J. Plankton Res. (2019) 41(3): 293–308

Journal of Plankton Research academic.oup.com/plankt

J. Plankton Res. (2019) 41(3): 293–308. First published online July 2, 2019 doi:10.1093/plankt/fbz015 Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 ORIGINAL ARTICLE Latitude, distance offshore and local environmental features as modulators of zooplankton assemblages across the NE Atlantic Shelves Province

ALVARO FANJUL1,*, ARANTZA IRIARTE2, FERNANDO VILLATE1, IBON URIARTE2, MIGUEL ARTIACH3, ANGUS ATKINSON4 AND KATHRYN COOK5 1department of plant biology and ecology (faculty of science and technology), university of the basque country (upv/ehu), po box 644, 48080 bilbao and research centre for experimental marine biology and biotechnology (plentzia marine station; pie-upv/ehu) areatza pasalekua plentzia-bizkaia, 48620, spain, 2department of plant biology and ecology (faculty of pharmacy), university of the basque country (upv/ehu), paseo de la universidad 7, 01006 gasteiz and research centre for experimental marine biology and biotechnology (plentzia marine station; pie-upv/ehu) areatza pasalekua, plentzia-bizkaia, 48620, spain, 3department of applied economics, econometrics and statistics, faculty of economics and business, university of the basque country (upv/ehu), avda. lehendakari aguirre, 83, 48015 bilbao, spain, 4plymouth marine laboratory, prospect place, the hoe, plymouth, pl13dh, united kingdom and 5marine laboratory, marine scotland science, scottish government, 375 victoria road, aberdeen ab11 9db, united kingdom

*corresponding author: [email protected]

Received June 14, 2018; editorial decision March 18, 2019; accepted March 28, 2019

Corresponding Editor: Marja Koski

Contribution of latitude, distance offshore and environmental factors to variations in zooplankton assemblages across the Northeast Atlantic Shelves Province, from the Bay of Biscay [Bilbao 35 (B35) and Urdaibai 35 (U35)] to the English Channel (Plymouth L4; L4) and the North Sea (Stonehaven; SH), were assessed mainly by redundancy analysis. For coarse zooplankton groups latitude explained the main between-site differences, and meroplankton contributed more than holoplankton. Latitudinal differences were best indicated by contrasting abundances of cirripede larvae and doliolids (most abundant at the lowest latitude sites) and bryozoan and polychaete larvae (most abundant at the highest latitude site). Doliolids were best indicators of temperature-mediated latitudinal differences. The interaction between latitude and distance offshore or salinity and phytoplankton biomass explained smaller percentages of the variability. The main differences in copepod and cladoceran genera reﬂected the oceanic inﬂuence, with highest presence of

available online at academic.oup.com/plankt © The Author(s) 2019. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: [email protected] JOURNAL OF PLANKTON RESEARCH VOLUME 41 NUMBER 3 PAGES 293–308 2019

Corycaeus and Oncaea at L4, likely related to the higher inﬂuence of off-shelf water intrusions, and neritic Acartia dominating at SH, U35 and B35. Podon and Evadne, which decreased from south to north, reﬂected latitude-related differences driven more by salinity than by temperature. Instances where a single species (e.g. Acartia clausi)dominated showed common relationships with temperature, consistent with a common thermal niche. Differences in co-generic species dominance between sites depicted the latitudinal gradient.

KEYWORDS: zooplankton community; spatial variations; latitude; shelf waters; North Atlantic Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 INTRODUCTION difficulties in getting so many different institutions and researchers involved in collaborative work with common Knowledge of the effect of environmental drivers on objectives and data analysis methodologies. marine plankton is essential to be able to predict the Several of these monitoring sites are located in the response of pelagic ecosystems to environmental change eco-geographical unit of the Northeast Atlantic Shelves (Pepin et al., 2015). Coastal plankton communities exhibit Province (NECS; Longhurst, 1998), where zooplankton high variability (Ribera d’Alcalà et al., 2004) because, in show a lack of coherence between sites in their interan- addition to larger-scale oceanographic and atmospheric nual variations and seasonal cycles, suggesting a higher forcing effects, they are also subject to smaller-scale influence of local factors over large-scale environmental effects due to interactions between water circulation and drivers (Fanjul et al., 2017, 2018).Theaimofthepresent bathymetry, benthic–pelagic interactions and terrestrial study was to (i) describe the main differences in the inputs of freshwater, nutrients and pollutants through zooplankton community structure between four selected rivers and estuarine plumes (Pepin et al., 2015). coastal sites in the NECS province and (ii) assess the con- Zooplankton play key roles in food webs and biogeo- tribution of latitude, distance offshore and environmental chemical cycles (Longhurst, 1991; Richardson, 2008; drivers in accounting for these differences. The selected Mackas and Beaugrand, 2010), and the study of coastal sites are more or less equidistant along a latitu- zooplankton communities of shelf ecosystems is crucial dinal gradient: Bilbao 35 (B35) and Urdaibai 35 (U35) because these are high productivity areas, supporting are the southernmost sites, located in the Bay of Biscay, over half of the world’s marine fisheries (Caddy et al., within the Lusitanian unit of the Eastern Atlantic warm 1998; Mossop, 2007). The composition of zooplankton temperate region; Stonehaven (SH) is the northernmost communities varies with latitude, when spanning dif- site, located in the North Sea within the cool-temperate ferent climatic zones (Xu et al., 2016). Inshore–offshore Boreal zone; and Plymouth L4 (L4) is the mid-latitude gradients along shelf waters also show corresponding site, located in the English Channel, in the southern part zooplankton community gradients (Roura et al., 2013), of the Boreal biogeographic unit, near the border with with water depth and intrusions of oceanic water onto the Lusitanian unit (Southward et al., 2005; Spalding et al., the shelf being very influential (Tremblay and Roff, 2007). In addition to latitude, B35 and U35 differ in 1983; Blachowiak-Samolyk et al., 2008; Dvoretsky and distance offshore from L4 and SH, and B35 differs from Dvoretsky, 2015; Pepin et al., 2015). Furthermore, there the other three sites in phytoplankton biomass. Empha- are also zooplankton community differences related to the sis was placed in establishing the relative importance level of nutrient enrichment and pollution of nearshore of latitude/temperature versus other local environmental coastal waters (Uriarte and Villate, 2004). drivers. For this purpose, we used a multivariate ordina- For the study of the effect of environmental factors on tion approach, which helps to summarize the variance shelf zooplankton communities it is, therefore, desirable of a wide range of zooplankton and environmental data. that the comparisons between sites of different charac- In order to better understand the effect of latitude, we teristics are made both within and across regions. Zoo- also compared the relationships between the variations of plankton are being regularly monitored at multiple fixed zooplankton abundance and temperature at the different sites in shelf waters around the world (Mackas and Beau- sites. grand, 2010; O’Brien et al., 2013). However, up to the present, these time series from fixed sites have been greatly underutilized for comparative purposes (e.g. Bonnet et al., METHODS 2007; Castellani et al., 2016; Fanjul et al., 2017, 2018). This may be due to difficulties in accessing archived zoo- Study area and data acquisition plankton data, to differences in sampling methodologies Mesozooplankton (>200 μm) abundance, water temper- and taxonomic discrimination between time series and to ature (WT), salinity (S) and chlorophyll a (Chl a)data

