SCI. MAR., 60 (Supl. 2): 117-140 SCIENTIA MARINA 1996 THE EUROPEAN ANCHOVY AND ITS ENVIRONMENT, I. PALOMERA and P. RUBIÉS (eds.)

The spawning environment of the Bay of Biscay anchovy (Engraulis encrasicolus L.)*

LORENZO MOTOS, ANDRÉS URIARTE and VICTOR VALENCIA AZTI. InstitutoTecnológico Pesquero y Alimentario, Av. Satrústegui 8, 20008 Donostia-San Sebastián, Basque Country, Spain.

SUMMARY: The spawning season of the Bay of Biscay anchovy population extends from March to August, although most of the reproductive effort is made during May-June. Spawning is triggered by the warming up of surface waters and coin- cides with the maximum rate of temperature rising and the onset of stratification. Preferential sea surface temperatures are in the range of 14 - 18° C. This population shows a relatively short peak spawning period that seems to be an adaptation to stable and predictable environmental conditions in the SE Bay of Biscay. At the beginning of the spawning season, usually in April, anchovy concentrate in the SE corner of the Bay of Biscay. Later, in May, anchovy spawning is prevalent in the SE Bay of Biscay. As the spawning season progresses, in June, distinct coastal and oceanic spawning areas appear. Simul- taneously, the geographic range of spawning extends north and westwards. The Bay of Biscay anchovy selects stable habi- tats related to river plumes, shelf edge fronts and oceanic eddies, where increased biological production potentially occurs. i) There are low salinity plumes related to main river estuaries (Gironde, Adour and Bidasoa); ii) structures related to shelf break fronts - French shelf to the North of 45 N and southern Bay of Biscay-; and iii) oceanic eddies of slope waters (SWOD- DIES), in oceanic sites outside the continental shelf. The population shows a density-dependent selection of the spawning habitat. The spawning areas associated with the Estuaries of Gironde, Adour and Bidasoa, in the SE region of the Bay of Biscay, seem to form a refuge basin favourable to the maintenance of this population of anchovy even in adverse environ- mental conditions.

Keywords: Anchovy, Engraulis encrasicolus, spawning, environment, Bay of Biscay.

RESUMEN: CONDICIONES AMBIENTALES DE LA PUESTA DE LA ANCHOA DEL GOLFO DE VIZCAYA (ENGRAULIS ENCRASICOLUS L.). – La puesta de la anchoa del Golfo de Vizcaya se extiende de Marzo a Agosto, aunque la mayor parte del esfuerzo repro- ductivo tiene lugar en Mayo y Junio. El calentamiento estacional de las aguas superficiales actúa como un impulsor de las actividades reproductivas de la anchoa, que coinciden con las tasas máximas de calentamiento y el comienzo de la estrati- ficación vertical, en un rango preferencial de 14 a 18°C. Esta población muestra un periodo de puesta relativamente corto, adaptado a las condiciones ambientales estables y predecibles, prevalentes año tras año en el Sudeste del Golfo de Vizcaya. Al comienzo de la estación de puesta, generalmente en abril, la anchoa se concentra en un área restringida de la región SE del Golfo. Posteriormente, en mayo, la puesta se extiende por toda esta región. A medida que la temporada avanza, en junio, se pueden distinguir áreas de puesta diferenciadas costeras y oceánicas, respectivamente, a la vez que se extiende el rango geográfico de puesta hacia el N y hacia el W. La anchoa del Golfo de Vizcaya parece seleccionar un hábitat relacionado a condiciones de estabilidad vertical y de producción biológica como son i) plumas de agua de baja salinidad provenientes de los estuarios más importantes del sur del Golfo; ii) estructuras relacionadas con frentes de talud y plataforma; y iii) remoli- nos oceánicos de aguas de talud. Esta población de anchoa muestra una selección del hábitat dependiente de la abundancia. El área de puesta asociado a los grandes estuarios del sur del Golfo de Vizcaya parecen conformar una cuenca favorable para su mantenimiento aun en condiciones ambientales adversas.

Palabras clave: Anchoa, Engraulis encrasicolus, puesta, ambiente, Golfo de Vizcaya.

*Received December 20, 1995. Accepted September 25, 1996.

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 117 INTRODUCTION anchovy eggs and larvae on the French side of the Bay of Biscay, with occasional coverage of the Iber- Timing of reproduction seems to be a mechanism ian shelf (Cantabrian Sea). Those authors consistent- to synchronize the occurrence of larval stages with ly found two main spawning areas in waters off the the optimal phase of the annual plankton cycle southern Bay of Biscay coast. One in shallow waters (Cushing, 1975) or, more generally, to conditions off the French coast and the other linked to the shelf favourable to survival of early life history (ELH) break in the southeastern tip of the Bay. They found stages either by food-chain mechanisms (starvation- that in summer (July), anchovy spawning ranges predation) (Lasker, 1978), spatial mechanisms from the southeastern region of the Bay to the north- (retention vs advection) (Iles and Sinclair, 1982) or ern-most areas, with a limit in Penmarch (Britanny). both. Selection of a specific spawning site is related They were not able to identify any clear relationship to good feeding conditions for adults (Blaxter and between oceanographic conditions and spawning Hunter, 1982). Several studies on the spatio-tempo- distribution and intensity. They concluded that ral distribution of anchovy eggs and larvae can be anchovy spawning might be related to geographic found in the literature. However, very little has been sites more than to environmental features. Santiago published regarding the relationship between Bay of and Eltink (1988) and Santiago and Sanz (1992) Biscay anchovy spawning and concurrent oceano- found a similar distribution pattern in different years. graphic features. Two main spawning areas were identified: a first one According to the literature, the temporal distrib- along the shelf break and a second one along the ution of the Bay of Biscay anchovy eggs and larvae French coast. Those authors suggested that according indicates that this species spawns from April to July- to different works (reviewed in Santiago and Eltink, August (Arbault and Lacroix-Boutin, 1977; Suau op. cit.), those areas might correspond to spawning and Vives, 1979; Villegas, 1979; Dicenta, 1984; of different fractions of the anchovy population. 2+ Solá, 1987; Santiago and Eltink, 1988; Valencia et year-old anchovy would spawn along the shelf break, al., 1989; Santiago and Sanz, 1992; Motos and Uri- whereas 1-year-old anchovy would spawn in coastal, arte, 1991, 1992, 1993). Studies of spawning sea- inshore areas of the Bay. Data from the fishery in the sonality based on the sexual maturity cycles on second fortnight of May 1988 showed that the com- anchovy adults yielded similar results (Fage, 1920, position of catches in areas related to the shelf break Arné, 1931; Furnestin, 1945; Andréu, 1950; Cort et (oceanic areas) presented a high proportion of 1- al., 1977; Cendrero et al., 1981; Sanz and Uriarte, year-old anchovy, suggesting that some spawning of 1989; Lucio and Uriarte, 1990; Junquera, 1991). A the youngest fraction of the population may take later maturation of 1-year-old anchovies during the place in the outer shelf or even beyond the shelf reproductive season has also been suggested. Sever- break (offshore) (Martín and Uriarte, 1989). In that al works mentioned the existence of two discrete respect, Cort et al. (1976) reported the presence of spawning periods in the Bay of Biscay anchovy. The small anchovy in oceanic areas in June, saying that youngest mature individuals (1-year old) spawn fishermen regarded it as an unusual event. mainly in June. The 2-year old and older individuals This paper analyses the spatio-temporal patterns spawn earlier, in late April-May (Fage, 1920 and of spawning of the Bay of Biscay anchovy popula- Arné, 1930 cited by Andréu, 1950; Anon., 1954; tion based on egg cruises carried out in May-June Cort et al., 1976; Cendrero et al., 1981; Santiago from 1989 to 1992 for assessment purposes (Motos and Eltink, 1988). However, Lucio and Uriarte and Santiago, 1990; Motos et al., 1991; Motos and (1990), studied the evolution of the Gonadosomatic Uriarte, 1991, 1992, 1993). The reproductive activi- Index (GSI) and the macroscopic maturation of ties of adult spawners were tracked through the dis- anchovies. They found that the smallest anchovies tribution and abundance of anchovy eggs. The envi- reach maturity somewhat later than bigger ronmental features where the eggs occured and anchovies but, nevertheless, they observed that most developed are described, and the potential implica- of the population is already fully mature during the tions for recruitment of this short period of the peak spawning time in May-June. anchovy life history are explored. The literature on Data on the spatial distribution of the Bay of Bis- the relationships between spawning of this popula- cay anchovy have been reported by several authors. tion and enviromental features has been scarce Arbault and Lacroix (1971) and Arbault and (Cendrero, 1994; Massé et al., 1995; Koutsikopou- Lacroix-Boutin (1977) studied the distribution of los and Le Cann, 1996). The implications for

118 L. MOTOS et al. recruitment of both the spawning habitat described TABLE 1. – Timing of anchovy egg cruises carried out from 1989 to 1992. herein and the subsequent oceanographic conditions experimented with anchovy early life history stages Year Period no. samples References have been analysed by Borja et al. (1996). 1989 10 - 21 May 337 Motos and Santiago, 1990 14 - 25 June 437 1990 4 - 15 May 525 Motos and Uriarte, 1991 METHODOLOGY 29 May - 15 June 536 1991 16 May - 7 June 538 Motos and Uriarte, 1992 Field work 1992 16 May - 13 June 651 Motos and Uriarte, 1993

The temporal distribution of anchovy eggs and larvae was studied from January 1987 to December spawning area of the anchovy population, and (2) to 1989. A cross-shelf transect located in front of San sample more intensively the main spawning centres Sebastian was sampled using a BONGO net month- (Motos, 1994). ly during this time period. Details on the sampling Egg sampling was carried out using a CALVET methodology are given by Solá et al. (1990). type “PAIROVET” net of 150 µm mesh (Smith et The spatial distribution of anchovy spawning al., 1985) in vertical hauls to a maximum depth of was studied through egg surveys carried out from 100 metres, or 5 metres above the sea bottom in 1989 to 1992 (Table 1). Attempts were made to sam- shallower waters. A lowering and hauling rate of 70 ple the potential spawning area of the Bay of Biscay m min-1 was intended, and the net was stopped dur- anchovy as derived from previous studies (Anon., ing 10 seconds at the maximum depth for stabiliza- 1988; Santiago and Eltink, 1988): i.e., the area tion. The net was equipped with digital flowmeters between the Spanish and French coasts, to the south (G.O. 2030). of 47° N Lat., and to the east of 5° W Long., reach- Samples were preserved immediately after col- ing up to 30 miles beyond the shelf break (Fig.1). lection in a 4 % borax buffered formaldehyde solu- The sampling scheme was focused to (1) enclose the tion in filtered sea water.

