Ecosystem relevance of variable jellyfish biomass in the Irish Sea between years, regions and water types

Estuarine, Coastal and Shelf Science 149 (2014) 302e312

Contents lists available at ScienceDirect

Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss

Ecosystem relevance of variable jellyfish biomass in the Irish Sea between years, regions and water types Thomas Bastian a, b, *, Martin K.S. Lilley c, 1, Steven E. Beggs d, Graeme C. Hays c, 2, Thomas K. Doyle a, 3 a Coastal and Marine Research Centre, Environmental Research Institute, Glucksman Marine Facility, University College Cork, Irish Naval Base, Haulbowline, Cobh, Co. Cork, Ireland b School of Biological, Earth and Environmental Sciences, University College Cork, Distillery Fields, North Mall, Cork, Ireland c Department of Biosciences, Swansea University, Singleton Park, Swansea SA2 8PP, Wales, United Kingdom d Fisheries and Aquatic Ecosystems Branch, Agri-Food and Biosciences Institute, Newforge Lane, Belfast BT9 5PX, Northern Ireland, United Kingdom

a r t i c l e i n f o

a b s t r a c t

Article history: Received 11 February 2014 Accepted 30 August 2014 Available online 16 September 2014

Monitoring the abundance and distribution of taxa is essential to assess their contribution to ecosystem processes. For marine taxa that are difficult to study or have long been perceived of little ecological importance, quantitative information is often lacking. This is the case for jellyfish (medusae and other gelatinous plankton). In the present work, 4 years of scyphomedusae by-catch data from the 2007e2010 Irish Sea juvenile gadoid fish survey were analysed with three main objectives: (1) to provide quantitative and spatially-explicit species-specific biomass data, for a region known to have an increasing trend in jellyfish abundance; (2) to investigate whether year-to-year changes in catch-biomass are due to changes in the numbers or in the size of medusa (assessed as the mean mass per individual), and (3) to determine whether inter-annual variation patterns are consistent between species and water masses. Scyphomedusae were present in 97% of samples (N ¼ 306). Their overall annual median catch-biomass ranged from 0.19 to 0.92 g m3 (or 8.6 to 42.4 g m2). Aurelia aurita and Cyanea spp. (Cyanea lamarckii and Cyanea capillata) made up 77.7% and 21.5% of the total catch-biomass respectively, but species contributions varied greatly between sub-regions and years. No consistent pattern was detected between the distribution and inter-annual variations of the two genera, and contrasting inter-annual patterns emerged when considering abundance either as biomass or as density. Significantly, A. aurita medusae were heavier in stratified than in mixed waters, which we hypothesize may be linked to differences in timing and yield of primary and secondary productions between water masses. These results show the vulnerability of time-series from bycatch datasets to phenological changes and highlight the importance of taking species- and population-specific distribution patterns into account when integrating jellyfish into ecosystem models. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Cnidaria scyphozoa spatial variations temporal variations quantitative distribution stratification western European shelf Irish Sea

1. Introduction


du Littoral Co ^te d'Opale, LOG UMR 8187, * Corresponding author. Universite Maison de la Recherche en Environnement Naturel, 32 avenue Foch, F-62930 Wimereux, France. E-mail addresses: [email protected] (T. Bastian), [email protected] (M.K.S. Lilley), [email protected] (S.E. Beggs), [email protected] (G.C. Hays), [email protected] (T.K. Doyle). 1 Queen Mary University of London, School of Biological and Chemical Sciences, Mile End Road, London E1 4NS, United Kingdom. 2 Centre for Integrative Ecology, School of Life and Environmental Sciences, Deakin University, Warrnambool, Victoria 3280, Australia. 3 Zoology, Ryan Institute & School of Natural Sciences, National University of Ireland, Galway. http://dx.doi.org/10.1016/j.ecss.2014.08.018 0272-7714/© 2014 Elsevier Ltd. All rights reserved.

Scyphomedusae (Cnidaria, Schypozoa; hereafter referred to as “jellyfish”) are receiving increased recognition as key components of marine ecosystems (Hay, 2006; Doyle et al., 2014). For years now, their role as predators (mostly of crustacean and gelatinous zooplankton, but also ichthyoplankton) has been under scrutiny (Purcell, 1997; Purcell and Arai, 2001), and it is now thought that their competition for food with planktivorous fish, and their predation on fish eggs and larvae, can be driving forces in the emergence of alternative states in altered ecosystems (Richardson et al., 2009; Utne-Palm et al., 2010). Meanwhile, evidence has been accumulating that the nature and the outcome of jellyfishefish interactions (i.e. whether positive or negative to fish populations)

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

depend on which species and which ontogenetic stages are involved (Brodeur, 1998; Lynam and Brierley, 2006; Masuda, 2009). In addition to their potential impacts on fish populations, several studies have highlighted how scyphomedusae blooms can affect the lowest levels of pelagic food webs (Titelman et al., 2006; Tinta et al., 2010) down to nutrient and carbon cycling (Pitt et al., 2005; Condon et al., 2011). They also contribute to bentho-pelagic coupling processes (Billett et al., 2006; West et al., 2008; Yamamoto et al., 2008). However, quantifying the actual contribution of jellyfish to these various processes remains challenging because broad-scale quantitative abundance data are scarce. Such datasets are indeed a prerequisite for (1) establishing the spatial and temporal overlaps of medusae with other components of the ecosystem, and of (2) reliably including jellyfish in numerical models which would allow reasonable extrapolations of findings from laboratory or mesocosms experiments to ecosystem level (Pauly et al., 2009; Purcell, 2009; Ruzicka et al., 2012). In the Irish Sea, the role of scyphomedusae is of primary interest as there has been an increasing trend in the mean overall jellyfish biomass in the western part of the region since at least 1994 (Lynam et al., 2011). The Irish Sea is a semi-enclosed sea between Ireland and Great Britain (Fig. 1). Based on its bathymetry, it can be subdivided into two regions. The region east of the Isle of Man is relatively shallow with depths less than 50 m, and is influenced by major estuarine inputs with the existence of a salinity front in Liverpool Bay (Dickson and Boelens, 1988). Conversely, the region west of the Isle of Man is characterised by a channel 100e150 m deep (running along a northesouth axis), and which becomes

303

seasonally stratified during spring and summer (Simpson and Hunter, 1974), leading to the formation of a cyclonic near-surface gyre (western Irish Sea gyre) (Hill et al., 1996). The eastern and western regions therefore present contrasting environments, within which ecological processes (e.g. primary production, fish spawning) present contrasting dynamics. In particular, seasonal production differs between mixed and stratified regions (Gowen et al., 1995), spawning of several fish and crustaceans species (including commercially important ones) concentrates in specific spawning grounds (Fox et al., 2000; Armstrong et al., 2001; Heffernan et al., 2004), and fish larvae are not randomly distributed (Dickey-Collas et al., 1996; Bunn et al., 2004). As regards to scyphomedusae, six different species can be found in the Irish Sea (Aurelia aurita, Cyanea capillata, Cyanea lamarckii, Rhizostoma octopus, Chrysaora hysoscella, and at times Pelagia noctiluca (Russell, 1970)), and previous analysis of stranding events around Ireland and Wales suggested that different species occur in different ecological regions within the Irish Sea (Houghton et al., 2007; Doyle et al., 2007). Developing knowledge of the spatial distribution of these species in Irish Sea waters is therefore necessary to establish and quantify the potential for competition with, and predation on, other species, and therefore better assess how the increasing trend in the overall jellyfish abundance (Lynam et al., 2011) may affect various ecological processes as shown by much jellyfish research conducted during the past decade. Aerial surveys have been used to describe discrete Rhizostoma octopus hotspots in coastal bays (Houghton et al., 2006a), while beach surveys and citizen science schemes have provided useful information on the seasonal occurrence of several species along the coastline (Houghton et al., 2007; Fleming et al., 2013; Pikesley et al., 2014). As regards to more offshore areas, surveys from ships of opportunity (Doyle et al., 2007) have confirmed the presence of Aurelia aurita and Cyanea capillata beyond immediate coastal waters and suggested that surface distribution patterns could be linked to variations of temperature and salinity (Bastian et al., 2011). However, quantitative information on the abundance and the distribution of these organisms in the main body of the Irish Sea are still missing. The present work explores four years of quantitative spatiallyexplicit bycatch data of Aurelia aurita and Cyanea spp. in the Irish Sea with three specific objectives: (1) to provide the first speciesspecific quantitative estimates of jellyfish biomasses in the Irish Sea and a description of their distributions; (2) to examine whether inter-annual variations of biomass are consistent across the Irish Sea; and (3) to check whether these are linked to variations in number of individuals or of the size of the individuals. Beyond the specific case of the Irish Sea, these last two points should provide further elements on the importance of considering regional and local scales when studying jellyfish temporal dynamics (Dawson et al., 2014), as well as on the vulnerability of bycatch datasets to phenological changes. 2. Methods 2.1. Data collection

Fig. 1. Study site and sampling stations in the Irish Sea.