294 A. FANJUL ET AL. MODULATORS OF ZOOPLANKTON ASSEMBLAGES ACROSS THE NE ATLANTIC SHELVES PROVINCE

Table 1: Main features of the sampling sites Characteristic B35 U35 L4 SH

Distance <1 <16.55 offshore (km) Water depth (m) mean 13.0 4.5 54.0 48.0 Stratification/mixing Partially mixed Mixed Transitionally Mixed/weak mixed/stratified stratification in in summer summer Salinity mean (range) 34.8 (32.9–35.5) 35.0 (30.3–35.6) 35.0 (34.0–35.4) 34.5 (33.8–34.9) Temperature (◦C) mean (range) 16.0 (11.3–23.7) 16.2 (10.8–24.9) 12.6 (7.6–19.9) 9.5 (4.5–13.9) Chlorophyll a mean (range) 2.19 (0.08–31.33) 0.82 (0.04–7.91) 1.24 (0.23–6.29) 1.29 (0.09–5.96) Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 (μg L−1) for the 1999–2013 period were obtained at the following month and that of the following month was performed four monitoring coastal sites that lie along a latitudinal to fill in occasional missing values (<5%) in the monthly gradient in the NECS (Longhurst, 1998): U35, B35, L4 data series. and SH (Fig. 1). The main features of these sites have been Identification of zooplankton was performed to the summarized in Table 1. B35 and U35 are nearshore, shal- lowest possible taxonomic level, which depended on the low sites located in close proximity (B35 at 43◦ 20.92N, expertise of the analysts involved. Only cladocerans and 3◦ 1.69W and U35 at 43◦ 24.27N, 2◦ 41.77W) in the copepods were identified to species or genera levels at all southeastern Bay of Biscay. These two sites differ in their four sites and for the entire period under study. Thus, trophic status, B35 being mesotrophic due to the influence in order to use the same taxonomic categories in the of the plume of the estuary of Bilbao (Ferrer et al., 2009), data analyses for the entire four time series, zooplankton while U35 is oligotrophic (sensu Smith et al., 1999). L4 data were grouped at two levels: (i) the herein termed (50◦ 15N, 4◦ 13W) is deeper and located in the western zooplankton group (ZG) level, which included six holo- English Channel, 6.5 km offshore. SH (56◦ 57.8N, 02◦ plankton categories (copepods, cladocerans, appendicu- 06.2W) has a water depth similar to that of L4, and larians, chaetognaths, siphonophores and doliolids) and it is located 5 km offshore in the northwest North Sea nine meroplankton categories (cirripede larvae, decapod (Fig. 1). Further information about the characteristics of larvae, gastropod larvae, bivalve larvae, polychaete larvae, these sites has been provided in Aravena et al. (2009), Bres- fish eggs and larvae, bryozoan larvae, echinoderm larvae nan et al. (2015)andAtkinson et al. (2015). Zooplankton and hydromedusae) and (ii) the holoplankton crustacean samplings were performed using 200 μm mesh size nets genera level, consisting of genera or genera-assemblages at all sites. Vertical hauls were carried out at L4 (50 m (exceptionally family) of copepods and cladocerans (cope- to surface, with WP2 nets) and SH (45 m to surface with pod and cladoceran genera; CCGen): Acartia, Centropages, bongo nets), whereas horizontal tows at mid-depth, below Temora, Oithona, Oncaea and Corycaeus (former genus that the halocline (when present), of a ring net were performed represents mainly the present genus Ditrichocorycaeus at the at B35 (∼4 m depth) and U35 (∼2 m depth). WT and four sites) genera, the ‘PCPC-calanus’ genera assemblage S were measured in situ, and water samples were taken (this includes Paracalanus, Clausocalanus, Pseudocalanus and for Chl a analysis. The values of these environmental Ctenocalanus) and the family Calanidae for the copepods, variables used in the present study correspond to surface and Evadne and Podon genera for the cladocerans. Per- ones in the case of L4 and SH and subsurface ones at B35 forming the data analyses at these two levels allowed (∼4 m depth) and U35 (∼2 m depth). Further information us to obtain information at the whole community level on sampling and analytical methods can be found in (even if it was based on coarse taxonomic differentiation) previous papers (Fanjul et al., 2017, 2018). and to obtain finer taxonomic level information for the crustacean holoplankton, which is the most abundant component of the community. Data pretreatment Prior to redundancy analyses, the zooplankton abundance data (individuals m−3) were transformed using log B35 and U35 were sampled monthly, whereas L4 and SH (x + 1; see ter Braak and Smilauer,ˇ 2002). were generally sampled weekly. For L4 and SH monthly data series were obtained by calculating the mean of all values for each month, in order to get homogeneity in Data treatment the data periodicity needed for comparative purposes. In order to assess the taxa that contributed most to Interpolation between the mean value of the previous between-site differences in mesozooplankton community

295 JOURNAL OF PLANKTON RESEARCH VOLUME 41 NUMBER 3 PAGES 293–308 2019 Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019

Fig. 1. Map of the study area showing the location of sampling sites.

and the environmental variables that best explained of the longest gradient can be used to assess if the these taxa variations, multivariate ordination analyses data are too heterogeneous and so too many species were performed using Canoco v. 4.55 (ter Braak and deviate from the assumed model of linear response Smilauer,ˇ 2002). First, taxa data alone were analyzed or not (Leps and Smilauer,ˇ 2003). Since the length of by Detrended Correspondence Analyses (DCA) in thelongestgradientwasinallcaseslessthan2,RDAs order to assess whether Canonical Correspondence (linear models) were selected (see Leps and Smilauer,ˇ Analysis or Redundancy Analysis (RDA), recommended 2003). Multicollinearity between explanatory variables for unimodal and linear relationships between taxa was checked by means of Variation Inﬂation Factor (VIF) and environmental variables, respectively (ter Braak analysis using vif function from the faraway R package and Smilauer,ˇ 2002), should be used. The lengths of (R version 3.5.2, 2018). Highly correlated variables (VIF gradients obtained in DCA are measures of the beta >4; Hair et al., 2010) were removed and not included diversity in community composition along the individual in RDA analysis. Taking this into account, in order to independent gradients (ordination axes), and the value perform the RDAs relevant water environment variables

296 A. FANJUL ET AL. MODULATORS OF ZOOPLANKTON ASSEMBLAGES ACROSS THE NE ATLANTIC SHELVES PROVINCE

routinely monitored at all sites, i.e. WT, Chl a and S and RESULTS site-specific features such as latitude and distance offshore Zooplankton community composition were used as explanatory variables. Separate redundancy analyses were conducted for the two taxonomic levels Figure 2 shows the percentage contribution of different tested (ZG and CCGen), but in both cases data pooled taxa to total ZG and CCGen abundances. At all sites for the four sites were run. Ordination along axis 1 copepods made up more than half of the total meso- reflected mainly seasonal variations in mesozooplankton zooplankton abundance (52.3–70.4%), and cirripede taxa for CCGen (spring–summer as opposed to winter). larvae were the second most abundant ZG (10–27.5%) In order to better assess differences attributable to site- at B35, U35 and L4, although at B35 copepods and specific characteristics, partial RDAs were performed cirripede larvae accounted for a markedly lower and Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 with month as covariable (thereby removing the effect higher percentage, respectively, than at U35 and L4. It is of months) and sites as supplementary variables for noteworthy that SH was characterized by a substantially both taxonomic levels tested. Monte Carlo tests were lower contribution of cirripede larvae and a higher contri- performed (499 permutations) under reduced model, bution of polychaete and bryozoan larvae than the other with unrestricted permutations and blocks defined by sites. Within CCGen, Acartia contributed most to total the covariables (ter Braak and Smilauer,ˇ 2002). In abundance (26.9–38.8%), followed by PCPC-calanus RDA forward selection of environmental variables (13.3–16.2%) at B35, U35 and SH. At L4, however, was performed, but only the marginal effects (lists of the contribution of Acartia was much lower (4.4%), and environmental variable in order of the variance they PCPC-calanus was the most abundant (22.3%) CCGen, explain singly) have been shown. followed by Oithona (11.4%) and Oncaea (10.4%) genera. Additionally, Spearman’s rank–order correlation anal- SH showed the lowest cladoceran contribution and was yses were performed between environmental factors and theonlysiteinwhichPodon was more abundant than scores on each axis, in order to better assess the environ- Evadne. mental factors that best explained the pattern shown by Information on the species composition and the relative each ordination axis. contribution to the total abundance of each CCGen at Latitude may be considered itself a surrogate for some each of the sites under study is available as supplementary other underlying mechanisms that are typically not well material in Fanjul et al. (2017). understood (Iken et al., 2010). Although temperature (both annual mean and range of variation) is linked to latitude, the combined effect of latitude and temperature Differences in zooplankton community and on species distribution and community changes is still environmental drivers unclear. For a more accurate analysis of the quantitative Results of the RDA for ZG revealed that the environmen- response of taxa to temperature in the latitudinal context, tal variables tested explained 16.2% of taxa variations, models of the relationship between temperature and and the main mode of variation (axis 1) explained 62.2% taxon abundance (log abundance +1) were obtained. of this taxa–environment relationship. Overall, latitude, For this purpose, polynomial orthogonal regression followed by distance offshore and WT were the factors analyses (to control for multicollinearity) were performed that best explained ZG variability (Table 2). Ordination on all taxa taking temperature up to grade 3 as the along axis 1 (Fig. 3A) evidenced differences between independent variable. Subsequent tests on significance sites, where doliolids and cirripede larvae were the taxa and constraints of equality of effects among different sites with the highest relationship with B35 and U35, and were performed in order to get the most representative bryozoan, polychaete and echinoderm larvae were more and parsimonious model. Models for each site, as well related to SH and L4. Variations of ZG taxa scores on as common models for the four sites, were tested for axis 1 appeared to be related primarily to latitude, and each taxon. In zero inflated models of the least frequent sequentially to a lesser extent to distance offshore, WT taxa, the high occurrence of zeros associated with high and S (Table 3). The second main mode of variation or low temperatures is not a problem for the correct (axis 2) accounted for 32.3% of the taxa–environment interpretation of the response to temperature. However, relationship that could be explained by the environmental when a high abundance of zero values is distributed more variables tested. The taxa that contributed most to this or less regularly along the entire range of temperatures second mode of variation were fish eggs and larvae, analyzed, the results should be taken with caution, since siphonophores and hydromedusae, that showed highest this would denote the relevance of factors different abundances at L4 and lowest ones at U35 and SH, in from temperature as responsible for the absence of contrast to gastropod larvae, that appeared to be more such taxa. linked to SH and U35 (Fig. 3A). Distance offshore and S

