentrations were high (>1.5 mg m-3) in sunounding open waters. Pigment levels in the southern Gulf fell between April and June but rose within NTS. In July, pigments within NTS declined from June levels but remained much higher than in most of the southern Gulf. Pigments had increased everywhere in the southern Gulf by August, and the increase within NTS was particularly prominent. Pigment concentrations fell again in September.

The spatial pattern remained roughly the same throughout May and September with a bi-modal signal, as shown in the three-year composite (Figure 5). Chlorophyll levels were two to three times higher inside NTS than outside. Within NTS, pigments were relatively high in the western region in a band extending from Richibucto Cape - Carey Point to Cape Tormentine - Tryon Head. Another band of relatively high chlorophyll was found in the eastern region, between Macquarrie Point and Cape Bear. Pigment levels were lowest in an eddy-like feature in the central strait.

3.2 Regionalization

Results of in situ measurements and satellite estimations following fishing regions (ICNAFNAFO), physical oceanographic features (Petrie et al., 1996), and satellite (CZCS)- derived regions are shown in the next sections.

3.2.1 NAFO regions

ICNAF/NAFO defined three fishing regions in the GSL (4T, 4S, and 4R) in relation to the distribution and migration patterns of commercial fishes (in particular cod). Our study area overlaps NAFO's 4T subdivision, specifically 4TL, 4TH, and 4TG (Figure 6).

The analysis of both in situ and satellite data following this regionalization showed that phytoplankton biomass appeared higher in NTS than in the outside waters (Figures 5 and 7). Of the two regions outside NST, the mean in situ values in the 4TG region were higher than those in the 4TL while the CZCS images showed higher mean values in the 4TL region (Figure 7). However, the in situ data from 4TG may not be representative of the region since most of the field sarnples were taken in the Cardigan River and St. Georges Bay.

Monthly pigment variations calculated from CZCS images (1979-1981) (Figure 4) for NAFO regions 4TL, 4TH, and 4TG and in situ measurements (1994-1995) (Figure 8) confirmed the

persistently higher chlorophyll pigment levels measured in NTS compared to surrounding southern Gulf waters. It is particularly evident in the 4TH region, where both satellite and in situ data sets showed sirnilar temporal evolution between June and September. However, the 4TG region was more homogeneous and had lower CZCS-derived chlorophyll concentrations.

3.2.2 Petrie regions

Petrie et al. (1996) divided the entire GSL into 21 horizontal subareas based primarily on the topography. Their main objective was the study of salinity, temperature, and density variabilities. Our study-area is included in .Petrie regions 11 (Shediac Valley), 14 (W Northumberland Strait), and 17 (E Northumberland Strait) (Figure 9). The upper layer (O to 30 m) of Petrie regions 11, 14, and 17 have yearly mean temperatures of 5.2, 8.0, and 6.8 OC, respectively, while the yearly mean salinities are 29.89,28.90, and 29.32, respectively (Petrie et al., 1996).

The in situ and satellite-derived pigments both differentiate between the western (subarea 14) and eastern (subarea 17) regions of NTS, with higher chlorophyll values in the former (Figure 10). Monthly pigment means calculated using satellite images (1979-1981) (Figure 4) and in situ measurements (1994-1995) in Petrie regions 11, 14, and 17 (Figure 11) allowed the identification of temporal patterns. This information suggested a bi-modal signal in CZCS-derived pigment levels in regions 14 and 17 with a clear minimum in July. Maximum in situ values were measured during September in both regions.

3.2.3 CZCS-derived regions

Five regions were identified by visual analyses of the monthly and three-year composites of CZCS-derived phytoplankton pigments (Figures 4 and 5). These areas were designated as NW (northwest), W (west), C (central), E (east), and SE (southeast) (Figure 12). Details of the geographical delimitation are in Table 3.

Satellite-derived and ship-board measurements show high pigment concentrations in the western strait (region W), a shallow area between Richibucto Cape - Carey Point and Cape Tormentine - Tryon Head (Figures 5 and 13). The eastern strait (region E), between Macquarrie Point and Cape Bear, also shows high mean pigment levels in satellite images, but lower than the western area.