FIG. 1. – Map of the Bay of Biscay with the main geographical refereces mentioned in the text.

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 119 Sea surface temperature was recorded at every RESULTS station and sea surface salinity at every third sta- tions. Salinity and temperature profiles by means of Seasonality SEABIRD ‘19’ or ‘25’ CTD casts were obtained in a representative subset of stations. Figure 2 shows the values of mean monthly egg and larval abundances averaged over the period of Sample processing study 1987-1989. In San Sebastián, anchovy spawn- ing continues from April to September with a peak In the laboratory, BONGO (temporal distribu- in May-June. In the same figure, the monthly rate of tion) and PAIROVET (spatial distribution) samples warming up of surface temperature is also shown. In were sorted for all fish eggs and larvae. Anchovy general, the timing of spawning in front of San eggs and larvae were identified and counted. Indi- Sebastian matches the months of substantial warm- viduals per haul were transformed to density either ing: e.g. April to July. Furthermore, for the period in numbers under 10 m2 of sea surface in the case of 1987 to 1989, maximum warming was recorded in BONGO samples or 0.05 m2 in the case of May (3.1 °C) and June (3.9 °C), coinciding with the PAIROVET samples (Smith and Richardson, 1977). months of peak spawning. Flowmeter readings were used for this purpose. A 100% filtration efficiency was assumed (Posgay et Spatial distribution and spawning environment al., 1980; Smith et al., 1985). When studying the temporal distribution of 1989 anchovy eggs and larvae, monthly values of egg and larval abundance were estimated. A monthly aver- In May, the spawning area was restricted to south age figure was set equal to the arithmetic mean of of 45° 45’ N and to east of 4° 30’ W (Fig. 3). Within the station values achieved each particular month. this area, two main spawning centres were identified: The data is presented on a logarithmic scale to show the area of shallow waters near the Gironde Estuary the temporal borders of spawning activity. Monthly and outer shelf and shelf break waters off the area sea surface temperature was estimated in a similar from the Spanish coast to Arcachon Bay (44° 40’N). way. Additional sea surface temperature was used Highest egg abundances were found in the Cap Bre- from the daily time series taken at San Sebastian’s ton canyon region and in waters off the Basque coun- Aquarium. try and Cantabria, up to 3° 15’ W.

FIG. 2. – Temporal evolution of anchovy egg abundance and surface temperature in front of San Sebastián (monthly averages for the period 1987-1989). SST is presented for the transect (FIELD SST) and for the San Sebastián Aquarium series (SST). The monthly rate of warming is extracted from the increase (in degrees centigrades) in mean surface temperature between the current and the previous month.

120 L. MOTOS et al. FIG. 3. – Distribution of anchovy eggs (top), surface temperature (midle) and surface salinity (bottom) during the cruise carried out in May (left) and June (right) 1989.

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 121 The northern spawning area was related to the French continental shelf , namely in waters off the Gironde estuary outflow, which this year extended Gironde Estuary and around 44°N in front of the well offshore over the platform. The 35 psu isoha- Adour river. In addition, scattered egg patches line reached well beyond the 100 m contour in this occurred along the continental shelf break, with area and enclosed most of this egg patch (Fig. 3). highest concentrations along the edge of the Spanish There, SST (Sea Surface Temperature) were over shelf, mainly in front of Cape Matxitxako. 15°C and vertical stratification was already apparent The scarce and scattered spawning centres found (Fig. 4a). A less important egg patch was located in in June over the French shelf occurred in an envi- Fer à Cheval, in waters of higher salinity, tempera- ronment characterised by a strong influence of low tures below 15° C and vertical profiles showing salinity and high temperature in surface waters (Fig. weak thermoclines (Fig. 4 b). 3). Surface salinities were below 35 psu all over the Much of the spawning in May was located in lat- French shelf, and even beyond it, from the Basque itudes to the south of Arcachon Bay. In this south- coast up to 47° N, affecting a wider area than in ernmost part of the Bay of Biscay, the major egg May. Spawning patches were located in very low concentrations were related to fresh water plumes salinity, generally below 34 psu, both in the north coming from the Adour and Bidasoa rivers. These and in the south. However, extended areas of low plumes reached areas well beyond the 200 m depth salinity, high temperature waters in front of the contour. Egg abundances were maxima in the outer Gironde Estuary were devoid of eggs. shelf and beyond it in the area of Cap Breton. There, Cantabrian sea waters were slightly colder, vertical TS profiles presented high surface tempera- showing an east-west cooling gradient. Anchovy ture, low surface salinity and marked thermoclines eggs were scarce and occurred in several patches, (Fig. 4d). either in very coastal or in oceanic waters beyond The spawning areas located along the 200 m the shelf (Fig. 3) . The southern “oceanic” spawning depth contour of the French shelf edge between 44° areas, as well as the egg patches located close to the and 44° 30’ N and in front of the Cantabric coast shelf edge at 47 º N show spawning concentrations presented surface salinities above 35.2 psu and SST not clearly linked to waters influenced by any river values between 16° and 17°C. There, vertical gradi- plume, but namely to the presence of well formed ents were weaker, although the strong surface warm- thermoclines (Fig. 4). ing up recorded during the cruise led to marked ther- moclines (Figs. 4c and e). 1990 The distribution pattern of eggs found in May dramatically changed in June (Fig. 3). Instead of a In May, anchovy eggs were abundant all along continuous spawning concentrated in southern Bis- the shelf from the Spanish Coast up to 46° 30’N, cay, several discrete egg patches were found dis- with density peaks in waters off the Gironde Estuary persed over the whole sampling area. Some egg area, from 45° to 46° 15’N (Fig. 5). In the southern patches were found close to the coasts, over the part of the Bay, egg densities over the shelf were

FIG. 4. – Summary of temperature and salinity profiles found in 1989. May: a) Gironde Estuary, b) Fer à Cheval, c) Slope in front of the French Coast, d) Cap Breton, and e) Cantabrian sea. June: f) Gironde Estuary, g) Cantabrian sea.

122 L. MOTOS et al. FIG. 5. – Distribution of anchovy eggs (top), surface temperature (midle) and surface salinity (bottom) during the cruise carried out in May (left and in June (right) 1990.

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 123 FIG. 6. – Temperature (bold lines) and Salinity profiles in 1990. May: a) coastal zone influenced by river plumes; b) oceanic area near Cap Breton. June: c-d-e-f) Stations located in a transect perpendicular to the Cantabrian coast at 4°W. Stations e and f are located close to the core of an SWOddie (see text). much lower or nil to the west of 3° 30’W, although were cooler due to doming of the seasonal thermo- distinct egg patches were still noticeable beyond the cline. Very high egg abundances (>1000 eggs/m2) shelf break all along the Cantabrian Sea. were found in the stations close to the core of the The coastal spawning centres were clearly asso- eddy. Vertical temperature and salinity profiles in ciated in May 1990, as they were in 1989, to waters stations inside the eddy were different from the pro- influenced by river outflows: in the north, the main files found in surrounding waters and show a egg concentrations were enclosed by waters with marked pycnocline at a depth of 15-20 m (Figs. 6e surface values of temperature over 17°C and salini- and f). A more scarce, uniform egg patch extended ty below 35 psu. In the south, they were enclosed by in this oceanic water further to the west, with the the 18°C isotherm and the 34.6 psu isohaline (Fig. borders being located in waters of surface tempera- 5). Vertical temperature and salinity profiles in this ture close to 15°C. area showed the characteristic strong stratification In June, the egg distribution showed an expan- of runoff water plumes (Fig. 6). sion of the spawning centers both to the north and to The extension of the influence of the continental the west of the borders of the spawning area found fresh waters over the Bay of Biscay was weaker in in May. In the northern shelf area (north of 45° N) May 1990 than that observed in 1989: high salinity (Fig. 5) anchovy eggs occurred in two distinct areas: waters (36 psu), were prevalent over the French a coastal area, in waters of less than 100 m depth, continental slope and outer shelf areas (Fig. 5). In and an oceanic area, along the shelf break. These addition, the 35 psu isohaline was restricted to the two areas were separated by an intermediate gap shelf area within 100 m water depth in the north and devoid of eggs. The distribution of anchovy eggs 200 m water depth in the south, except for the extended more to the north than in May. In the region of Cap Breton canyon, where low salinity region to the south of 44° 30’N, the biggest abun- waters extended beyond the shelf break. Neverthe- dances occurred in coastal areas and decreased sea- less, with the exception of the neighbouring Cap- ward. Anchovy eggs in coastal areas were relatively Ferret, anchovy spawning extended over most of abundant to the west of 3° 30’, showing a similar the shelf from 47°N to waters off the Cantabrian westward extension of spawning. In general, Sea. Important egg concentrations were found far anchovy eggs still appeared in the borders of the away from the coast in the latter region. One peak sampling area, namely in the northern shelf break, was centred in 44° N and 3° 50’ W, coinciding with and in Central Cantabrian waters. an anticyclonic eddy detected during the cruise The major coastal spawning centres were associ- (Fig. 5) and already described by Pingree and Le ated with waters at low surface salinity (below 35 Cann (1992). Surface temperatures of outer eddy psu) (Fig. 5). Those patches were mainly restricted waters reached over 17°C, whereas core waters to the shelf. They were located within the 100 m

124 L. MOTOS et al. FIG. 7. – Distribution of anchovy eggs (top), surface temperature (midle) and surface salinity (bottom) during cruises carried out in May-June 1991 (left) and 1992 (right).