297 JOURNAL OF PLANKTON RESEARCH VOLUME 41 NUMBER 3 PAGES 293–308 2019 Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019

Fig. 2. Relative abundance (+ standard error), expressed as percentage of total zooplankton, of ZG (upper panel) and CCGen (lower panel) taxa at the four sampling sites under study. were the factors that best correlated with taxa scores on Oithona, which were those that best correlated to distance axis 2, followed by Chl a and latitude (Table 3), suggesting offshore, in opposition to Oncaea, Corycaeus, Podon and that these two last variables had key contributions to Evadne, which were the taxa that best correlated to S explain the differences in taxa abundance between L4 (Fig. 3B). and SH and between B35 and U35, as shown by the ordination of variables and sites along axis 2.For CCGen the environmental variables tested explained 23.2% of Relationships between taxa abundance taxa variations, and the main mode of variation (axis 1) and WT explained 70% of this taxa–environment relationship The models fitted for the relationship between the abun- (Fig 3B). Distance offshore and latitude, followed by dance of ZG and WT are shown in Fig. 4. Among the WT and Chl a, were the environmental factors that taxa that showed a relationship with latitude in the RDA, explained the largest proportion of CCGen variability doliolids evidenced a pattern of linear increase in abun- (Table 2). Site scores on axis 1 showed mainly differences dance with increasing WT at the four sites under study, between the CCGen at L4 and the rest of sites tested. and they showed the same or very similar quantitative The taxa that contributed most to these differences (abundance) response to temperature at all sites. Thus, were Corycaeus and Oncaea, which were most related to a global model could be fitted to log (abundance +1) L4, in contrast to Acartia,mostabundantattherest data pooled for all sites. Cirripede larvae did not show of sites. Although distance offshore appeared as the the same relationship with WT along the entire range main environmental factor related to these differences of temperatures, instead, a second-degree polynomial in CCGen between sites, all the other variables tested curve could be fitted for all sites, which showed minima were also strongly related (significant correlations to at intermediate temperatures ∼15◦C. Bryozoan larvae P < 0.001 level; Fig 3B and Table 3). Axis 2 accounted showed no significant differences in abundance over a for 25% of the taxa–environment relationship tested large range of WTs, but they also showed increases in for CCGen. According to correlation analyses, latitude abundance at temperatures below 1◦C at SH and L4. and WT were the factors that best correlated with Although global models for cirripede and bryozoan larvae this second axis, although distance offshore and S abundances could be fitted for data pooled for the four were also significant (Table 3). Taxa with the highest sites, such models could not account for the between-site contributions to axis 2 were Calanidae, Temora and differences in abundance for a given WT. In the caseof

298 A. FANJUL ET AL. MODULATORS OF ZOOPLANKTON ASSEMBLAGES ACROSS THE NE ATLANTIC SHELVES PROVINCE

Table 2: Marginal effects of environmental variables for ZG and CCGen. Variables with signiﬁcant effects in bold ZG CCGen

Variable Lambda1 F P-value Lambda1 F P-value

Latitude 0.10 111.84 0.002 0.09 125.02 0.002 Distance 0.04 54.60 0.002 0.13 138.59 0.002 WT 0.01 12.56 0.002 0.001 14.40 0.002 Chl a 0.01 7.15 0.002 0.00 3.08 0.038 Sal 0.00 3.29 0.004 0.01 1.72 0.090 Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 Distance, distance offshore; WT, water temperature; Sal, salinity; Chl a, concentration of chlorophyll a.

Fig. 3. RDA triplot for ZG (A) and CCGen (B). Taxa are shown by thin arrows, explanatory variables by thick arrows and sites by triangles. Acar, Acartia; Appe, appendicularians; Biva, bivalve larvae; Bryo, bryozoans; Cala, Calanidae; Cent, Centropages; Chae, chaetognaths; Cirr, cirripede larvae; Clad, cladocerans; Cope, copepods; Cory, Corycaeus; Deca, decapod larvae; Doli, doliolids; Echi, echinoderm larvae; Evad, Evadne; Fish, ﬁsh eggs and larvae; Gast, gastropod larvae: Hydr, hydromedusae: Oith, Oithona;Onca,Oncaea; PCPC, PCPC-calanus; Podo, Podon;Poly,polychaete larvae; Siph, siphonophores; Temo, Temora. Environmental variable abbreviations as in Table 2.

Table 3: Correlation coefficients (with P-values in parentheses) between environmental factors and scores on axis 1 and axis 2 for ZG and CCGen. Signiﬁcant correlations in bold

ZG CCGen

Axis 1 Axis 2 Axis 1 Axis 2

Latitude −0.828 (<0.001) −0.183 (<0.001) 0.332 (<0.001) 0.637 (<0.001) Distance −0.746 (<0.001) 0.299 (<0.001) 0.679 (<0.001) 0.378 (<0.001) WT 0.575 (<0.001) 0.120 (0.001) −0.176 (<0.001) −0.551 (<0.001) Sal 0.219 (<0.001) 0.264 (<0.001) 0.133 (<0.001) −0.322 (<0.001) Chl a −0.062 (0.098) 0.196 (<0.001) 0.137 (<0.001) −0.018 (0.638)

Distance, distance offshore; WT, water temperature; Sal, salinity; Chl a, concentration of chlorophyll a. polychaete larvae, no significant response to temperature degree polynomials could be fitted for the abundance at any of the sites under study was found, and in the of echinoderm larvae at SH and L4 respectively, but no case of echinoderm larvae, no common model for the significant responses to WT could be observed at B35 and four sites under study was obtained. Second and third U35.

299 JOURNAL OF PLANKTON RESEARCH VOLUME 41 NUMBER 3 PAGES 293–308 2019 Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019

Fig. 4. Models of WT versus log (taxon abundance +1) for ZG. WT (temp) in ◦C and abundance in individuals m−3.