Between 2007 and 2010, scyphomedusae were caught as bycatch during the Agri-Food and Biosciences Institute (AFBI) of Northern Ireland annual Methot Isaacs Kidd-net Survey (NI-MIK), targeting juvenile gadoid fish, on board the RV Corystes. The survey follows a fixed station stratified design, taking place at the end of May and early June (during the period prior to settlement of pelagic juvenile gadoids) across the northern part of the Irish Sea (north of 53.25 N) (Fig. 1). Annually the western Irish Sea (i.e. west of 4.75 W) is sampled twice and the eastern Irish Sea once, but dates

304

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

and the number of stations varied slightly as a result of logistical or mechanical constraints and weather conditions (Table 1). Sampling was by a 5 m2 Methot trawl equipped with a modified Isaac-Kidd depressor (MIK, 5 mm mesh size, see Methot, 1986), towed at 3 knots during the hours of darkness in a doubleeoblique profile through the water column to within 4 m of the seabed. A second profile was taken at shallow stations to ensure a tow duration of ~15 min. A mechanical flowmeter (General Oceanics Model 2030R) was attached to the centre of the MIK net frame, recording volume filtered during each tow. The catch was sorted immediately with scyphozoan jellyfish identified to species or genus level, and the total wet mass and the number of individuals per species recorded. Although large Cyanea capillata and Cyanea lamarckii can be easily distinguished by their colours (C. capillata is red-brown, C. lamarckii is blue), small medusae of both species can have a transparent yellowish coloration. Species identification has therefore to be based on the number of tentacles and the presence or absence on the muscle folds of pit-like intrusions from the gastrovascular cavity (Russell, 1970; Holst and Laakmann, 2014). Unfortunately, time and logistical constraints during the cruise did not allow for reliable observations of these characteristics, and results are therefore reported at the genus level. Large catches (>15 kg) were subsampled and genus-specific biomass and numbers of individuals present in the catch were raised to total catch. 2.2. Jellyfish abundance For each station, the catch biomass (wet mass) and the number of individuals were standardised by the volume filtered (m3). These volume standardised biomass (g m3) or number of individuals (ind. m3) were used for all analyses. For comparisons with other studies, an overall median biomass density was also calculated for each year: the biomass density (g m2) was calculated for each sampling station by multiplying the volume standardised biomass (g m3) by the depth at which the net was cast (m); the median for all stations of the second leg of a given year was then taken as the average annual biomass density. Medians were preferred to means as the data were not normally distributed (see Section 2.4). Throughout the surveys, where time allowed, individuals of each species were weighed and the bell diameter measured to construct bell diameter e wet mass relationships. Cyanea spp. were measured across their widest point, while Aurelia aurita bell diameters were estimated based on a centre to widest point measurement. A. aurita were only measured where net damage to their fragile manubrium was minimal, with 17 individuals outside of statistical 95% prediction intervals removed to minimise erroneous measurements. Blue-coloured Cyanea sp. were used to derive a specific relationship for Cyanea lamarckii. Twenty-one Chrysaora hysoscella were recorded sporadically. All species data were fitted with two parameter power law relationships typical of growth rates, using the Least Squares method.

Table 1 Sampling effort in the Irish Sea between 2007 and 2010. Number of valid jellyfish sampling stations (n) and dates (in dd/mm) between which sampling took place in the eastern and western Irish Sea each year. Note that the western Irish Sea was sampled twice a year (Leg 1 and Leg 2). Year

Leg 1

Leg 2

West 1

2007 2008 2009 2010

East

West 2

Dates

n

Dates

n

Dates

n

29/05e01/06 27/05e01/06 26/05e30/05 11/05e15/05

24 27 25 30

03/06e08/06 01/06e06/06 02/06e06/06 09/06e11/06

27 26 26 14

10/06e14/06 06/06e10/06 06/06e11/06 05/06e09/06

25 23 30 29

In the absence of exhaustive measurement data for each catch, an estimated average mass (wet mass) per individual (g ind.1) was calculated for each genus at each station, by dividing the catch biomass by the total number of individuals in the catch. Resampling of 19 to 29 western Irish Sea stations between the two legs of the survey allowed the calculation of genus-specific population growth rates. These instantaneous growth rates were calculated assuming constant non-linear growth (Busacker et al., 1990; Lilley et al., 2014):

m ¼ ln ðbt2 =bt1 Þ=ðt2  t1 Þ where m is the instantaneous growth rate (d1), bt2 the mean biomass in the western Irish Sea during leg2, bt1 the mean biomass in the western Irish Sea during leg1, and (t2  t1) the mean number of days separating successive samples of the western Irish Sea stations between leg 1 and 2 (range 7e27 days).

2.3. Attributing catches to water types The Irish Sea Pilot project (Golding et al., 2004) identified 4 eco-regions or ‘water column marine landscapes' for the Irish Sea, defined by salinity (high/low) and stratification (mixed/stratified). Environmental data collected during the 2008e2010 surveys were used to identify those water-masses corresponding to each catch and test for significant differences in individual wet mass of Aurelia aurita and Cyanea spp. medusae between these 4 water types (SHS for stratified and high salinity; MHS for mixed and high salinity; SLS for stratified and low salinity, and MLS for mixed and low salinity). Temperature (±0.001  C) and salinity (±0.01) profiles were collected by a CTD profiler (SeaBird e SBE19plus) mounted on a GULF-VII high speed plankton sampler (Nash et al., 1998) deployed in daytime. Deployment was the same as with the MIK-net, with a second profile in water depth <30 m. CTD data were averaged by 0.5 m-depth bins. For each station and parameter, a composite profile was calculated using the mean of the downward and upward profiles for each 0.5 m-depth bin. A running median smoother was applied to the resulting profile (window width ¼ 5 data points) and a visual inspection was used to validate the quality of each profile. Simpson's stratification index (f) was calculated for each station (Simpson, 1981). Following Dickey-Collas et al. (1996), values of f  10 and f  20 were used to discriminate between mixed and stratified waters (intermediate waters were disregarded) and near surface salinity (2e3 m depth) of 34.0 was used as the limit between high and low salinity waters (Golding et al., 2004). The sampling stations for deployment of MIK-nets (scyphomedusae samples) and CTD casts were not identical (Fig. 1). 48 stations were common to both sampling gears and matching jellyfish catches with CTD casts was straightforward (n ¼ 48 records). 69 jellyfish sampling stations were localised at mid-distance of 2 CTD sampling stations; in this case the jellyfish catch was associated with the mean value from the two corresponding CTD profiles. In other cases (n ¼ 95), the MIK net station was much closer to a single CTD station than another one; in these cases the value from the closest CTD profile was directly match to the jellyfish record (distances: 90th percentile ¼ 6.5 km, max distance ¼ 21.7 km). Finally, 25 jellyfish catches could not be satisfactorily associated with a CTD profile (e.g. sampling stations outside sampling grid of CTD, or failure of the CTD) and were excluded from the environmental analysis. Elapsed time between the jellyfish sampling and corresponding CTD cast ranged from <1 h to 4.5 days (median ¼ 12.5 h). Only data less than 48 h apart were considered for analysis.

Table 2 Catches of Aurelia aurita and Cyanea spp. in the Irish Sea between 2007 and 2010. Each year, the western region was sampled twice (see Table 1 for details). “Frequency pres.” is the frequency of occurrence of each species in the sampling events of each region; “% total catch” is the contribution of each species to the overall wet mass of scyphomedusae.