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 125 depth contour in the north and reached the shelf For instance, coastal waters of the western Cantabri- break in the south. an Sea showed surface temperatures between 13.5º A SST peak (16°C) was found in waters off the and 14ºC and eggs occured as far as the influence of northernmost shelf break area, which was coincident the continental fresh waters was noticeable (Fig. 7). with the spawning centres located along the shelf This influence decreased to the west of 4ºW, result- break from ‘Fer á Cheval’ towards the north. ing in the occurrence of higher salinity, lower tem- In the Cantabrian Sea, anchovy eggs occurred perature (below 14ºC) surface waters. These fea- over and beyond the shelf, with abundance maxima tures matched with the end of the anchovy spawning in waters below 18°C and salinity about 35.5 psu. An area in coastal waters of the Cantabrian sea and oceanic patch extended seaward at 4° W, in the same coincided with an intrusion of more saline waters to area where an oceanic eddy was observed in May. subsurface layers (Fig. 8d). A transition area between the northern and south- 1991 ern peak spawning regions was located over the French continental shelf at latitudes 44º 40’- 44°50’. Two major spawning centers appeared in 1991 In that area, sea surface waters were cooler and (Fig. 7). The highest egg concentrations were found much more saline than in peak spawning areas, in the Cap Breton canyon area, close to the shelf either to the north or to the south. The influence of break. A second abundance peak was located in the freshwater run-off was reduced at this latitude (Figs. Gironde Estuary region, between 45° and 46° N, 7 and 8c). An intrusion of central North Atlantic within the 50-60 m depth zone. This egg patch waters was apparent in the area, reaching up to 30 m reached the shelf break in the area within latitudes below surface (Valencia, 1994). 45° and 45° 30’N (Fer à Cheval). Another notice- Finally, it should be noted that some spawning able egg patch occurred in oceanic waters of the areas were not strongly influenced by continental Cantabrian Sea between Cape Matxitxako (2° 45’W) freshwater, although sometimes weak haloclines and Cape Ajo (3° 30’W). appeared (SSS [Sea Surface Salinity] above 35psu). Spawning in the southern area around Cap Bre- Examples of such spawning areas can be found in ton occurred at SST ranging from 14º to 18ºC, with oceanic waters off the Cantabrian shelf (at 3ºW) and in maximum spawning activities enclosed by the 15º the northern area over the shelf edge around Fer à and 17ºC isotherms. Spawning in the northern area cheval. In both cases, a mixed water layer (temperature around the Gironde Estuary occurred at SST ranging about 14ºC) from the surface down to 20 m was found, from 14.5º to 16ºC during the first part of the survey, followed by a sharp thermocline (Figs. 8c and d). whilst it increased to 16º-17ºC in the second part. Again, these two major spawning areas presented 1992 relatively low surface salinities (Fig. 7) indicating the influence of continental fresh water run-off. A major spawning centre was located in the These spawning areas were characterized by Gironde Estuary area, between latitudes 45° and 46° strong vertical gradients in temperature and salinity, N and within the 80 m isobath (Fig. 7). Another rel- resulting from the spring warming of sea surface evant spawning centre occurred over the French waters on areas under the influence of the outflows continental shelf to the North of the Cap Breton from the Garonne, Adour and Bidasoa rivers (Figs. canyon area, mainly in water depths from 80 m to 8a and b). These characteristics were restricted to 200 m. Other less abundant egg patches were found surface waters. Temperatures between 12º and along the shelf break to the North of Cap Ferret (Fer 13.5ºC were recorded at depths around 20 m in all à Cheval). Finally, a less important spawning centre areas of the southeastern part of the Bay of Biscay, was identified along the Cantabrian sea. It extended whereas salinities ranged between 35 and 35.5 psu between 3° and 4° W along the coast and to the lim- at the same depth. Neither the warming nor the its of the sampling area (5° W and 44° 30’N) in freshwater influence were noticeable at 50 m depth, oceanic waters. where temperatures ranged between 11.7º and The hydrological features of the two major 12.1ºC and salinities around 35.60 -35.65 psu. spawning areas located over the French shelf clear- The limits of the spawning areas generally corre- ly reflect the influence of continental freshwater in sponded to areas of low SST (lower than 13.5º or their upper layers, mainly coming from the 14ºC) and/or strong intrusion of high salinity waters. Gironde and Adour Estuaries respectively (Fig. 7).

126 L. MOTOS et al. FIG. 8. – Distribution of temperature (top panel) and salinity (bottom panel) in transects placed in the main spawning centers found in May- June 1991. a) Gironde estuary, b) Cap Breton canyon. c) in front of Arcachon Bay. d) Along the shelf edge of the Cantabrian sea. Egg abundances along the transects (numbers per 0.05 m2) are also shown.

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 127 FIG. 9. – Vertical distribution of temperature (top panels) and salinity (bottom panels) in May-June 1992. a) in front of the Gironde estuary. b) canyon of Cap Breton. c) in front of Arcachon Bay, accross the Cap Ferret canyon. d) in front of Santander, accross an oceanic SWOddy. Egg abundances along the transects (numbers per 0.05 m2) are also shown.

128 L. MOTOS et al. Similar to previous years, a low egg abundance gap was found in 1992 between the northern and southern spawning areas of the Bay of Biscay. It usually occured between 44º30’ and 45ºN Lat, where eggs were scarce except in the shallowest waters. It is noteworthy that high salinity subsurface waters are prevalent there, penetrating from the area of Cap Ferret canyon (Fig. 9c).

DISCUSSION

Spatial and temporal dynamics of spawning of the Bay of Biscay anchovy population.

Seasonality

FIG. 10. – Temperature and salinity profiles within the spawning The yearly evolution of SST in front of San center located in Fer á Cheval in May-June 1992. Abundances of Sebastián showed extreme SST values, both in win- -2 up to 1000 eggs m were recorded in the area. ter and in summer. They seem to be caused by a strong continental influence derived from the topog- These two coastal spawning areas present strongly raphy of the inner Bay of Biscay beyond the lack of warmed surface waters (17º-18.5ºC) and salinities strong upwelling events. An apparent SST gradient below 35.2 psu and even lower at peak spawning was described along the Cantabrian sea in summer, centers. These structures are clearly shown in ver- from the colder waters of the Galician upwelling tical TS profiles perpendicular to the coast in front system to the warm waters off the Basque Country of the Adour and Gironde estuaries (Figs. 9a and (Boucher, 1985). The latter are a consequence of the b). The influence of freshwater in the upper layers entrainment of water masses in the southeastern cor- was noticeable in the southern region, even in ner of the Bay due to the relaxation of wind and cur- oceanic spawning areas in front of Cap Breton rent regimes in this season. Winter satellite infrared canyon. The spawning areas located in other imagery shows the inversion of this situation, with oceanic waters such as Fer à cheval and offshore relatively warmer waters off Galicia and in the west- waters in front of Santander, presented distinct ern Cantabrian Sea than in the inner eastern hydrological features. SST ranged between 16º and Cantabrian Sea (Urrutia, 1990). These relatively 17.5º C and surface salinity was above 35 psu, warmer and more saline surface waters displace clearly showing that the continental runoff influ- along the shelf break off the Northern Iberian penin- enced these areas very little. sula coast, ocassionaly reaching Cap Ferret canyon Spawning in the oceanic area off southwestern (44º30’). This is considered a manifestation of the Biscay took place at SST 16º-16.5ºC, with strong poleward geostrophical current running northwards gradients in depth but without sharp thermoclines. along the European slope and was called “Corriente Surface salinities ranged between 35 and 35.5 psu de Navidad” due to its season of occurrence (Pin- (Fig. 9d). This spawning centre seems to be associ- gree and Le Cann, 1992). The biological conse- ated with an oceanic gyre. quences of this current in the Bay of Biscay are still Spawning in northwestern areas occurred in poorly understood and should become a major some particular cases (Fig. 9a), associated with research objective in the near future. Finally, cool marked haloclines, however the most common fea- freshwater plumes coming from water runoff also tures prevalent in northern oceanic spawning areas may contribute to maintain eastern Cantabrian Sea were SST values between 16° and 18ºC and the waters colder in the winter season (Valencia et al., presence of a noticeable thermocline (Fig. 10). 1989; Urrutia, 1990). There, the surface mixed layer (16.5º-17ºC) was The seasonality of anchovy spawning found in about 20 m thick, and was followed by a sharp ther- San Sebastián (eastern Cantabrian waters) confirm mocline. previous results from studies made in the same area

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 129 TABLE 2. – Summary of bibliographical information on anchovy spawning seasonality based on egg and larval temporal occurrence in the Bay of Biscay and adjacent waters.

