nena masthead
NENA Home Staff & Editors For Readers For Authors

Seasonal Abundance of the Ctenophore Mnemiopsis leidyi in Relation to Water Temperature and Other Zooplankton in the Thames River Estuary, Connecticut
Lucy S. Vlietstra

Northeastern Naturalist, Volume 21, Issue 3 (2014): 397–418

Full-text pdf (Accessible only to subscribers. To subscribe click here.)

 

Access Journal Content

Open access browsing of table of contents and abstract pages. Full text pdfs available for download for subscribers.



Current Issue: Vol. 30 (3)
NENA 30(3)

Check out NENA's latest Monograph:

Monograph 22
NENA monograph 22

All Regular Issues

Monographs

Special Issues

 

submit

 

subscribe

 

JSTOR logoClarivate logoWeb of science logoBioOne logo EbscoHOST logoProQuest logo

Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 397 2014 NORTHEASTERN NATURALIST 21(3):397–418 Seasonal Abundance of the Ctenophore Mnemiopsis leidyi in Relation to Water Temperature and Other Zooplankton in the Thames River Estuary, Connecticut Lucy S. Vlietstra* Abstract - I examined seasonal patterns in the abundance of the ctenophore Mnemiopsis leidyi (Atlantic Comb Jelly) to determine whether unusually early ctenophore blooms occur outside the range in which they were originally reported (Narragansett Bay, RI) and whether ctenophore abundance is correlated with sea surface temperature (SST) or with declines in zooplankton prey. Sampling was conducted from 23 April to 16 November 2004 in the Thames River Estuary, CT. Adult (≥1 cm) ctenophores first appeared in the estuary on 20 May 2004, with a seasonal peak in mean (± SD) density (9.9 ± 2.5 individuals m-3, n = 7 stations) and biovolume (26.6 ± 9.9 ml m-3, n = 7 stations) on 9 July 2004. Adult Atlantic Comb Jelly abundance was positively correlated with SST, but only during the spring and early summer. Food availability also likely played an important role in ctenophore-population growth early in the year. However, I observed a delay between copepod density and ctenophore biovolume, possibly due to unusually cold winter temperatures that reduced ctenophore survivorship in the months preceding the study. Ctenophore diets in 2004 consisted primarily of copepods. I observed an inverse correlation between ctenophore biovolume and copepod density, but my data suggest that adult Atlantic Comb Jellies consumed only 0.0–2.2% of the copepod standing stock per day. Overall, my results are consistent with the hypothesis that early summer ctenophore blooms occured over a relatively large spatial scale; however, trophic impacts in the Thames River Estuary in 2004 appeared to be relatively low. Introduction Mnemiopsis leidyi Agassiz (Atlantic Comb Jelly) is a seasonally abundant lobate ctenophore native to estuaries and shallow marine habitats off the Atlantic coast of North America, ranging from New England to the Florida Keys and into the Gulf of Mexico (Purcell et al. 2001). Seasonal patterns in ctenophore abundance vary with latitude, with annual maxima (blooms) in the northern part of the range occurring during summer, historically during August–October (Deason 1982). However, since the mid-1980s, Atlantic Comb Jelly blooms have been occurring earlier in the year. The results of studies conducted in Narragansett Bay, RI, for example, show a consistent shift in the timing of ctenophore blooms from late summer–early fall to June and early July, especially in shallow inshore waters (Beaulieu et al. 2013, Costello et al. 2006, Sullivan et al. 2001). Such a shift in the timing of Atlantic Comb Jelly blooms could have serious implications for estuarine food webs because ctenophores are voracious *Department of Science, US Coast Guard Academy, 15 Mohegan Avenue, New London, CT 06320; Lucy.S.Vlietstra@uscga.edu. Manuscript Editor: Tom Trott Northeastern Naturalist 398 L.S. Vlietstra 2014 Vol. 21, No. 3 consumers of copepods, fish eggs, and possibly fish larvae (Cowan and Houde 1993, Kremer 1979, Purcell et al. 1994). Because fish production in New England usually peaks during spring and early summer, high abundances of ctenophores in early summer could reduce fish populations through predation on icthyoplankton or competition with larval fishes (Purcell et al. 2001, Sullivan et al. 2001). Evidence of ctenophores exerting top-down control in local food webs has been documented in Narragansett Bay, RI, where high Atlantic Comb Jelly abundance was followed by low copepod densities and high values of chlorophyll-a (Costello et. al 2006, Deason and Smayda 1982). One hypothesis for the cause of early ctenophore blooms involves warm spring temperatures associated with positive phases of the North Atlantic Oscillation (NAO). Hawk (1998) and Sullivan et al. (2001) described a warming trend in waters off Rhode Island over the last several decades, during which time the NAO shifted from a negative phase to a positive one. During positive phases, currents bring relatively warm water to the northern Atlantic Coast of North America (Ottersen et al. 2001). This situation could account for a recent shift in the timing of seasonal ctenophore blooms because water temperature appears to be an important factor limiting the onset of ctenophore blooms in New England and elsewhere along the Atlantic Coast (Purcell et al. 2001, Sullivan et al. 2001). If early summer Atlantic Comb Jelly blooms are associated with a large-scale atmospheric–oceanographic coupling mechanism, such as the NAO, early blooms would be expected in other estuaries in the northern native range. I conducted this study to determine whether Atlantic Comb Jelly blooms also occured during early summer in a small estuary located approximately 70 km west of Narragansett Bay, on the northern shore of Long Island Sound. In addition, I sought to define the relationship, if any exists, between sea surface temperature (SST) and Atlantic Comb Jelly abundance in the estuary as well as the potential impact of the ctenophore bloom on the abundance of local prey populations, specifically copepods, fish eggs, and fish larvae. Field-Site Description I conducted this study from 23 April to 16 November 2004 in the Thames River Estuary, CT (41°21'14.8''N, 72°6'3.6''W; Fig. 1). The Thames River headwaters are located at the junction of the Yantic and