All scyphomedusae n stations Frequency pres. Total catch (kg) Aurelia aurita Frequency pres. % total catch g m3 Mean Median SD Maximum 3 ind. 100 m Mean Median SD Maximum Cyanea spp. Frequency pres. % total catch g m3 Mean Median SD Maximum ind. 100 m3 Mean Median SD Maximum

West 2

East

West 2 and East

2007

2008

2009

2010

Overall

2007

2008

2009

2010

Overall

2007

2008

2009

2010

Overall

2007

2008

2009

2010

Overall

24 0.96 103

27 0.93 21

25 0.92 87

30 0.87 94

106 0.92 305

25 1.00 81

23 0.96 197

30 1.00 305

29 0.83 187

107 0.97 770

27 1.00 97

26 0.92 115

26 1.00 331

14 1.00 56

93 0.98 599

52 1.00 178

49 0.94 312

56 1.00 636

43 0.95 243

200 0.98 1369

0.70 0.97 1.21 0.005 2.29 7.24 1.86 0.04 3.47 12.01

0.80 0.48 0.31 0.16 0.45 1.65 0.10 0.06 0.12 0.46

0.83 0.95 1.58 0.24 2.82 10.32 0.70 0.12 1.38 5.47

0.87 0.83 2.36 0.19 7.26 38.25 1.51 0.14 5.07 27.48

0.76 0.86 1.42 0.38 2.84 14.52 0.68 0.17 1.38 6.71

0.81 0.83 1.45 0.21 4.29 38.25 0.78 0.12 2.86 27.48

0.85 0.60 0.43 0.06 1.19 6.09 0.50 0.06 1.41 7.11

0.54 0.32 0.23 0.0023 0.80 3.94 0.35 0.02 1.05 4.07

0.96 0.97 3.49 0.29 5.87 25.63 3.72 0.40 5.13 17.76

0.64 0.09 0.10 0.010 0.28 1.08 0.10 0.04 0.17 0.63

0.76 0.71 1.18 0.049 3.47 25.63 1.27 0.06 3.18 17.76

0.83 0.54 0.37 0.10 0.91 6.09 0.31 0.06 1.03 7.11

0.67 0.72 0.86 0.023 2.11 10.32 0.52 0.02 1.21 5.47

0.91 0.91 2.89 0.20 6.60 38.25 2.52 0.24 5.18 27.48

0.72 0.69 0.99 0.139 2.41 14.52 0.49 0.09 1.16 6.71

0.79 0.78 1.32 0.13 3.92 38.25 1.01 0.09 3.02 27.48

1.00 0.49 0.34 0.23 0.42 1.85 1.11 0.25 2.42 11.88

0.91 0.05 0.08 0.05 0.11 0.53 0.18 0.14 0.14 0.51

0.93 0.17 0.37 0.09 0.73 2.94 0.87 0.16 1.79 8.20

0.90 0.13 0.22 0.08 0.33 1.42 0.16 0.12 0.13 0.56

0.93 0.16 0.26 0.08 0.48 2.94 0.58 0.16 1.55 11.88

1.00 0.41 0.40 0.24 0.43 1.62 1.79 1.12 2.23 9.44

0.85 0.63 0.48 0.03 1.49 7.39 3.56 0.4746 10.70 52.87

0.88 0.02 0.09 0.06 0.10 0.43 0.59 0.40 0.63 2.70

1.00 0.91 1.06 0.96 1.07 4.44 0.83 0.502 0.85 2.63

0.92 0.28 0.43 0.14 0.96 7.39 1.81 0.564 5.84 52.87

1.00 0.44 0.37 0.24 0.42 1.85 1.46 0.60 2.33 11.88

0.88 0.26 0.29 0.048 1.10 7.39 1.97 0.163 7.91 52.87

0.91 0.09 0.24 0.06 0.55 2.94 0.74 0.25 1.38 8.20

0.93 0.31 0.49 0.205 0.77 4.44 0.38 0.171 0.58 2.63

0.93 0.22 0.34 0.09 0.74 7.39 1.15 0.24 4.18 52.87

0.83 0.73 0.71 0.15 2.18 10.57 0.30 0.10 0.81 3.92

0.52 0.68 0.11 0.007 0.18 0.49 0.08 0.01 0.13 0.42

0.60 0.91 0.84 0.16 1.93 8.61 0.37 0.08 0.97 4.75

0.96 0.19 0.18 0.09 0.23 0.99 0.43 0.25 0.60 2.51

0.89 0.30 0.05 0.02 0.08 0.39 0.23 0.107 0.36 1.36

0.84 0.09 0.08 0.05 0.10 0.36 0.22 0.13 0.23 0.81

0.67 0.03 0.02 0.01 0.03 0.12 0.13 0.0752 0.15 0.55

0.66 0.85 0.73 0.029 1.87 10.57 0.71 0.04 2.07 12.01 0.83 0.12 0.08 0.03 0.14 0.99 0.24 0.1093 0.37 2.51

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

West 1

305

306

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

Fig. 2. Distribution of volume standardised biomass (g m3) of Aurelia aurita and Cyanea spp. in the Irish Sea in early June 2007e2010.

2.4. Statistical analysis All statistics were performed on both volume-standardised biomass data (g m3) and density data (ind.$100 m3). Since the data were not normally distributed (ShapiroeWilk normality test, p ¼ 0.05 threshold), non-parametric statistics were used. For each species, inter-annual abundance differences were tested using KruskaleWallis tests, followed by a Steel-Dwass post-hoc test when appropriate. Similar procedures were used to test for differences between sub-regions (i.e. eastern vs. western halves of the basin), years (within the sub-regions), and average size (g ind.1) between sub-regions and the 4 water types. Species distributions of Aurelia aurita and Cyanea spp. were
r-von Mises compared using the non-parametric modified Crame statistic proposed by Syrjala (Syrjala, 1996), which tests for differences in spatial-distributions between populations, and is

particularly adapted to species that exhibit aggregative behaviour (Syrjala, 1996; Brodeur et al., 2002). The null hypothesis of the test is that the normalized distributions of the two populations (in our case species) are the same. The Syrjala-test is sensitive to differences in distribution, but is insensitive to differences in abundance between the two populations. Statistical significance was determined by a randomization test (1000 iterations). The Syrjala-test was also used to test for significant inter-annual differences in the distribution of each species. In this case, for each paircomparison, the samples were restricted to stations sampled in both years. Since some studies have suggested significant interactions between Aurelia aurita and Cyanea capillata (Båmstedt et al., 1997; Hansson, 1997), the possible existence of an association between A. aurita and Cyanea spp. abundances was investigated through Spearman's rank correlations.

