Abstract
Whether a perceived increase in the abundance of jellyfishes is related to changing marine environments has been considered primarily using large-scale analyses of multi-species assemblages. Yet jellyfish blooms—rapid increases in the biomass of pelagic coelenterate species—are single-species demographic events. Using published and new genetic analyses and population surveys, we investigate whether there may be a critical knowledge gap between the scales of recent analyses and the scales of natural phenomena. We find that scyphomedusae may show population genetic structure over scales of tens to hundreds of kilometers, that environments vary regionally and locally, and that populations of medusae can display uncorrelated dynamics on these scales. These findings suggest genetic differences between populations and/or environmental differences between sites are important determinants of population dynamics in these jellyfishes. Moreover, the local abundance of medusae may be most strongly correlated with preceding rather than current local environmental conditions, indicating there is a cumulative time-course to the formation of ‘blooms’. Broad-scale macro-ecological analyses will need to build from coordinated, long-term, fine-grained studies to synthesize, rather than mask, population-level phenomena in larger-scale analyses.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Whether a perceived increase in the abundance of jellyfishes is a consequence of changing marine environments often has been framed as a question requiring very large-scale analyses of multiple species (Purcell 2005; Richardson et al. 2009; Brotz et al. 2012; Condon et al. 2012, 2013). Yet an increase in the number or biomass of a pelagic coelenterate species—a ‘jellyfish bloom’—is a “characteristic of … populations” (Graham et al. 2001:199), i.e., a demographic phenomenon (Lucas and Dawson 2014). While ‘large’ and ‘population’ scales are not necessarily mismatched analytically, they may be empirically, and it is timely to ask whether a critical gap in knowledge exists between the scales of recent analyses and the scales of natural phenomena.
Jellyfish blooms have been categorized either as a ‘true bloom’ if resulting from production and growth of new individuals, or as an ‘apparent bloom’ if resulting from redistribution of existing individuals (Graham et al. 2001). Although an apparent bloom most often will be preempted elsewhere by a true bloom, the distinction between ‘true’ and ‘apparent’ blooms is of some importance because it indicates the location, and possibly also timing, of the cause(s) of a jellyfish bloom (Lucas and Dawson 2014). The cause(s) of a true bloom occur(s), or occurred, at the site of the bloom; the cause(s) of an apparent bloom include(s), or included, event(s) elsewhere. For most blooms in most places, this spatio-temporal series of events remains largely unknown, but is essential to understanding the scales, timing, and causes of jellyfish blooms and whether or not they are ‘invasive’ sensu Valéry et al. (2008, 2009; see also “outbreak” sensu Richardson et al. 2009).
An allied issue is identifying the geographic scale at which jellyfish blooms occur. Elevated biomass of planktonic cnidarians has been reported across localized fronts (e.g., Brodeur et al. 1997; Purcell et al. 2000), within and between coastal lakes and fjords (Hamner and Hauri 1981; Lucas and Lawes 1998; Pitt and Kingsford 2000, 2003; Youngbluth and Båmstedt 2001), across bays and estuaries (e.g., Graham et al. 1992; Lucas et al. 1997), and in coastal shelf and midwater regions (e.g., Raskoff 2001; Suchman and Brodeur 2005; Decker et al. 2013); these studies often report temporal heterogeneity in abundance too. The spatial scales over which such variation in abundance reflects particular demographic events or demes, however, is uncertain. For example, information on the whereabouts of polyps and subsequent movements of ephyrae and medusa are for the most-part lacking.
Uncertainty about the provenance and extent of jellyfish abundances introduces ambiguity into efforts to describe jellyfish demographics and identify the causes of fluctuations. Analyses at single locations may not capture the full extent or time-course of demographic events nor discern among factors acting at larger scales (Decker et al. 2013; Lee et al. 2013; Lucas and Dawson 2014). Analyses that have aggregated data at the level of large marine ecosystems (LMEs; Brotz et al. 2012) or used primarily polyphyletic groups (e.g., Condon et al. 2013) to discern global trends, adopt shortcomings of the original studies and additionally may mask finer-scale spatial, temporal, and taxonomic variation in the occurrence of jellyfishes (and of other taxa in those datasets such as salps). Oversimplistic aggregation of data thus risks dissociating patterns of abundance from underlying biological, spatial, and temporal causes. Influences on marine and coastal systems that have been hypothesized to drive jellyfish dynamics—such as eutrophication, coastal development, alteration of large-scale ecosystem dynamics by over-fishing, inter-regional species introductions, and global climate change (Richardson et al. 2009; Duarte et al. 2012; Purcell 2012; Bayha and Graham 2014)—do vary spatially and temporally in their occurrences, intensities, and frequencies (Halpern et al. 2008, 2009). Even ‘global’ climate change has regionally differentiable environmental effects (Falvey and Garreaud 2009), and the global biodiversity crisis manifests through regional and local events (Opgenoorth and Faith 2013). The scales, strengths, transformations, and interactions of drivers of population dynamics of jellyfishes are uncertain.
Designing analyses that match the scales of pelagic coelenterate population dynamics with diverse putative causes thus remains largely unaddressed. Here, we explore a method that combines approaches from population genetics and population biology to identify genetic units, their spatial extent, their environments, and to determine whether distinct populations respond differently or similarly to which kinds of environmental variation (Lucas and Dawson 2014; see also e.g., Etterson and Shaw 2001; Both et al. 2006; Lehtonen et al. 2009; Pespeni et al. 2013). Our exploration is preliminary, limited to the five case studies—Aurelia aurita, Catostylus mosaicus, Chrysaora melanaster, Mastigias papua, and Rhizostoma octopus—for which both population biology and population genetic data currently are available through a mixture of published and new analyses. In each case, we wish to answer the following four questions. Is there significant population genetic structure? On what spatial scales and how strongly are these jellyfish subdivided? Do genetically distinct populations have similar demographics? And, conversely, do all locations inhabited by the same population show the same dynamics? Based on the answers to those questions, we consider how to integrate the scale of jellyfish population dynamics analytically with the scales of putative causes, including global change.
Methods
Population genetics
Collection and phylogeographic analyses of C. mosaicus, M. papua and R. octopus are described by Dawson (2005a), Dawson and Hamner (2005), and Lee et al. (2013), respectively. Here, we summarize the subset of findings relevant to the current study (1) because they were used to decide the study design and (2) to distinguish this information from new results of this study.