Monthly means of satellite (1979-1981) (Figure 4) and in situ (1994-1995) pigments in CZCS- derived regions NW, W, C, E, and SE (Figure 14) confirm the bi-modal signal in chlorophyll concentrations. Remote sensing images also showed high pigment levels during April, which was not sampled in situ.

4. DISCUSSION

This report uses both ship-based and CZCS data to describe the spatio-temporal variability of phytoplankton pigments in the southern GSL, in particular the Northumberland Strait region.

Both data sets suggested higher accumulation of algal biomass in the strait than in other areas of the southern Gulf. The western side of the strait showed higher accumulations of biomass than the eastern side.

The CZCS data suggested smaller scale features than those inferred from in situ measurements that can be linked to bathymetric and physical factors and may need to be considered in future sampling. In NTS, two areas with high phytoplankton pigment concentrations were identified around an eddy-like feature. These two areas seemed to be coupled with the bathymetry and frontal features at both ends of the strait. The high pigment values in August seemed to correspond with the highest temperatures reported in NTS. During July, the lowest salinities could be associated with strong freshwater run-off flows and may be responsible for the lowest chlorophyll concentrations measured and calculated in the southern Gulf (Petrie et al., 1996; Waite et al., 1997).

4.1 NAFO regions

Analyses using NAFO subdivisions were useful for showing that NTS has more phytoplankton biomass than the surrounding water (Figures 5 and 7). Monthly data captured the bi-modal cycle, with high pigment concentrations during early spring and late surnmer-early fall. (Figures 4 and 8). However, remote sensing images indicated that the 4TH region (Figure 6) seemed to contain more than one phytoplankton pigment region (Figures 4 and 5). In addition, CZCS pigment levels in the 4TG and 4TL region (Figure 6) were low compared to in situ means (Figures 5 and 8). In particular, 4TG region comprises a wider area than the field sarnpling station grid (Figures 3 and 6). The majority of the field samples were taken in the Cardigan River and St. Georges Bay. Consequently, the in situ data may not be representative of the 4TG region, where low pigment concentrations characterized the CZCS images.

4.2 Petrie regions

The CZCS-derived (Figures 4 and 5) and ship-board (Figures 10 and 11) values for the subareas suggested by Petrie et al. (1996) (Figure 9) show differences between the western and eastern NTS subareas. However, the fine-scale features observed with satellite images were not resolved in the Petrie subareas. The higher pigment concentrations found with both the ship-board and remote sensing methods in the western NTS area seem to be linked with the warmer and fresher waters characterizing this region (Petrie et al., 1996; Waite et al., 1997).

Specific discrepancies were observed. The spatial analysis suggests an underestimation of satellite-derived pigments (Figure 10). Despite the apparent similarity in subarea I l , the region was poorly sampled by ship-board measurements (Figure 3), with only six stations in June 1995 (Figure 11) in the area from the eastern tip of the Gaspé Peninsula to Miramichi Bay (Figure 9).

A research mission during August and September 1995 in the GSL (Gilbert et al., 1997) measured chlorophyll values in Petrie regions 11, 14, and 17 similar to those estimated by the CZCS. Both in situ and satellite pigment values measured in the Petrie regions revealed

differences in the strait from West to east. Because of the irregular distribution of the in situ sampling, the finer-scale structures noted in the CZCS imagery were not observed.

4.3 S atellite-derived regions

Grouping CZCS data into five boxes (Figure 12) confirms the visual impression of higher pigment values at both ends of NTS with lower levels in the center (Figures 4 and 5). This pattern, however, is not evident in the in situ data from the eastern areas (Figure 13). Nevertheless, the satellite-derived subdivisions seem to correspond to physical and biological features previously described in the literature. For example, in the NW region (Figure 12), the satellite images (Figures 4 and 5) clearly show a region with low to medium pigment values, suggesting the influence of eddy circulation. Lauzier (1965) identified a cyclonic eddy at the northern entrance of the strait that corresponds to this pigment structure, with a southerly drift along the New Brunswick coast and a northeasterly drift along the P.E.I. coast (Figure 2), in accord with the large-scale cyclonic motion in the Magdalen Shallows. In addition, Koutitonsky and Bugden (1991), using numerical models, showed that predominant winds from the southwest may also generate anticyclonic eddy motion in the Miscou-Shediac Valley region off the Miramichi Estuary.