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

Statistical analyses and computing were conducted in R (R Development Core Team, 2011). The ‘oce’ package was used to calculate densities from CTD profiles (Kelley, 2011). The Syrjala test was computed using the ‘ecespa’ package (De la Cruz, 2008). Mapping of results was done in Q-GIS 2.0.1-Dufour (http://qgis. osgeo.org/en/site/). 3. Results 3.1. Overall catch summary Between 2007 and 2010, 306 valid tows were conducted with scyphozoan jellyfish present in 95.4% of tows (Table 2, Fig. 2). The mean volume filtered (X ± SD) was 4671 ± 1789 m3 (range 1412e11 720 m3). Depth in the sampling area was between 16 and 130 m (mean ¼ 61.5 ± 30.7 m). The summary of catches for the entire survey (leg 1 and 2) each year is presented in Table 2. Only data from stations sampled from the eastern region and the second sampling of the western region were analysed. Between 2007 and 2010 the annual median (±IQR, inter-quartile range) total jellyfish volume-standardised catch biomass were: 0.37 ± 0.86; 0.19 ± 0.63; 0.52 ± 2.57; 0.92 ± 0.98 g m3 (Table 2); leading to median biomass densities of: 18.6 ± 34.87; 8.16 ± 50.18; 23.5 ± 128.36; and 42.4 ± 75.79 g m2 (or t km2). Overall (4 years combined), Aurelia aurita represented 77.7% of the catches (wet mass) and Cyanea spp. (Cyanea lamarckii and Cyanea capillata) 21.5%, but the contribution of each species varied greatly between regions and between years (Table 2). The long term overall median biomass densities of A. aurita and Cyanea spp. were 5.7 g m2 (range of annual median 0.75e12.0) and 5.0 g m2 (range 2.6e13.2) respectively. The only other scyphozoan caught was the compass jellyfish (Chrysaora hysoscella), but it was only marginally present (16 individuals in 2007; 1 in 2009, and 1 in 2010), except in 2008 when 131 individuals were caught (but at only 2 stations). 3.2. Inter-annual variations of abundances and distributions of A. aurita and Cyanea spp. Catches of Aurelia aurita and Cyanea spp. were marked by significant inter-annual variability (A. aurita: KruskaleWallis H ¼ 16.994, df ¼ 3, p < 0.001; Cyanea spp.: H ¼ 25.838, df ¼ 3, p < 0.001). Abundances (catch biomass and densities) of Cyanea spp. were significantly higher in 2007 than in 2008 and 2009, whereas A. aurita was significantly more abundant in 2009 than in any other year (Steel-Dwass post-hoc, p < 0.05). In addition the distribution of each species across the study area was significantly different from one year to the next (Syrjala tests on density data, p < 0.05). There was no consistent direct relationship (positive or negative association) between the two species. The catch-biomass (g m3) of A. aurita was positively correlated to Cyanea spp. in 2007 and 2008 (Spearman's rank correlation, r2007 ¼ 0.230, r2008 ¼ 0.416, p < 0.05), however no significant correlations were found in 2009 or 2010. Catch densities (ind.$100 m3) were never significantly correlated (p > 0.05). The spatial distribution of the two species varied significantly in every year except in 2009 (both catch-biomass and densities) and 2007 (catch-biomass only) (Syrjala test p < 0.05).

(H ¼ 33.24, df ¼ 3, p < 0.001, with p < 0.05 for 2007e2008, 2008e2009, 2009e2010, and 2007e2009 Steel-Dwass post-hoc pairwise comparisons) but not in the west (H ¼ 3.139, df ¼ 3, p ¼ 0.37). The northern half of the eastern Irish Sea exhibited particularly marked inter-annual variations of abundance for this species, with large quantities caught there in 2009 compared to 2008 and 2010 (Fig. 2 and Figure A-1). In fact, the median density (ind.$100 m3) of A. aurita in the eastern Irish Sea in 2009 was >22 and 9.3 times higher than the median density of 2008 and 2010 respectively (see Table 2). For Cyanea spp. most inter-annual differences of biomass (g m3) were also significant in the eastern part of the Irish Sea (H ¼ 33.136, df ¼ 3, p < 0.001, all p-values of Steel Dwass post-hoc pairwise comparisons < 0.05, except for 2008e2009); but in the west, only catches from 2007 were significantly different (higher) from those of 2008 (H ¼ 8.663, df ¼ 3, p ¼ 0.034, Steel-Dwass p < 0.05 only for 2007e2008 comparison). It is of note that, despite significant biomass variations in the East, no significant density (ind.$100 m3) differences were found in that region (H ¼ 6.548, df ¼ 3, p ¼ 0.088). Therefore, the high 2010 biomass was not due to higher numbers, but to individuals being larger on average. Conversely, in the west, the higher biomass of 2007 in comparison with 2008 corresponded to higher densities (KruskaleWallis on Cyanea spp. densities: H ¼ 13.501, df ¼ 3, p ¼ 0.003, SteeleDwass post-hoc tests <0.05 for 2007e2008 and 2007e2010 pair-wise comparisons). Insufficient sample size prevented comparison of distribution patterns within each sub-region using the Syrjala statistics as was done for distributions across the entire Irish Sea. 3.4. Mass, size, survey timing and environmental parameters Time constraints during the surveys meant that it was not possible to weigh and measure every jellyfish caught individually, however the bell diameter to wet mass relationships for Irish Sea jellyfish and power law relationships were calculated from several fresh individuals caught during the surveys (Table 3 and Figure A2). All relationships accounted for 91e95% of the variability and ranged over at least a 20 cm bell diameter range. For Cyanea spp., a common relationship is presented, as well as species-specific ones (Table 3 and Figure A-2). For both Aurelia aurita and Cyanea spp. the average mass per individual (g ind.1) was greater to a highly significant level in the western sub-region compared with the east, suggesting that individuals in the west were larger on average than in the east (Wilcoxon rank-sum test, A. aurita: W ¼ 657, p < 0.001, medianwest ¼ 214.7 g ind.1, medianeast ¼ 68.4 g ind.1, nwest ¼ 86, neast ¼ 69; Cyanea spp.: W ¼ 2,728, p < 0.001, medianwest ¼ 41.0 g ind.1, medianeast ¼ 15.5 g ind.1, nwest ¼ 100, neast ¼ 87) (Fig. 3a and c). When considering water types, it appeared that A. aurita were bigger on average in stratified high salinity waters (medianSHS ¼ 280.0 g ind.1) than in other water types (medianMHS ¼ 163.16 g ind.1, medianSLS ¼ 135.9 g ind.1, Table 3 Wet mass (WW, g) to bell diameter (BD, cm) power relationships (WW ¼ a.BDb) for 4 common Scyphozoan species in the Irish Sea. n

3.3. Inter-regional variations of A. aurita and Cyanea spp. abundances Year-to-year variations in the abundances of Aurelia aurita and Cyanea spp. differed between the eastern and the western parts of the Irish Sea (Fig. 2 and Figure A-1). The abundance (g m3) of A. aurita significantly differed from one year to the next in the east

307

Aurelia aurita Cyanea capillata Cyanea lamarckii Cyanea spp. Chrysaora hysoscella

245 947 1161 2108 21

Size range (cm)

Coefficients

min.

max.

a

b

r2

5 0.1 1.5 0.1 2

36 40 21 40 24

0.271 0.066 0.074 0.055 0.035

2.341 2.836 2.764 2.888 3.057

0.940 0.921 0.949 0.927 0.913

308

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

medianMLS ¼ 75.5 g ind.1, H ¼ 24.74, df ¼ 3, p < 0.001, Steel-Dwass p < 0.05 only for SHS-MHS, SHS-SLS, and SHS-MLS pairwise comparisons) (Fig. 3b). A similar pattern was observed in Cyanea spp. with individuals being larger on average in stratified high salinity waters (medianSHS ¼ 35.0 g ind.1) than in other water masses (medianMHS ¼ 26.9 g ind.1, medianMLS ¼ 13.1 g ind.1, and medianSLS ¼ 33.1 g ind.1), but the differences were not statistically significant (H ¼ 6.35, df ¼ 3, p ¼ 0.09) (Fig. 3d). Assuming a constant non-linear increase in mass between the two legs of the survey in the western Irish Sea, Aurelia aurita instantaneous growth rates ranged between 0.006 and 0.068 day1 (in 2009 and 2010 respectively). For Cyanea sp. it ranged from 0.022 to 0.085 d1 (in 2007 and 2008). Notably the first western survey in 2010 was around two weeks earlier than normal, with the second legs at the regular time. The estimated average mass per individual of the two species reflects this earlier sampling time with slightly smaller initial Cyanea sp. individuals, and considerably lighter

A. aurita in the 2010 first western survey than in other years, although the overall catch biomass was typical (Table 2 and Figure A-3). 3.5. Environmental context As regards to the distribution of near-surface temperatures, salinity, and limits of mixed and stratified waters during sampling in early June 2008, 2009 and 2010, the eastern and western parts of the Irish Sea exhibited contrasting inter-annual variability (Fig. 4). For example, temperature was higher in the west in 2010 compared with 2009, but in the east, the differences of SST were not as marked. Conversely, there was a marked difference in salinity in the east between 2009 and 2010, but it was only moderate in the west (Fig. 4). These variations of temperature and salinity influenced the delimitation of mixed and stratified waters (Simpson Stratification Index > 20). The apparent increase in the stratified area during the 2010 campaign in the western part of the Irish Sea can be linked to the more pronounced gradient of near surface temperatures and a wider distribution of near surface salinities >34 in that region in 2010. 4. Discussion

Fig. 3. Variations in the estimated average mass per individual of Aurelia aurita and Cyanea spp. in the Irish Sea between regions (a,c) and water types (b,d). The definition of the different water types was based on Simpson's Stratification Index and the nearsurface salinity in early June each year; see material and methods for details and Fig. 4 for distribution of water masses. SHS: stratified high salinity, MHS: mixed high salinity, SLS: stratified low salinity, and MLS: mixed low salinity waters. Boxplots delimits 1st and 3rd quartiles, with thick horizontal lines indicating the median and black filled circle the arithmetic mean. Whiskers delimit the values within 1.5*IQR, open circles shows outliers. Horizontal lines on top of boxes shows significancy of Wilcoxon Rank Sum tests (a. and c.) or Steel Dwass post-hoc tests (b. and d.) at levels of p < 0.05 (*), 0.01 (**), 0.001 (***).