300 A. FANJUL ET AL. MODULATORS OF ZOOPLANKTON ASSEMBLAGES ACROSS THE NE ATLANTIC SHELVES PROVINCE Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019

Fig. 5. Models of WT versus log (taxon abundance +1) for CCGen. WT (temp) in ◦C and abundance in individuals m−3.

At the coinciding range of WTs, a larger number of larvae showed a pattern of decreasing abundance with ZG were more abundant at SH (bryozoan, polychaete, increasing latitude (most abundant at B35 or U35 echinoderm, gastropod and bivalve larvae) and L4 than at L4 and SH) at the range of WTs at which (siphonophores, hydromedusae, copepods and ﬁsh eggs they were present at all sites, but their abundance was and larvae) than at the rest of sites. Only cirripede much higher at B35 than at U35. Appendicularians

301 JOURNAL OF PLANKTON RESEARCH VOLUME 41 NUMBER 3 PAGES 293–308 2019

and cladocerans did not show differences in abundance coastal doliolids appeared to be more related to temper- between sites within the same range of temperatures, but ature than to Chl a concentration, since at the same tem- appendicularian abundance was found to be independent perature similarly high abundances were observed at both of WT at all sites, while cladocerans showed a common the mesotrophic and oligotrophic sites of the southern pattern of variation with WT at all sites, with the Bay of Biscay and minimum values in the northern North optimum temperature lying between 17 and 20◦C. Sea site. This agrees well with the finding by Deibel and The models of the relationship between CCGen abun- Lowen (2012) that doliolids operate at a generation time dance and WT (Fig. 5) showed that Corycaeus had the fixed primarily by temperature and secondarily by food clearest warm water affinity and the log (abundance +1) concentration. It is interesting to note, also, that the rela- versus WT relationship could be fitted to linear models of tionship between doliolid abundance and temperature is Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 common slope for the four sites under study. Oncaea also very similar at all sites, reinforcing the view that latitudinal showed a clear increase of abundance with increasing differences in density are linked primarily to differences in temperature at the southernmost sites but it showed no temperature. relation to temperature at L4 and SH. However, both The relationship with latitude did not seem to genera were much more abundant at L4 than at the other be temperature-mediated in the same way for the sites at any temperature. In fact, L4 was the site where meroplankton taxa that contributed most to the main the largest number of CCGen was most abundant at a mode (axis 1) of zooplankton variability because the given temperature. PCPC-calanus and Oithona were two WT versus abundance models for cirripede, bryozoan and such CCGen. The abundance of Oithona was not related polychaete larvae showed large between-site differences at to WT at any site, and for PCPC-calanus different non- a given temperature. The higher abundance of cirripede linear models were fitted at different latitudinal locations, larvae at B35 and U35 may be related primarily to the fact showing two optima, one at about 7◦C and another one that these sites are shallower and closer to shore than L4 at >20◦C. For the rest of CCGen taxa studied, the abun- and SH. Continental shelf benthos generally decreases dance versus WT relationships were similar at the four in abundance with increasing depth (Rex et al., 2006; sites under study. Among them, Temora showed the most Nephin et al., 2014). In the case of barnacles, many of thermophilic behavior, followed by Centropages, Podon and them form thick belts in intertidal rocky shores. Intertidal Evadne. Calanidae and Acartia had the lowest temperature barnacle larvae tend to be more abundant in nearshore optimum. For common ranges of temperature, Centropages waters and they can become very rare or absent in areas was the only CCGen that showed a pattern of increase >5 km offshore (Shanks and Shearman, 2009). However, in abundance with latitude. Temora and Calanidae were our data showed noticeable differences in cirripede larvae much more abundant at SH and L4 than at B35 and abundance between B35 and U35, this suggesting the U35. Podon was less abundant at U35 than at the other influence of a factor linked to trophic condition, since Chl sites and Acartia was the only CCGen that showed lowest a concentrations were much lower at U35 than at B35. abundances at L4 as compared to the rest of sites for any Bryozoan and polychaete larvae were the ZG that given temperature. showed the highest association with high latitude/low temperature sites, the densities of both taxa being highest at SH and lowest at U35 and B35 for most of the year. DISCUSSION Furthermore, bryozoan larvae were most abundant in winter–early spring at L4 and SH and later in spring at Zooplankton differences at coarse group B35andU35(Fanjul et al., 2018). Accordingly, bryozoan level (ZG) larvae were also found to peak in early spring in Gal- The main between-site differences in ZG assemblages way Bay (Irish coast; Byrne, 1995) and in winter in a showed a gradient from the southernmost stations to fjord in high-Arctic Svalbard (Stübner et al., 2016). As for the northernmost one, represented by a closer associ- cirripedes, the large between-site differences in bryozoan ation of groups such as doliolids and cirripede larvae and polychaete larvae densities at a given temperature to B35 and U35 and bryozoan and polychaete larvae suggest an additional site effect unrelated to temperature. to SH. Mesoscale patchy spatial variations in doliolid Regarding bryozoans, low temperature seems to nega- abundance in shelf waters are often related to intru- tively affect the growth rate of bryozoans, with higher sions of nutrient-rich water and associated phytoplank- latitude bryozoans tending to grow relatively more slowly ton biomass increases (Deibel and Paffenhoffer, 2009; (Smith and Lawton, 2010). Trophic condition does not Liao et al., 2013; Villate et al., 2014). However, in our seem to be relevant to account for between-site differences study area, spanning a significant latitudinal and temper- in bryozoan larvae abundance either. Most bryozoan ature range, between-site differences in the abundance of larvae are suggested to have little or no dependence on

302 A. FANJUL ET AL. MODULATORS OF ZOOPLANKTON ASSEMBLAGES ACROSS THE NE ATLANTIC SHELVES PROVINCE

phytoplankton as food (Stübner et al., 2016), and this (1947) found that ophiopluteii of Ophiothrix fragilis were would agree with bryozoan larvae maxima not coinciding the commonest echinoderm larvae in the inshore waters with the main phytoplankton spring or summer maxima of Plymouth in the 1940s. In agreement with this, in at any of the sites under study. In any case, the latitude- the L4 time series analyzed in the present work, within related differences in bryozoan larvae observed in this the echinoderm larvae that have been identiﬁed to a study seem to be supported by the increasing contribution coarse taxonomic level, ophiopluteii are the most abun- of bryozoan to benthic communities toward the Arctic dant ones (Fanjul et al., 2017). These high echinoderm region, where they are often the dominant component larval abundances are, in turn, in accordance with the in hard substrate and phytal habitats (Bader and Schäfer, occurrence of high-density aggregations in the seabed