Two clades, representing subspecies of C. mosaicus occur in southeastern Australia (Fig. 1) and within each of these clades there is statistically significant finer scale phylogeographic structure. For example, within the “Central” clade, comparison of Botany Bay and Lake Illawarra (NLI) leads to estimation of ΦST = 0.393, a high degree of differentiation reflecting only a single shared allele (of nine total in both populations).
a Catostylus mosaicus inhabit coastal lagoons and estuaries along the coast of southeastern Australia. b C. mosaicus are highly structured genetically, with distinct clades north versus south of Cape Howe, and distinct populations within each region; see Dawson (2005a, b, c) for details. Colors represent different locations, circles represent different alleles, the area of circles is proportional to the abundance of the allele, and each line of unit length between any symbol represents a single mutation. c The population dynamics of geographically proximate populations in Botany Bay (NBB) and NLI—here shown as changes in biomass—are uncorrelated; see Pitt and Kingsford (2000) for details. Different colors are used to represent locations inhabited by genetically distinct (ΦST ≥ 0.2) populations. The area of a bubble is proportional to the abundance of medusae
Mastigias papua in Palau form multiple geographically isolated populations within marine lakes (Fig. 2). Genetic comparisons among the three lakes considered here, for which there also are abundance data, reveal values of ΦST ≥ 0.94, reflecting zero alleles being shared between any pair of populations.
a Mastigias papua inhabit coastal coves and marine lakes in Palau. b Animals in coves are genetically similar, whereas marine lakes harbor genetically distinct populations; see Dawson and Hamner (2005) for details. c The dynamics of populations in three lakes—OTM, NLK, and GLK—are weakly correlated (r = −0.266 ≤ r ≤ 0.113). d Abundances of M. papua in these three lakes show different strengths and timing of correlations with local environmental variation. The meaning of symbols, shapes, and colors is as described in Fig. 1
Rhizostoma octopus in the Irish Sea and adjacent areas has generally high haplotype diversity (h ≥ 0.667) except in Tremadog Bay (h = 0.396) which also had lowest nucleotide diversity. There are three genetically distinct groups: (1) La Rochelle, (2) Tremadog Bay, and (3) Gormanstown, Rosslare, and Camarthen Bay (Fig. 3).
a Rhizostoma octopus inhabit coastal embayments and open coastal areas in the northeastern Atlantic and adjacent seas and b are genetically connected or isolated as a function of landscape distance; see Lee et al. (2013) for details. c Annual abundances of medusae in four locations suggest mixed levels of spatial and temporal synchrony. The meaning of symbols, shapes, and colors is as described in Fig. 1
In addition, we conducted comparable analyses of new samples of Aurelia aurita and C. melanaster for this study. Oral arm tissue was clipped from 85 A. aurita at three locations—Horsea Lake, Beaulieu River, and Southampton Water (Fig. 4)—in southern England and two sites in the Irish Sea, all within its native range (Dawson et al. 2005). [To the best of our knowledge, all populations studied here are within species’ native ranges Dawson 2005a, b)]. A ~25 mm3 piece of bell tissue was cut from 125 whole C. melanaster collected by surface and midwater trawls at eight stations south of 60°N during annual surface and midwater trawl surveys by NOAA Alaska Fisheries Science Center (see below for more details; Fig. 5). All tissue samples were preserved individually in 1–2 ml of 95 % ethanol, labeled, refrigerated, and later shipped to the University of California, Merced for storage at −15 °C and analyses.
a Aurelia aurita inhabit coastal waters in the Irish Sea, and estuaries and a man-made lake in southern England. b A. aurita in the southern England sites, Southampton Water and Horsea Lake, are genetically similar. c Annual abundances of medusae in these two locations are negatively correlated. d Despite differences in the population dynamics between sites, a strong correlation between medusae abundance and prey abundance 2–3 months prior emerges at both locations. The meaning of symbols and shapes is as described in Fig. 1; shades of blue are used to distinguish study sites that, in this case, are genetically similar. Scale bar: three miles is approximately 5 kilometers
a Chrysaora melanaster across in the southeastern Bering Sea are b genetically undifferentiated; colors are used to indicate different sites only. c Site-specific abundances of medusae vary by several orders of magnitude across locations within years, and mean annual abundances differ by up to an order of magnitude. The maximum linear distance separating locations is ~500 km. Scale bar three miles is approximately 5 kilometers
DNA was extracted from the ethanol-preserved tissues using a CTAB-Phenol/Chloroform based protocol (Dawson et al. 1998). Cytochrome c oxidase subunit I (COI) was amplified using LCOjf (5′-ggtcaacaaatcataaagatattggaac-3′) and HCO2198 (Folmer et al. 1994) in PCRs consisting of six steps of 94 °C for 8 min, 51 °C for 2 min, 72 °C for 2 min, 94 °C for 4 min, 52 °C for 2 min, 72 °C for 2 min, then 33 cycles of 95 °C/45 s, 52 °C/45 s and 72 °C/60 s, followed by an extension step at 72 °C for 10 min; the reaction was terminated by cooling to 4 °C. Amplified COI fragments were purified using Qiagen (Valencia, CA) PCR clean-up columns, and prepared for sequencing in both directions using the PCR primers at the UC Berkeley DNA Sequencing Facility. Contigs were assembled and electropherograms were checked visually in Sequencher v4.8 (GeneCodes, Ann Arbor, Michigan, USA), misreads corrected, and poorly resolved terminal portions of sequences discarded. The identities of sequences were confirmed by BLAST searching sequences in GenBank (Altschul et al. 1997), verifying the existence of open reading frames in amino acid translations using the invertebrate mitochondrial code in Sequencher, and including sequences in preliminary phylogenetic analysis (e.g., with sequences from Bayha et al. 2010). Each species’ COI sequences were aligned using ClustalX (Jeanmougin et al. 1998) and alignments checked by eye using amino acid translated sequences. All terminal positions with missing data were excluded from subsequent analyses.
Gene and nucleotide diversity were calculated in Arlequin v.3.5.1.2 (Excoffier and Lischer 2010). To estimate population differentiation, we calculated ΦST values (10,100 permutations) using Arlequin. Significance levels were corrected for multiple tests using the sequential Bonferroni procedure (Rice 1989). Haplotype networks were reconstructed using statistical parsimony in TCS (Clement et al. 2000) or, for C. melanaster, using uncorrected pairwise sequence differences calculated in Arlequin as input for Haplotype Viewer (http://www.cibiv.at/~greg/haploviewer).
Population biology: abundance, biomass, and physical environment
Aurelia aurita, Southampton Water and Horsea Lake
Population dynamics of A. aurita were assessed in two locations in the Solent estuarine system on the south coast of the UK: (1) the productive estuary Southampton Water (50°51.36′N, 1°20.88′W) and (2) the semi-enclosed man-made Horsea Lake (50°49.32′N, 1°6.24′W). These two locations are within ~50 km of each other. Sampling in Southampton Water took place every 2 weeks between December 1992 and January 1995, and between February 1994 and January 1995 in Horsea Lake.
Collection of A. aurita medusae and environmental data (e.g., temperature, salinity) are fully described by Lucas and Williams (1994) for Southampton Water and by Lucas (1996) for Horsea Lake. Briefly, double-oblique hauls were made using 210 and 300 µm mesh cod-end plankton nets. A. aurita medusae were enumerated and bell diameter (BD) measured using a ruler. A YSI multiparameter probe was used to measure temperature, salinity and oxygen at 1 m depth intervals from the surface to 5 m.
Catostylus mosaicus, southeastern Australia
The population dynamics of C. mosaicus were assessed at Botany Bay (33°59′S, 151°11′E) and Lake Illawarra (NLI; 34°31′S, 150°50′E). Botany Bay was sampled approximately every 4–6 weeks from March 1996 until July 1998 and NLI was sampled approximately every 6–8 weeks from January 1996 to May 1998. Abundance of C. mosaicus was assessed using visual counts. The total numbers of medusae >50 mm BD were counted from the bow of a boat travelling at 10 kmh−1 along transects 1,500 m long, 3 m wide and 1 m deep. Numbers of medusae were counted along six transects at each of six sites within each location (i.e. n = 36). Current speeds at all locations were low and would have had a negligible effect on the length of transects. Counts were only made when the wind speed was <15 knots (kt) because wind-generated turbulence appeared to cause medusae to move deeper in the water column. At each location and time, the BD of 100 medusae was measured in situ by a diver at each of two sites (i.e. n = 200) to characterize the size distribution of the population.