Shetucket Rivers in Norwich, CT, approximately 25 km upstream of Long Island Sound. I sampled in the lower reaches of the estuary, where the Thames River is up to 15 m deep and 0.5–1.5 km wide and covers an area of approximately 10.8 km2. Tides are semi-diurnal, and the water column has a well-defined salt-wedge structure for most of the year. Surface and bottom salinity typically range from 5 to 20 ppt (parts per thousand) and from 25 to 33 ppt, respectively, with the pycnocline at 2–4 m depth. The Thames River Estuary supports a wide variety of fishes, including Centropristis striata L. (Black Sea Bass), Morone saxatilis (Walbaum) (Striped Sea Bass), Stenotomus chrysops (L.) (Scup), Anchoa mitchilli (Valenciennes) (Bay Anchovies), Peprilus triacanthus Peck (Butterfish), and several flounders, including both Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 399 Pseudopleuronectes americanus (Walbaum) (Winter Flounder) and Paralichthyes dentatus (L.) (Summer Flounder) (K. Mrakovcich, US Coast Guard Academy, New London, CT, unpubl. data). Methods I measured ctenophore abundance in the water column approximately every 2 weeks, with replicate (n = 2) vertical-net tows conducted at 7 fixed stations (stations 0–6; Fig. 1). Sampling was omitted at some stations in the first 2 weeks of the study while the protocol was established and occasionally thereafter when maritime traffic restricted access to portions of the estuary. Tows were made with a 63-mm-mesh plankton net lowered to a depth of 6 m (after Sullivan et al. 2001). Immediately following each tow, I recorded the number of Atlantic Comb Jellies caught, and measured the oral-aboral diameter to the nearest 0.1 cm. When I collected too many ctenophores for individual measurement, I visually assessed the Figure 1. Map of Thames River Estuary showing the location of sampling stations 0–6 (●). USCGA = campus of US Coast Guard Academy. Northeastern Naturalist 400 L.S. Vlietstra 2014 Vol. 21, No. 3 number in each size class (estimated to the nearest 0.1 cm). I gently transferred a subsample of individuals representing the range of sizes observed to a separate container for biovolume measurements in the laboratory. I measured the displacement volume of individual ctenophores and developed an equation for the length–volume relationship for the purpose of calculating biovolume of the total population. This relationship was y = 0.0011 x2.57, where x was oral-aboral length (r2 = 0.90, n = 79). I characterized ctenophore abundance in terms of density (individuals m-3) and biovolume (ml m-3), and calculated density from the total number of individuals collected in each tow divided by the volume of water (m3) sampled by the tow. The volume (V) was calculated with the equation, V = π r2 * z, where r was the radius of the net, and z was the distance of the tow. Biovolume was calculated from the total volume of ctenophore biomass (ml) in each tow divided by V. I averaged density and biovolume across replicate tows at each station. To derive abundance estimates for the study site as a whole, I averaged mean ctenophore density and biovolume station-1 across sampling stations. This yielded a grand mean (± standard error [SE]) ctenophore abundance at the study site on each sampling date, with standard error representing variance among sampling-station means. Immediately after collecting samples, I inspected them to detect ctenophores with intact gut contents. I removed these individuals from the sample and stored them in a separate container for a period of 0.5–1.5 h until I could get them to the laboratory, where I inspected prey items inside the gut with a dissecting microscope and identified each item to the lowest possible taxonomic level. Because some prey species were likely digested or expelled during transport (Granhag et al. 2011), these measures were not intended for quantitative analysis but to qualitatively identify important prey species. I ranked each species by its frequency of occurrence (FO) in gut samples to determine relative dietary importance. After sorting Atlantic Comb Jellies in the field, I preserved the remaining plankton sample in 80% ethyl alcohol. Beginning on 7 May 2004, I used a Folsom splitter to subsample preserved samples from Stations 0, 2, 3, 5, and 6, and transferred them to a counting wheel. I identified every member of the mesoplankton (0.2–2.0 mm) and macroplankton (2–20 mm) in the subsample, including holoplankton and meroplankton, to the lowest possible taxonomic level. The total number of individuals of each taxon was back-calculated from the number of splits used to derive the subsample. I determined seasonal patterns in non-gelatinous zooplankton density (individuals m-3) by dividing the number of individuals of each species in whole samples by V and averaging replicate tows per station. Those values were averaged across all stations to obtain a grand mean (± SE) zooplankton abundance for the study site as a whole, with standard error representing variance among samplingstation means. Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 401 I characterized impacts of Atlantic Comb Jelly consumption on lower trophic levels by the correlation, if any existed, between ctenophore biovolume and the ambient density of the single species most frequently observed in ctenophore diets, which was copepods (see Results). I used ctenophore biovolume for this analysis because prey consumption has been shown to vary with ctenophore length (Granhag et al. 2011), making biovolume a better indicator of consumption potential than ctenophore density. I performed correlation analysis on raw values of ctenophore biovolume (larvae and adults combined) and copepod density collected in the same tow. Because neither parameter fit a normal distribution with equal variance even after log-transformation was applied, the correlation was examined with a Spearman rank test. I further characterized trophic impacts by estimating the percentage of copepod standing stock consumed by Atlantic Comb Jellies on a daily basis using the relationship between copepod density and copepod-consumption rate by adult comb jellies in air-saturated water as defined by Decker et al. (2004): y = 0.05x - 0.07, where x is ambient copepod density (mean number of copepods