The highest pigment concentrations were found in the shallow W region (Figure 12) according to both satellite images (Figures 4 and 5) and in situ measurements (Figures 13 and 14). Citarella (1980) also found high phytoplankton biomass in the northwestern basin of NTS compared to other Gulf regions. The temporal variations reported by Citarella (1980) showed low values during surnmer but higher during fall, which are similar to the variations we observed (Figure 14). The amphidromic point (Godin, 1979) and tidal mixing (Pingree and Griffiths, 1980) in this region may be responsible for high nutrient concentrations and thus for the high phytoplankton pigment production.

There are few phytoplankton pigment data for the central (C) and eastern (E) regions (Figure 12). Region C, located between the two highest CZCS-derived pigment areas, has a deeper basin (20- 25 m). There may be less mixing in this region, which is necessary to sustain phytoplankton production, or the pigment maximum may be situated below the maximum Secchi disk penetration depth and therefore below CZCS penetration. Region E includes the other shallow area within NTS, where numerical models of the M2 tide have suggested tidal mixing (Pingree and Griffiths, 1980). Satellite pigment data characterize this region as having the second largest phytoplankton biomass, but in situ measurements here were the lowest (Figure 13). It is important to note that ship-board sampling was scant in both regions (C and E) during al1 research missions (Figures 3 and 14).

Region SE (Figure 12) had the lowest CZCS pigments (Figures 4 and 5) whereas in situ concentrations were medium to high (Figure 13). The reason for these differences was probably the location of sea sampling stations (Figure 3). In agreement with our satellite data, Hargrave et al. (1985) observed high levels of phytoplankton biomass in St. Georges Bay during spring (April) but low pigment values in June-July (Figure 14). These authors suggested that nutrients are depleted rapidly following the spring bloom and nitrogen appears to be limiting for

phytoplankton growth. Finally, we observed persistently low to medium pigment concentrations in the southeastem entrance of NTS (Figures 4 and 5), where a cyclonic eddy has been reported by Koutitonsky and Bugden (1991).

In addition to the detailed synoptic view of the spatial phytoplankton pigment distribution in the southem Gulf of St. Lawrence, this study emphasizes the predominance of high pigment values in fall, both using ship-board (1994-1995) and satellite (1979-1981) measurements. Harrison and Sameoto (1996), using CZCS-derived pigment production data from 1979, suggested that the spring bloom dominated the production cycle on the Labrador and Newfoundland shelves but that the fall bloom dominated further south, m the Grand Bmks, Smim Shelf, and Gulf of Maine. Fuentes-Yaco et al. (1997b), using remote sensing images, showed the influence of freshwater runoff during spring and the seasonal westerly wind variability during August and September on phytoplankton pigment distribution in the Gulf. Finally, Fuentes-Yaco (1997) also suggested the importance of late summer-early fall storms in the western Gulf as a possible mechanism for controlling the pigment distribution during this season.

5. CONCLUSIONS AND RECOMMENDATIONS

Both ship-board and CZCS data show similar broad synoptic views of the spatial distribution of phytoplankton pigments in the southern Gulf of St. Lawrence. The use of CZCS images gives a more detailed view, suggesting five regions based on the visual analysis of remote sensing images. The phytoplankton pigment concentrations evaluated in these regions seemed to vary temporally in agreement with in situ measurements according to a bi-modal cycle, with high and comparable pigment levels during both spring and late summer-early fall.

Several recornrnendations arise from this preliminary analysis:

The links between in situ phytoplankton pigments (chlorophyll a plus phaeophytin a), physical characteristics (Secchi depth, salinity, temperature, density, mixed-layer depth), and nutrients [NO2 -t NO3, PO4, Si(OH)4] should be studied in relation to the regions identified using CZCS-derived phytoplankton pigment measurements.

The regionalization proposed from this remote sensing study should be confirmed or modified using new satellites (SeaWIFS) and with sensors of higher spatial resolution (CASI).