AUTHOR(S) AREA TEMPORAL RANGE PEAK TEMPERATURE PREFERENCES (°C)

Arbault and Lacroix, 1971, 1977 Bay of Biscay Spring-Summer Spring (May) 14°-19° Dicenta, 1984 Basordas October-July April 14°-17° Suau and Vives, 1979 Lemoniz April-September May- July no data Valencia et al. San Sebastián April-September May -July 14°-18° Solá, 1987 Santander May-August July no data Villegas, 1989 Asturias June no data Ferreiro, 1985 Vigo (W Galicia) April-September July-August 14°-18° Ré, 1984 Portuguese estuaries February-July April 13.8°-15.3°

(Table 2). The spring warming of sea surface waters of zooplankton biomass. The occurrence of anchovy largely influenced the onset of the spawning season eggs clearly matches the annual onset of stratifica- in anchovy (Furnestin, 1945; Cort et al., 1979). The tion. This shows that the spawning strategy of present work shows clear evidence of this. The first anchovy leads anchovy larvae to develop in a occurrences of anchovy eggs and larvae in large strongly stratified environment and relatively high abundance in the plankton starts when surface tem- zooplankton abundances (mesozooplankton biovol- perature values rise above 14°C (Fig. 2). After- umes above 20 ml/m2, Valencia et al., op.cit.). The wards, local spawning peaks are recorded at SST intensity of stratification and its variability can have from 16° to 18.5°C. Anchovy spawning still pro- important consequences in larval survival and future ceeds at higher temperatures but at lower levels. recruitment strength (Borja et al., 1996). Anchovy seem to diminish their spawning activities The findings of this work are consistent with the when the rate of surface water warming decreases conclusions reached by Junquera (1991) when and it stops when this rate become nil or negative. In studying the spawning of anchovy in the Cantabri- the southernmost part of the Bay of Biscay, the an Sea and Galicia based on GSI data and the ther- decrease in the warming up rate starts in July, mal regime in the spawning area. Different authors becomes nil in August and is negative from Septem- have found relationships between the process of ber onwards (Fig. 2). gonadal maturation and concurrent thermal regime. Solá et al. (1990) produced times series analysis According to them, a longer spawning season is of anchovy egg and larval abundances and SST for expected in tropical areas, whereas it will be short- Santander, in mid-Cantabrian waters, and Vigo, in er in temperate areas with strong seasonal temper- Galicia (NW Spain). The shortest yearly range of ature changes (Fage, 1911, in Junquera, op.cit., and SST was recorded at Vigo, in the west, and the Arné, 1931). This author found a shorter spawning largest in San Sebastián, in the east, whereas San- period in the southernmost area of the Bay of Bis- tander represent a medium situation. A stronger, cay, whereas it was progressively longer towards faster spring warming up, in addition to a wider SST the western Cantabrian with maximum duration in annual range, was recorded in waters off San Galicia. This is interpretated according to Alheit Sebastián (eastern Cantabrian), when comparing to (1989), who explains partial spawning as an evolu- locations situated to the west (Santander and Vigo). tive adaptation to unstable and not predictable This feature seems to favour massive spawning for environments, e. g. upwelling areas. This repro- this species during May and June, when the warm- ductive modality allows anchovy to search for ing up rate is at its maximum. The abundance of time-space environmental windows with anchovy eggs and larvae is positively correlated to favourable conditions for egg and larval survival the warming up rate of surface waters and to the (Cury and Roy, 1989). In this interpretation, the zooplankton biovolume (Valencia et al., 1989). shorter spawning season recorded in southeastern Apparently that anchovy spawning reached a peak Bay of Biscay is a consequence of rather stable and during the transition between winter (cold, vertical- predictable climatic regime and oceanographic fea- ly homogeneous waters) and summer (warmer sur- tures prevalent in the area. This is also suggested face waters and vertically stratified waters) condi- by the low concentration of anchovy eggs usually tions. This period coincides with maximum values found in Galicia (Solá et al., 1990) which indicates

130 L. MOTOS et al. that spawning normally takes place there in small and Lacroix, 1971; Arbault and Lacroix-Boutin, schools scattered in time and space looking for 1977; Santiago and Eltink, 1988; Massé, 1996; San- favourable environmental windows. tiago and Sanz, 1992b), allow us to give a detailed Our results show that anchovy concentrate to description of the spatial patterns of spawning in spawn in the southeastern corner of the Bay of Bis- the Bay of Biscay anchovy population. cay. Mass spawning is favoured by the sudden spring The spawning season starts at the beginning of warming usually recorded in the area. The spring spring, i.e. at the end of March or beginning of warming and consequent stratification and stabilisa- April. Then, anchovy concentrate in a relatively tion of the water column are also enhanced by a restricted area of southeastern Biscay, from the strong vertical gradient density caused by extended Cantabrian coast to 46°N (Solá et al., 1990; Massé, river plumes all over the region. This sudden warm- 1996) (Fig. 11a). At that time, surface temperature is ing acts as a trigger for spawning activities towards close to 11°-12°C in the area although it rapidly maximum annual rates (Furnestin, 1945). The transi- increases up to around 14°C. Spawning starts at low tion from winter low temperature conditions towards rates and in areas restricted to south of 45°. Its tim- highest summer temperatures happens quickly during ing can change according to prevalent thermal con- May and June (Fig. 2), i.e. the peak spawning months ditions. for anchovy. For the Bay of Biscay, the southeastern In May, the spawning area extends over most of corner (to the south of 44°30’) has the highest warm- the southern Bay of Biscay, between the Spanish ing rates during spring and the maxima of SST values and French coasts and to the south of 46° 30’ N lat- during summer months, for the whole area. itude and to the east of 4° W longitude. No relevant Similar results have been found for other spawning activities can be recorded at this time in anchovy populations. Comparison between subpop- the area (Fig. 11b). Within the spatial range of ulations of Californian anchovy, inhabiting a lati- spawning, two main spawning centres consistently tude range from 30° to 40°N, and Oregon anchovy appear. The coastal region in front of the Gironde inhabiting the region from 40° to 50°N, (Richard- estuary, in the north (45° - 46°N), and the shelf, son, 1981) showed that the former practically repro- shelf break and oceanic regions close to the Cap duce all the year round whereas the latter spawn in a Breton Sea Canyon, in the south. short two-month period coincident with the spring As the reproductive season progresses, the warming and the beginning of summer. In addition, spawning area splits into two regions: coastal and the environmental conditions where they live are shelf break oceanic. In June, an area empty of eggs very different and, simultaneously, rather parallel to can be found in the French shelf between the coastal the contrasting situation found for the Bay of Biscay and the shelf break spawning regions (Fig. 11c). At anchovy and the anchovy population in Galicia. Cal- the end of June, and during the remainder of the ifornian anchovy inhabit a highly variable upwelling summer period, spawning centres can be found scat- system whereas Oregon anchovy live in a habitat tered in the Bay of Biscay. In northern areas, they which is more stable and predictable. can reach waters off Brittany, both close the coast These characteristics are also found in anchovy and near the shelf break (Arbault and Lacroix- populations inhabiting waters off Catalonia (NW Boutin, 1977). In waters off the Cantabrian Sea, Mediterranean sea) (Palomera, 1992). Two spawn- scattered spawning centres are found all along the ing centres were identified in the region: one to the shelf (Villegas, 1979; Solá et al., 1990). north, in the southern Gulf of Lions, where the In summary, three main centres are located in spawning season is shorter. Another centre is locat- coastal areas. The Gironde Estuary spawning area ed in the south, close to the Ebro river mouth, where is situated over the continental shelf from the coast there is a longer spawning season. The author to a depth of 80-100 m, in latitudes between 45º explains this difference by the different thermal and 46º N. The Adour-Bidasoa spawning area is regimes found in the two regions. situated around the Cap Breton canyon region to the south of 44º15’N, generally over the continen- Spatial distribution tal shelf. The Cantabrian coastal area is situated over the Spanish continental shelf. It usually The results obtained in these surveys, and reaches to the west up to 4º W, in an area influ- described herein, in addition to previous and con- enced by the freshwater outflow of the many small current information from other sources (Arbault Cantabrian rivers.

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 131 FIG. 11. – Scheme of the spatial distribution of anchovy spawning. a) beginning of the reproductive season in April; b) early peak in May; c) late peak in June.