L-1) and y is the number of copepods consumed per unit time (number of copepods h-1 ml ctenophore- 1). Values of y were multiplied by adult ctenophore biovolume, divided by mean copepod density, and multiplied by 100 to derive the % copepod standing stock consumed by ctenophores at each station per day. I made calculations for each station sampled on a given day and averaged these values to estimate the mean percent (± SE) copepod standing stock consumed in the estuary as a whole. Because copepods likely have a patchy distribution in the water column, individual ctenophores may encounter densities much different from mean values used for x (i.e., mean ambient copepod density). However, because consumption rates were calculated over a relatively large spatial scale, x was a useful estimate of ctenophore- prey availability. Prey consumption by larval ctenophores, which specialize on microphytoplankton, microzooplankton, and copepod nauplii (Sullivan and Gifford 2004, Waggett and Sullivan 2006), were not included in the analysis for the purpose of comparison with other studies. I acquired surface and mid-water temperature (°C) and salinity (ppt) for the period 23 April–November 2004 from a mooring station maintained by the University of Connecticut (http://sounddata.uconn.edu/) adjacent to station 3 (Fig. 1). Surface temperatures were measured at 0.3 m depth; mid-water temperatures were measured at 4.4 m depth. Following Sullivan et al. (2001), I derived daily water temperatures by averaging hourly data. Water temperature and salinity were not available from the mooring on the final sampling date, 16 November 2004, so I measured hydrographic parameters at midday with a hand-held YSI 6500 sonde. Correlations between ctenophore abundance (both density and biovolume) at station 3 and SST were examined with a Spearman rank test. In addition, water column stability (E) was calculated according to Knauss (1997): E = α (Ts - Tb) / D + β (Ss - Sb) / D, where α is 1.5; β is -7.6; Ts is surface temperature, Tb is mid-water temperature, Ss Northeastern Naturalist 402 L.S. Vlietstra 2014 Vol. 21, No. 3 is surface salinity, Sb is mid-water salinity, and D is depth. Following Costello et al. (2006), the water column was considered mixed when E > 2 x 10-4 m-1. Because winter SST may influence overwintering survivorship in ctenophores, and therefore population growth in the spring (Costello et al. 2006, Shiganova et al. 2001), I evaluated winter conditions prior to the study to determine whether they were unusually warm or cold. Data for this comparison were obtained from the National Oceanic and Atmospheric Administration (NOAA) weather station located on the east bank of the Thames River, adjacent to station 2 (station #461490, 41°21'6.0''N, 72°5'3.8''W, http://tidesandcurrents.noaa.gov). To characterize daily SST leading up to the study, I averaged hourly data every 24 h from 1 November 2003 to 31 March 2004. I also used this method to calculate daily mean SST during 1 November–31 March in each of the 8 previous winters (1995–1996 to 2002– 2003). Long-term (1995–2003) means ± 95% CI were plotted on the same graph as daily SST in winter 2003–2004 for visual comparison. Results I sampled in the Thames River Estuary every 10 ± 5 days (mean ± SD) for a total of 21 days during the study. On most days, I made replicate tows (n = 2) at each of the 7 stations; exceptions are noted in Table 1. I first observed Atlantic Comb Jelly larvae (<1 cm) and adults (≥1 cm) in the estuary on 12 May and 20 May 2004, respectively. Adult densities remained relatively low (0.0–0.2 individuals m-3) until 16 June 2004, when abundance began to rise sharply, peaking on 9 July 2004 (9.9 ± 1.0 ind m-3, n = 7 stations). I observed a smaller peak (8.3 ± 2.0 individuals m-3, n = 7 stations) on 5 September 2004 (Fig. 2a). Adult biovolume also showed a bimodal pattern, with a maximum value (26.6 ± 3.7 ml m-3, n = 7 stations) observed on 9 July 2004 and a smaller peak (17.2 ± 4.7 ml m-3, n = 7 stations) on 5 September 2004 (Fig. 2b). Seasonal patterns in adult ctenophore abundance were generally consistent among sampling stations, with two peaks evident at each station during the study (Figs. 3d–f, 4d–f). Larvae were generally 2–3 times more abundant than adults in the estuary, with peak densities measured on 28 October 2004 (77.3 ± 33.4 ind m-3, n = 7 stations, Fig. 2a). On that date, ctenophore larvae were 16 times more abundant than adults, with the highest densities at downstream stations (stations 0–2; Fig. 3a). Despite their high densities, larvae usually accounted for only 5–25% of total Atlantic Comb Jelly biovolume, with the greatest larval biovolume (2.2 ± 0.4 ml m-3, n = 7 stations) observed on 30 June 2004 (Fig. 2b). Ctenophore diet I examined gut contents from 21 Atlantic Comb Jellies collected 20 May–16 July 2004; individuals were 0.4–5.7 cm in length. I found copepods (all life stages) in 87% of the samples (FO = 0.87; Table 1), making them the most widespread prey item found in ctenophore diets. Most adult copepods were members of Order Calanoida, mainly Acartia (Acanthacartia) tonsa Dana (FO = 0.74), but Centropages spp. (FO = 0.09) and Temora longicornis (O.F. Müller) (FO = 0.04) were also Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 403 Table 1. Mean ± SE abundance of adult and larval Atlantic Comb Jellies and copepods (all age classes) in the Thames River Estuary, 2004. Sample size (n) is number of stations sampled day-1. Sampling was not conducted at stations 0, 5, and 6 on April 23; stations 0, 1, 5, and 6 on 7 May; stations 0 and 6 on 12 May; station 2 on 22 July; and station 0 on 16 October. We conducted only 1 tow at stations 1–4 on 23 April, stations 2–4 on 7 May, station 1 on 16 October, and station 2 on 28 October and 16 November. Copepods were not sampled on 16 October. Larvae (<1 cm) Adults (≥1 cm) Copepods Date n Density (individuals m-3) Biovolume (ml m-3) Density (individuals m-3) Biovolume (ml m-3) n Density (individuals m-3) 23 April 4 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0 − 7 May 3 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 2 1335 ± 97 12 May 5 0.2 ± 0.2 <0.1 ± <0.1 0.0 ± 0.0 0.0 ± 0.0 3 8576 ± 5322 20 May 7 0.2 ± 0.5 <0.1 ± <0.1 0.2 ± 0.2 0.7 ± 0.7 5 20,386 ± 5300 26 May 7 