2005). around the British Isles (particularly on the western side) Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 As for bryozoan larvae, polychaete larvae also showed of the ophiuroid species O. fragilis and Ophiocomina nigra in decreasing densities from north to south. Accordingly, areas of moderate to strong current (Aronson, 1989). No an increased abundance of benthic polychaetes from the information on the type of echinoderm larvae is avail- southern to the northern North Sea was also reported able for SH, and the benthic community at Stonehaven by Quiroz-Martinez et al. (2011), although regional pat- has been little studied. Although ophiopluteii were more terns of benthic polychaete distribution in continental abundant than echinopluteii at B35 and U35, in the Abra shelves seem to be mainly related to bottom water stability, bay where B35 is located, the sea urchin Paracentrotus lividus local distribution of sediment types and depth (Flint and was suggested to be the most abundant echinoderm in Rabalais, 1980; Quiroz-Martinez et al., 2011). The lack the 1980s, with maximum densities up to 20 individuals of a significant relationship between polychaete larvae m−2 (Arteche-Irueta, 1987). Regarding ophiuroids, on abundance and temperature at any site means that no the Basque coast Ophiotrix fragilis and Amphipholis squamata local thermal optima could be determined for them. have been observed in intertidal areas and Ophioderma However, highest levels of polychaete larvae occur at longicauda and O. nigra in deeper areas (Ibañez-Artica, L4 once temperature reaches 13–14◦C(Highfield et al., 2018), but no such dense aggregates as in waters around 2010), and their annual maxima varies from winter at B35 the British Isles have been reported. The comparison of and U35 to summer at SH (Fanjul et al., 2018). Taking studies carried out at different sites of the Basque coast data pooled for the four sites, the highest abundances also corroborates the offshore–inshore decrease in the were found at an intermediate range of temperatures contribution of echinoderm larvae to the total zooplank- (11–13◦C). ton (Villate et al., 2004). Cladocerans were one of the taxa for which a common Overall, despite their lower contribution to total model of abundance with temperature for the four sites zooplankton abundance, meroplankton contributed under study could be fitted. This model explained the more than holoplankton to between-site differences between-site differences in abundance quite well. Clado- in zooplankton structure. This seems to be due to cerans were also the ZG with the highest correlation with the widespread expatriation of planktonic species, in Chl a and S. They are filter-feeders feeding mainly on contrast to the smaller spatial scale resolution of benthic phytoplankton (Brown et al., 1997), and Chl a concen- habitat/communities, which are spatially constrained tration was suggested to be the most important factor not only by water column features, but also by seafloor determining the spatial distribution of cladocerans in features, thus showing higher spatial heterogeneity than shelf waters of the South China Sea (Xiong et al., 2012). planktonic ones (Costello, 2009; Guarinello et al., 2010). Chl a-driven differences in cladoceran abundance can It has to be noted that differences in sampling methods help to explain the lower density of cladocerans at the (vertical versus horizontal towing) could potentially oligotrophic U35 site in relation to the mesotrophic B35. introduce some bias due to differences in the vertical The positive relationship with S in the present study was positioning of zooplankton, particularly of meroplankton due to SH showing somewhat lower S values and low- (Ayata et al., 2011), but this is expected to be small because est cladoceran abundances. This relation has also been of the shallowness of B35 and U35 sites. observed at local scales, since in estuaries of the Basque coast penetration of cladocerans in low S waters is also limited (Villate et al., 2017). Zooplankton differences at CCGen level In addition to latitudinal differences, echinoderm lar- The main between-site differences in the CCGen vae showed high correlation with distance offshore, and assemblage explainable by the environmental factors this is because they were much more abundant at the under study were those between L4 and the rest of deeper/more offshore sites (L4 and SH) than at the shal- sites. Although these differences appeared primarily lower sites close to the coast (U35 and B35). Lebour related to distance offshore, the main taxaOncaea ( and

303 JOURNAL OF PLANKTON RESEARCH VOLUME 41 NUMBER 3 PAGES 293–308 2019

Corycaeus) responsible for such differences showed higher the case for total cladocerans in the analysis of the ZG. correlation with S. This result highlights the importance Lowest Podon and Evadne abundances at SH were mainly of local features in driving zooplankton compositional due to the lower abundance of the species P. intermedius differences, as was also found by Pepin et al. (2015), who and Evadne nordmanni, which was not balanced by the showed mesoscale features to be major factors shaping the abundance of Podon leuckartii, the dominant Podon species zooplankton structure within each of three large marine at the highest latitude SH site. The distribution of this ecosystems (Newfoundland Shelf, Gulf of St. Lawrence latter species seems to be determined also by temperature, and Scotian Shelf). Together with the higher abundance since P. leuckartii has a higher affinity for cold waters than of Corycaeus and Oncaea, and to a lesser extent also of P. intermedius (Onbé, 1999; Viñas et al., 2007) and has not

Calanidae at L4, the dominance of Acartia at B35, U35 been identified in the zooplankton of the western English Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 and SH was the main responsible for such differences. A Channel or Bay of Biscay sites under study. Furthermore, likely explanation for the substantially higher abundance at SH, unlike at the rest of the sites in the present work, the of Corycaeus and Oncaea atL4,evenwhencomparedto cladoceran species Pleopis polyphemoides is present, which SH which is also a deeper and more offshore site than has been found to be a more euryhaline species than P. B35 and U35, is that L4 is affected by intrusions of intermedius (Viñas et al., 2007). saltier off-shelf water from the Atlantic that bring those For some CCGen, the models of the relationship taxa. In fact, Corycaeus anglicus (Ditrichiocorycaeus anglicus), between taxon abundance and temperature showed a which is the dominant Corycaeus species at L4, is found common pattern for the four sites under study, which to be a good indicator of Atlantic oceanic water inflow accounted for the between-site differences in the seasonal to the North Sea through the English Channel (Bonnet pattern of those taxa. This was the case for Acartia and Frid, 2004), and Oncaea has also been found to be (almost exclusively A. clausi at all four sites), which showed representative of cross-shelf intrusions on the continental highest abundances at intermediate temperatures within shelf of northeastern Florida (Paffenhofer et al., 1984). the whole range of temperatures registered at the four In contrast, Acartia clausi (the dominant Acartia at the sites, but whose maxima coincided with the highest four sites under study) is a neritic species (Wootton and summer temperatures at the northernmost SH site and Castellani, 2017). Therefore, CCGen ordination along were closer to the lowest temperatures in early spring axis 1 seemed to represent differences between sites at the southernmost B35 and U35 sites. This common associated with the neritic/oceanic nature of zooplankton temperature optimum across a large latitudinal gradient taxa. Other studies have also highlighted the influence is important because it suggests the lack of temperature of inputs of oceanic water on the zooplankton species adjustment. Constancy of thermal niche is an important, composition of shelf waters (Pedersen et al., 2000; but rarely tested assumption in many species distribution Beare et al., 2002). models that are predicated on a fixed thermal niche Differences explained by the higher contribution of (Beaugrand et al., 2014). Instead, the phenology of A. Calanidae, Oithona and Temora to the zooplankton at SH clausi was found to be particularly temperature-sensitive and L4, as opposed to the higher contribution of Podon at L4 (Atkinson et al., 2015). We realize that we have and Evadne at L4, U35 and B35, were related to a large not estimated thermal niches, mainly because this study extent to distance offshore and S, respectively. These has not been conducted at the species level, but our species have been found to show vertical segregation in data are consistent with A. clausi having a fixed thermal the water column that might contribute to explain the niche. Therefore, adjustments in seasonal timing, to differences between inshore and offshore sites because of occur at more suitable seasonal temperatures, may be the close positive relation between distance offshore and a mechanism by which this particular species maintains a depth (r = 0.95, P < 0.001) at the studied sites. Calanus fixed thermal niche. Further studies should be conducted, helgolandicus, Oithona similis and Temora longicornis, the main though, to provide more evidence on the constancy of this Calanidae, Oithona and Temora species, respectively, at species thermal niche. L4 and SH, are from intermediate or deep layers over For other CCGen the models also showed a very similar the shelf (Vives, 1980; Villate, 1994; although Halvorsen shape of the temperature versus abundance curve at all et al., 1999 reported T. longicornis as a surface species four sites, but between-site differences in abundance for a in shelf waters in northern Norway). Evadne nordmanii given temperature. This may be attributed to between- and Podon intermedius, the main Evadne and Podon species, site compositional differences at the species level. For respectively, at L4, B35 and U35 are mostly found at example, the abundance of Temora was much higher at shallower depths than the former copepod species (Vives, L4 and SH than at U35 and B35, and the abundance 1980; Villate, 1994). However in this case, the between- of Centropages was highest in the North Sea site, inter- site differences were mostly associated with S, as was mediate at the English Channel site and lowest at the