The mode of the size distribution >50 mm BD was converted to wet weight (WW, grams) using WW = 0.006 × BD2.426 (Pitt and Kingsford 2003). The modal WW was then converted to carbon mass (mg) using the average C:WW ratio of 0.00947 for scyphozoan medusae (Lucas et al. 2011). The average carbon biomass density (mg C m−3) per location was then estimated by multiplying the carbon biomass density by the average biomass density (number m−3). Abundance and biomass per cubic meter are correlated (p = 0.30, p = 0.022, n = 58). Contemporaneous environmental data were unavailable.
Chrysaora melanaster, Bering Sea
Annual surface and midwater trawl surveys targeting forage fish were conducted by the NOAA Alaska Fisheries Science Center from 2004–2012 during August to October across the eastern Bering Sea. The sampling grid for the survey covered stations off the coast of Alaska, 159–174°W longitude and 54.5–64°N latitude. The surveys were conducted aboard contracted fishing vessels and the NOAA ship R/V Oscar Dyson. All tows were standardized for gear (50 m wide by 18 m deep) and duration of tow (30 min at 3.5–5 kt, covering 2.8–4.6 km). Jellyfish samples were collected using a midwater rope trawl, Model 400 (B.C. Cantrawl Pacific Limited). The rope trawl was 198 m long with nonuniform hexagonal mesh in the wings and body and a 1.2 cm mesh liner in the codend. All sampling was performed during daylight hours. Every medusa caught in the cod-end of the trawl net was sorted to species then enumerated.
Physical and biological oceanographic data were collected at each trawl station immediately prior to deploying the net. Vertical profiles of temperature, salinity, oxygen, chl a fluorescence, light attenuation (beam c), and photosynthetically available radiation (PAR) were measured with a Sea-Bird Electronics Inc. Model 25 or Model 9 conductivity-temperature-depth profiler with ancillary sensors. We considered values averaged throughout the water column and also bottom temperature (5–10 m above the seafloor); the latter is correlated with medusae abundance in north and central (but not south) areas of the eastern Bering Sea (Decker et al. 2013). Water samples for nutrients (N, P, Si) and chl a (total and size fractionated) were collected at various depths using 5-L Niskin bottles. Data were subsampled for only the eight locations at which medusa were collected for genetic analyses.
Mastigias papua, Palau
Changes in the abundance of M. papua medusae were measured in three stratified marine lakes: Ongeim’l Tketau (OTM; 07°09.830′N 134°22.500′E), Ngermeuangel Lake (NLK; 07°19.400′N 134°30.507′E), and Goby Lake (GLK; 07°18.910′N 134°30.133′E). Physical characteristics of these lakes are described by Hamner and Hamner (1998). From December 2000 to January 2004, we quantified the abundance of medusae monthly using a 0.5 m diameter, 1 mm mesh net hauled vertically from the bottom of the mixolimnion to the surface. Net hauls were distributed evenly across the lake at 15 stations, and the set of 15 stations was sampled as many (two or three) times as possible during the course of 1 day. Each medusa was measured to the nearest 0.5 cm BD. Values reported are the mean estimated abundance of medusae in the lake each day surveyed.
Monthly, for the same period, we measured temperature, salinity and oxygen concentration at 1 m intervals from the lake surface to the bottom of the mixolimnion, at two sites per lake, with hand-held YSI 85 or Hydrolab Quanta meters. For each lake and sample date, we calculated the mean mixolimnion temperature (MMT; Martin et al. 2006), the mean mixolimnion salinity (MMS), and mean mixolimnion dissolved oxygen (MMO) also from 0 to 12 m.
Rhizostoma octopus, Irish Sea and adjacent areas
Estimates of jellyfish abundance were made during aerial surveys from an altitude of 152 m at a constant speed of 185 km h−1 (100 kt). Numbers of jellyfish observed within a 250 m observation strip each side of the aircraft were tallied every 5 min by each observer and combined to give an estimate of abundance (detailed methods are described in Houghton et al. 2006). Aggregations were mapped in the Irish Sea (51°12′N–55°04.8′N, 003°W–008°34.8′W) from June to October. Maximal abundance recorded on individual surveys was used to describe the prevalence of blooms each year.
We obtained sea surface temperature data for each 1° grid-square encompassing each bloom site from the International Comprehensive Ocean–Atmosphere Data Set (http://rda.ucar.edu/datasets/ds540.1/). This widely used data-set is formed by merging many national and international data sources that contain measurements and visual observations from ships (merchant, navy, research), moored and drifting buoys, coastal stations, and other marine platforms and thus provides comparable data among sites.
Synthesis
For each species, we ran parametric correlation analyses (Pearson’s r) comparing changes in abundances among genetically distinct populations (i.e., ΦST ≥ ~0.2, which under a suite of simplifying assumptions approximates ≤1 migrant per generation) in Statistica v.7.1 (StatSoft Inc. 2006). For the datasets for which we also have environmental data, we investigated (1) contemporary correlations in a variable across sites, (2) contemporary correlations among all variables within each site, and (3) correlations between abundances of medusae in each genetically distinct population and (a) contemporary local environment and (b) local environment in preceding months. Positive autocorrelation within datasets will result in lower effective sample sizes (Fortin and Dale 2005:222) and lower power. We calculated the power (Fisher’s z) for the relevant correlation coefficient r and sample size n of all statistical tests using the power analysis tool in Statistica v.7.1. Sequential Bonferroni adjustment (Rice 1989) was used to correct the α-level for multiple tests using the same dataset within species; our primary interest was in comparing relative effect sizes.
Results
Population genetics
Previously published genetic analyses of A. aurita (sensu stricto), C. mosaicus, and M. papua relevant to this study are summarized in the “Methods” section (see also Figs. 1, 2, 3). The results of new analyses of A. aurita and C. melanaster follow.
Two genetically distinct groups of A. aurita occurred within our samples, one occupied the Irish Sea sites and another occurred in southern England (i.e. Southampton Water, Beaulieu River, and Horsea Lake); between-groups ΦST ≥ 0.28 (p < 0.05), and within-groups ΦST ≤ 0.11 (p > 0.05). Although haplotypes in these two populations occurred primarily in distinct clans (or ‘haplogroups’ Fig. 4), some haplotypes within Southampton Water and Horsea Lake were nested within haplotypes from the Irish Sea. Sequences are available in GenBank KJ026272–KJ026356.
Chrysaora melanaster has high haplotype richness (h = 0.9824 ± 0.0063), modest differentiation among alleles (π = 0.0064 ± 0.0036), and no phylogeographic structure in the southeastern Bering Sea (Fig. 5). Comparisons among all samples of C. melanaster medusae yield ΦST < ~0.09 (statistically undifferentiable from ΦST = 0). Sequences are available in GenBank KJ026148–KJ026271 (Figs. 2, 3, 4).