Ecosystem models can provide important guidance for marine management, particularly on the interactions between species. However, the gelatinous zooplankton are either mostly ignored or poorly represented by these models. Pauly et al. (2009) highlighted the need for jellyfish biomass to be recorded during fisheries stock assessments and routine time-series, so that critical parameters such as annual biomass can be estimated. Hence we analysed four years of species-specific jellyfish density data for the Irish Sea and showed that the abundances and distributions of Aurelia aurita and Cyanea spp. varied between years and sub-regions (Table 2, Fig. 2 and Figure A-1), suggesting the existence of species-specific and regional-specific inter-annual dynamics, as documented elsewhere (Suchman and Brodeur, 2005; Lynam et al., 2005; Brodeur et al., 2008; Eriksen et al., 2012). Specifically, our results show that the annual median scyphozoan biomass density varied ~5-fold, and that Aurelia aurita and Cyanea spp. were present both inshore and offshore annually from early-May onwards. With values ranging 8.6e42.4 g m2 (wet mass), the overall jellyfish biomass density in the Irish Sea seem to be greater than in several other regions where fisheries bycatch has been recorded e.g. the Sea of Okhotsk (0.166e1.27 g m2), and the Barents Sea (0.78 g m2) (Zavolokin, 2010; Eriksen et al., 2012); however, the relatively coarse fisheries nets used in the latter studies may underestimate the catch of smaller or more fragile gelatinous species compared to the MIK net used in the present study (Lilley et al., 2011). In the South East Bering Sea where a MIK net was used an abundance of 114.9 g m2 was reported in 1999 (Brodeur et al., 2002). In early summer (MayeJune), A. aurita biomass is approaching, but has not reached, its peak (Bastian et al., 2011), while most Cyanea are still in their juvenile stages and hard to distinguish between the species (Holst and Laakmann, 2014). Therefore, we may have underestimated the maximum biomass in the Irish Sea, especially as these figures do not take into account the large, but localised, aggregations of Rhizostoma octopus in distinct coastal hotspots not sampled by the NI-MIK survey (Houghton et al., 2006b; Lilley et al., 2009). This also implies that the relative abundance contribution of each species to total jellyfish catch may evolve during the season. Our results provide, nonetheless, the first available quantitative data on scyphomedusae biomass for the Irish Sea. A basic extrapolation of the above annual median biomass densities to the surface area of the Northern Irish Sea (53e55 N,

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

309

Fig. 4. Near surface temperature ( C) and salinity, and stratification index in the Irish Sea in early June 2008e2010. For stratification index, only the 10 and 20 isolines used to delimit stratified and mixed waters are drawn. All isolines were generated after triangular interpolation (grey crosses mark the positions of CTD sampling stations).

calculated to be 33.96$109 m2 after projection of the British National Grid) lead to total average biomass estimates ranging from 292$106 to 1440$106 kg wet mass (extrapolation of 1st quartiles instead of median: 52.7$106 and 225.2$106 kg). By comparison, in 2007 the total Irish Sea biomass of cod (Gadus morhua) and herring (Clupea harengus) were estimated at 3.6$106 kg and 39.7$106 kg respectively (ICES, 2009, 2013). However, and importantly, such calculations do not take into account the high spatial heterogeneity in jellyfish biomass distribution (Table 2, Fig. 2, Figure A-1) and should therefore be regarded with caution. Also, our sampling method does not take into account a possible heterogeneous vertical distribution of medusae (e.g. Barz and Hirche, 2007), which may further bias biomass estimates. Regarding morphometric relationships (Table 3), variation between studies was observed, particularly for larger medusae (Figure A-2). For example, it seems that Aurelia aurita medusae above 20 cm in bell diameter could be expected to achieve a greater mass in both Norwegian and Baltic waters than in the Irish Sea (Båmstedt et al., 1994). In a same way, Norwegian Cyanea capillata measured by Båmstedt et al. (1994) were heavier than Irish Sea

individuals, whereas the use of the Hay et al. (1990) equation from North Sea Cyanea spp. would appear to underestimate the mass of Irish Sea Cyanea spp. For Chrysaora hysoscella, a comparison of our equation with a previously published equation from Houghton et al. (2007) indicated that stranded individuals appear to be heavier for a comparable bell diameter. This may be due to a retractation of the bell after the individual has stranded. Our Irish Sea equation was, however, similar to the relationship measured for this species in summertime in the Benguela upwelling (Buecher et al., 2001). As regards to growth rates, for A. aurita (0.01e0.07 d1) they were close to those of wild populations in Kerteminde fjord (max 40 mm BD, 0.02e0.09 d1) (Olesen et al., 1994) and comparable with the larger animals observed in Kiel €ller, Bight (max ¼ 197 mm BD, mean 0.02, max ¼ 0.18 d1) (Mo 1980) and the Sea of Japan (0.05e0.08 d1) (Uye and Shimauchi, 2005). Growth of the Cyanea species has been less well studied, but Brewer (1989) calculated rates of 0.02e0.16 d1. This is also in agreement with our Irish Sea population (0.02e0.08 d1). While an important focus of this study was to investigate the distribution of scyphomedusae biomass, note must be taken of

310

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

the possible impacts that such variable biomass may have on commercially important species, and in particular in the eastern part of the Irish Sea where major spawning and nursery grounds of many fish species occur (Fox et al., 1997, 2009; Bunn et al., 2004; Heffernan et al., 2004). Indeed, several diet and feeding studies have demonstrated that fish eggs and larvae can contribute to the diet of Cyanea capillata and Aurelia aurita (Titelman and Hansson, 2006; Barz and Hirche, 2007). The most € ller (1984) who extreme example probably came from Mo observed up to 49 larvae in a single 68 mm A. aurita. In the Irish Sea, fish spawning peaks in late winter e early spring (late February to early April), with larvae peaking one month later (Bunn et al., 2004). Therefore, temporal overlaps between jellyfish and fish larvae and juveniles would appear to be stronger than between jellyfish and fish eggs. However, the synchronising of juvenile fish and jellyfish may not be detrimental, as some juvenile gadoid species are known to associate with scyphomedusae and with Cyanea capillata in particular (Nagabhushanam, 1959; Russell, 1970; Purcell and Arai, 2001). Therefore, exactly which species in the Irish Sea ultimately benefit or lose out as a result of the presence of scyphomedusae remains unknown for now, and the response will most likely depend on the life-stages (Purcell and Arai, 2001; Masuda, 2009) and regions (Eriksen et al., 2012) considered. Importantly, another exploited species that may be affected by high scyphomedusae abundance is the Norway lobster (Nephrops norvegicus). On several occasions during the survey, hundreds of Nephrops zoea and juveniles were observed stuck within the tentacles of Cyanea spp. (pers. obs.). Although these may have got entangled in the net, it is likely Nephrops larvae are part of the diet of Cyanea spp. As Nephrops sustain the most valuable fishery in the Irish Sea (Tingley, 2006), whether or not this predation pressure by jellyfish significantly affects the recruitment of Nephrops deserves further investigation. The large inter-annual and regional variations in jellyfish biomass such as observed in the present study may also have local knock-on effects for jellyfish predators and scavengers (e.g. variation in prey densities, less jellyfish falls) (Doyle et al., 2014). Investigating the precise ecological mechanisms underlying the observed inter-annual and inter-regional variations of jellyfish abundance was beyond the scope of the present work. Nevertheless, Lynam et al. (2011) demonstrated that the overall annual mean catch of scyphomedusae in the western half of the Irish Sea is influenced by (1) the quantity of food available to the previous generation of medusae and (2) the environmental conditions that can influence polyp growth and the production of ephyrae. In particular, sea surface temperature from the previous 18 months was positively correlated with the average catch of scyphomedusae, while spring precipitation (a proxy for freshwater runoffs and therefore inshore salinity) was negatively correlated to it. The eastern region of the Irish Sea is subject to greater variations in temperature and salinity than the western part of the Irish Sea because it is much shallower and under a noticeably stronger influence of river runoffs (Dickson and Boelens, 1988). This was confirmed to a certain extent by our data (Fig. 4), and may therefore explain the contrasting interannual variation patterns observed in scyphomedusae abundance between the eastern and western part of the region. As more data become available for the Eastern Irish Sea, it will be of value to formally test whether local variations in the 3 parameters identified from analysing the Western Irish Sea time series (Lynam et al., 2011), can also explain local variations in jellyfish biomass observed in the East. Integrating species-specific considerations may also further improve this existing model as we observed that the year-to-year variations of abundance and distribution of Aurelia aurita were not the same as for Cyanea spp.