0.2 ± 0.2 <0.1 ± <0.1 0.0 ± 0.0 0.0 ± 0.0 5 12,478 ± 2399 3 Jun 7 0.3 ± 0.3 <0.1 ± <0.1 0.2 ± 0.2 0.2 ± 0.2 5 14,037 ± 1654 10 Jun 7 0.2 ± 0.2 <0.1 ± <0.1 0.0 ± 0.0 0.0 ± 0.0 5 9374 ± 1245 16 Jun 7 2.5 ± 1.3 0.2 ± 0.1 1.2 ± 0.5 6.3 ± 5.0 5 12,624 ± 1793 24 Jun 7 5.7 ± 1.6 0.5 ± 0.2 2.0 ± 0.8 7.0 ± 6.1 5 17,080 ± 5388 30 Jun 7 24.9 ± 3.3 2.2 ± 0.4 7.6 ± 1.3 9.5 ± 2.5 5 11,497 ± 1030 9 Jul 7 21.1 ± 2.4 2.2 ± 0.5 9.9 ± 1.0 26.6 ± 3.7 5 10,433 ± 3598 16 Jul 7 16.0 ± 4.2 1.7 ± 0.4 6.6 ± 1.7 9.4 ± 3.1 5 9791 ± 1374 22 Jul 6 6.1 ± 2.5 0.8 ± 0.3 5.5 ± 2.2 10.5 ± 5.4 4 7742 ± 1409 6 Aug 7 1.9 ± 0.9 0.2 ± 0.1 4.2 ± 1.2 12.2 ± 4.0 5 12,090 ± 3264 17 Aug 7 5.9 ± 1.2 0.4 ± 0.1 4.6 ± 0.9 17.0 ± 7.2 5 3842 ± 1056 28 Aug 7 11.1 ± 2.1 1.4 ± 0.4 4.0 ± 1.6 13.2 ± 7.1 5 5616 ± 1397 5 Sep 7 22.8 ± 8.8 1.6 ± 0.7 8.3 ± 2.0 17.2 ± 4.8 5 6480 ± 1365 25 Sep 7 43.2 ± 16.1 1.9 ± 0.2 5.5 ± 1.9 6.8 ± 3.0 5 8428 ± 690 16 Oct 6 8.6 ± 2.5 1.2 ± 0.5 3.4 ± 0.9 4.7 ± 2.9 0 − 28 Oct 7 77.3 ± 33.4 1.8 ± 0.4 4.9 ± 1.6 4.9 ± 1.9 5 2084 ± 411 16 Nov 7 34.7 ± 15.3 0.4 ± 0.1 1.2 ± 0.4 5.6 ± 2.5 5 1706 ± 341 Northeastern Naturalist 404 L.S. Vlietstra 2014 Vol. 21, No. 3 common. Copepod prey also included adult harpacticoid copepods (FO = 0.13) and unidentified copepidites and nauplii (FO = 0.52). Other common prey items were: barnacle nauplii (FO = 0.70), bivalve veligers (FO = 0.48), and cladocerans (FO = 0.48) (Table 2). Less common were mysid shrimp, crab zoea, and tunicate larvae (Table 2). I observed neither fish eggs nor fish larvae in gut samp les. Non-gelatinous zooplankton Non-gelatinous mesozooplankton and macrozooplankton in the water column consisted mainly of crustaceans (Fig. 5), which represented 81% of all Figure 2. Mean (± SE) (a) density and (b) biovolume of adult (≥1 cm) and larval (<1 cm) Atlantic Comb Jellies in the Thames River Estuary, 23 April-16 November 2004. Standard error bars represent variation among mean ctenophore abundance at each sampling station. Sample sizes are shown in Table 1. Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 405 non-gelatinous zooplankton collected in nets. The most abundant crustaceans were copepods and barnacle larvae (15% and 81% of crustaceans, respectively). Molluscs (12% overall) were also relatively abundant, specifically bivalve veligers (76% of molluscs), as were polychaete larvae (6% overall). Icthyoplankton represented 0.1% (by number) of all zooplankton counted. Fish eggs were collected sporadically during 7 May–28 August, with a maximum density of 37 ± 29 eggs m-3 Figure 3. Mean (± SE) density of larval (<1 cm) Atlantic Comb Jellies collected in replicate (n = 2) tows conducted at (a) stations 0-2, (b) stations 3 and 4, and (c) stations 5 and 6. Mean (± SE) density of adult (≥1 cm) Atlantic Comb Jellies collected in replicate tows at (d) stations 0–2; (e) stations 3 and 4, and (f) stations 5 and 6. Note the different y-axis scales used for larvae and adults. Northeastern Naturalist 406 L.S. Vlietstra 2014 Vol. 21, No. 3 (n = 5 stations) on 3 June. I observed fish larvae in samples collected 26 May–30 June, with a maximum density of 15 ± 10 individuals m-3 (n = 5 stations) on 24 June. Densities of non-gelatinous zooplankton (all species) were greatest during 20 May–30 June 2004 (Fig. 5a). Copepods, specifically, were most abundant on 20 May 2004 (20,384 ± 5300 individuals m-3, n = 5 stations), after which time they declined to 12,478 ± 2399 individuals m-3 (n = 5 stations, Fig. 6). Copepod densities Figure 4. Mean (± SE) biovolume of larval (<1 cm) Atlantic Comb Jellies collected in replicate (n = 2) tows conducted at (a) stations 0–2, (b) stations 3 and 4, and (c) stations 5 and 6. Mean (± SE) biovolume of adult (≥1 cm) Atlantic Comb Jellies collected in replicate tows at (d) stations 0-2, (e) stations 3 and 4, and (f) stations 5 and 6. Note the different y-axis scales used for larvae and adults. Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 407 remained relatively stable until 17 August 2004, when they declined to 3842 ± 1046 individuals m-3 (n = 5 stations). On 25 September 2004, densities increased to 8428 ± 690 individuals m-3 (n = 5 stations) before decreasing again in late October (Fig. 6). Overall, I detected a negative correlation between Atlantic Comb Jelly biovolume (all sizes) and copepod density measured in the same tow (rs = -0.31, df = 145, P < 0.001). I estimated that, on average, adult Atlantic Comb Jellies in the Thames River Estuary consumed 0.0–2.8% of the standing stock of copepods day-1 (n = 19 days; Fig. 7). Prey-consumption rates were highest during the ctenophore bloom on 9 July 2004, when adult Atlantic Comb Jellies consumed between 2.8 ± 0.5% (n = 5 stations) of the copepod standing stock (Fig. 7), with the greatest portion (4.1%) consumed at station 5. Correlation with water temperature During the study, SST at station 3 was 9.7–22.8 °C and surface salinity was 6–24 ppt (Fig. 8a, b). Mid-water temperature and salinity ranged from 7 °C to 21 °C and from 28 ppt to 30 ppt, respectively (Fig. 8a, b). Water column stability (E) exceeded the threshold for stratification for most of the study, which is consistent with the salt-wedge structure of the water column. Exceptions occurred on 18 and 29 September 2004, when E < 2 x 10-4 m-1, which indicated that the water column had become mixed to a depth of at least 4.4 m (Fig. 8c). Table 2. Frequency of occurrence (FO) of prey items found in the guts of Atlantic Comb Jellies (n = 21) collected in the Thames River Estuary, May–June 2004. Specimen FO Phylum Annelida Polycheate larvae (spinoid) 0.35 Phylum Mollusca Gastropod larvae 0.35 Bivalve larvae (veliger) 0.48 Phylum Arthropoda Copepoda 0.87 Calanoida (adult) 0.74 Acartia spp. 0.74 Centropages spp. 0.09 Temoral longicornis 0.04 Harpacticoida 0.13 Unid. copepidites and nauplii 0.52 Cladocera 0.48 Podon spp. 0.48 Evadne spp. 0.09 Mysida 0.13 Decapoda (crab zoea) 0.17 Cirripedia (barnacle nauplii and cypris) 0.70 Phylum Chordata Actinula larvae 0.09 Northeastern Naturalist 408 L.S. Vlietstra 2014 Vol. 21, No. 3 At station 3, both density and biovolume of adult Atlantic Comb Jellies were positively correlated with SST (density: rs = 0.36, df = 38, P = 0.021; biovolume: rs = 0.40, df = 38, P = 0.010). However, the correlation was not always strong; SST increased from 14 °C to 17 °C during April-early May, but Atlantic Comb Jelly density remained equal or close to 0.0 individuals m-3 until mid-June. In addition, ctenophore density declined when SSTs reached their highest values in mid-summer (16 Jul–17 Aug 2004). Because factors controlling population growth may differ from factors controlling population decline (e.g., temperature vs. food), ctenophore abundance and SST measured prior to the 9 July bloom (23 April–9 July Figure 5. Mean (± SE) density of all non-gelatinous meso- and macrozooplankton collected in vertical-net tows, 7 May–16 November 2004. Standard error bars represent variance among sampling stations. Note differences in scale between (a) and (b). Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 409 Figure 6. Mean (± SE) copepod density and mean (± SE) Atlantic Comb Jelly biovolume (all sizes) among sampling stations in the Thames River Estuary, 2004. Sample sizes are given in Table 1. Figure 7. Mean (± SE) % of copepod (all life stages) standing stock consumed by Atlantic Comb Jellies in the Thames River Estuary, 7 May–16 November 2004. Zeros represent days when adult ctenophores were not collected in net tows. Sample sizes are provided in Table 1. Northeastern Naturalist 410 L.S. Vlietstra 2014 Vol. 21, No. 3 Figure 8. Daily (a) sea surface temperature (°C), (b) salinity (ppt), and (c) water column stability index (E) in the Thames River Estuary, 23 April–16 November 2004. Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 411 2004) were analyzed (post hoc) separately from those measured after the bloom (16 July–16 November 2004). Correlations were still positive for the periods before the bloom (density: rs = 0.59, df = 18, P = 0.006; biovolume: rs = 0.56, df = 18, P = 0.010) but not afterwards (density: rs = -0.10, df = 18, P = 0.677; biovolume: rs = 0.003, df = 18, P = 0.990) (Fig 9). Likewise, larval ctenophore abundance at station 3 was not correlated with SST during the study (density: rs = 0.18, df = 38, P = 0.267; biovolume: rs = 0.25, df = 38, P = 0.125) unless I analyzed values prior to 9 July separately. In that case, larval abundance was positively correlated with SST (density: rs = 0.57, df = 18, P = 0.009; biovolume: rs = 0.60, df = 18, P = 0.005; Fig. 9). Larval abundance was not correlated with SST after the bloom (density: rs = -0.25, df = 18, P = 0.285; biovolume: rs = -0.15, df = 18, P = 0.516; Fig. 9). Figure 9. Sea surface temperature (SST) at station 3 and the (a) density and (b) biovolume of adult Atlantic Comb Jellies at station 3 on and before 9 July 2004 (closed circles) and after 9 July 2004 (open circles). SST at station 3 and the (c) density and (d) biovolume of larval Atlantic Comb Jellies at station 3 on and before 9 July 2004 (closed circles) and after 9 July 2004 (open circles). Trend lines show significant (P < 0.05) correlations. Northeastern Naturalist 412 L.S. Vlietstra 2014 Vol. 21, No. 3 During the winter months preceding the study, November 2003–March 2004, SST in the Thames River Estuary ranged from 0.9 °C to 14.4 °C (Fig. 10). Overall, temperatures were cool relative to average winter temperatures during 1995– 2003. The December-February period was especially cold; daily SST was up to 3.9 °C below the long-term mean. Occasionally, SST exceeded average winter temperatures for periods lasting a few days, but below-average temperatures were more common. Discussion The seasonal patterns I observed in adult Atlantic Comb Jelly abundance in the Thames River Estuary are consistent with the hypothesis that early summer ctenophore blooms occur outside the range in which they were originally reported in the northeast. Both their date of first appearance (20 May 2004) and peak abundance (9 Jul 2004) coincide with early blooms previously described in Narragansett Bay, RI (Beaulieu et al. 2013; Sullivan et al. 2001, 2008). Similar phenological shifts in the northern part of the range have since been described by McNamara et al. (2010), who found that Atlantic Comb Jelly blooms in two Long Island estuaries occurred 2–3 months earlier in 2006 than blooms in the 1970s and 1980s. In addition, Condon and Steinberg (2008) found that blooms in the York River Estuary, Chesapeake Bay, VA occurred 1 month earlier than they did in the late 1960s. Without historical records from the Thames River Estuary, I could not determine whether the early Figure 10. Mean daily SST in the Thames River Estuary, November-March from 1995–96 to 2002–03 (solid line, n = 8) and in 2003–04 (white circles). Dotted lines represent the 95% confidence interval around the long-term (1995–2003) mean. Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 413 July bloom I observed represents a temporal shift compared to past decades, but the pattern supports the idea that mechanisms promoting early summer ctenophore blooms operate on a relatively large spatial scale. In general, factors driving the onset of ctenophore blooms are not fully understood, but the results of this study and others suggest that water temperature plays an important role. Water temperature appears to influence overwintering survivorship in ctenophores as well as metabolic rates and egg production (Condon and Steinberg 2008; Costello et al. 2006; Kremer 1979, 1994). Overall, both adult and larval Atlantic Comb Jelly abundances in the Thames River Estuary were positively correlated with SST during periods of population growth in early summer, before the first adult bloom, suggesting that rising water temperature stimulates reproduction. However, the relationship between temperature and ctenophore abundance was not always linear. For example, when adult Atlantic Comb Jellies first appeared in the estuary on 20 May 2004, both adults and larvae were scarce until 16 June 2004 despite an 8 °C (from 14 to 22 °C) increase in SST during that time. Because the minimum temperature threshold for ctenophore egg production in native habitats is 10 °C (Costello et al. 2006) and larvae hatch 20–24 h after eggs are produced (Purcell et al. 2001), the lack of population growth in the spring probably did not reflect unfavorable temperatures for reproduction. Laboratory studies have shown that egg production in Atlantic Comb Jelly is highly sensitive to prey density, and