Population biology: abundance, biomass, and physical environment: and synthesis
Catostylus mosaicus, southeastern Australia
The maximum density of C. mosaicus medusae differs in Botany Bay and NLI, as does the timing of maximum and minimum densities. Changes in biomass of C. mosaicus medusa in Botany Bay and NLI do not co-vary (Pearson’s r = −0.004, p = 0.991, n = 13; Fisher’s z = 0.05; Fig. 1c). Environmental data were unavailable.
Mastigias papua, Palau
The monthly mean abundance of M. papua medusae varies almost threefold among marine lakes in Palau. Changes in abundances of M. papua medusae are not significantly correlated among lakes: GLK and NLK (r = 0.113, p = 0.498, n = 38; z = 0.10), GLK and OTM (r = −0.208, p = 0.209, n = 38; z = 0.24), or NLK and OTM (r = −0.266, p = 0.106, n = 38; z = 0.37).
Environmental characteristics show differing degrees of contemporary correlation among lakes; salinity is highly correlated across all lakes (0.89 ≤ r ≤ 0.94, p < 0.001, n = 38), temperature is moderately to strongly correlated with the stronger correlations between closer lakes (0.59 ≤ r ≤ 0.81, p < 0.001, n = 38; 0.98 ≤ z ≤ 1.00), while dissolved oxygen and Secchi disc depth generally are not correlated among lakes [DO: −0.08 ≤ r ≤ 0.06, p > 0.6 (z ≤ 0.08), except NLK vs OTM r = 0.54, p ≤ 0.001 (z = 0.95), n = 38; Secchi: −0.01 ≤ r ≤ 0.29, p ≥ 0.07 (z ≤ 0.07), n = 38].
Within GLK, temperature, salinity, and DO were cross-correlated (magnitude r ≥ 0.41, p ≤ 0.011, n = 38; z ≥ 0.74) but Secchi disk depth was uncorrelated (−0.04 ≤ r ≤ 0.33, p ≥ 0.04, n = 38; z ≤ 0.54). Within NLK, all environmental characteristics were consistently modestly cross-correlated (magnitude r = 0.40–0.61, p ≤ 0.013, n = 38; 0.72 ≤ z ≤ 0.99). Within OTM, temperature, salinity, DO, and Secchi disk depth were all uncorrelated (−0.19 ≤ r ≤ 0.38, p ≥ 0.02, n = 38; 0.21 ≤ z ≤ 0.67).
There were no significant correlations between contemporary medusae abundances and environmental parameters in GLK (−0.02 ≤ r ≤ 0.34, p ≥ 0.04, n = 38; 0.05 ≤ z ≤ 0.56) nor OTM (−0.35 ≤ r ≤ 0.04, p ≥ 0.03, n = 38; 0.06 ≤ z ≤ 0.59), but in NLK temperature, salinity and DO were each significantly correlated with medusae abundance (magnitude r ≥ 0.45, p ≤ 0.005, n = 38; z ≥ 0.83). In the majority of cases medusae abundance was more strongly correlated with some aspect of preceding rather than contemporary environment (Fig. 2d).
Rhizostoma octopus, Irish Sea and adjacent areas
Annual abundance of R. octopus medusae varies approximately tenfold within and among sites. Changes in abundances of R. octopus medusae at Camarthen Bay and Tremadog, the two sites for which sufficient data exist, are uncorrelated (r = 0.15, p = 0.74, n = 7; z = 0.06). Visual inspection suggests abundances at other sites also are uncorrelated (Fig. 3). In contrast, environmental temperature is highly correlated among sites (0.76 ≤ r ≤ 0.96, p ≤ 0.001, n = 50; z ≤ 1.00). Medusae abundance and local temperature are uncorrelated at Camarthen (r = −0.57, p = 0.18, n = 7; z = 0.30) and at Tremadog (r = 0.67, p = 0.10, n = 7; z = 0.43); sample sizes were too small at Rosslaire and La Rochelle to include these locations in comparisons.
Aurelia aurita, Southampton Water and adjacent waters
The maximum density of A. aurita medusae differed approximately fivefold between Horsea Lake and Southampton Water; changes in abundance in the two locations are modestly negatively but not significantly correlated (r = −0.47, p = 0.106, n = 5; z = 0.14). Considering the same five contemporaneous sample dates, environmental variation at the two sites is highly correlated in terms of temperature (r = 0.99, p = 0.001, n = 5; z = 0.99) and uncorrelated in terms of salinity (r = −0.74, p = 0.16, n = 5; z = 0.35) and non-medusoid zooplankton abundance (r = −0.61, p = 0.58, n = 5; z = 0.20).
Within sites, more data were comparable. At Horsea Lake, temperature, salinity, DO, zooplankton abundance, and chl a concentration are statistically uncorrelated (range: −0.60 ≤ r ≤ 0.34, p ≥ 0.07); sample sizes were ≥19 for all comparisons except any with DO (n = 9–10, leading to low power: z = 0.49 for the strongest correlation). In Southampton Water, temperature was modestly correlated with salinity, DO, and chl a only prior to Bonferroni correction (magnitude r = 0.25–0.52, 0.02 ≤ p ≤ 0.037, 19 ≤ n ≤ 72).
Considering all data available, correlations between contemporary medusae abundance and environmental characteristics at Horsea Lake were non-significant (−0.26 ≤ r ≤ 0.40, p ≥ 0.25, ntemp = 24, nsal = 21, nzoo = 23, nDO = 9, nChla = 23). Correlations between contemporary medusae abundance and environmental characteristics in Southampton Water were non-significant (magnitude r ≤ 0.25, p ≥ 0.29, nsal = 72, nzoo = 23, nDO = 19, nChla = 19) except with temperature (r = −0.32, p = 0.006, n = 74; z = 0.80).
Considering the subset of years and months for which at least 4 contiguous months of data existed, slightly stronger relationships were evident, most notably in Horsea Lake, but medusae abundance usually was more strongly correlated with preceding than contemporary environment (Fig. 4d). Considering both sites, the most consistent and strong correlation was between medusae abundance and zooplankton density 2–3 months prior.
Chrysaora melanaster, Bering Sea
In the Bering Sea, C. melanaster are patchily distributed and thus abundance varies over 1,000-fold among sites within years; mean abundance of C. melanaster varies sixfold inter-annually (Fig. 5c). Correlations among environmental characteristics, considering all times for the locations for which genetic data are available, were weak (−0.12 ≤ r ≤ 0.14; n = 55–60 for each pairwise comparison; 0.14 ≤ z ≤ 0.19). Correlations among medusae abundance and contemporaneous co-located environment characteristics also were weak (−0.15 ≤ r ≤ 0.20; n = 54 for each pairwise comparison; 0.19 ≤ z ≤ 0.31). Mean and bottom water temperatures were not significantly correlated (r = 0.05, p = 0.741, n = 54; z = 0.065) and bottom temperature was not significantly correlated with medusae abundance (r = −0.15, p = 0.267, n = 54; z = 0.19) in this dataset; results for other variables are shown in Table 1.