304 A. FANJUL ET AL. MODULATORS OF ZOOPLANKTON ASSEMBLAGES ACROSS THE NE ATLANTIC SHELVES PROVINCE

Bay of Biscay sites. The dominant Temora at L4 and level, the main differences showed an association with the SH is T. longicornis, and the dominant Centropages at SH latitudinal gradient, but only some holoplanktonic ZG is C. hamatus, which are neritic boreal, cold-temperate abundances evidenced temperature-mediated latitudinal species (Colebrook, 1964; Halsband-Lenk et al., 2002), differences, the warm water affinity doliolids being the whereas at U35 and B35 the dominant species is T. best indicators of latitudinal differences based on temper- stylifera and at L4, U35 and B35 it is C. typicus,which ature. Cladocerans were also a group for which the abun- are southern, warm-temperate or intermediate latitude dance versus temperature model explained the between- species, respectively (Colebrook, 1964; Halsband-Lenk site differences relatively well, although these were also et al., 2002). The higher abundance for a given temper- affected by trophic condition, and their abundances did ature of T. longicornis and C. hamatus in boreal regions not show a clear-cut latitudinal gradient throughout the Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 than of T. stylifera and C. typicus, respectively, in warm year. Bryozoan, polychaete, cirripede and echinoderm temperate regions, that has been observed in the present larvae abundances showed marked latitudinal differences, work, has also been reported elsewhere (Halsband-Lenk but these differences did not seem to be primarily linked et al., 2004). In the case of PCPC-calanus, very similar to differences in temperature with latitude. Cirripede temperature versus abundance models could be fitted to and echinoderm larvae, for example, were more affected data from U35 and B35 but no common model could by local features such as water depth, distance offshore be fitted for the four sites. PCPC-calanus abundance and, in the case of cirripedes, also of phytoplankton peaked both at low (i.e. ∼7◦C at L4 and SH) and high availability. temperatures (i.e. 19–21◦C at L4, B35 and U35), and At the level of CCGen, other local factors appeared to decreased at intermediate temperatures (i.e. 10–15◦C). have a greater influence than latitude in between-site dif- However, peaks at low temperature correspond mainly ferences, where Corycaeus and Oncaea seemed to be the best to the cold-water species Pseudocalanus elongatus that is the indicators of off-shelf water intrusions from the Atlantic most abundant species within this group of genera at the at L4. Copepods of the family Calanidae and the genera highest latitude SH site, and peaks at high temperature Temora and Oithona together with the cladoceran genera corresponded to Paracalanus parvus, a neritic warm-water Podon and Evadne were useful indicators of distance off- species that was the most abundant PCPC-calanus species shore that might be associated to bathymetry differences. at the lowest latitude B35 and U35 sites (Fanjul et al., When the genera were dominated by a single species 2017). The model at L4 reflected the similar relative (e.g. A. clausi), data were consistent with a fixed thermal abundance of P.parvus and P.elongatus at this intermediate niche, the species adjusting its seasonal timing at different latitude site. latitudes so as to occur at more suitable temperatures. All this suggests that a predominant latitudinal mode Overall, local conditions played a significant role in of zooplankton variation, more in agreement with that shaping the zooplankton community structure within observed for ZG, might have been obtained from the NECS, particularly at the finer taxonomic CCGen level. comparison of copepod and cladoceran assemblages, if Therefore, mesoscale variability in environmental condi- species instead of genera could have been used. However, tions should not be overlooked in studies of biogeographic this was not possible in the present study because indi- variability in zooplankton community structure. viduals of some genera were not distinguished to species level at all the four sites and/or throughout the entire time series. FUNDING In addition, it has to be born in mind that temporal Spanish Ministry of Economy and Competitiveness variations in zooplankton abundance can also be influ- (CGL2013-47607-R); Natural Environment Research enced by other factors, such as mortality through preda- Council’s (NERC) National Capability (to L4 time series tion (Irigoien and Harris, 2003; Hirst et al., 2007; Corn- and A.A.) and we would like to thank all the ship crew well et al., 2018), but unfortunately this is a variable that and scientists in providing these data; Department for it is not routinely measured in zooplankton monitoring Environment, Food and Rural Affairs (NE/L003279/1 to programs. A.A.) Marine Ecosystems Research Programme. Marine Scotland Science data were collected under Scottish CONCLUSION Government Service Level Agreement ST03p. At the level of coarse ZG, meroplankton contributed more than holoplankton to the between-site differences in REFERENCES zooplankton structure, which may be related to the higher Arteche-Irueta, I. 1987. Resultados Programa Abra: Equinodermos. degree of expatriation of plankton. At this taxonomic Kobie XVI: 307–314 (In Spanish).

305 JOURNAL OF PLANKTON RESEARCH VOLUME 41 NUMBER 3 PAGES 293–308 2019

Aravena, G., Villate, F., Uriarte, I., Iriarte, A. and Ibañez, B. Colebrook, J. M. (1964) Continuous plankton records: a principal com- (2009) Response of Acartia populations to environmental vari- ponent analysis of the geographical distribution of zooplankton. Bull. ability and effects of invasive congenerics in the estuary of Mar. Ecol., 6, 78–100. Bilbao, Bay of Biscay. Estuar. Coast. Shelf Sci., 83, 621–628 Cornwell, L. E., Findlay, H. S., Fileman, E. S., Smyth, T. J., Hirst, A. doi:10.1016/j.ecss.2009.05.013. G., Bruuns, J. T., McEvoy, A. J., Widdicombe, C. E. et al. (2018) Aronson, R. B. (1989) Brittlestar beds: low-predation anachronisms in Seasonality of Oithona similis and Calanus helgolandicus reproduction the British Isles. Ecology, 70, 856–865 doi:10.2307/1941354. and abundance: contrasting responses to environmental variation at Atkinson, A., Harmer, R. A., Widdicombe, C. E., McEvoy, A. J., Smyth, a shelf site. J. Plankton Res., 40, 295–310 doi:10.1093/plankt/fby007. T. J., Cummings, D. G., Somerﬁeld, P. J., Maud, J. L. et al. (2015) Costello, M. J. (2009) Distinguishing marine habitat classiﬁcation con- Questioning the role of phenology shifts and trophic mismatching cepts for ecological data management. Mar. Ecol. Prog. Ser., 397,

