Differences in the size distributions of C. kilmeri among seep sites are indicative of variation in demographic rates, individual growth rates, or both. The low abundance of large size classes at the Clam Flat site and bimodal character of sizes at Clam Field and Clam Flat suggest that the age structures of these populations are unstable, with episodic recruitment pulses which dominate size distributions producing peaks in juvenile or adult sizes. Alternatively, stable, bimodal size distributions may arise from changes in rates of growth or mortality with age (Barry & Tegner 1990). In addition, temporal changes in the habitability of seep environments related to sulfide availability may exert strong influence on the stability of vesicomyid populations.
The ratio of mortality-to-growth for C. kilmeri is low at all sites, suggesting that these populations are limited principally by recruitment and have low rates of mortality. An index of mortality to growth developed by Brey & Gage (1997) as ΔZ/K = log (Zmeasured/Zpredicted), where log (Zpredicted = 0.339 + 1.037( log (Kmeasured))), was negative for each site (Clam Field = −0.38; Clam Flat = −0.27, Mt Crushmore = −0.45; K = 0.245 for all sites). Negative values of ΔZ/K are generally found in populations with a lower than average ratio of mortality-to-growth, while those with high mortality rates (e.g. heavily fished species) typically have positive ΔZ/K indices (Brey & Gage 1997). Combined with size–frequency data from each site, the ΔZ/K indices are consistent with population dynamics dominated by limited or variable recruitment rates or both, but generally low rates of mortality. Thus, predation rates may be quite low in these populations, even though they appear to be quite vulnerable to some predators (e.g. octopus, decapod crustaceans). Rates of predation may be influenced by several factors, including sulfidic sediments and tissues which may inhibit predators, or individual size. Predation may be the highest for small, thin-shelled juveniles, then decline as individuals grow to a larger size with thicker, stronger shells. It is not possible to discriminate between larval supply and early post-settlement survival as controls on recruitment rates.
Growth rate variability
Growth rates reported here for C. kilmeri are generally similar to those estimated for other bivalve taxa from cold seep and hydrothermal vent environments. Radionuclide dating techniques were used to estimate growth in C. magnifica at the Galapagos Spreading Center (Turekian & Cochran 1981) and the East Pacific Rise hydrothermal vents (Turekian et al. 1983). Lutz et al. (1988) also modeled growth of C. magnifica at these sites. Growth of C. magnifica varied 15-fold (2.7–40 mm·year−1) between these sites, suggesting a large impact of site to site environmental variation (i.e. sulfide availability). In addition, the authors reported considerable variation in growth rate within individual clams, suggesting that both spatial and temporal variation in energy availability may modulate vesicomyid growth. Growth by chemosynthetic mytilids at Gulf of Mexico seeps, which are similar in size to C. kilmeri, varies considerably (c.−1 to 18 mm·year−1) depending on the availability of methane and perhaps sulfide (Nix et al. 1995; Smith et al. 2000; Bergquist et al. 2004), with growth coefficients from K = 0.02 to 0.21. Although limited tag–recapture data prevented a full characterization of C. kilmeri growth at each site, the non-significant ANCOVA comparing growth among sites, and results of size–frequency analyses indicated fairly similar growth at all stations. Little information is available presently concerning variation in growth within individuals.
Growth by chemosynthetic taxa from vent and seep environments overlaps most closely with heterotrophic, shallow-water taxa found in some of the most productive ocean habitats. For example, growth rates of two shallow-water venerids similar in size to seep vesicomyids, Callista chione (K = 0.24; c. 15 mm·year−1, S∞ = 93 mm) from the Aegean Sea (Metaxatos 2004), and Eurhomalea exalbida (K = 0.15, S∞ = 74 mm) from the shallow subtidal in Ushuaia Bay in the Beagle Channel (Lomovasky et al. 2002), both overlap the values typical for chemosynthetic bivalves.
In contrast, growth rates reported for most chemosynthetic metazoans including vesicomyids are far greater than non-chemosynthetic, deep-sea species, due presumably to the severe limitations on energy availability for bathyal and deeper taxa. Even though few estimates of growth rates and production are available, deep-sea bivalves are small and considered to be slow-growing (Allen 1979; Oliver & Allen 1980). The deep-sea clam Tindaria callistiformis from 3800 m depth in the North Atlantic reportedly requires 100+ years (SD = 38 years) to reach a maximum size near 8 mm (Turekian et al. 1975). Gage (1994) estimated much more rapid growth for the bathyal nuculanid Ledella pustulosa (c. 1 mm·year−1), though it was still much lower than C. kilmeri. A sympatric species, Yoldiella jeffreysi, had size–frequency distributions similar to L. pustulosa, suggesting a similar pattern of growth. Thus, these nuculanoid protobranchs appear to grow to adult size in only a few years.
Production of seep/vent communities
Production by chemosynthetic assemblages at hydrothermal vent and cold seep habitats is expected to be higher than for heterotrophic, deep-sea benthos dependent upon organic input from surface waters. Reports of secondary production at bathyal depths support this idea. Production by a bivalve-dominated sediment community in the Rockall Trough (2900 m) was reported as c. 0.12 g wet weight m−2·year−1 (Gage 1992). Gage (2003) also measured the productivity of a bathyal (700–1000 m) brittle star (Ophiocten gracilis) assemblage (c. 0.06 g·AFDM·m−2·year−1) off Scotland. Both rates were far lower than estimated in this study for C. kilmeri (294 g·AFDM·m−2·year−1 = 2,292 g·wet wt.·m−2·year−1) aggregations, which are 4 orders of magnitude more productive than the Rockall Trough benthos, and 3 orders more than the slope ophiuroid assemblage. However, because C. kilmeri may dominate productivity in the seep community to a greater extent than Ophiopectin in the bathyal community, this comparison probably overestimates the ratio of seep to bathyal ophiuroid community production.