The principal spawning centres in oceanic waters ning of the spawning period, and disperse after- are located near the continental slope in the north. wards to occupy the whole Bay of Biscay area (Cort They occur in the area called Fer a Cheval (45ºN) et al., 1976; Arbault and Lacroix-Boutin, 1977; San- when spawning at maximum intensity starts, in May, tiago and Eltink, 1988; Uriarte and Motos 1991, and extend northwards along the slope as the season 1992). Whereas some anchovy usually remain in the proceeds and surface waters warm up. In southern initial concentration area, the population area regions, two main oceanic areas occur: e.g., one expands and the anchovy shoals colonize sorround- along the French shelf near Cap Breton canyon, and ing areas suitable for spawning and/or feeding. one in the Cantabrian sea to the west of 3° W. The latter progressively moves westwards as the season Spawning environment advances. Anchovy catch data from the purse-seine fleet shows that mature anchovy shoals move every Pelagic fishes use to spawn in areas of substan- year from the bottom of the Bay (Cap Breton tial biological production to ensure adequate adult canyon) west and northwestwards into oceanic areas feeding (Blaxter and Hunter, 1982). The persistence of the Cantabrian Sea. Scattered catches are record- in the aerial association of spawning of pelagic fish- ed there at the end of May and June (Uriarte and es with enhanced production is a well-documented Motos, 1992). fact. Different ocean processes can give rise to bio- In summary, the spatial evolution of anchovy logical production and have been pointed out as spawning shows a pattern of concentration/expan- likely to govern much of pelagic fish reproductive sion. Anchovies concentrate in the spawning success (Bakun, 1993; Lima and Castello, 1995). grounds to begin reproductive activities at the begin- Transport and advection processes determine the

132 L. MOTOS et al. displacement of eggs and larvae into favourable or ative effect when acting above a certain intensity by unfavourable areas and the establishment of a larval disruption of food particle aggregations and disper- retention mechanism (Bailey, 1981; Bakun and Par- sion of eggs and larvae to offshore areas of low rish, 1982; Sinclair, 1988). Processes related to nutritional content (Lasker, 1978; Cury and Roy, water column stability may allow the generation and 1989). Mesoscale eddies can play a role in the estab- maintenance of fine-scale food particle aggregations lishment of either larval retention or larval disper- required for the success of first feeding larvae sion (Iles and Sinclair, 1982; Fiedler, 1986), and (Lasker, 1975, 1981; Peterman and Bradford, 1987; they can simultaneously modify the biological Shelton and Hutchings, 1990). Coastal upwellings processes of surrounding oceanic waters (Pingree may favour larval survival through enhancement of and Le Cann, 1992) generating plankton aggrega- biological production. However, it may have a neg- tions (Owen, 1982). Thermohaline fronts may sup-

FIG. 12. – Satellite infrared image showing the situation found in 24 May 1991 (Met Mar nº 154, 1er 1992).

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 133 FIG. 13. – Synoptic description of the main oceanographical features linked to anchovy spawning. U=Upwelling, P=River Plumes, E=Eddies, SF1 and SF2=shelf edge fronts. B=’bottom cold water (<12°C) tongue’. port high zooplankton productivity and concentra- cruise (Fig. 7) illustrates the association between tion (Smith and Lane, 1991) and are related to both variables, temperature and salinity, which are important fish spawning centres (Palomera, 1992; themselves connected to the main spawning centres Sabatés, 1990). of the Bay of Biscay anchovy (Fig. 7a). Some oceanographic features were persistent The hydrological conditions around the plumes year after year, no matter the level of egg abundance are typical of anchovy spawning. The main concen- found. This shows that anchovy spawning is carried trations of anchovy eggs are normally found there. Its out in a consistent geographical pattern linked to influence sometimes reaches the continental slopes particular environmental mesoscale features. These and oceanic waters beyond the slopes, as it was features are colonized by different components of noticeable in 1989 (Figs. 3 and 7) and in 1991 (Fig. the anchovy spawning population inhabiting either 13). The boundaries of these coastal spawning areas coastal or oceanic areas. are characterised by either the absence of continental The main oceanographic feature related to the freshwater influence or, sometimes, the occurrence of anchovy spawning centres identified in the cruises is waters of high salinity in surface layers. the occurrence of low salinity (32.00-35.00 psu) The association between anchovy spawning and river plumes originating in the most important estu- river plumes have already been described in several aries of the southern Bay of Biscay (Gironde, Adour species of Clupeoids and, in particular, Engraulids and Bidasoa). Surface waters in those plumes pre- (Richardson, 1981; Palomera, 1989; Lima and sent high rates of warming during the spring transi- Castello, 1995; Anon., 1994). This association tion passing from 12° -13° C to 16° -18° C in short results from the persistence of these structures and periods, and leading to strong, although shallow (< the higher levels of biological production connected 20-30 m) thermoclines. Fig. 12 shows a satellite to them. The outflow of rivers into coastal seas pro- thermal picture taken on 24 May 1991. It can be vides nutrients directly from terrigenous sources, seen there that differential surface warming affects through upwelling driven by vertical entrainment of the plumes after a short period of intense heating. deeper ocean water into the offshore flowing plume, The distribution of surface salinity found in the and by a permanent distortion of the seabed (Anon.,

134 L. MOTOS et al. 1994). In addition, the river outflow constitutes a should be related to mesoscale oceanographic fea- boundary between warm and cool water and tures that enhance local biological production: a) between saline and fresher water that may enhance slope fronts, as it is the structure of relatively cool water column stability, enrich nutrients or concen- waters distributed along the french shelf break to trate food particles at the interface. The small-scale the north of 45°N, which is caused by internal mechanisms by which these boundaries enhance the waves; b) slope fronts separating coastal shelf survival of larval phases of small pelagic species are waters from oceanic waters in the southernmost still poorly understood and are the focus of broad area of the Bay, which is often associated with international investigations (Anon., op.cit.). intrusions of high salinity waters of Atlantic origin, In addition to the effect of river plumes mentioned this seasonal event takes place during the transition above, they enhance also vertical stabilisation and between mixing and stratification phases and biological production in certain hydroclimatic condi- favours the increment of biological production in tions (Borja et al., 1996). Weak to medium upwelling slope areas (Fernández, 1989); and c) oceanic gyres conditions are common features when first quadrant of slope waters (SWODDIES), namely the one winds prevail in the area. This can ease the penetra- placed around 4°W. Those are stable structures of tion of more saline waters towards the coast under the up to one or more years of persistency. They may offshore driven river plumes, and a parallel uplift of have a potential of retention in the inner part and an the thermocline and a stretching of the mixed surface agitation effect in the outer part which biological layer (Fig. 8a). The consequences of this event are an implications can be important (Pingree and Le increase in the stability of the water column and a Cann, 1992). The signification of these structures potential increase in the biological production of sub- concerning anchovy spawning and subsequent surface layers by the movement of nutrients to the recruitment has not been studied in this population. lower part of the photic zone. Fiedler (1986) studied the effect caused by an anti- Finally, it is worth mentioning the presence of a cyclonic gyre that trapped and displaced more than persistent feature found at medium depth shelf bot- 400 km tropical waters for Californian anchovy toms of the northern Bay of Biscay (to the North of spawning. Anchovy eggs and larvae were displaced 46°N) throughout the year (Vincent and Kurc, 1969). from their usual development and feeding areas. Anchovy spawners seems to avoid this area in May- However, this author did not find any apparent June. According to Le Cann (1982, in Koutsikopou- effect of this event on the recruitment success. Year los and Le Cann, 1996), freshwater discharge inhibits class strength produced that year did not differ from the formation of this cold bottom water tongue in the average recruitments of neighbouring years. areas of influence of the river plumes. It can be Several studies were carried out in May 1987 and hypothesised that the Garonne river plume can 1988 to define the thermal conditions associated impede this cold water mass to reach more southern with large anchovy spawning using vertical profiles latitudes, favouring in this way the best conditions for of temperature (Uriarte, 1988; Lucio et al., 1989). anchovy spawning in the area. Most of the anchovy caught in those years was of The oceanic spawning areas do not usually show big size (average weight between 25 and 40 g) and any noticeable influence of continental freshwater the fishing grounds coincided with the oceanic runoff, although weak haloclines (with surface spawning areas found in concurrent egg cruises and salinity above 35 psu) can often occur. High num- those found from 1989 to 1992. The vertical profiles bers of anchovy eggs can sometimes be found in iso- of the fishing grounds in May 1987 and May 1988 lated low salinity water lenses occurring far away showed SST values between 14º and 17ºC with well from the coast (Fig. 8a). On the contrary, high salin- formed thermoclines at 20-30 m. Consequently, it ity water intrusions in coastal surface areas usually can be considered that the conditions associated lack anchovy eggs (Fig. 8c). with the oceanic spawning areas in the period 1989- All these areas have the common feature of 1992 are usual in many other years. NACW (North Atlantic Central Water) type water The usual spawning periods and areas of masses with high salinity (>35 psu), surface tem- anchovy coincide with minimum values of residual peratures lower than on the coast and thermoclines currents in the Bay of Biscay. Le Cann (1994) weaker but deeper (20-30 m). Spawning takes place showed that the intensity of geostrophic currents in in a range of surface temperatures between 14° and the Bay of Biscay peaks in winter in the south and in 17.5°C. On the other hand, spawning in these areas summer in the north. Tidal currents are very weak in

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 135 the south of the Bay (Koutsikopoulos and Le Cann, year-old anchovy occur together with the biggest 1- 1996). Wind intensity decreases markedly at the year-old anchovy individuals. Older anchovies are beginning of the anchovy spawning season and the usually prevalent in oceanic spawning areas where prevalent direction changes at the same time from spawners are larger than in coastal areas, although the third to the first quadrant (Borja et al., 1996). All all age classes are usually represented. The terms these changes lead to a minimisation of surface cur- coastal and oceanic spawning areas have already rents in the southern area of the Bay of Biscay which been used to describe the geographical distribution is considered a ‘slack-water zone’ (Pingree and Le of the Bay of Biscay anchovy spawning and their Cann, 1992) at this time of the year. All this data typical association with small and big anchovies, provide evidence that the spawning area chosen by respectively (Santiago and Eltink, 1988; Motos et the anchovy population is relatively stable in the al., 1991; Motos and Uriarte, 1991). The present peak spawning period in comparison to neighbour- study confirms previously reported geographical ing areas and supports the ocean stability hypothesis spawning areas and provides a more detailed of Lasker (1975) for recruitment success of the Bay description of the relationships between the environ- of Biscay anchovy. ment linked to different components of the anchovy population. Spatial components of anchovy spawners Annual variability of the anchovy The egg distribution found in the surveys is con- spawning environment. sistent with data on the spatial structure of adult anchovy in the spawning grounds. Coastal and Neither the spatial distribution of spawning nor oceanic regions correspond to spawning areas of the oceanographic conditions were the same during two different components of the anchovy popula- the years of study. Substantial differences were tion, respectively. Big anchovy, aged 2 or more found, for instance, in the influence of continental years and the bigger 1-year-old anchovy, mainly freshwater. The cruises performed in 1989 and 1991 spawn in waters close to the shelf break or beyond revealed a larger extension of the surface influence it. Small anchovy, mainly 1-year-old anchovy, of the river plumes than in 1990 or 1992. The rea- spawns in shallow waters close to the coast (Cort et sons for this may primarily be related to the hydro- al., 1979; Cendrero et al., 1981; Santiago and climatological regime affecting the region. Among Eltink, 1988; Uriarte and Santiago, 1990; Motos and the factors driven by this regime, present and recent Uriarte,1991; Massé, 1996; Anón., 1994). In years (previous winter) atmospheric temperatures are of good recruitment, as happened in 1988 (Santiago important. The mild winters of 1989 and 1990 and Sanz, 1992) and mainly in 1990 and in 1992, 1 (Valencia, 1993) were especially relevant in this year-old anchovy were also dominant in oceanic regard. Also, both the level of rain and the cumulat- areas close to or beyond the shelf break. ed rainfall anomalies, that were especially scarce in Spawning associated with river plumes of strong the period from 1988-1990 (Valencia, op.cit), and thermohaline stratification is mainly carried out by their relationship with the level of river run-off 1-year-old individuals. However, these structures (Valencia and Borja, in press) need be considered. are also prevalent in southern areas where 2 or more Other main factors influencing oceanographic con-