This is likely to be linked to A. aurita, Cyanea lamarckii and Cyanea capillata having distinct ecological differences (Russell, 1970; Doyle et al., 2007): for example, the growth and strobilation of their polyps are not affected the same way by variations of temperature and salinity (laboratory studies suggested that cold winter temperatures inhibit the strobilation of A. aurita polyps whereas it promotes strobilation in C. capillata) (Holst and Jarms, 2010; Holst, 2012). The other main finding of our analysis was that, at the time of the survey, individuals of both species were on average twice as heavy (as indicated by the average mass per individual) in the west than in the east (Fig. 3). This might be linked to phenological differences in the timing of the primary production season in different regions of the Irish Sea. Coastal areas in the western part of the basin (east coast of Ireland) are known to have a production season starting early in the year (MarcheApril) followed by production in more offshore regions (Gowen et al., 1995, 1999), while the spring production seems to occur slightly later in Liverpool Bay (eastern Irish Sea) (Foster et al., 1982). Although Gowen et al. (2000) found no differences in the timing of the spring bloom in the eastern and western part of the Irish Sea; analysis of Continuous Plankton Recorder programme data did suggest the existence of a slight temporal shift (<1 month) in the spring increase of the colour index (an index of phytoplankton abundance) of the two regions (Edwards and Johns, 2004). Therefore, if jellyfish production (i.e. strobilation of polyps) or growth is synchronised with the spring bloom (to increase chances of food availability to ephyrae and following pelagic stages), younger medusae (i.e. smaller and lighter) might be expected in the eastern region at the time of the survey in comparison with the western region. Furthermore, the stratified region of the western Irish Sea supports a higher daily primary production and zooplankton biomass (or standing stock) than mixed regions (Fogg et al., 1985; Williams et al., 1994; Gowen et al., 1995). This may provide better feeding (and therefore growth) conditions for scyphomedusae, which may explain why larger Aurelia aurita were found in stratified than in mixed waters (Fig. 3). Such links between water stratification and food availability have previously been suggested for Irish Sea sprat larvae (Sprattus sprattus, Coombs et al., 1992). Finally, the observed spatial heterogeneity in the average size of medusae highlights an important potential source of bias for jellyfish by-catch time series analysis: the risk that even a slight change in the timing of the production season may affect the reliability/comparability of the record from one year to another. It is of concern that, under current climate change scenarios, phenological changes are to be expected, and therefore this issue may be exacerbated (Edwards and Richardson, 2004). In the Irish Sea, recent simulations forecasted that the time at which annual maximum and minimum temperatures are reached may shift of up to two weeks (Olbert et al., 2012). This would affect both the biomass estimates of gelatinous zooplankton and the life-stages observed by annual surveys of 0-group fish. 5. Conclusions The present work suggests that any interaction between scyphomedusae and any other components of the ecosystem of the Irish Sea would not be exerted homogenously across the region or from one year to the next. Importantly, this is likely to be also the case elsewhere (Dawson et al., 2014), and any future efforts at numerically modelling jellyfishefish interactions, or jellyfish contribution to other ecological processes, should not only consider the ‘overall jellyfish biomass’ of a system, but also species-specific distribution patterns.

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

Acknowledgements AFBI's Northern Irish Sea Methot-Isaacs Kidd-net Survey for pelagic juvenile gadoids is funded by the Department of Agriculture and Rural Development (DARD) of Northern Ireland. TB, TKD, and GCH were supported by the EcoJel project, funded through the INTERREG IVa programme of the European Regional Development Fund (ERDF). MKSL was supported by a NERC Studentship to GCH and projects funded through the French National Research Agency (Ecogely project ANR-10-PDOC-005-01 and ANR-12-EMMA-0008). Many thanks are expressed to the scientists and crew involved in the surveys, with special thanks to Damien Haberlin for his help in collecting the data in 2010. We are grateful to Prof. John Davenport and Dr Rob McAllen for their valuable feedback, and to Prof. Mike Elliott, Dr Jon Houghton and other anonymous reviewers for their constructive comments. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ecss.2014.08.018. References Armstrong, M.J., Connolly, P., Nash, R.D.M., Pawson, M.G., Alesworth, E., Coulahan, P.J., Dickey-Collas, M., et al., 2001. An application of the annual egg production method to estimate the spawning biomass of cod (Gadus morhua L.), plaice (Pleuronectes platessa L.) and sole (Solea solea L.) in the Irish Sea. ICES J. Mar. Sci. 58, 183e203. Båmstedt, U., Ishii, H., Martinussen, M.B., 1997. Is the scyphomedusa Cyanea capillata (L.) dependent on gelatinous prey for its early development? Sarsia 82, 269e273. Båmstedt, U., Martinussen, M.B., Matsakis, S., 1994. Trophodynamics of two scyphozoan jellyfishes, Aurelia aurita and Cyanea capillata, in western Norway. ICES J. Mar. Sci. 51, 369e382. Barz, K., Hirche, H.-J., 2007. Abundance, distribution and prey composition of scyphomedusae in the southern North Sea. Mar. Biol. 151, 1021e1033. Bastian, T., Haberlin, D., Purcell, J.E., Hays, G.C., Davenport, J., McAllen, R., Doyle, T.K., 2011. Large-scale sampling reveals the spatio-temporal distributions of the jellyfish Aurelia aurita and Cyanea capillata in the Irish Sea. Mar. Biol. 158, 2639e2652. Billett, D.S.M., Bett, B.J., Jacobs, C.L., Rouse, I.P., Wigham, B.D., 2006. Mass deposition of jellyfish in the deep Arabian Sea. Limnol. Oceanogr. 51, 2077e2083. Brewer, R.H., 1989. The annual pattern of feeding, growth, and sexual reproduction in Cyanea (Cnidaria, Scyphozoa) in the Niantic River estuary, Connecticut. Biol. Bull. 176, 272e281. Brodeur, R.D., 1998. In situ observations of the association between juvenile fishes and scyphomedusae in the Bering Sea. Mar. Ecol. Prog. Ser. 163, 11e20. Brodeur, R.D., Beth, M., Ciannelli, L., Purcell, J.E., Bond, N. a., Stabeno, P.J., Acuna, E., et al., 2008. Rise and fall of jellyfish in the eastern Bering Sea in relation to climate regime shifts. Prog. Oceanogr. 77, 103e111. Brodeur, R.D., Sugisaki, H., Hunt, G.L.J., 2002. Increases in jellyfish biomass in the Bering Sea: implications for the ecosystem. Mar. Ecol. Prog. Ser. 233, 89e103. Buecher, E., Sparks, C., Brierley, A. s., Boyer, H., Gibbons, M.J., 2001. Biometry and size distribution of Chrysaora hysoscella (Cnidaria, Scyphozoa) and Aequorea aequorea (Cnidaria, Hydrozoa) off Namibia with some notes on their parasite Hyperia medusarum. J. Plankton Res. 23, 1073e1080. Bunn, N., Fox, C.J., Nash, R.D.M., 2004. Spring Plankton Surveys of the Eastern Irish Sea in 2001, 2002 and 2003: Hydrography and the Distribution of Fish Eggs and Larvae. CEFAS Science Series Data Report 42, Lowesoft, UK. Busacker, G.P., Adelman, I.R., Goolish, E.M., 1990. Chapter 11-growth. In: Schreck, C.B., Moyle, P.B. (Eds.), Methods in Fish Biology. American Fisheries Society, Bethesda, Maryland, pp. 363e387. Condon, R.H., Steinberg, D.K., del Giorgio, P.A., Bouvier, T.C., Bronk, D.A., Graham, W.M., Ducklow, H.W., 2011. Jellyfish blooms result in a major microbial respiratory sink of carbon in marine systems. Proc. Natl. Acad. Sci. 108, 10225. Coombs, S.H., Nichols, J.H., Conway, D.V.P., Milligan, S., Halliday, N.C., 1992. Food availability for sprat larvae in the Irish Sea. J. Mar. Biol. Assoc. U. K. 72, 821e834. Dawson, M.N., Cieciel, K., Decker, M.B., Hays, G.C., Lucas, C.H., Pitt, K.A., 2014. Population-level perspectives on global change: genetic and demographic analyses indicate various scales, timing, and causes of scyphozoan jellyfish blooms. Biol. Invasions. http://dx.doi.org/10.1007/s10530-014-0732-z. De la Cruz, M., 2008. Metodos para analizar datos puntuales. In: Maestre, F.T., Escudero, A., y Bonet, A. (Eds.), Introduccion al Analisis Espacial de Datos en Ecologia y Ciencias Ambientales: Metodos y Aplicaciones. Asociacion Espanola de Ecologia Terrestre, Universidad Rey Juan Carlos y Caja de Ahorros del Mediterraneo, Madrid, pp. 76e127.