in a planktonic food web. Prog. Oceanogr., 137, 498–512 doi:10.1016/ 253–268. Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 j.pocean.2015.04.023. Deibel, D. and Paffenhoffer, G. A. (2009) Predictability of patches of Ayata, S., Stolba, S., Comtet, T. and Thiébaut, É. (2011) Meroplankton neritic salps and doliolids (Tunicata, Thaliacea). J. Plankton Res., 31, distribution and its relationship to coastal mesoscale hydrological 1571–1579 doi:10.1093/plankt/fbp091. structure in the northern Bay of Biscay (NE Atlantic). J. Plankton Res., 33, 1193–1211 https://doi.org/10.1093/plankt/fbr030. Deibel, D. and Lowen, B. (2012) A review of the life cycles and life- history adaptations of pelagic tunicates to environmental conditions. Bader, B. and Schäfer, P. (2005) Bryozoans in polar latitudes: Arctic ICES J. Mar. Sci., 69, 358–369 doi:10.1093/icesjms/fsr159. and Antarctic bryozoan communities and facies. Denisia 16zugleich Kataloge der OÖ. Landesmuseen Neue Serie, 28, –263, 282. Dvoretsky, V. G. and Dvoretsky, A. G. (2015) Regional differences of mesozooplankton communities in the Kara Sea. Cont. Shelf Res., 105, Beare, D. J., Batten, S., Edwards, M. and Reid, D. G. (2002) Prevalence 26–41 doi:10.1016/j.csr.2015.06.004. of boreal Atlantic, temperate Atlantic and neritic zooplankton in the North Sea between 1958 and 1998 in relation to temperature, Fanjul, A., Villate, F., Uriarte, I., Iriarte, A., Atkinson, A. and Cook, K. salinity, stratification intensity and Atlantic inflow. J. Sea Res., 48, (2017) Zooplankton variability at four monitoring sites of the North- 29–49 doi:10.1016/S1385-1101(02)00131-4. east Atlantic shelves differing in latitude and trophic status. J. Plankton Res., 39, 891–909 doi:10.1093/plankt/fbx054. Beaugrand, G., Goberville, E., Luczak, C. and Kirby, R. R. (2014) Marine biological shifts and climate. Proc. R. Soc. B, 281, http://dx. Fanjul, A., Iriarte, A., Villate, F., Uriarte, I., Atkinson, A. and Cook, K. doi.org/10.1098/rspb.2013.3350. (2018) Zooplankton seasonality across a latitudinal gradient in the Northeast Atlantic Shelves Province. Cont. Shelf Res., 160, 49–62 Blachowiak-Samolyk, K., Søreide, J. E., Kwasniewski, S., Sundfjord, A., doi:10.1016/j.csr.2018.03.009. Hop, H., Falk-Petersen, S. and Hegseth, E. N. (2008) Hydrodynamic control of mesozooplankton abundance and biomass in northern Ferrer, L., Fontán, A., Mader, J., Chust, G., González, M., Valencia, Svalbard waters (79–81◦N). Deep Sea Res. II, 55, 2210–2224 doi:10. V., Uriarte, A. and Collins, M. B. (2009) Low-salinity plumes in the 1016/j.dsr2.2008.05.018. oceanic region of the Basque Country. Cont. Shelf Res., 29, 970–984 doi:10.1016/j.csr.2008.12.014. Bonnet, D. and Frid, C. (2004) Seven copepod species considered as indicators of water-mass influence and changes: results from a Flint, R. W. and Rabalais, N. N. (1980) Polychaete ecology and Northumberland coastal station. ICES J. Mar. Sci., 61, 485–491 niche patterns: Texas continental shelf . Mar. Ecol. Prog. Ser., 3, doi:10.1016/j.icesjms.2004.03.005. 193–202. Bonnet, D., Harris, R., Lopez-Urrutia, A., Halsband-Lenk, C., Greve, Guarinello, M. L., Shumchenia, E. J. and King, J. W. (2010) Marine W., Valdes, L., Hirche, H. J., Engel, M. et al. (2007) Comparative habitat classification for ecosystem-based management: a proposed seasonal dynamics of Centropages typicus at seven coastal monitoring hierarchical framework. Environ. Manag., 45, 793–806 doi:10.1007/ stations in the North Sea, English Channel and Bay of Biscay. Prog. s00267-010-9430-5. Oceanogr., 72, 233–248 doi:10.1016/j.pocean.2007.01.007. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E. and Tatham, Bresnan,E.,Cook,K.B.,Hughes,S.L.,Hay,S.J.,Smith,K.,Walsham, R. L. (2010) Multivariate Data Analysis7th edn, Pearson, New York, P. and Webster, L. (2015) Seasonality of the plankton community at p. 734. an east and west coast monitoring site in Scottish waters. J. Sea Res., Halsband-Lenk, C., Hirche, H. J. and Carlotti, F. (2002) Temperature 105, 16–29 doi:10.1016/j.seares.2015.06.009. impact on reproduction and development of congener copepod pop- Brown, M. R., Jeffrey, S. W., Volkman, J. K. and Dunstan, G. A. (1997) ulations. J. Exp. Mar. Biol. Ecol., 271, 121–153 doi:10.1016/S0022- Nutritional properties of microalgae for mariculture. Aquaculture, 151, 0981(02)00025-4. 315–331 doi:10.1016/S0044-8486(96)01501-3. Halsband-Lenk, C., Carlotti, F. and Greve, W. (2004) Life-history Byrne, P. (1995) Seasonal composition of meroplankton in the strategies of calanoid congeners under two different climate regimes: Dunkellin estuary, Galway Bay. PROC. R. IR. ACAD. B COOK , 95B, a comparison. ICES J. Mar. Sci., 6, 709–720 doi:10.1016/j.icesjms. 35–48. 2004.03.020. Caddy, J. F., Carocci, F. and Coppola, F. (1998) Have peak fishery Halvorsen, E., Tande, K. S. and Høisæter, T. (1999) Physical and production levels been passed in continental shelf areas? Some per- biological factors influencing the seasonal variation in distributrion spectives arising from historical trends in production per shelf area. of zooplanton across the shelf at Nordvestbaskem, northern J. Northwest Atl. Fish. Sci., 23, 191–219. Norway, 1994. Sarsia, 84, 279–292 doi:10.1080/00364827.1999. Castellani, C., Licandro, P., Fileman, E., di, I. and Mazzocchi, M. G. 10420432. (2016) Oithona similis likes it cool: evidence from two long-term time Highfield, J. M., Eloire, D., Conway, D. V. P., Lindeque, P. K., Attrill, series. J. Plankton Res., 38, 703–717 doi:10.1093/plankt/fbv104. M. J. and Somerfield, P.J. (2010) Seasonal dynamics of meroplankton

306 A. FANJUL ET AL. MODULATORS OF ZOOPLANKTON ASSEMBLAGES ACROSS THE NE ATLANTIC SHELVES PROVINCE

assemblages at station L4. J. Plankton Res., 32, 681–691 doi:10.1093/ Quiroz-Martinez, B., Schmitt, F. G., Dauvin, J.-C., Dewarumez, J.-M., plankt/fbp139. Foveau, A. and Garcia, C. (2011) Regional patterns of continen- Hirst, A. G., Bonnet, D. and Harris, R. P.(2007) Seasonal dynamics and tal shelf polychaete diversity: examples for the North Sea, English mortality rates of Calanus helgolandicus over two years at a station in Channel, Irish Sea and outer Bristol Channel areas. Ital.J.Zool., 78, the English Channel. Mar. Ecol. Prog. Ser., 340, 189–205 doi:10.3354/ 324–332 doi:10.1080/11250003.2011.591183. meps340189. R Development Core Team (2018). R: a language and environment Ibañez-Artica, M. (2018) ZoologíaEnciclopedia Auñamendi. retrieved for statistical computing. R Foundation for Statistical Computing, on 03 May 2018 http://aunamendi.eusko-ikaskuntza.eus/es/ Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project. zoologia/ar-147554/. org. Rex, M. A., Etter, R. J., Morris, J. S., Crouse, J., McClain, C. R., Iken, K., Konar, B., Benedetti-Cecchi, L., Cruz-Motta, J. J., Knowlton,