TABLE 3. –Environmental parameters potentially influencing the extension of the river plumes in the region during the study period. a) area to the south of 46° 30’ N and to the east of 4° W with sea surface salinity below 35 psu as recorded during each survey; b) discharge volume of the Garonne River, c) Raining at San Sebastián, and d) northern component of the wind during cruises.

YEARS 1989 1990 1991 1992 Average(1) Month May June May June May June May June* May June

Surface <35 psu (1) 1.70 3.30 2.36 1.00 4.25 1.64 Garonne outflow (2) 947 394 676 478 1048 639 718 2047 1177 842 Raining at San Sebastián (3) 81 36 114 123 126 38 87 178 102 54 Northern wind component (4) 315 928 450 585 1561 812 198 790 631 779

(1) Index of the surface enclosed by the 35 psu isohaline. The index is scaled to the lowest value (June 1990). (2) The average concerning the Garonne river otuflow refers to the period 1959-92 (m2 s-1). (3) Most rain occurred after the cruise in 1992 (mm) (4) Mean intensity of northern coind component at San Sebastián along the month (km s-1 100)

136 L. MOTOS et al. ditions in the region are prevalent winds and in recruitment success (Lluch-Belda et al., 1989). In induced vertical and horizontal water transport, fact, the strength of the anchovy year classes pro- which for instance, strongly influence the horizontal duced from the spawning seasons analyzed in this and vertical extension of generally warmer low work (1989-1992) showed enormous fluctuation salinity waters. Finally, gale episodes are important (Motos, 1994). The 1990 spawning resulted in the as well. For instance, short period mixing events minimum recruitment level, whereas the 1991 were recorded in April-May 1991, which led to tem- spawning resulted in the maximum level, i.e. the porary homogenisation of the water column. years of minimun and maximun influence of the Table 3 presents several environmental parame- river plumes and northern component of the winds, ters that may have determined the extension of the respectively. influence of the river plumes in the region during the study period. There, the area with sea surface salin- Spatial distribution of the anchovy ity below 35 psu as recorded during each egg survey spawning population within the whole sampling area is shown. The table also shows values for the discharge of the Garonne The spatial distribution pattern of anchovy spawn- River, the evapotranspiration balance at San ing can change depending on the level of adult abun- Sebastián and the cumulative intensity of the north- dance. In years with high adult abundances (1990, ern component of the wind in the concurrent months 1992), eggs cover most of the sampling area, with of the surveys. peak intensity centres in the usual places said above. The situation in spring 1990 was conditioned by On the contrary, in years of low adult abundance the 1989-90 winter drought, the low discharge of (1989, 1991), eggs occur in discrete aggregations rivers and the weak northern winds. On the contrary, centred on the principal spawning centres, such as the 1989 and 1991 showed important river discharges in Gironde estuary and Cap Breton canyon regions. May, concurrently with or followed by relevant Empty or low abundance areas occur between these northern winds. The situation found in 1992 shows aggregations. Within these aggregations local egg an intermediate stage. The distribution of salinity densities can be of a magnitude comparable to those found in these cruises show the action of northern found in years of high adult abundance. winds on the expansion of the river discharges over In years of high adult abundance, spawning the shelf. The resultant extension of the river plumes invariably occurred in almost every coastal place is a main factor contributing to the increasing of ver- subject to the influence of continental freshwater. tical stability in the area, as discussed above. The However, this was not so for all the oceanic areas widest extension was recorded in 1991 (Fig. 7) fol- suitable for anchovy spawning: e.g. moderate warm- lowed by June 1989 (Fig. 3), whereas the narrowest ing, a thick mixed surface layer and a noticeable was found in June 1990 (Fig. 5). The situation found thermocline. It seems that the potential oceanic area in a given year can vary from May to June depend- of spawning for big anchovies was larger than the ing on the hydroclimatic regime. In 1989, combined areas actually occupied. From information about the low Garonne outflow and high northern wind com- fishing areas of the Spanish purse seine fleet (Uri- ponent led to an important extension of low salinity arte and Motos, 1992, 1993), evidence was found waters. However, in 1990 a combination of similar that big/old anchovy move across most of the poten- outflow values and lower northern wind component tial oceanic spawning area in the south of the Bay of in June led to the minimum extension of the river Biscay during May. In the period 1989-1992, these plumes. The wind regime seems to be the main fac- southern oceanic areas were among the most impor- tor influencing the extension of the river plume tant fishing areas for the Spanish purse seine fleet. influence over the area, with the river outflow also Spatial distribution patterns with abundance max- important. The northern component of the winds ima in the centre and gradual decline towards the were also very high both in May and in June 1991, borders of the habitat are a common feature in vege- the year of maximum extension. tal and animal species (Brown, 1984). Arbault and Consequently, the spawning environment of the Lacroix-Boutin (1977) indicated that anchovy anchovy population may vary over a wide range of spawning can occur in discrete patches or in a con- temperature and salinity conditions both within and tinuum, depending on the year. This view is con- between seasons. This high frequency environmen- firmed by our results. The pattern of spatial variation tal variability is responsible for interannual variation in abundance will depend on the number and kind of

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 137 environmental factors affecting the niche of the The processing work was done AZTI’s staff, species and its pattern of spatial variation (Brown, including José Antonio Alvarez, Iñaki Artetxe, op.cit.). For a given period (years) under low vari- Maria José Costa, Luis Cuesta, Inmaculada Martin, ability environmental conditions, the variation in the Iñaki. Rico, Nerea Uriondo and Asier Zamakona extension of the spawning area of a particular species (technicians); Joseba Antxustegi and Juan Pedro will depend on the abundance of genitors. In years of Santiago (fishermen), Jose María Emazabal and low abundance - as in 1989 and 1991 -eggs occurred Carlos Erauskin. The authors appreciate the collab- spatially restricted to the main spawning centres, oration of Concha Franco, Amor Solá, Ana Lago de whereas in years of higher abundance - as in 1988, Lanzós and Alberto García (IEO) who were in 1990 and 1992 - eggs occurred over most of the charge of egg cruises carried out on board R/V study area, although abundance maxima were locat- ‘INVESTIGADOR’ in 1990 and ‘CORNIDE DE ed in the same main spawning centres. SAAVEDRA’ in 1991 and 1992. The captains and This pattern had already been reported in species crew of this R/V and that of the ‘IBAIZABAL DOS’ similar to anchovy. Murphy (1977) indicated that the tug ship greatly eased the work at sea. José Manuel spatial range of pelagic spawning fishes (Clupeoids) Cabanas (IEO Vigo) provided us with the CTD sta- shrinks as population abundance declines and tions collected during the ‘INVESTIGADOR’ expands as population abundance expands. MacCall cruise of April-May 1990. Finally, the authors (1990) studied Californian sardine and anchovy pop- aknowledge the support and ideas received from a ulation, and found that for both populations the distri- number of colleagues working together with us on bution area shrinks in periods of scarce abundance. anchovy research activities. To all of them we Then, spawning is practically restricted to the more express our gratitude. favourable spawning sites, which maintain minimal sustaining conditions for these populations. On the contrary, in periods of high abundance the population REFERENCES extends over most of its potential distribution, sug- gesting a habitat selection dependant on population Andréu, B. – 1950. Sobre la maduración sexual de la anchoa size (MacCall, op.cit.). According to the basin theory (Engraulis encrasicholus L.) de la costa norte de España. Datos biológicos y biométricos. Publ. Inst. Biol. Apl. 7(1950):7-36. of this author, Clupeoid populations present optimal Alheit, J. – 1989. Comparative spawning biology of anchovies, sar- areas or basins for reproduction. The spawning areas dines and sprats. Rapp. P.-V. Réun. Cons. Int. Explor. Mer 191: 7-14, 1989. associated with the Estuaries of Gironde, Adour and Anon. – 1954. Anchois. Rev. Trav. Inst. Pêches Marit., 18: 2-4. Bidasoa, in the southeastern region of the Bay of Bis- Anon. – 1988. Report of an ad hoc Working Group on the descrip- tion of the fisheries directed at anchovy in ICES Divisions VIIIb cay, seem to constitute a refuge basin favourable for and VIIIc. Nantes, 25-29 January 1988 (mimeo). the maintenance of this anchovy population. This Anon. – 1994. Small Pelagic Fishes and Climate Change Program. Globec Report Series. La Paz, México. basin presents particular environmental features that Arbault, S. and N. Lacroix. – 1971. Aires de ponte de la sardine, du ensures the continuation of the anchovy population. sprat et de l’anchois dans le golfe de Gascogne et sur le plateu celtique. Resultats de 6 années d’étude. Rev. Trav. Inst. Pêches Low frequency, decadal environmental changes may Marit., 35: 35-56. have restricted the anchovy population to this most Arbault, S. and N. Lacroix-Boutin. – 1977. Oeufs et larves de Clu- peides et Engraulides dans le Golfe de Gascogne (1969-1973). favourable habitat (Junquera, 1986). However, fish- Distribution des frayeres. Relations entre les facterus du milieu et ing may have influenced the shrinking of the popula- la reproduction. Rev. Trav. Inst. Pêches Marit., 41: 227-254. Arné, P. – 1931. Contribution à l’étude de l’anchois: Engraulis tion habitat (Radovich, 1982) by maintaining high encrasicolus (L.) du Golfe de Gascogne. Rev. Trav. Office des exploitation levels in periods of decreased stock pro- Pêches Maritimes 4: 153-181. Bailey, K.M. – 1981. Larval transport and recruitment of Pacific duction (Cort et al., 1976, Igelmo et al., 1984; Uriarte hake Merluccius productus. Mar. Ecol. Progr. Ser. 6: 1-9. and Astudillo, 1987). Bakun, A. – 1993. The California Current, Benguela Current, and southwestern Atlantic Shelf ecosystems: a comparative approach to identifying factores regulating biomass yields. In: K. Sherman, L.M. Alexander and B. Gold, (eds.): Large marine ecosystems-stress, mitigation and sustainability, pp. 199-244. ACKNOWLEDGEMENTS Am. Assoc. Adv. Sci., Washington, D.C. Bakun, A. and R.H. Parrish. – 1982. Turbulence, transport and pelagic fish in the California and Peru current systems. Calif. This work was supported by the European Union Coop. Oceanic Fish. Invest. Rep., 23: 99-112. Blaxter, J.H.S. and J.R. Hunter. – 1982. The Biology of Clupeoid FAR and AIR program (project MA-151-1 in 1989- fishes. Advances in Marine Biology, 20: 1-223. 1990 and project MA 2.495 in 1991-1992) and the Borja, A., A. Uriarte, V. Valencia, L. Motos, and Ad. Uriarte. – 1996. Relationships between anchovy (Engraulis encrasicolus) Department of Industry, Agriculture and Fisheries of recruitment and the environment in the Bay of Biscay. Sci. Mar. the Basque Government. (this volume).