The grid of stations in future oceanographic research missions should be modified to adequately sarnple the different subareas identified using CZCS images.

6. ACKNOWLEDGEMENTS

The authors thank M. Pierre Larouche (Remote Sensing Laboratory, Maurice Lamontagne Institute, DFO) and Dr. Yves Gratton (INRS-Océanologie), who kindly allowed the use of their computer equipment. Thanks are expressed to Drs. M. Gosselin and M. Starr for their valuable cornments during the review of this document.

7. REFERENCES

Arnone, R.A., and P.E. Laviolette. 1984. A method of selecting optimal Angstrom coefficients to obtain quantitative ocean color data from Nimbus-7 CZCS. Proceedings of SPIE, Ocean Optics 7,489: 187-194.

Babin, M., J.-C. Therriault, L. Legendre and A. Condal. 1993. Variations in the specific absorption coefficient for natural phytoplankton assemblages: impact on estimates of primary production. Limnol. Oceanogr. 38: 154-177.

Citarella, G. 1980. Production phytoplanctonique dans le bassin nord-ouest du Détroit de Northumberland (NW Atlantique). Bot. Mar. 23: 173-177.

Côté, B., M. 1. El-Sabh and R. De la Durantaye. 1986. Biological and physical characteristics of a frontal region associated with the arrival of spring freshwater discharge in the southwestern Gulf of St. Lawrence. In: The role of freshwater outflow in coastal marine ecosystems. Edited by S. Skreslet. NATO AS1 Series, Vol. G7, Springer-Verlag Inc., Berlin, pp. 261-269.

de Lafontaine, Y., S. Demers and J. Runge. 1991. Pelagic food web interaction and productivity in the Gulf of St. Lawrence: A perspective. In: The Gulf of St. Lawrence: small ocean or big estuary? Edited by J.-C. Therriault. Can. Spec. Publ. Fish. Aquat. Sci. 113: 99-124.

Fuentes-Yaco, C. 1997. Télédétection de la biomasse phytoplanctonique dans le golfe du Saint- Laurent, Canada: analyse des données du capteur Coastal Zone Color Scanner, de 1979 à 1981. Ph. D. Thesis, Université du Québec à Rimouski, Rimouski, Québec, Canada, 267 PP.

Fuentes-Yaco, C., P. Larouche, A. Vézina, C. Vigneau and M. Gosselin. 1995. Catalogue of phytoplankton pigment images from the Gulf of St. Lawrence: Coastal Zone Color Scanner data from 1979 to 1981. Can. Data Rep. Hydrogr. Ocean Sci. 135: v + 91 pp.

Fuentes-Yaco, C., A.F. Vézina, P. Larouche, C. Vigneau, M. Gosselin and M. Levasseur. 1997a. Phytoplankton pigment in the Gulf of St. Lawrence, Canada, as determined by the Coastal Zone Color Scanner. Part 1. Spatio-temporal variability. Cont. Shelf Res. 17: 1421-1439.

Fuentes-Yaco, C., A.F. Vézina, P. Larouche, Y. Gratton and M. Gosselin. 1997b. Phytoplankton pigment in the Gulf of St. Lawrence, Canada, as determined by the Coastal Zone Color Scanner. Part 2. Multivariate analysis. Cont. Shelf Res. 17: 1441-1459.

Gilbert, D., A.F. Vézina, B. Pettigrew, D.P. Swain, P.S. Galbraith, L. Devine and N. Roy. 1997. État du golfe du Saint-Laurent: conditions océanographiques en 1995. Rapp. tech. can. hydrogr. sci. océan. 191: xii + 113 pp.

Godin, G. 1979. La marée dans le Golfe et l'Estuaire du Saint-Laurent. Naturaliste cm., 106: 105- 121.

Gordon, H.R., D.K. Clark, J.W. Brown, O.B. Brown, R.H. Evans and W.W. Broenkow. 1983. Phytoplankton pigment concentrations in the Middle Atlantic Bight: comparison of ship determination and CZCS estimates. Appl. Opt. 22: 20-36.