adults stop producing eggs after only a few days without food (Purcell et al. 2001, Reeve et al. 1989). In the Thames River Estuary, the main food of Atlantic Comb Jellies was copepods and other crustaceans, which were not locally abundant until 20 May 2004. This date occurred 3 weeks after SSTs exceeded the minimum threshold for egg production and coincided with the first appearance of adult ctenophores in the estuary. This observation is important because it suggests that both SST and food availability must be adequate before habitats become suitable for ctenophores in the spring. However, if water temperature and food availability were the only factors limiting spring population growth in ctenophores, one would expect a larger increase in ctenophores following the peak in copepod abundance. Instead, ctenophores remained scarce for the next 4 weeks. Such a delay may be due, in part, to the time required for foraging adult ctenophores to produce eggs, and for eggs to mature into larvae and eventually into adults. However, that idea is not consistent with the fact that adult Atlantic Comb Jellies typically experience high fecundity, and larvae reach sexual maturity at a relatively young age (14 days; Reeve and Walter 1978 reviewed by Costello et al. 2012). Alternatively, delayed ctenophore production in the spring of 2004 may be related to the unusually low surface temperatures observed in the winter months preceding the study. Shiganova et al. (2001) found that Atlantic Comb Jellies in the Black Sea were smaller and less abundant in spring months following cold winters. In addition, individuals were practically absent from inshore waters, where winter temperatures had reached their lowest levels. SSTs in the Thames River Estuary during the months preceding the study were below average for most of the winter, Northeastern Naturalist 414 L.S. Vlietstra 2014 Vol. 21, No. 3 on some days dropping to 0–4 °C, temperatures at which Atlantic Comb Jellies do not survive (Costello et al. 2012, Shiganova et al. 2001). If ctenophore populations in the Thames River Estuary were reduced by low temperatures during the winter of 2003–2004, time may be required for ctenophores to disperse into the estuary from surrounding areas. Such source–sink dynamics appear to influence the timing of Atlantic Comb Jelly population growth and abundance in nearby Narragansett Bay, RI (Beaulieu et al. 2013, Costello et al. 2006). Ctenophores there overwinter in inshore areas, where water warms rapidly in the spring, favoring early reproduction. Such sites serve as source regions, supplying adults to cooler regions of the bay where advective currents are stronger and ctenophore densities are reduced. For comparison, the annual range of SST in the Thames River Estuary in 2004 (1–24 °C) more closely resembled temperatures in sink regions of Narragansett Bay during 2003 (Fox Island, Dutch Island: 2–24 °C) than they did temperatures in the source region (Greenwich Cove: -1–27 °C; Costello et al. 2006). I did not measure tidal currents in this study, but ctenophore abundance could be further inhibited if strong currents in the spring flushed individuals out of the estuary and into Long Island Sound. In addition to the Atlantic Comb Jelly bloom in early July, a smaller abundance peak occurred on 5 September 2004. Ctenophore blooms during August–October were typical in New England prior to the mid-1980s (Sullivan et al. 2001), and blooms are still observed during this time in the central waters of Narragansett Bay, RI (Beaulieu et al. 2013). It is possible that local biophysical processes independent of large-scale currents continue to operate in the Thames River Estuary leading to a fall spike in ctenophore production. For example, a weakened thermocline in autumn coupled with wind mixing may bring nutrients to surface waters, promoting primary production and energy transfer to organisms at higher trophic levels, including copepods, which, in turn, are consumed by ctenophores. This idea is consistent with the observation that the water-column stability index (E) decreased in late August and September 2004, indicating intermittent mixing of the water column, while copepod and Atlantic Comb Jelly densities increased. In addition, preliminary results from research conducted at the study site in 2003 and 2011 indicate that surface concentrations of chlorophyll-a and net primary production peaked during mid-August and early September (S. Wainright, US Coast Guard Academy, New London, CT, unpubl. data). Seasonal abundance of larval ctenophores followed a different pattern than that of adults. Larval densities were variable during the study, with exceptionally high densities on 28 October 2004. At that time, SST was around 13 °C, dropping to <10 °C within two weeks. Sullivan et al. (2001) also reported a marked pulse in larval ctenophore density during October in Narragansett Bay, RI. Because the pulse in larval abundance occured just before water temperatures dropped below levels required to sustain population growth, the reason for an autumn pulse in ctenophore reproduction is unclear. One possibility is that reproduction during late autumn is an adaptive strategy for ensuring genetic representation in overwintering populations that inhabit shallow areas in estuarine habitats. Overwintering populations may be Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 415 an important source of individuals initiating production in the spring (Costello et al. 2006). Potential Ecological Impacts Concern over early-summer Atlantic Comb Jelly blooms in New England largely involves the potential for ctenophores to reduce fish production by consuming large amounts of ichthyoplankton and outcompeting young fish for copepod prey (Purcell et al. 2001). Such potential is clearly evident in Chesapeake Bay, VA, where Atlantic Comb Jelly blooms coincide with spawning in Anchoa mitchilli Valenciennes (Bay Anchovies ) and where ctenophores consume 20–100% of the eggs and larvae produced (Cowen and Houde 1993, Purcell et al. 2001). Using similar calculations based on laboratory-derived consumption rates (Monteleone and Duguay 1988) and observed ctenophore densities (300 individuals m-3), Sullivan et al. (2001) estimated that in some parts of Narragansett