Discussion
Whether a perceived increase in the abundance of jellyfishes (Condon et al. 2012) is correlated with changing marine environments (Halpern et al. 2008, 2009) has been considered generally in studies taking large-scale or global perspectives (e.g., Richardson et al. 2009; Brotz et al. 2012; Condon et al. 2012, 2013; Duarte et al. 2012; Purcell 2012) and occasionally including polyphyletic datasets (e.g., Condon et al. 2013, and to a lesser extent Duarte et al. 2012). Century-long global trends across such datasets are weak to non-existent (e.g., Condon et al. 2013); statistically significant trends (or trends in which there is high confidence)—increasing and decreasing—appear to emerge in recent decades globally (Condon et al. 2013; see also Brotz et al. 2012), in a minority of subregions (Brotz et al. 2012), and in a minority of individual datasets (Condon et al. 2013). The apparent long-term trends in jellyfish abundance are superimposed on decadal periodicity (Purcell 2012; Condon et al. 2013; Decker et al. 2014) and considerable inter-annual heterogeneity. Thus, the challenge for pelagic coelenterate ecology is how to extract general relationships representing causes and effects from complex natural data when links between inferred patterns and actual population dynamics and causes will be tenuous if care is not taken to address the appropriate scales of possible drivers and demographics. For example, 24 of the 37 long-term datasets considered by Condon et al. (2013), are multi-species datasets which may be driven by a single species, by different species at different times, or by all species in concert. On the other hand, five of the 13 single-species datasets considered the same species—Pelagia noctiluca—in different regions of the Mediterranean but did not assess whether these represented one (see Stopar et al. 2010) or more independent populations, although four show no long-term trend and the fifth showed a significant increase. Thus aggregate analyses may mask considerable variation attributable to functional and habitat differences among species or populations (see also Hallett et al. 2004), while failing to recognize the extent of populations could artificially inflate sample sizes.
Indeed, our initial exploration of the population biology and population genetics of five scyphozoans indicates that large-scale analyses that aggregate across locations and/or taxa will often mask considerable population-level variation. Of our five case-studies, three reveal significant population genetic structure on spatial scales of tens-to-hundreds of kilometers (C. mosaicus, M. papua, R. octopus). Of the four case-studies that also include abiotic data, temperatures and salinities across sites often were moderately to strongly correlated, consistent with regional climatic drivers, but other environmental parameters usually were weakly correlated across sites, suggesting locally determined dynamics. Correlations between abundances of conspecific medusae in genetically distinct populations and/or different locations were always weak and as often negative as positive. Moreover, the dynamics of populations may differ between species inhabiting exactly the same location (e.g., M. papua cf. Aurelia sp. 4 in OTM, Palau; this study cf. Patris et al. 2011).
Correctly identifying populations, therefore, enables us to more precisely posit links between cause and effect. For example, relatively cool, saline, well-oxygenated waters may favor high contemporary abundances of M. papua medusae in NLK, a relationship that does not emerge strongly in other lakes in Palau (but see Martin et al. 2006). More notable, perhaps, is that even though contemporary M. papua dynamics were uncorrelated across lakes in Palau, comparisons of timelines indicated that temperature is positively correlated with medusae abundance 5 or 6 months later, following a cooling trend, in both NLK and OTM (Fig. 2d). Similarly, even though the strength and timing of environmental factors and A. aurita dynamics varied across locations in southern England, a single factor—zooplankton abundance—was strongly implicated as acting in the same manner and with similar lag across locations (Fig. 4d). Jellyfish abundance also lags oceanographic conditions, in this case in the preceding year, in the eastern Bering Sea, where population size is influenced at least in part by advection across a broad shelf (Brodeur et al. 2008; Decker et al. 2013).
Our study indicates that genetic differences between populations and/or environmental differences between sites strongly influence population dynamics in these jellyfishes. Despite regional climatic forcing of local environmental characteristics such as temperature and salinity, presumably by seasonal and inter-annual variation in insolation and precipitation (e.g., by El Niño–La Niña; Martin et al. 2006), non-linear location-specific responses in these or other environmental characteristics appear to affect ecosystem dynamics substantially. Such population-level dynamics are the most direct indicators of the proximate mechanisms via which higher-level macro-ecological or regional patterns (e.g., Decker et al. 2014) must be mediated (e.g., see Hallett et al. 2004). To understand the effects of global change, therefore, it may be necessary to study local processes at many locations (for similar sentiments, see also e.g., Falvey and Garreaud 2009; Opgenoorth and Faith 2013) over appropriate intervals and durations.
We do not suggest that strong inferences can be made about specific causes of jellyfish population dynamics or their generality from this study. We have considered only five case studies, those for which some relevant data exist (see also Kogovšek et al. 2010; Stopar et al. 2010), which represents only a tiny percentage of pelagic coelenterate species, populations and locations. Also, mostly the original projects were of short duration, employed different study designs, and sometimes were of low power (e.g., our results for OTM cf. Martin et al. 2006). Moreover, all of our inferences are correlational and we have not considered laboratory or field-based manipulations that have the potential to establish cause and effect; larger-scale and longer analyses that might clarify the relative influence of regional climate signals also were not included. Environmental conditions are well known to influence demographic processes and our ability to show statistical links between the two is influenced by the temporal and spatial scales over which both are measured (e.g., Hallett et al. 2004), the relative strengths of intrinsic and extrinsic drivers (e.g., Grenfell et al. 1998), and can depend upon sufficient data and techniques (Hinder et al. 2012).
We do suggest that this population-level approach—which identifies the natural modules that permit independent tests of cause and effect—is generalizable and in fact necessary for studying jellyfish dynamics in changing coastal systems. Each population can be studied individually, in detail, and because a population study is a natural module it can be replicated and used in more complex studies that bridge scales. Such replicates are the foundation of purposefully designed ‘natural experiments’ that can test linkages between life-history stages, environments, places and times (see Lucas and Dawson 2014) using, for example, nested hierarchical designs and meta-analyses. This approach is essential because, like species-level responses to climate change which are heterogeneous (e.g., Parmesan and Yohe 2003; Perry et al. 2005; Parmesan 2006), it is now clear that population dynamics in jellyfishes can be influenced strongly by local processes despite global trends in climate change (as is the case in other taxa: Etterson and Shaw 2001; Both et al. 2006; Lehtonen et al. 2009; Pespeni et al. 2013). The five case-studies presented here, and other phylogeographic analyses (e.g., Dawson and Jacobs 2001; Holland et al. 2004; Dawson 2005c; Govindarajan et al. 2005), indicate that population genetic differentiation within jellyfish species may be commonplace. Biogeographic and oceanographic analyses show considerable variation in coastal environments and communities (Legaard and Thomas 2006, 2007; Blanchette et al. 2008). This combination of genetic and environmental differentiation provides opportunity for local adaptation (Sanford and Kelly 2011) which may manifest as idiosyncratic population dynamics. [We note also that local adaptation can manifest in fewer differences than might be expected, due to counter-gradient variation (Dawson and Martin 2001)]. Such local variation means that large-scale analyses which aggregate across locations and/or taxa can mask direct causal links and therefore alone will be insufficient to identify the causes of long-term trends. If we wish to understand past and present changes, and to evaluate risks in the future, we need to design studies that reveal the scales of population genetic and demographic variation and to study dynamics across relevant scales (Harley et al. 2006; Franks and Hoffmann 2012) and timelines. As time-series of jellyfish populations increase in number and duration, our ability to discern causal links between population dynamics and environmental conditions over various scales will improve while also dealing with issues such as temporal auto-correlation within the time series (Hinder et al. 2012).