Hargrave, B.T., G.C. Harding, K.F. Drinkwater, T.C. Lambert and W.G. Harrison. 1985. Dynarnics of the pelagic food web in St. Georges Bay, southern Gulf of St. Lawrence. Mar. Ecol. Prog. Ser. 20: 221-240.

Harrison, W.G., and T. Platt. 1995. The state of the pelagic ecosystem: a synoptic assessment based on satellite remote-sensing. Fisheries Oceanography Cornrnittee, Working Paper 95/13.

Harrison, W.G., and D. Sameoto. 1996. Incorporating ecosystem information into the fisheries assessment process: Can we develop a quantitative "Plankton index"? Fisheries Oceanography Cornmittee, Working Paper 961 12.

Holligan, P.M., T. Aanip and S.B. Groom. 1989. The North Sea Satellite Color Atlas. Cont. Shelf Res. 9: 665-765.

Koutitonsky, V.G., and G.L. Bugden. 1991. The physical oceanography of the Gulf of St. Lawrence: A review with emphasis on the synoptic variability of the motion. In: The Gulf of St. Lawrence: small ocean or big estuary? Edited by J.-C. Theniault. Can. Spec. Publ. Fish. Aquat. Sci. 113: 57-90.

Kruse, F.A., and G.L. Raines. 1984. A technique for enhancing digital colour images by contrast stretching in Munsel colour space. Proceedings of the International Symposium on Remote Sensing for Exploration Geology, Colorado Springs, Colorado, April 16-19, p. 755-760.

Lauzier, L.M. 1965. Drift bottle observations in Northumberland Strait, Gulf of St. Lawrence. J. Fish. Res. Board Can. 22: 353-368.

Morel, A., and L. Prieur. 1977. Analysis of variations in ocean color. Limnol. Oceanogr. 22: 709- 722.

Pingree, R.D., and D.K. Griffiths. 1980. A numerical mode1 of the M2 tide in the Gulf of St. Lawrence. Oceanol. Acta, 3: 221-225.

Parsons, T.R., Y. Maita and C.M. Lalli. 1984. A manual of chernical and biological methods for seawater analysis. Pergamon Press, New York, 173 p.

Petrie, B., K. Drinkwater, A. Sandrom, R. Pettipas, D. Gregory, D. Gilbert and P. Sekhon. 1996. Temperature, salinity and sigma-t atlas for the Gulf of St. Lawrence. Can. Tech. Rep. Hydrogr. Ocean Sci. 178: v + 256 pp.

Waite, L.E., J.C. Smith, P. Cormier and K. Pauley. 1997. Biological, chemical and physical oceanographic conditions in the Southem Gulf of Saint Lawrence, 1994. Can. Data Rep. Fish Aquat. Sci. 1027: viii + 439 pp.

Yentsch, C.S., and D.W. Menzel. 1963. A method for the determination of phytoplankton chlorophyll and phaeophytin by fluorescence. Deep-Sea Res. 10: 221-23 1.

Table 1. Dates and number of remote sensing images.

Month Dates Number of 1979 1980 198 1 images

A P ~ 25 2 , 4 11 4

May 19, 20, 23 20,2 1, 30 25 7

Jun 8, 17, 19, 6 7 , 13, 2, 8, 13,24, 20,26 22,23,29 27,30

17

Jul 4,7, 10, 19,

1,4, 18 2,3,9, 17, 21,22 24,26

15

18,31 1,20,22,

Aug 28,30 20, 25, 30 10

Sep 5, 16 9, 13 10,27 6

Number of images

19 21 19 59

Table 2. Research missions in the southern Gulf of St. Lawrence between May 1994 and October 1995.

SURVEY AREA or SURVEY NUMBER DURATION

SHIP NAME begin end

Miramichi Bay, N. B. - 25-May-94 Miramichi Bay, N. B. - 22-Jun-94 Miramichi Bay, N. B. - 21-Jul-94 - 22-Jul-94 Miramichi Bay, N. B. - 23-Aug-94 Miramichi Bay, N. B. - 28-Sep-94

Bouctuche Bay, N. B. Bouctuche Bay, N. B.

Cardigan, P.E.I. Cardigan, P.E.I.