138 L. MOTOS et al. Boucher, J. – 1985. Caractéristiques physiques et biologiques du Lima I.D. and J.P. Castello. – 1995. Distribution and abundance of domaine pélagique. In: Laubier and Monniot (ed.): Peuple- south-west Atlantic anchovy spawners (Engraulis anchoita) in ments profonds du Golfe de Gascogne, IFREMER, 1985. relation to oceanographic processes in the southern Brazilian Brown, J.H. – 1984. On the relationship between abundance and shelf. Fish. Oceanogr. 4: 1-16. distribution of species. Am. Nat., 124: 255-279. Lluch-Belda, D., R.J.M. Crawford, T. Kawasaki, A.D. MacCall, Cendrero, O. (ed.). – 1994. Improvement of stock assessment by R.H. Parrish, R.A. Schwartzlose and P.E. Smith. – 1989. direct methods, its application to the anchovy (Engraulis World-wide fluctuations of sardine and anchovy stocks: the encrasicholus) in the Bay of Biscay. Final report of the EC regime problem. S. Afr. J. Mar. Sci. 8: 195-205. FAR Project 1991-1993, 99 pp. Contract No. MA 2495 EF Lucio, P., I. Martin, L. Motos, J. Santiago and A. Uriarte. – 1989. (mimeo). Informe técnico de la pesquería de anchoa en el año 1988. Ser- Cendrero, O., J.L. Cort and E. de Cardenas. – 1981. Revision de vicio Central de Publicaciones del Gobierno Vasco. Gasteiz- algunos datos sobre la biología de la anchoa (Engraulis encra- Vitoria, 1988. sicholus L.) del mar Cantábrico. Bol. Inst. Esp. Ocean., 220. Lucio, P. and A. Uriarte. – 1990. Aspects of the reproductive biol- Cort, J.L., O. Cendrero and E. de Cárdenas. – 1977. Nouvelle infor- ogy of the anchovy (Engraulis encrasicolus L.) during 1987 mation sur la pêche espagnole a l’anchois du Golfe de and 1988 in the Bay of Biscay. ICES C.M. 1990/H:27 Gascogne. CIEM Doc. CM 1977 / J:6. MacCall, A.D. – 1990. Dynamic geography of marine fish populations. Cort, J.L., O. Cendrero and E. de Cárdenas. – 1979: Nuevos datos Washington Sea Grant. University of Washington Press, Seattle. sobre la anchoa del Cantábrico. Información Pesquera. Inst. Martín, I . and A. Uriarte. – 1989. Stock assessment of the Bay of Esp. Ocean., 9: 1-9. Biscay anchovy and catch predictions for 1989. ICES CM Cort, J. L., O. Cendrero and X. Iribar. – 1976. La anchoa, Engraulis 1989/H:24. encrasicholus (L.) del Cantábrico. Resultados de las campañas Massé, J. – 1996. Acoustic observations in the Bay of Biscay: de 1974, 1975 y 1976. Bol. Inst. Espa. Ocean., 220: 1-34. schooling, vertical distribution, species assemblages and behav- Cury, P. and C. Roy. – 1989. Optimal environmental window and iour. Sci. Mar. (this volume) pelagic fish recruitment success in upwelling areas. Can. J. Massé, J., C. Leroy , D. Halgand and P. Beillois. – 1995. Fish. Aquat. Sci., 46: 670-680. Anchovy adult and egg distributrion in the Bay of Biscay in Cushing, D.H. – 1975. Marine ecology and fisheries. Cambridge relation to environmental conditions, as observed during University Press. acoustic surveys (1990, 1991, 1992). Actas del IV coloquio Dicenta, A. – 1984. Aportación al conocimiento de ictioplancton de Internacional sobre Oceanografía del Golfo de Vizcaya: 281- la costa vasca. Bol. Inst. Esp. Oceanog., 1: 94-105 283. Santander. Fage, L. – 1920. Engraulide, Clupeidae. Rep. Dan Oceanogr. Motos, L. – 1994. Estimación de la biomasa desovante de la Exped. 1908-10 Medit, adjac. Seas, biol. A9:140 pp. población de anchoa del Golfo de Vizcaya Engraulis encrasico- Fernández, E. – 1989. Sistemas frontales y de Afloramiento en el lus, a partir de su producción de huevos. Bases metodológicas Cantábrico occidental: relación con la estructura de las redes y aplicación. Ph.D. thesis, Univ. País Vasco. tróficas planctónicas. In: Urrutia J. and A. Rallo, (eds.): Motos, L. and J. Santiago. – 1990. An egg production estimate of Oceanografía del Golfo de Vizcaya. pp. 191-223. Universidad biomass of the Bay of Biscay anchovy (Engraulis encrasicolus, del País Vasco, San Sebastián. L.) in 1989. ICES C.M. 1990/H:26. Ferreiro, M. J. – 1985. El ictioplancton de la ría de Vigo. Ph. D. the- Motos, L., J. Santiago and A. Uriarte. – 1991. Assessment of sis. Univ. Santiago de Compostela. spawning biomass of the Bay of Biscay anchovy population, Fiedler, P.C. – 1986. Offshore entrainment of anchovy spawning Engraulis encrasicholus L., using the daily egg production habitat, eggs and larvae by a displaced eddy in 1985. CalCOFI method. ICES CM 1991/H:33. Rep. XXVII: 144-152. Motos, L. and A. Uriarte. – 1991. An egg production estimate of Furnestin, J. – 1945. Norte preiminaire sur l’anchois, Engraulis biomass of the Bay of Biscay anchovy in 1990. ICES CM encrasicolus (L.), du golfe de Gascogne. Rev. Trav. Off. Sci. 1991/H:32. Tech. Pêches Marit. 13: 197-209. Motos, L. and A. Uriarte. – 1992. An egg production estimate of Igelmo, A., X. Iribar, J. Legarra and S. Lerga. – 1984. Inventario de biomass of the Bay of Biscay anchovy in 1990. ICES CM artes de pesca en Euskadi. Eusko Jaurlaritza-Gobierno Vasco. 1992/H:32. Consejería de Comercio, Pesca y Turismo. Motos, L. and A. Uriarte. – 1993. An egg production estimate of Iles, T.D. and Sinclair, M. – 1982. Atlantic herring: stock discrete- biomass of the Bay of Biscay anchovy (Engraulis encrasicho- ness and abundance. Science, 215: 627-633. lus, L.) in 1992. ICES C.M., 1993/H:26. Junquera, S. – 1986. Pêche de l’anchois (Engraulis encrasicolus L.) Murphy, G.I., 1977. Clupeoids. In: Fish Population Dynamics. J.A. dans le Golfe de Gascogne et sur le littoral Atlantique de la Gal- Gulland (ed.), pp 283-308. Jonh Wiley, London. ice depuis 1920, variations quantitatives. Rev. Trav. Inst. Pèch- Owen, R.W. – 1981. Patterning of flow and organisms in the larval es Marit. 48: 133-142. anchovy environment. In: . G. D. Sharp (ed.): Workshop on the Junquera, S. – 1991. Estudio de la diversidad poblacional de la Effects of Environmental Variation on the Survival of Larval anchoa, Engraulis encrasicolus L. 1758, (Pisces, Engraulidae) Pelagic Fishes. IOC Work. Rep. 8: 167-200. mediante el análisis canónico de caracteres morfométricos y Owen, R.W. – 1982. Microscale plankton patchiness in the larva parámetros biológicos. Ph. D. thesis, Universidad de Oviedo. anchovy environment. Rapp. P.-v. Réun. Cons. int. explor. mer Koutsikopoulos C. and B. Le Cann. – 1996. Physical processes and 178: 364-368. hydrological structures related to the Bay of Biscay anchovy. Palomera, I. – 1989. Primeras fases del desarrollo de la anchoa, Sci. Mar., (this volume). (Engraulis encrasicolus) en la Costa Catalana. Ph. D. thesis, Lasker R. – 1975. Field criteria for Survival of anchovy larvae; the Univ. Barcelona. relation between inshore chlorophyll maximum layers and suc- Palomera, I. – 1992. Spawning of anchovy, Engraulis encrasicolus cessful first feeding. Fish. Bull., 73: 453-462. L., in the Northwestern Mediterranean relative to hydrographic Lasker, R. – 1978. The relation between oceanographic conditions features in the region. Mar. Ecol. Prog. Ser., 79: 215-223. and larval anchovy food in the California current: identification Peterman, R.M. and M.J. Bradford. – 1987. Wind speed and mor- of factors contributing to recruitment failure. Rapp. P.-v. Réun. tality rate of a marine fish, the northern anchovy (Engraulis Cons. int. Explor. Mer, 173: 212-230. mordax). Science, 235: 354-355. Lasker, R. – 1981. Factors contributing to variable recruitment of Pingree, R.D. and B. Le Cann. – 1992. Three anticyclonic Slope the northern anchovy (Engraulix mordax) in the California cur- Water Oceanic Eddies (SWODDIES) in the southern Bay of rent: contrasting years, 1975 through 1978. Rapp. P.-v. Reún. Biscay in 1990. Deep-Sea Res., 39: 1147-1175. Cons. int. Explor. Mer., 178: 375-388. Posgay, J.A., R. R. Marak and R.C. Hennemuth. – 1968. Development Le Cann, B. – 1982. Evolution annuelle de la structure and test of new zooplankton samplers. ICNAF Res. Doc., 85. hydrologique du plateau continental au sud de la Bretagne: Radovich, J. – 1982. The collapse of the California sardine fishery: what Modelisation numerique. Thèse 3e cycle en Océanographie have we learned. Calif. Coop. Oceanic Fish. Invest. Rep. 23: 56-78. physique, Univ. Bretagne Occidentale. Ré, P. – 1984. Ictioplancton da regiao central da costa portuguesa Le Cann, B. – 1994. Circulation dans le Golfe de Gascogne. Actas e do estuario do tejo. Ecologia da postura e da fase planctóni- del IV coloquio Internacional sobre Oceanografía del Golfo de ca de Sardina pilchardus (Walbaum, 1792) e de Engraulis Vizcaya: 217-234. Santander. encrasicolus (Linné, 1758). Ph. D. thesis, Univ. Lisboa.