Bay, RI, Atlantic Comb Jellies consume as much as 111–1500% of fish eggs present. Where ctenophore densities were lower (7.5 individuals m-3), consumptions rates ranged between 2.5 and 35% of eggs present per day. However, we found no evidence in the Thames River Estuary to suggest that Atlantic Comb Jellies were a major consumer of icthyoplankton during 2004. Our net tows yielded fish eggs only occasionally during May-August, and our May-June samples contained fish larvae at least 4–8 weeks before adult Atlantic Comb Jelly became locally abundant. In addition, I found no ichthyoplankton in Atlantic Comb Jelly gut contents. In contrast, copepods were an important food source for Atlantic Comb Jelly in the Thames River Estuary during 2004, having been consumed by 87% of ctenophores examined. Yet it remains unclear whether ctenophore predation depressed copepod populations to an appreciable extent. The inverse correlation between copepod density and ctenophore biovolume suggests that ctenophores could reduce prey availability to other consumers, such as local fishes. However, research by Kremer (1979) serves to emphasize that correlations may not indicate cause-effect relationships. Specifically, Kremer (1979) observed a similar correlation but calculated that during the 1970s, ctenophores in Narragansett Bay, RI, consumed 5–10% of the copepod standing stock per day. This level of consumption could not account for the 20–25% day-1 copepod decline observed in late summer. Instead, predation pressure from other estuarine consumers, such as fish and other zooplankton, and possibly reduced phytoplankton concentrations were also thought to play a role in the decline (Kremer 1979). In the Thames River Estuary in 2004, my calculations suggest that ctenophores also consumed a relatively small portion (0.0–2.2%) of copepods on a daily basis. This figure is much lower than some estimates of copepod consumption by ctenophores in Narragansett Bay, RI. The findings from a different study from the 1970s estimated that Atlantic Comb Jellies consumed approximately 20% of the copepod standing stock per day and noted that ctenophore blooms were followed by rapid declines in zooplankton (Deason 1982, Deason and Smayda 1982). By the 2000s, ctenophore densities in Narragansett Bay had increased from earlier decades, and Northeastern Naturalist 416 L.S. Vlietstra 2014 Vol. 21, No. 3 Atlantic Comb Jellies were considered responsible for drastically reducing the dominant copepod, Acartia tonsa, during blooms (as cited in Costello et al. 2006). Although ctenophores in the Thames River Estuary appeared to consume a relatively small portion of the copepod standing stock in 2004, caution should be used when drawing conclusions about consumption rates calculated at too fine a resolution. Recent studies based on ctenophore-gut contents observed in the field suggest that laboratory-derived feeding rates could underestimate feeding activity in nature by up to 300–400% (Madsen and Riisgård 2010, Purcell 2009). One observation supporting the idea that ctenophores had a relatively small impact on copepod densities in the Thames River Estuary in 2004 is that ctenophore densities there were much lower than in other estuaries where copepod densities were comparable but trophic impacts were more pronounced. For example, the maximum density of adult Atlantic Comb Jellies in the Thames River Estuary was 9.7 individuals m-3, whereas maximum densities in Narragansett Bay, RI during 2002–2003 were 350–450 individuals m-3 (Costello et al. 2006). In Chesapeake Bay, VA, maximum densities during 2003–2006 were >100 to >400 individuals m-3 (Condon and Steinberg 2008). In addition, copepod densities in the Thames River Estuary increased over a period of 6 weeks in August–September concurrent with ctenophore abundance reaching a second annual peak. This pattern suggests that, while ctenophore predation may have dampened copepod growth rates during this time, predation pressure was not strong enough to prevent copepods from achieving positive population growth. I did not specifically investigate the reasons for low ctenophore densities in the Thames River Estuary during 2004, but the situation could have involved several ecological factors, such as prey availability, predation pressure, and water conditions, including the effects of low winter temperatures previously described (Costello et al. 2006, Purcell et al. 2001, Purcell and Decker 2005). Perhaps ctenophore densities were higher and trophic impacts greater in previous or subsequent years. Comparative studies of ctenophore-population dynamics in the same estuary in additional years and in estuaries of varying size, hydrographic conditions, and food-web structure could be useful in identifying factors that influence geographic variation in ctenophore density, and thus the potential for early ctenophore blooms to impact local fish production. In conclusion, early ctenophore blooms in New England and the mid-Atlantic appear to take place on a relatively large spatial scale, a pattern consistent with the hypothesis that phenological shifts in ctenophore production are due to largescale physical processes, such as the NAO, affecting SST along the North Atlantic coast. However, SST accounted for relatively little seasonal variation in ctenophore abundance, and further research into both physical and biological factors limiting the seasonal onset of ctenophore production is needed to better understand reasons behind seasonal patterns. In the Thames River Estuary, ctenophore densities in 2004 appeared low compared to other, larger estuaries where impacts on organisms at lower trophic levels, including copepods, appeared more substantial. However, increases in the abundance of gelatinous consumers concurrent with large-scale Northeastern Naturalist Vol. 21, No. 3 L.S. Vlietstra 2014 417 climate change have been widely documented in bays, coasts, and oceans worldwide (Mills 2001, Richardson et al. 2009). If similar increases in ctenophore biomass develop in small New England estuaries, the ecological impacts of Atlantic Comb Jelly blooms in the Thames River Estuary may become more pronounced. Acknowledgments This study was funded by a grant from the Quebec–Labrador Foundation Atlantic Center for the Environment and by a postdoctoral teaching fellowship from the United States Coast Guard Academy (USCGA). I thank the University of Connecticut for granting permission to use their mooring data, and Laura Gibbons (USCGA ’09) for assisting with field-work. I am also grateful for assistance with mapping from Sabrina Bateman, and for valuable feedback on the manuscript provided by Sam Wainright, Mark Patrick, Thomas Trott, and two anonymous reviewers. Literature Cited Beaulieu, W.T., J.H. Costello, G. Klein-Macphee, and B.K. Sullivan. 