The generic claim that jellyfish are on the rise due to global change (e.g., Jackson 2008; Richardson et al. 2009) has long been considered oversimplistic (Mills 2001; Purcell 2005; Daryanabard and Dawson 2008; Condon et al. 2012; Gibbons and Richardson 2013; Decker et al. 2014; Lucas and Dawson 2014). Refocusing attention away from whether “the numbers of gelatinous zooplankton [are] rising in the world’s oceans” (Condon et al. 2012) and toward discovering which species are on the rise, where, and why is overdue (Gibbons and Richardson 2013; Lucas and Dawson 2014); here we emphasize the importance of discovering which populations are on the rise (or otherwise), where, and why.
How to reframe the discussion and avoid pitfalls may in part be guided by reference to related fields. For example, fisheries biology has long tackled the difficult problem of identifying marine populations (Carvalho and Hauser 1995; Stokstadt 2010; Limborg et al. 2012), epidemiology and phylodynamics explore the origins, extent, causes, and severity of outbreaks (e.g., Grenfell et al. 2004), invasion biology’s subject matter is ‘outbreaks’ and additionally provides cautionary tales about rhetoric and threshold criteria (Valéry et al. 2008, 2009; Davis et al. 2011), and biogeography is tackling integration of spatial and temporal scales (Dawson and Hortal 2012). Providing mechanistic answers requires the integration of long-term quantitative ecology—particularly population biology—and population genetics (Lucas and Dawson 2014) coupled with abiotic and biotic environmental measurements. Determining the causes of variation in the abundances of jellyfishes arguably is first the purview of ecological genetics (Urban 2008; Franks and Hoffmann 2012), which can provide information at the appropriate scales and resolution, and subsequently suited to macro-ecological meta-analyses of larger-scale and global patterns. A population perspective on global change encourages a productive balance between collection of detailed data and analyses of ‘big data’, and perhaps suggests a need for coordinated experimental networks (e.g., Adler et al. 2011) to complement sustained observational time-series (Fig. 5).
Conclusion
Scyphozoan populations span various spatial scales, from hundreds-of-meters to hundreds-of-kilometers, at least. The temporal dynamics of these populations may not be synchronized, even in closely related geographically adjacent populations. Discovering the causes of spatio-temporal variation in jellyfish abundances therefore requires studies of populations, from the smallest to largest scales. While recent aggregate analyses have attempted to investigate global patterns, they have not addressed the scales on which the causes of population dynamics occur and from which any larger patterns must emerge. The consequent gaps in knowledge, assumptions, and biases within those larger-scale analyses need to be redressed.
The key challenge is to integrate the scales of jellyfish population dynamics analytically with the scales of putative causes, from proximate events to global change. To meet this challenge, we need long-term studies of jellyfish densities and distributions replicated at the level of populations that are designed explicitly to address multiple multi-scale phenomena. Meta-analyses of population studies, which preserve the integrity of component modules as they are assembled into hierarchical datasets, present an unexplored opportunity.
References
Adler PB, Seabloom EW, Borer ET, Hillebrand H, Hautier Y et al (2011) Productivity is a poor predictor of plant species richness. Science 333:1750–1753
Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25:3389–3402
Bayha KM, Graham WM (2014) Nonindigenous marine jellyfish: invasiveness, invisibility and impacts. In: Pitt KA, Lucas CH (eds) Jellyfish blooms. Springer, Berlin, pp 45–77
Bayha KM, Dawson MN, Collins AG, Barbeitos MS, Haddock SHD (2010) Evolutionary relationships among scyphozoan jellyfish families based on complete taxon sampling and phylogenetic analyses of 18S and 28S ribosomal DNA. Integr Comp Biol 50:436–455
Blanchette CA, Miner CM, Raimondi PT, Lohse D, Heady KEK, Broitman BR (2008) Biogeographical patterns of rocky intertidal communities along the Pacific coast of North America. J Biogeogr 35:1593–1607
Both C, Bouwhuis S, Lessells CM, Visser ME (2006) Climate change and population declines in a long-distance migratory bird. Nature 441:81–83
Brodeur RD, Wilson MT, Napp JM (1997) Distribution of juvenile Pollock relative to frontal structure near the Pribilof Islands, Bering Sea. In: Forage fishes in marine ecosystems, American Fisheries Society, Lowell Wakefield Fisheries Symposium Series, no. 14, pp 573–589
Brodeur RD, Decker MB, Ciannelli L, Purcell JE, Bond NA, Stabeno PJ, Acuna E, Hunt GL Jr (2008) The rise and fall of jellyfish in the Bering Sea in relation to climate regime shifts. Prog Oceanogr 77:103–111
Brotz L, Cheung WWL, Kleisner K, Pakhomov E, Pauly D (2012) Increasing jellyfish populations: trends in large marine ecosystems. Hydrobiologia 690:3–20
Carvalho GR, Hauser L (1995) Molecular genetics and the stock concept in fisheries. In: Carvalho GR, Pitcher TJ (eds) Molecular genetics in fisheries. Springer, Berlin, pp 55–79
Clement M, Posada D, Crandall KA (2000) TCS: a computer program to estimate gene genealogies. Mol Ecol 9:1657–1660
Condon RH, Graham WM, Duarte CM, Pitt KA, Lucas CH, Haddock SHD, Sutherland KR, Robinson KL, Dawson MN, Decker MB, Mills CE, Purcell JE, Malej A, Mianzan H, Uye S-I, Gelcich S (2012) Questioning the rise of gelatinous zooplankton in the world’s oceans. Bioscience 62:160–169
Condon RH, Duarte CM, Pitt KA, Robinson KL, Lucas CH, Sutherland KR, Mianzan HW, Bogeberg M, Purcell JE, Decker MB, Uye S-I, Madin LP, Brodeur RD, Haddock SHD, Malej A, Parry GD, Eriksen E, Quiñones J, Acha M, Harvey M, Arthur JM, Graham WM (2013) Recurrent jellyfish blooms are a consequence of global oscillations. Proceedings of the National Academy of Sciences USA, vol 110, pp 1000–1005. www.pnas.org/cgi/doi/10.1073/pnas.1210920110
Daryanabard R, Dawson MN (2008) Jellyfish blooms: Crambionella orsini (Scyphozoa, Rhizostomeae) in the Gulf of Oman, Iran, 2002–2003. J Mar Biol Assoc UK 88:477–483
Davis MA, Chew MK, Hobbs RJ, Lugo AE, Ewel JJ et al (2011) Don’t judge species on their origins. Nature 474:153–154
Dawson MN (2005a) Incipient speciation of Catostylus mosaicus (Scyphozoa, Rhizostomeae, Catostylidae), comparative phylogeography and biogeography in south-east Australia. J Biogeogr 32:515–533. doi:10.1111/j.1365-2699.2004.01193.x