THE SPAWNING OF THE BAY OF BISCAY ANCHOVY 139 Richardson, S.L. – 1981. Spawning biomass and early life of north- Uriarte, A. – 1988. Informe Técnico de la Pesquería de la anchoa ern anchovy, Engraulis mordax, in the northern subpopulatin en el año 1987. Servicio central de publicaciones del Gobierno off Oregon and Washington. Fish. Bull., 78: 855-876. Vasco, Vitoria, 32 pp. Sabatés, A. – 1990. Distribution pattern of larval fish populations in Uriarte, A. and A. Astudillo. – 1987. The anchovy in the Bay of bis- Northwestern Mediterranean. Mar. Ecol. Prog. Ser. 59: 75-82. cay: new data and analysis of the fishery 1974-1987. ICES C. Santiago, J. and Eltink, A. – 1988. Distribution and abundance of M. 1987/ H:20. anchovy eggs in the Bay of Biscay in May 1987 in comparison Uriarte, A. and L. Motos. – 1990. Informe técnico de la pesquería to 1983 and 1986. ICES, C.M. 1988/H:9 de anchoa en el año 1989. Servicio Central de Publicaciones Santiago, J. and A. Sanz. – 1992. Egg production estimates of the del Gobierno Vasco. Gasteiz-Vitoria, 1989. Bay of Biscay anchovy, Engraulis encrasicholus (L.), spawn- Uriarte, A. and L. Motos. – 1991. Informe técnico de la pesquería ing stock in 1987 and 1988 (Estimaciones de la producción de de anchoa en el año1990. Servicio Central de Publicaciones del hevos de l stock reproductor de anchoa, Engraulis encrasicho- Gobierno Vasco. Gasteiz-Vitoria, 1991. lus (L.), del golfo de Vizcaya en 1987 y 1988). Bol. Inst. Esp. Uriarte, A. and L. Motos. – 1992. Informe técnico de la pesquería Oceanogr. 8: 225-230. de anchoa en el año1991. Servicio Central de Publicaciones del Sanz, A. and A. Uriarte. – 1989. Reproductive cycle and batch Gobierno Vasco. Gasteiz-Vitoria, 1992. fecundity of the Bay of Biscay anchovy (Engraulis encrasicho- Uriarte, A. and L. Motos. – 1993. Informe técnico de la pesquería lus) in 1987. CalCOFI Rep., 30: 127-135 de anchoa en el año1992. Servicio Central de Publicaciones del Shelton, P.A., and Hutchings, L. – 1990. Ocean stability and Gobierno Vasco. Gasteiz-Vitoria, 1993. anchovy spawning in the southern Benguela Current region. Uriarte, A. and J. Santiago. – 1990. Conversion of the spawning Fish. Bull. U.S. 88: 323-338. stock biomass into numbers at age. ICES C. M. 1990/H:26. Sinclair, M. – 1988. Marine Populations: an essay on population Urrutia, J. – 1990. Los ríos como agente de fertilización costera. In: regulation and speciation. University of Washington Press, Urrutia and A. Rallo (eds.): Oceanografía del Golfo de Vizcaya. Seattle, WA. VIII cursos de verano en San Sebastián, Servicio Editorial Uni- Smith, P.E., W. Flerx and R.H. Hewitt. – 1985. The CalCOFI Ver- versidad del Pais Vasco. tical Egg Tow (CalVET) Net. In: R. Lasker (ed.): An egg pro- Valencia, V. – 1993. Estudio de la variacion temporal de la hidro- duction method for estimating spawning biomass of pelagic grafía y el plancton en la zona nerítica frente a San Sebastián fish: Application to the northern anchovy, Engraulis mordax, entre 1988-1990. Informes Técnicos/Gobierno Vasco, Dpto. de pp. 27-32. U.S. Dep. Commer., NOAA Tech. Rep. NMFS 36. Agricultura y Pesca, 52: 1-105. Smith, P.E. and P.V.Z. Lane. – 1991. The jet off Punta Arena, Cal- Valencia, V. – 1994. Oceanographical conditions during the ifornia: its role in aspects of secondary production in the cope- anchovy DEPM cruises in 1991 and 1992. In: Cendrero, O. pod Eucalanus californicus Johnson. J. Geophys. Res. 96: (ed.): Improvement of stock assessment by direct methods, its 14849-14858. application to the anchovy (Engraulis encrasicholus) in the Bay Smith, P.E. and S.L. Richardson. – 1977. Standard techniques for of Biscay. Final report of the EC FAR Project 1991-1993, 99 pelagic fish eggs and larval surveys. FAO Fish. Tec. Papers, pp. Contract No. MA 2495 EF (mimeo). 175, 100 pp. Valencia, V., A. Borja and J. Franco. – (in press). Estudio de las varia- Solá, A. – 1987. Yearly variation of the icthyoplanckton of com- ciones a corto y largo término de varios parámetros oceanográfi- mercial interest species in the Santander coast (North of Spain). cos y meteorológicos de interés para las pesquerías del Golfo de ICES C.M. 1987/L:32 Vizcaya. Inf. Técn. Dep. Agric. y Pesca Gob. Vasco., Vitoria. Solá, A. and C. Franco. – 1985. Contribución al estudio del ictio- Valencia, V., L. Motos and J. Urrutia. – 1989. Estudio de la plancton de especies de interés comercial de la plataforma Can- variación temporal de la hidrografía y el plancton en la zona tábrica. Bol. Inst. Esp. Oceanog., 2:1-14 nerítica frente a San Sebastián. Resultados abril 1986-diciem- Solá, A., L. Motos, C. Franco and A. Lago de Lanzos. – 1990. Sea- bre 1987. Informes Técnicos S.I.O., 20 sonal occurrence of pelagic fish eggs and larvae in the Cantabri- Villegas, M.L. – 1979-1980. Larvas y postlarvas de sardina y an Sea (VIIIc) and Galicia (IXa) from 1987 to 1989. ICES C.M. anchoa del mar Cantábrico. Rev. Fac. Cienc. Univ. Oviedo (Ser. 1990/H:25. Biología) 20-21: 177-187. Suau, P. and F. Vives. – 1979. Ictioplancton de las aguas del Can- Vincent, A. and G. Kurc. – 1969. Hydrologie, variations saison- tábrico frente a Punta Endata (N. de España). Inv. Pesq., 43: nières de la situation thérmique du Golfe de Gascogne en 1967. 723-736 Rev. Trav. Inst. Pêches Marit., 33: 79-96.

140 L. MOTOS et al.