2013. Seasonality of the ctenophore Mnemiopsis leidyi in Narragansett Bay, Rhode Island. Journal of Plankton Research 35:785–791. Condon, R.H., and D.K. Steinberg. 2008. Development, biological regulation, and fate of ctenophore blooms in the York River Estuary, Chesapeake Bay. Marine Ecology Progress Series 369:153–168. Costello, J.H., B,K. Sullivan, D.J. Gifford, D. van Keuren, and L.J. Sullivan. 2006. Seasonal refugia, shoreward thermal amplification, and metapopulation dynamics of the ctenophore, Mnemiopsis leidyi in Narragansett Bay, Rhode Island. Limnology and Oceanography 51:1819–1831. Costello, J.H., K.M. Bayha, H.W. Mianzan, T.A. Shiganova, and J.E. Purcell. 2012. Transitions of Mnemiopsis leidyi (Ctenophore: Lobata) from a native to an exotic species: A review. Hydrobiologia 690:21–46. Cowan, J.H., Jr., and E.D. Houde. 1993. Relative predation potential of scyphomedusae, ctenophores, and planktivorous fish on icthyoplankton in Chesapeake Bay. Marine Ecology Progress Series 95:55–65. Deason, E.E. 1982. Mnemiopsis leidyi (Ctenophora) in Narragansett Bay, 1975–1979: Abundance, size composition and estimation of grazing. Estuarine and Coastal Shelf Science 15:121–134. Deason, E.E., and T.J. Smayda. 1982. Ctenophore–zooplankton–phytoplankton interactions in Narragansett Bay, Rhode Island. Journal of Plankton Research 4:219–236. Decker, M.B., D.L. Breitburg, and J.E. Purcell. 2004. Effects of low dissolved oxygen on zooplankton predation by the ctenophore Mnemiopsis leidyi. Marine Ecology Progress Series 280:163–172. Granhag, L., L.F. Møller, and L.J. Hansson. 2011. Size-specific clearance rates of the ctenophore Mnemiopsis leidyi based on in situ gut-content analysis. Journal of Plankton Research 33:1043–1052. Hawk, J.D. 1998. The role of the North Atlantic Oscillation in winter climate variability as it relates to the winter–spring bloom in Narragansett Bay. M.Sc. Thesis. University of Rhode Island, Kingston, RI. 148 pp. Knauss, J.A. 1997. Introduction to Physical Oceanography, Second Edition. Prentice Hall, Inc., NJ. 309 pp. Northeastern Naturalist 418 L.S. Vlietstra 2014 Vol. 21, No. 3 Kremer, P. 1979. Predation by the ctenophore Mnemiopsis leidyi in Narragansett Bay, Rhode Island. Estuaries 2:97–105. Kremer, P. 1994. Patterns of abundance for Mnemiopsis in US coastal waters: A comparative overview. ICES Journal of Marine Science 51:347–354. Madsen, C.V., and H.U. Riisgård. 2010. Ingestion-rate method for measurement of clearance rates of the ctenophore Mnemiopsis leidyi. Aquatic Invasions 5:357–361. McNamara, M.E., D.J. Lonsdale, and R.M. Cerrato. 2010. Shifting abundance of the ctenophore Mnemiopsis leidyi and the implications for larval bivalve mortality. Marine Biology 157:401–412. Mills, C. 2001. Jellyfish blooms: Are populations increasing globally in response to changing ocean conditions? Developments in Hydrobiology 155:55–68. Monteleon, D.M., and L.E. Duguay. 1988. Laboratory studies of predation by the ctenophore Mnemiopsis leidyi on the early stages in the life history of the Bay Anchovy, Anchoa mitchilli. Journal of Plankton Research 10:359–372. Ottersen, G., B. Planque, A. Belgrano, E. Post, P.C. Reid, and N.C. Stenseth. 2001. Ecological effects of the North Atlantic Oscillation. Oecologia 128:1–14. Purcell, J.E. 2009. Extension of methods for jellyfish and ctenophore trophic ecology to large-scale research. Hydrobiologia 616:23–50. Purcell, J.E., and M.B. Decker. 2005. Effects of climate on relative predation by scyphomedusae and ctenophores on copepods in Chesapeake Bay during 1987–2000. Limnology and Oceanography 50:376–387. Purcell, J.E., D.A. Nemazie, S.E. Dorsey, E.D. Houde, and J.C. Gamble. 1994. Predation mortality of Bay Anchovy, Anchoa mitchilli, eggs and larvae due to scyphomedusae and ctenophores in Chesapeake Bay. Marine Ecology Progress Series 114:47–58. Purcell, J.E., T.A. Shiganova, M. B. Decker, and E.D. Houde. 2001. The ctenophore Mnemiopsis in native and exotic habitats: US estuaries versus the Black Sea basin. Hydrobiologia 451:145–176. Reeve, M.R., and M.A. Walter. 1978. Nutritional ecology of ctenophores: A review of recent research. Advances in Marine Biology 15:249–287. Reeve, M.R., M.A. Syms, and P. Kremer. 1989. Growth dynamics of a ctenophore (Mnemiopsis) in relation to variable food supply. I. Carbon biomass, feeding, egg production, growth and assimilation efficiency. Journal of Plankton Research 11:535–552. Richardson, A.J., A. Bakun, G.C. Hays, and M.J. Gibbons. 2009. The jellyfish joyride: Causes, consequences and management responses to a more gelatinous future. Trends in Ecology and Evolution 24:312-322. Shiganova, T.A., Z.A. Mirzoyan, E.A. Studenikina, S.P. Volovik, I. Siokou-Frangou, S. Zervoudaki, E.D. Christou, A.Y. Skirta, and H.J. Dumont. 2001. Population development of the invader ctenophore Mnemiopsis leidyi, in the Black Sea and in other seas of the Mediterranean basin. Marine Biology 139:431–445. Sullivan, L.J., and J. Gifford. 2004. Diet of the larval ctenophore, Mnemiopsis leidyi A. Aggassiz (Ctenophora, Lobata). Journal of Plankton Research 26:417–431. Sullivan, B.K., D. Van Keuren, and M. Clancy. 2001. Timing and size of blooms of the ctenophore Mnemiopsis leidyi in relation to temperature in Narragansett Bay, RI. Hydrobiologia 451:113–120. Sullivan, B.K., D. J. Gifford, J.H. Costello, and J.R. Graff. 2008. Narragansett Bay ctenophore– zooplankton–phytoplankton dynamics in a changing climate. Pp. 485–498, In A. Desbonnet and B.A. Costa-Pierce (Eds.). Science for Ecosystem-based Management, Springer, New York, NY. 570 pp. Waggett, R.J., and L.J. Sullivan. 2006. Feeding efficiency of the larval ctenophore Mnemiopsis leidyi A. Agassiz (Ctenophora, Lobata). Journal of Plankton Research 28:719–723.