311

Dickey-Collas, M., Gowen, R.J., Fox, C.J., 1996. Distribution of larval and juvenile fish in the western Irish Sea: relationship to phytoplankton, zooplankton biomass and recurrent physical features. Mar. Freshw. Res. 47, 169e181. Dickson, R.R., Boelens, R.G.V., 1988. The Status of Current Knowledge on Anthropogenic Influences in the Irish Sea. ICES Cooperative Research Report 155, Copenhagen, Denmark. Doyle, T.K., Hays, G.C., Harrod, C., Houghton, J.D.R., 2014. Ecological and societal benefits of jellyfish. In: Pitt, K.A., Lucas, C.H. (Eds.), Jellyfish Blooms. Springer, Netherlands, pp. 105e127. Doyle, T.K., Houghton, J.D.R., Buckley, S.M., Hays, G.C., Davenport, J., 2007. The broad-scale distribution of five jellyfish species across a temperate coastal environment. Hydrobiologia 579, 29e39. Edwards, M., Johns, D.G., 2004. SAHFOS/Irish Sea Pilot. Plankton Communities initial report. In Irish Sea Pilot e Report on the development of a Marine Landscape classification for the Irish Sea. JNCC, pp. 98e109. Report 346. Edwards, M., Richardson, A.J., 2004. Impact of climate change on marine pelagic phenology and trophic mismatch. Nature 430, 881e884. Eriksen, E., Prozorkevich, D., Trofimov, A., Howell, D., 2012. Biomass of scyphozoan jellyfish, and its spatial association with 0-group fish in the Barents Sea. PLoS One 7, e33050. Fleming, N.E.C., Harrod, C., Houghton, J.D.R., 2013. Identifying potentially harmful jellyfish blooms using shoreline surveys. Aquac. Environ. Interact. 4, 263e272. Fogg, G.E., Egan, B., Floodgate, G.D., Jones, D.A., Kassab, J.Y., Lochte, K., Rees, E.I.S., et al., 1985. Biological studies in the vicinity of a shallow-sea tidal mixing front VII. The frontal ecosystems. Philosophical transactions of the Royal Society of London. Series B Biol. Sci. 310, 555e571. Foster, P., Voltolina, D., Beardall, J., 1982. A seasonal study of the distribution of surface state variables in Liverpool Bay. IV. The spring bloom. J. Exp. Mar. Biol. Ecol. 62, 93e115. Fox, C.J., Dickey-Collas, M., Winpenny, A.J., 1997. Spring Plankton Surveys of the Irish Sea in 1995: the Distribution of Fish Eggs and Larvae. Cefas Science Series Technical Report No. 104, Lowestoft, UK. Fox, C.J., McCloghrie, P., Nash, R.D.M., 2009. Potential transport of plaice eggs and larvae between two apparently self-contained populations in the Irish Sea. Estuar. Coast. Shelf Sci. 81, 381e389. Fox, C.J., O'Brien, C.M., Dickey-Collas, M., Nash, R.D.M., 2000. Patterns in the spawning of cod (Gadus morhua L.), sole (Solea solea L.) and plaice (Pleuronectes platessa L.) in the Irish Sea as determined by generalized additive modelling. Fish. Oceanogr. 9, 33e49. Golding, N., Vincent, M.A., Connor, D.W., 2004. Irish Sea Pilot-report on the Development of a Marine Landscape Classification for the Irish Sea. Report 346. JNCC. Gowen, R.J., McCullough, G., Kleppel, G.S., Houchin, L., Elliott, P., 1999. Are copepods important grazers of the spring phytoplankton bloom in the western Irish Sea? J. Plankton Res. 21, 465e483. Gowen, R.J., Stewart, B.M., Mills, D.K., Elliott, P., 1995. Regional differences in stratification and its effect on phytoplankton production and biomass in the northwestern Irish Sea. J. Plankton Res. 17, 753e769. Gowen, R., Mills, D., Trimmer, M., Nedwell, D., 2000. Production and its fate in two coastal regions of the Irish Sea: the influence of anthropogenic nutrients. Mar. Ecol. Prog. Ser. 208, 51e64. Hansson, L.J., 1997. Capture and digestion of the scyphozoan jellyfish Aurelia aurita by Cyanea capillata and prey response to predator contact. J. Plankton Res. 19, 195e208. Hay, S., 2006. Marine ecology: gelatinous bells may ring change in marine ecosystems. Curr. Biol. 16, R679eR682. Hay, S.J., Hislop, J.R.G., Shanks, A.M., 1990. North-Sea scyphomedusae; summer distribution, estimated biomass and significance particularly for 0-group gadoid fish. Neth. J. Sea Res. 25, 113e130. Heffernan, O.A., Danilowicz, B.S., Milligan, S.P., 2004. Determination of speciesspecific spawning distributions of commercial finfish in the Irish Sea using a biochemical protein-based method. Mar. Ecol. Prog. Ser. 284, 279e291. Hill, A.E., Brown, J., Fernand, L., 1996. The western Irish Sea gyre: a retention system for Norway Lobster. Oceanol. Acta 19, 357e368. Holst, S., 2012. Morphology and development of benthic and pelagic life stages of North Sea jellyfish (Scyphozoa, Cnidaria) with special emphasis on the identification of ephyra stages. Mar. Biol. 159, 2707e2722. Holst, S., Jarms, G., 2010. Effects of low salinity on settlement and strobilation of scyphozoa (Cnidaria): is the lion's mane Cyanea capillata (L.) able to reproduce in the brackish Baltic Sea? Hydrobiologia 645, 53e68. Holst, S., Laakmann, S., 2014. Morphological and molecular discrimination of two closely related jellyfish species, Cyanea capillata and C. lamarckii (Cnidaria, Scyphozoa), from the northeast Atlantic. J. Plankton Res. 36, 48e63. Houghton, J.D.R., Doyle, T.K., Davenport, J., Hays, G.C., 2006a. Developing a simple, rapid method for identifying and monitoring jellyfish aggregations from the air. Mar. Ecol. Prog. Ser. 314, 159e170. Houghton, J.D.R., Doyle, T.K., Davenport, J., Lilley, M.K.S., Wilson, R.P., Hays, G.C., 2007. Stranding events provide indirect insights into the seasonality and persistence of jellyfish medusae (Cnidaria: Scyphozoa). Hydrobiologia 589, 1e13. Houghton, J.D.R., Doyle, T.K., Wilson, M.W., Davenport, J., Hays, G.C., 2006b. Jellyfish aggregations and leatherback turtle foraging patterns in a temperate coastal environment. Ecology 87, 1967e1972. ICES, 2009. Cod in division VIIa (Irish sea). In: ICES Advice 2009. ICES, Copenhagen, Denmark. Book 5. ICES, 2013. Herring in division VIIa North of 52 300 N (Irish sea). In: ICES Advice 2013. ICES, Copenhagen, Denmark. Book 5.