Dawson MN (2005b) Five new subspecies of Mastigias (Scyphozoa, Rhizostomeae, Mastigiidae) from marine lakes, Palau, Micronesia. J Mar Biol Assoc UK 85:679–694
Dawson MN (2005c) Cyanea capillata is not a cosmopolitan jellyfish: morphological and molecular evidence for C. annaskala and C. rosea (Scyphozoa: Semaeostomeae: Cyaneidae) in south-eastern Australia. Invertebr Syst 19:361–370. doi:10.1071/is03035
Dawson MN, Hamner WM (2005) Rapid evolutionary radiation of marine zooplankton in peripheral environments. Proc Natl Acad Sci USA 102:9235–9240
Dawson MN, Hortal J (2012) A cure for seeing double? Convergence and unification in biogeography and ecology. Front Biogeogr 4:3–6
Dawson MN, Jacobs DK (2001) Molecular evidence for cryptic species of Aurelia aurita (Cnidaria, Scyphozoa). Biol Bull 200:92–96. doi:10.2307/1543089
Dawson MN, Martin LE (2001) Geographic variation and ecological adaptation in Aurelia (Scyphozoa, Semaeostomeae): some implications from molecular phylogenetics. Hydrobiologia 451:259–273. doi:10.1023/a:1011869215330
Dawson MN, Raskoff KA, Jacobs DK (1998) Field preservation of marine invertebrate tissue for DNA analyses. Mol Mar Biol Biotechnol 7:145–152
Dawson MN, Gupta AS, England MH (2005) Coupled biophysical global ocean model and molecular genetic analyses identify multiple introductions of cryptogenic species. Proc Natl Acad Sci USA 102:11968–11973
Decker MB, Liu H, Ciannelli L, Ladd C, Cheng W, Chan KS (2013) Linking changes in eastern Bering Sea jellyfish populations to environmental factors via non-linear time series models. Mar Ecol Prog Ser 494:179–189
Decker MB, Cieciel K, Zavolokin A, Lauth R, Brodeur RD, Coyle KO (2014) Population fluctuations of jellyfish in the Bering Sea and their ecological role in this productive shelf ecosystem. In: Pitt K, Lucas C (eds) Jellyfish blooms. Springer, Berlin, pp 153–183
Duarte CM, Pitt KA, Lucas CH, Purcell JE, Uye S-I et al (2012) Is global ocean sprawl a cause of jellyfish blooms? Front Ecol Environ. doi:10.1890/110246
Etterson JR, Shaw RG (2001) Constraint to adaptive evolution in response to global warming. Science 294:151–154
Excoffier L, Lischer HEL (2010) Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Resour 10:564–567
Falvey M, Garreaud RD (2009) Regional cooling in a warming world: recent temperature trends in the southeast Pacific and along the west coast of subtropical South America (1979–2006). J Geophys Res 114:D04102. doi:10.1029/2008JD010519
Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol 3:294–299
Fortin MJ, Dale MRT (2005) Spatial analysis: a guide for ecologists. Cambridge University Press, Cambridge
Franks SJ, Hoffmann AA (2012) Genetics of climate change adaptation. Annu Rev Genet 46:185–208
Gibbons MJ, Richardson AJ (2013) Beyond the jellyfish joyride and global oscillations: advancing jellyfish research. J Plankton Res 35:929–938
Govindarajan AF, Halanych KM, Cunningham CW (2005) Mitochondrial evolution and phylogeography in the hydrozoan Obelia geniculata (Cnidaria). Mar Biol 146:213–222
Graham WM, Field JG, Potts DC (1992) Persistent “upwelling shadows” and their influence on zooplankton distributions. Mar Biol 114:561–570
Graham WM, Pagès F, Hamner WM (2001) A physical context for gelatinous zooplankton aggregations: a review. Hydrobiologia 451:199–212
Grenfell BT, Wilson K, Finkenstädt BF, Coulson TN, Murray S, Albon SD, Pemberton JM, Clutton-Brock TH, Crawley MJ (1998) Noise and determinism in synchronized sheep dynamics. Nature 394:674–677
Grenfell BT, Pybus OG, Gog JR, Wood JLN, Daly JM, Mumford JA, Holmes EC (2004) Unifying the epidemiological and evolutionary dynamics of pathogens. Science 303:327–332
Hallett TB, Coulson T, Pilkington JG, Clutton-Brock TH, Pemberton JM, Grenfell BT (2004) Why large-scale climate indices seem to predict ecological processes better than local weather. Nature 430:71–75
Halpern BS, Walbridge S, Selkoe KA, Kappel CV, Micheli F, D’Agrosa C, Bruno JF, Casey KS, Ebert C, Fox HE et al (2008) A global map of human impact on marine ecosystems. Science 319:948–952. doi:10.1126/science.1149345
Halpern BS, Kappel CV, Selkoe KA, Micheli F, Ebert CM et al (2009) Mapping cumulative human impacts to California Current marine ecosystems. Conserv Lett 2:1–11
Hamner WM, Hamner PP (1998) Stratified marine lakes of Palau (Western Caroline Islands). Phys Geogr 19:175–220
Hamner WM, Hauri IR (1981) Long-distance horizontal migrations of zooplankton (Scyphomedusae: Mastigias). Limnol Oceanogr 26:414–423
Harley CDG, Hughes AR, Hultgren KM, Miner BG, Sorte CJB, Thornber CS, Rodriguez LF, Tomanek L, Williams SL (2006) The impacts of climate change in coastal marine systems. Ecol Lett 9:228–241. doi:10.1111/j.1461-0248.2005.00871.x
Hinder SL, Gravenor MB, Edwards M, Ostle C, Bodger OG, Lee PLM, Walne AW, Hays GC (2012) Changes in marine dinoflagellate and diatom abundance under climate change. Nat Clim Chang 2:271–275
Holland BS, Dawson MN, Crow GL, Hofmann DK (2004) Global phylogeography of Cassiopea (Scyphozoa: Rhizostomeae): molecular evidence for cryptic species and multiple invasions of the Hawaiian Islands. Mar Biol 145:1119–1128. doi:10.1007/s00227-004-1409-4
Houghton JDR, Doyle T, Davenport J, Hays GC (2006) Developing a simple, rapid method for identifying and monitoring jellyfish aggregations from the air. Mar Ecol Prog Ser 314:159–170
Jackson JBC (2008) Ecological extinction and evolution in the brave new ocean. Proc Natl Acad Sci USA 105:11458–11465
Jeanmougin F, Thompson JD, Gouy M, Higgins DG, Gibson TJ (1998) Multiple sequence alignment with ClustalX. Trends Biochem Sci 23:403–405
Kogovšek T, Bogunovic B, Malej A (2010) Recurrence of bloom-forming scyphomedusae: wavelet analysis of a 200-year time series. Hydrobiologia 645:81–96
Lee PLM, Dawson MN, Neill SP, Robins PE, Houghton JDR, Doyle TK, Hays GC (2013) Identification of genetically and oceanographically distinct blooms of jellyfish. J R Soc Interface 10:20120920. doi:10.1098/rsif.2012.0920
Legaard KR, Thomas AC (2006) Spatial patterns in seasonal and interannual variability of chlorophyll and sea surface temperature in the California Current. J Geophys Res Oceans 111:C06032. doi:10.1029/2005JC003282
Legaard KR, Thomas AC (2007) Spatial patterns of intraseasonal variability of chlorophyll and sea surface temperature in the California Current. J Geophys Res Oceans 112:C09006. doi:10.1029/2007JC004097
Lehtonen PK, Laaksonen T, Artemyev AV, Belskii E, Both C et al (2009) Geographic patterns of genetic differentiation and plumage colour variation are different in the pied flycatcher (Ficedula hypoleuca). Mol Ecol 18:4463–4476