Johnson, N. A., Stuart, C. T., Deming, J. W. et al. (2006) Global Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 A., Pohle, G., Mead, A., Miloslavich, P. et al. (2010) Large-scale spatial bathymetric patterns of standing stock and body size in the deep-sea distribution patterns of echinoderms in Nearshore rocky habitats. benthos. Mar. Ecol. Prog. Ser., 317,1–8doi:10.3354/meps317001. PLoS One, 5, e13845 doi:10.1371/journal.pone.0013845. Ribera d’Alcalà, M., Conversano, F., Corato, F., Licandro, P. and Calanus Irigoien, X. and Harris, R. P. (2003) Interannual variability of Mangoni, O. (2004) Seasonal patterns in plankton communities in helgolandicus Fish. Oceanogr. in the English Channel. , 12, 317–326 a pluriannual time series at a coastal Mediterranean site (gulf of doi:10.1046/j.1365-2419.2003.00247.x. Naples): an attempt to discern recurrences and trends. Sci. Mar., 68, Lebour, M. V.(1947) Notes on the inshore plankton of Plymouth. J. Mar. 65–83 doi:10.3989/scimar.2004.68s165. Biol. Assoc. UK , 26, 527–547 doi:10.1017/S0025315400013734. Richardson, A. J. (2008) In hot water: zooplankton and climate change. Leps,J.andSmilauer,ˇ P.(2003) Multivariate Analysis of Ecological Data Using ICES J. Mar. Sci., 65, 279–295 doi:10.1093/icesjms/fsn028. CANOCO, Cambridge University Press, New York, p. 269. Roura, A., Álvarez-Salgado, X. A., González, Á. F.,Gregori, M., Rosón, Liao, Z.-H., Hsieh, H.-Y. and Lo, W.-T. (2013) Influence of monsoon- G. and Guerra, Á. (2013) Short-term meso-scale variability of meso- driven hydrographic features on thaliacean distribution in waters zooplankton communities in a coastal upwelling system (NW Spain). around Taiwan, western North Pacific Ocean. Zool. Stud., 52,49 Prog. Oceanogr., 109, 18–32 doi:10.1016/j.pocean.2012.09.003. doi:10.1186/1810-522X-52-49. Shanks, A. L. and Shearman, R. K. (2009) Paradigm lost? Cross- Longhurst, A. R. (1991) Role of the marine biosphere in the global shelf distributions of intertidal invertebrate larvae are unaffected by carbon cycle. Limnol. Oceanogr., 36, 1507–1526 doi:10.4319/lo.1991. upwelling or downwelling. Mar. Ecol. Prog. Ser., 385, 189–204 doi:10. 36.8.1507. 3354/meps08043. Longhurst, A. R. (1998) Ecological Geography of the Sea, Academic Press, Smith, A. M. and Lawton, E. I. (2010) Growing up in the temper- San Diego, 398 pp. ate zone: age, growth, calcification and carbonate mineralogy of Melicerita chathamensis (Bryozoa) in southern New Zealand. Palaeogeogr. Mackas, D. L. and Beaugrand, G. (2010) Comparisons of zooplankton Palaeoclimatol. Palaeoecol., 298, 271–277 doi:10.1016/j.palaeo.2010.09. time series. J. Mar. Syst., 79, 286–304 doi:10.1016/j.jmarsys.2008.11. 033. 030. Smith, V. H., Tilman, G. D. and Nekola, J. C. (1999) Eutrophication: Mossop, J. (2007) Protecting marine biodiversity on the continental shelf impacts of excess nutrient inputs on freshwater, marine, and terres- Ocean Dev. Int. Law beyond 200 nautical miles. , 38, 283–304 doi:10. trial ecosystems. Environ. Pollut., 100, 179–196 doi:10.1016/S0269- 1080/00908320701530474. 7491(99)00091-3. Nephin, J., Juniper, S. K. and Archambault, P. (2014) Diversity, abun- Southward, A. J., Langmead, O., Hardman-Mountford, N. J., Aiken, J., dance and community structure of benthic macro- and megafauna Boalch, G. T., Dando, P. R., Genner, J., Joint, I. et al. (2005) A review on the Beaufort shelf and slope. PLoS One, 9, e101556 doi:10.1371/ of long-term research in the western English Channel. In Southward, journal.pone.0101556. A. J., Tyler, P. A., Young, C. M. and Fuiman, L. A. (eds.), Advances in O’Brien, T. D., Wiebe, P. H. and Falkenhaug, T. (2013) ICES zooplank- Marine BiologyVol. 47, Elsevier Academic Press, San Diego CA, USA, ton status report 2010/2011. ICES Cooperative Research Report, 318, pp. 3–84 328 pp. 208 pp. Spalding, M. D., Fox, H. E., Allen, G. R., Davidson, N., Ferdaña, Z. Onbé, T. (1999) Ctenopoda and Onychopoda (=Cladocera). In A., Finlayson, M., Halpern, B. S., Jorge, M. A. et al. (2007) Marine Boltovskoy, D. (ed.), South Atlantic Zooplankton, Backhuys Publishers, ecoregions of the world: a bioregionalization of coastal and shelf Leiden, The Netherlands, pp. 797–813. areas. BioScience, 57, 573–583. Paffenhofer, G.-A., Wester, B. T. and Nicholas, W. D. (1984) Zooplank- Stübner, E. I., Søreide, J. E., Reigstad, M., Marquardt, M. and ton abundance in relation to state and type of intrusions onto the Blachowiak-Samolyk, M. (2016) Year-round meroplankton dynamics southeastern United States shelf during summer. J. Mar. Res., 42, in high-Arctic Svalbard. J. Plankton Res., 38, 522–536 doi:10.1093/ 995–1017 doi:10.1357/002224084788520819. plankt/fbv124. ˇ Pedersen, O.-P., Tande, K. S., Timonin, A. and Semenova, T. (2000) TerBraak,C.J.F.andSmilauer, P. (2002) CANOCO Reference manual A possible connection between hydrography and the distribution of and CanoDraw for Windows User’s guide: Software for Canonical Community Calanus finmarchicus on the Norwegian midshelf in 1997. ICES J. Mar. Ordination (version 4.5), Microcomputer Power 500 pp, Ithaca, NY, Sci., 57, 1645–1655 doi:10.1006/jmsc.2000.0980. USA. Pepin, P., Johnson, C. L., Harvey, M., Casault, B., Chassé, J., Tremblay, M. J. and Roff, J. C. (1983) Community gradients inthe Colbourne, E. B., Galbraith, P.S., Hebert, D. et al. (2015) A multivari- Scotian shelf zooplankton. Can. J. Fish. Aquat. Sci., 40, 598–611 doi:10. ate evaluation of environmental effects on zooplankton community 1139/f83-079. structure in the western North Atlantic. Prog. Oceanogr., 134, 197–220 Uriarte, I. and Villate, F. (2004) Effects of pollution on zooplankton doi:10.1016/j.pocean.2015.01.017. abundance and distribution in two estuaries of the Basque coast

307 JOURNAL OF PLANKTON RESEARCH VOLUME 41 NUMBER 3 PAGES 293–308 2019

(Bay of Biscay). Mar. Pollut. Bull., 49, 220–228 doi:10.1016/j. Viñas, M. D., Ramirez, F. C., Santos, B. A. and Marrari, M. (2007) marpolbul.2004.02.010. Spatial and temporal distribution patterns of Cladocera in the Villate, F. (1994) Temporal variability of the spatial distribution of argentine sea. Hydrobiologia, 594, 59–68 doi:10.1007/s10750-007- the zooplankton community in a coastal embayment of the Basque 9077-2. Country in relation to physical phenomena. Hydrobiologia, 288, 79–95 Vives, F. (1980) Los copépodos de las aguas neríticas de las doi:10.1007/BF00007128. costas de Vizcaya durante 1976. Invest. Pesq., 44, 313–330 Villate, F.,Uriarte, I., Irigoien, X., Beaugrand, G. and Cotano, U. (2004) (In Spanish). Chapter 16: zooplankton communities. In Borja, A. and Collins, M. Wootton, M. and Castellani, C. (2017) Crustacea: Copepoda. In (eds), Oceanography and Marine Environment of the Basque Country.Elsevier Castellani, C. and Edwards, M. (eds.), Marine Plankton. A Practical Oceanography Series, Oxford, pp. 395–424. Guide to Ecology, Methodology, and Taxonomy, Oxford University Press,

Villate, F., Uriarte, I., Olivar, M. P., Maynou, F., Emelianov, M. and Oxford, pp. 267–380. Downloaded from https://academic.oup.com/plankt/article-abstract/41/3/293/5522253 by guest on 18 October 2019 Ameztoy, I. (2014) Mesoscale structure of microplankton and meso- Xiong, L. L., Yin, J. Q., Huang, L. M., Li, K. Z. and Lian, S. M. (2012) plankton assemblages under contrasting oceanographic conditions in Seasonal and spatial variations of cladocerans on the northwest the Catalan Sea (NW Mediterranean). J. Mar. Syst., 139,9–26doi:10. continental shelf of the South China Sea. Crustaceana, 85, 473–496 1016/j.jmarsys.2014.05.004. doi:10.2307/23212676. Villate, F., Iriarte, A., Uriarte, I. and Sanchez, I. (2017) Seasonal Xu, J. Y., Xu, Z. L. and Gao, Q. (2016) Distribution of dominant and interannual variability of mesozooplankton in two contrasting zooplankton species along a latitudinal gradient in China Sea during estuaries of the Bay of Biscay: relationship to environmental factors. spring. J. Ocean U. China, 15, 502–508 doi:10.1007/s11802-016- J. Sea Res., 130, 189–203 doi:10.1016/j.seares.2017.05.002. 2871-6.

308