312

T. Bastian et al. / Estuarine, Coastal and Shelf Science 149 (2014) 302e312

Kelley, D., 2011. Package 'oce': Analysis of Oceanographic data, R package.
ry, A., Lilley, M.K.S., Ferraris, M., Elineau, A., Berline, L., Cuvilliers, P., Gilletta, L., Thie et al., 2014. Culture and growth of the jellyfish Pelagia noctiluca in the laboratory. Mar. Ecol. Prog. Ser. 510, 265e273. Lilley, M.K.S., Beggs, S.E., Doyle, T.K., Hobson, V.J., Stromberg, K.H.P., Hays, G.C., 2011. Global patterns of epipelagic gelatinous zooplankton biomass. Mar. Biol. 158, 2429e2436. Lilley, M.K.S., Houghton, J.D.R., Hays, G.C., 2009. Distribution, extent of inter-annual variability and diet of the bloom-forming jellyfish Rhizostoma in European waters. J. Mar. Biol. Assoc. U. K. 89, 39e48. Lynam, C.P., Brierley, A.S., 2006. Enhanced survival of 0-group gadoid fish under jellyfish umbrellas. Mar. Biol. 150, 1397e1401. Lynam, C.P., Hay, S.J., Brierley, A.S., 2005. Jellyfish abundance and climatic variation: contrasting responses in oceanographically distinct regions of the North Sea, and possible implications for fisheries. J. Mar. Biol. Assoc. U. K. 85, 435e450. Lynam, C.P., Lilley, M.K.S., Bastian, T., Doyle, T.K., Beggs, S.E., Hays, G.C., 2011. Have jellyfish in the Irish Sea benefited from climate change and overfishing? Glob. Change Biol. 17, 767e782. Masuda, R., 2009. Ontogenetic changes in the ecological function of the association behavior between jack mackerel Trachurus japonicus and jellyfish. Hydrobiologia 616, 269e277. Methot, R.D., 1986. Frame trawl for sampling pelagic juvenile fish, California Cooperative Oceanic Fisheries Investigations, pp. 267e278. €ller, H., 1980. Scyphomedusae as predators and food competitors of larval fish. Mo Meeresforschung 28, 90e100. €ller, H., 1984. Reduction of a larval herring population by jellyfish predator. Mo Science 224, 621. Nagabhushanam, A.K., 1959. Studies on the Biology of the Commoner Gadoids in the Manx Area, with Special Reference to Their Food and Feeding Habits. University of Liverpool, UK. PhD Thesis. Nash, R.D.M., Dickey-Collas, M., Milligan, S.P., 1998. Descriptions of the Gulf VH/ PRO-NET and MAFF/Guildline unencased high-speed plankton samplers. J. Plankton Res. 20, 1915e1926. Olbert, A.I., Dabrowski, T., Nash, S., Hartnett, M., 2012. Regional modelling of the 21st century climate changes in the Irish Sea. Cont. Shelf Res. 41, 48e60. Olesen, N.J., Frandsen, K., Riisgard, H.U., 1994. Population-dynamics, growth and energetics of jellyfish Aurelia aurita in a shallow fjord. Mar. Ecol. Prog. Ser. 105, 9e18. Pauly, D., Graham, W., Libralato, S., Morissette, L., Palomares, M.L.D., 2009. Jellyfish in ecosystems, online databases, and ecosystem models. Hydrobiologia 616, 67e85. Pikesley, S.K., Godley, B.J., Ranger, S., Richardson, P.B., Witt, M.J., 2014. Cnidaria in UK coastal waters: description of spatio-temporal patterns and inter-annual variability. J. Mar. Biol. Assoc. U. K. 1e8. http://dx.doi.org/10.1017/ S0025315414000137. Pitt, K.A., Koop, K., Rissik, D., 2005. Contrasting contributions to inorganic nutrient recycling by the co-occurring jellyfishes, Catostylus mosaicus and Phyllorhiza punctata (Scyphozoa, Rhizostomeae). J. Exp. Mar. Biol. Ecol. 315, 71e86. Purcell, J.E., 1997. Pelagic cnidarians and ctenophores as predators: selective predation, feeding rates, and effects on prey populations. Ann. l’Institut Oceanogr. 73, 125e137.

Purcell, J.E., 2009. Extension of methods for jellyfish and ctenophore trophic ecology to large-scale research. Hydrobiologia 616, 23e50. Purcell, J.E., Arai, M.N., 2001. Interactions of pelagic cnidarians and ctenophores with fish: a review. Hydrobiologia 451, 27e44. R Development Core Team, 2011. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Richardson, A.J., Bakun, A., Hays, G.C., Gibbons, M.J., 2009. The jellyfish joyride: causes, consequences and management responses to a more gelatinous future. Trends Ecol. Evol. 24, 312e322. Russell, F.S., 1970. The Medusae of the British Isles. In: Pelagic Scyphozoa, vol. II. Cambridge University Press, Cambridge, UK. Ruzicka, J.J., Brodeur, R.D., Emmett, R.L., Steele, J.H., Zamon, J.E., Morgan, C.A., Thomas, A.C., et al., 2012. Interannual variability in the Northern California current food web structure: changes in energy flow pathways and the role of forage fish, euphausiids, and jellyfish. Prog. Oceanogr. 102, 19e41. Simpson, J.H., 1981. The shelf-sea fronts e implications of their existence and behavior. Phil. Trans. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci. 302, 531. Simpson, J.H., Hunter, J.R., 1974. Fronts in the Irish sea. Nature 250, 404e406. Suchman, C.L., Brodeur, R.D., 2005. Abundance and distribution of large medusae in surface waters of the northern California current. Deep Sea Res. Part II: Top. Stud. Oceanogr. 52, 51e72. Syrjala, S.E., 1996. A statistical test for a difference between the spatial distributions of two populations. Ecology 77, 75e80. Tingley, D., 2006. Northern Ireland Fleet Futures Analysis (2004e2013) e Methodology and Results. Final report. CEMARE, Portsmouth, UK. Tinta, T., Malej, A., Kos, M., Turk, V., 2010. Degradation of the Adriatic medusa Aurelia sp. by ambient bacteria. Hydrobiologia 645, 179e191. Titelman, J., Hansson, L.J., 2006. Feeding rates of the jellyfish Aurelia aurita on fish larvae. Mar. Biol. 149, 297e306. Titelman, J., Riemann, L., Sornes, T.A., Nilsen, T., Griekspoor, P., Båmstedt, U., 2006. Turnover of dead jellyfish: stimulation and retardation of microbial activity. Mar. Ecol. Prog. Ser. 325, 43e58. Utne-Palm, A.C., Salvanes, A.G.V., Currie, B., Kaartvedt, S., Nilsson, G.E., Braithwaite, V.A., Stecyk, J.A.W., et al., 2010. Trophic structure and community stability in an overfished ecosystem. Science 329, 333e336. Uye, S., Shimauchi, H., 2005. Population biomass, feeding, respiration and growth rates, and carbon budget of the scyphomedusa Aurelia aurita in the Inland Sea of Japan. J. Plankton Res. 27, 237e248. West, E.J., Welsh, D.T., Pitt, K.A., 2008. Influence of decomposing jellyfish on the sediment oxygen demand and nutrient dynamics. Hydrobiologia 616, 151e160. Williams, R., Conway, D.V.P., Hunt, H.G., 1994. The role of copepods in the planktonic ecosystems of mixed and stratified waters of the European shelf seas. Hydrobiologia 292, 521e530. Yamamoto, J., Hirose, M., Ohtani, T., Sugimoto, K., Hirase, K., Shimamoto, N., Shimura, T., et al., 2008. Transportation of organic matter to the sea floor by carrion falls of the giant jellyfish Nemopilema nomurai in the Sea of Japan. Mar. Biol. 153, 311e317. Zavolokin, A.V., 2010. Distribution and abundance dynamics of jellyfish in the Sea of Okhotsk. Russ. J. Mar. Biol. 36, 157e166.