Limborg MT, Helyar SJ, De Bruyn M, Taylor MI, Nielsen EE, Ogden R, Carvalho GR, Consortium FPT, Bekkevold D (2012) Environmental selection on transcriptome-derived SNPs in a high gene flow marine fish, the Atlantic herring (Clupea harengus). Mol Ecol 21:3686–3703
Lucas CH (1996) Population dynamics Aurelia aurita (Scyphozoa) from an isolated brackish lake, with particular reference to sexual reproduction. J Plankton Res 18:987–1007
Lucas CH, Dawson MN (2014) What are jellyfish and thaliaceans and why do they bloom? In: Pitt KA, Lucas CH (eds) Jellyfish blooms. Springer, Berlin, pp 9–44
Lucas CH, Lawes S (1998) Sexual reproduction of the scyphomedusa Aurelia aurita in relation to variable food supply. Mar Biol 131:629–638
Lucas CH, Williams JA (1994) Population dynamics of the scyphomedusa Aurelia aurita in Southampton Water. J Plankton Res 16:879–895
Lucas CH, Hirst AG, Williams JA (1997) Plankton dynamics and Aurelia aurita production from two contrasting ecosystems: comparisons and consequences. Estuar Coast Shelf Sci 45:209–219
Lucas CH, Pitt KA, Purcell JE, Lebrato M, Condon RH (2011) What’s in a jellyfish? Proximate and elemental composition and biometric relationships for use in biogeochemical studies. Ecology 92:1704
Martin LE, Dawson MN, Bell LJ, Colin PL (2006) Marine lake ecosystem dynamics illustrate ENSO variation in the tropical western Pacific. Biol Lett 2:144–147
Mills CE (2001) Jellyfish blooms: are populations increasing globally in response to changing ocean conditions? Hydrobiologia 451:55–68
Opgenoorth L, Faith DP (2013) The intergovernmental science-policy platform on biodiversity and ecosystem services (IPBES), up and walking. Front Biogeogr 5:207–211
Parmesan C (2006) Ecological and evolutionary responses to recent climate change. Annu Rev Ecol Evol Syst 37:637–669
Parmesan C, Yohe G (2003) A globally coherent fingerprint of climate change impacts across natural systems. Nature 421:37–42
Patris SW, Dawson MN, Bell LJ, Martin LE, Colin PL, Ucharm G (2011) Ongeim’l Tketau. Coral Reef Research Foundation, Koror
Perry AL, Low PJ, Ellis JR, Reynolds JD (2005) Climate change and distribution shifts in marine fishes. Science 308:1912–1915
Pespeni MH, Sanford E, Gaylord B, Hill TM, Hosfelt JD et al (2013) Evolutionary change during experimental ocean acidification. Proc Natl Acad Sci USA 110:6937–6942
Pitt KA, Kingsford MJ (2000) Geographic separation of stocks of the edible jellyfish Catostylus mosaicus (Rhizostomeae) in New South Wales. Mar Ecol Prog Ser 196:143–155
Pitt KA, Kingsford MJ (2003) Temporal and spatial variation in recruitment and growth of medusae of the jellyfish, Catostylus mosaicus (Scyphozoa: Rhizostomeae). Mar Freshw Res 54:117–125
Purcell JE (2005) Climate effects on formation of jellyfish and ctenophore blooms: a review. J Mar Biol Assoc UK 85:461–476
Purcell JE (2012) Jellyfish and ctenophore blooms coincide with human proliferations and environmental perturbations. Annu Rev Mar Sci 4:209–235
Purcell JE, Brown ED, Stokesbury KDE, Haldorson LH, Shirley TC (2000) Aggregations of the jellyfish Aurelia labiata: abundance, distribution, association with age-0 walleye pollock, and behaviors promoting aggregation in Prince William Sound, Alaska, U.S.A. Mar Ecol Prog Ser 195:145–158
Raskoff KA (2001) The impact of El Niño events on populations of mesopelagic hydromedusae. Hydrobiologia 451:121–129
Rice WR (1989) Analyzing tables of statistical tests. Evolution 43:223–225
Richardson AJ, Bakun A, Hays GC, Gibbons MJ (2009) The jellyfish joyride: causes, consequences and management responses to a more gelatinous future. Trends Ecol Evol 24:312–322
Sanford E, Kelly MW (2011) Local adaptation in marine invertebrates. Annu Rev Mar Sci 3:509–535
StatSoft Inc (2006) Electronic statistics textbook. StatSoft, Tulsa
Stokstadt E (2010) To fight illegal fishing, forensic DNA gets local. Science 330:1468–1469
Stopar K, Ramšak A, Trontelj P, Malej A (2010) Lack of genetic structure in the jellyfish Pelagia noctiluca (Cnidaria: Scyphozoa: Semaeostomeae) across European seas. Mol Phylogenet Evol 57:417–428. doi:10.1016/j.ympev.2010.07.004
Suchman CL, Brodeur RD (2005) Abundance and distribution of large medusa in surface waters of the northern California Current. Deep-Sea Res II 52:51–72
Urban MC (2008) Ecological genetics. In: Encyclopedia of life sciences (ELS). Wiley, Chichester. doi:10.1002/9780470015902.a0021214
Valéry L, Fritz H, Lefeuvre J-L, Simberloff D (2008) In search of a real definition of the biological invasion phenomenon itself. Biol Invasions 19:1345–1351
Valéry L, Fritz H, Lefeuvre J-L, Simberloff D (2009) Invasive species can also be native …. Trends Ecol Evol 24:585
Youngbluth MJ, Båmstedt U (2001) Distribution, abundance, behavior and metabolism of Periphylla periphylla, a mesopelagic coronate medusa in a Norwegian fjord. Hydrobiologia 451:321–333
Acknowledgments
We thank Keith M. Bayha and Sarah Abboud who sequenced Aurelia aurita and Chrysaora melanaster, and Coral Reef Research Foundation who conducted the surveys of Mastigias papua and collected matching environmental data in Palau. We also thank the scientific staff from the BASIS project and the fishing crews of the F/V Sea Storm and F/V Northwest Explorer F/V Epic Explorer, R/V Oscar Dyson, and F/V Bristol Explorer for their considerable efforts and technical assistance in all aspects of the field surveys, without whose help the Bering Sea work would have been impossible. Two anonymous reviewers helpfully critiqued an earlier version of the paper allowing us to refine its message. This research was supported in part by National Science Foundation Grant no. DEB-07-17071 and presented as a work-in-progress at the MOLTOOLS workshop held in Lecce in September 2012; support for participation by MND in the MOLTOOLS workshop was provided by the European Community’s Seventh Framework Programme (FP7/2011-2015) for the project Vectors of Change in Oceans and Seas Marine Life, Impact on Economic Sectors (VECTORS). Use of trade names does not imply endorsement by the National Marine Fisheries service, NOAA.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dawson, M.N., Cieciel, K., Decker, M.B. et al. Population-level perspectives on global change: genetic and demographic analyses indicate various scales, timing, and causes of scyphozoan jellyfish blooms. Biol Invasions 17, 851–867 (2015). https://doi.org/10.1007/s10530-014-0732-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10530-014-0732-z