CSS Ophilio 94-0 1 19-Jul-94 - 28-Jul-94

CSS Navicula 94-02 3-Aug-94 - 17-Aug-94

CSS Navicula 94-03 06-Sep-94 - 10-Sep-94

CSS Navicula 94-04 0 1 -Nov-94 - 06-Nov-94

CSS Ophilio 95-01 05-Jun-95 - 12-Jun-95

CSS Calanus 95-02 28-Jun-95 - 06-Jul-95

CSS Navicula 95-03 29-Jun-95 - 08-Jul-95

CSS Navicula 95-04 8-Aug-95 - 15-Aug-95

CSS Navicula 95-05 07-Sep-95 - 14-Sep-95

CSS Navicula 95-06 1 1 -0ct-95 - 19-0ct-95

Table 3. Local names and coordinates of the five regions in the study area identified using the CZCS images.

CZCS REGION

LOCAL LAT. LONG. NAME (ON) ("W)

N W (Northwest)

W (West)

C (Central)

E (East)

SE (Southeast)

Miscou Point (N.B.) North Cape (P.E.I.) Carey Point (P.E.I.) Richibucto Cape (N.B.)

Richibucto Cape (N.B.) Carey Point (P.E.I.) Tryon Head (P.E.I.) Cape Tormentine (N.B.)

Cape Tormentine (N.B.) Tryon Head (P.E.I.) Rice Point (P.E.I.) Macquarrie Point (N.S .)

Macquarrie Point (N.S .) Rice Point (P.E.I.) Cape Bear (P.E.I.) Arisaig Point (N.S.)

Arisaig Point (N.S.) Cape Bear (P.E.I.) East Point (P.E.I.) Black Point (N.S.)

Figure 1. The southern Gulf of St. Lawrence. PEI: Prince Edward Island; 1 : Miscou Point; 2: Mirarnichi Bay; 3: North Cape; 4: Richibucto Cape; 5: Carey Point; 6: Cape Tormentine; 7: Tryon Head; 8: Rice Point; 9: Cape Bear; 10: Macquarrie Point; 1 1: Arisaig Point; 12: St. Georges Bay; 13: Black Point; 14: East Point.

Figure 2. Residual surface circulation in Northumberland Strait inferred from drift bottles (from Lauzier, 1965).

Figure 3. Sampling locations for research missions between July 1994 and October 1995. Survey numbers are given in Table 2.

Figure 4. CZCS composite images of phytoplankton pigments (mg m") from April to September (1979-198 1) in the southem Gulf of St. Lawrence.

Figure 5. Global average of CZCS composite images of phytoplankton pigments (mg m") (1979- 1981) in the southern Gulf of St. Lawrence.

I GLOBAL AVERAGE (1979-1981) Phytoplankton Pigrnt (mg m**3)

Figure 6. NAFO regions 4TL, 4TH, and 4TG

CZCS in situ

64

Figure 7. Mean of CZCS (1979-198 1) and in situ (1994-1995) pigments (mg m-3), standard enors, and number of data for NAFO regions 4TL, 4TH, and 4TG.

in situ

APR MAY JUN JUL AUG SEP

Figure 8. Monthly mean of CZCS (1979- 198 1) and in situ (1 994- 1995) pigments (mg m-3), standard errors, and number of data for NAFO regions 4TL, 4TH, and 4TG.

Figure 9. Petrie regions 1 1 (Shediac Valley), 14 (W Northumberland Strait), and 17 (E Northumberland Strait).

CZCS in situ

Figure IO. Mean of CZCS (1979-1981) and in situ (1994-1995) pigments (mg m"), standard errors, and number of data for Petrie regions 1 1, 14, and 17.

CZCS

W." ,

APR MAY JUN JUL AUG SEP

3.0 -

Figure 1 1. Monthly mean of CZCS (1 979- 198 1) and in situ (1 994- 1995) pigments (mg m-3), standard errors, and number of data for Petrie regions 1 1, 14, and 17.

2.5 -

in situ

in situ

0.0 I APR MAY JUN JUL AUG SEP

Figure 14. Monthly mean of CZCS (1979-198 1) and in situ (1994- 1995) pigments (mg m"), standard enors, and number of data for CZCS-derived regions NW,W, C, E, and SE.