A more conservative estimate of the ratio of seep/non-seep production can be made by comparing production by C. kilmeri to Brey and Gerdes’ (1998) characterization of shallow to abyssal macrobenthic productivity. Brey and Gerdes related production by macrobenthic communities to water depth as log(production) = 1.273–0.419(depth + 1). Using this relationship, the expected productivity by typical heterotrophic benthos at the average depth of the three seep study sites (846 m) is 1.1 gC·m−2·year−1 = c. 2.4 g·AFDM·m−2·year−1, or >100 times lower than estimated for C. kilmeri. Alternatively, we can estimate heterotrophic benthic production from the sinking particulate flux (c. 14 gC·m−2·year−1; Pilskaln et al. 1998) multiplied by the dynamic ecotrophic efficiency (e) estimated near 0.25 (Brey 2001), yielding a benthic productivity near 3.6 gC·m−2·year−1, still only 1.5% of production estimated for C. kilmeri.
Although secondary production by seep vesicomyids is clearly high in the context of heterotrophic bathyal macrobenthos, it appears comparable to production in other seep and vent assemblages, based on the high rates of growth reported among chemosynthetic taxa, ranging from seep mytilids (Nix et al. 1995; Bergquist et al. 2004) to Riftia sp. from hydrothermal vents (Roux et al. 1989; Lutz et al. 1994). Production by very slow-growing vestimenterans (Fisher et al. 1997; Bergquist et al. 2000) is low, but may nonetheless exceed that of heterotrophic deep-sea communities.
Somatic production by vesicomyids is also generally higher than that of most heterotrophic, macroinvertebrate assemblages in marine benthic habitats. Cusson & Bourget's (2005a) review of benthic production focused principally on coastal (<50 m depth) environments and reported average ranges in productivity of 0.9–4.0 g·AFDM·m−2·year−1 among major invertebrate taxa. Molluscan assemblages had the highest rates of production (4.0 g·AFDM·m−2·year−1), but these average rates were still c. 60 times lower than estimated for C. kilmeri.
Intertidal bivalve communities along productive shores appear to have rates of production comparable with seep and vent communities. Somatic production in intertidal mussel communities is reported to be c. 640 g·AFDM·m−2·year−1 (Cusson & Bourget 2005b), exceeding that of most nearshore benthos, but comparable with production by C. kilmeri. Production by surf clams (Donax serra) along South African shores (167–637 g·AFDM·m−2·year−1; Laudien et al. 2003) is also similar to our estimates for a bathyal chemosynthetic vesicomyid. Although comparable in production, vesicomyid assemblages explored in Monterey Bay were far less extensive than these productive shallow water bivalve assemblages.
P/B ratios for C. kilmeri (0.14–0.42 year−1) were low compared with most non-chemosynthetic taxa surveyed (see Cusson & Bourget 2005a), including molluscs (P/B c. 1.77). However, this low P/B ratio is consistent with the global patterns identified by Cusson & Bourget (2005a), showing a negative correlation of P/B with life span and body mass, and positive relationship with biomass density. Calyptogena kilmeri has large body mass compared with typical marine bivalves, with a moderately high maximum age expected to be near 15–30 years, based on growth and mortality estimates. In addition, the biomass density of C. kilmeri (c. 705–2059 g·AFDM·m−2) is far greater than the average molluscan taxon (c. 2.24 g·AFDM·m−2; Cusson & Bourget 2005a). Calyptogena kilmeri biomass also exceeds the benthic biomass of all macrobenthic communities of comparable depth reported by Brey & Gerdes (1997), and is more similar to shallow-water mytilids (773.7 g·AFDM·m−2; Cusson & Bourget 2005b). Relatively low P/B ratios and high rates of production (apparently due to large body size) are probably typical of many densely aggregated chemosynthetic assemblages.
The population dynamics of C. kilmeri in Monterey Bay chemosynthetic assemblages appear to include relatively rapid rates of population turnover (i.e. replacement by death and recruitment), in contrast with initial observations of growth and succession in seep communities. Although rapid colonization, growth, and succession are typical for hydrothermal vent communities (Smith & Hessler 1987; Shank 1999), the growth rates and population dynamics of some of the first seep megafauna studied were slow, particularly seep vestimentiferans with very slow growth and low rates of mortality (e.g.Lamellibrachia luymesi; 0.02–0.12% per year) (Bergquist et al. 2000, 2003; Cordes et al. 2005). Vesicomyids, like many other chemosynthetic bivalves, are intermediate in the spectrum of growth and population replacement observed in hydrothermal vent and seep systems, with fairly high growth and mortality rates.
Olu et al. (1996) proposed that vesicomyids are initial colonists of sulfidic seep sediments off Barbados, later replaced by chemosynthetic mytilid mussels, although the successional mechanisms remain unclear. Because methanotrophic or sulfide-oxidizing mytilids have not been discovered in northeastern Pacific seep communities, likely due to limited methane availability near the seabed because of limited fluxes, rapid mixing, or both (Barry et al. 1997), successional pathways for seep communities in this area may be shorter than for seep environments drawing from a larger potential species pool. Succession may thus be controlled by geochemical limits on the availability of reduced inorganic compounds (i.e. sulfide & methane). Because local fluid chemistry may not support large vestimentiferan or mytilid populations, vesicomyids may be released from competition with these potentially effective late-successional taxa. Progress in understanding the physiology, growth, and production of chemosynthetic species has been rapid during the past two decades. Complementary studies of chemosynthetic taxa will broaden the demographic context of these studies to include processes influencing reproduction, colonization, and recruitment, enabling a more comprehensive perspective concerning the development and regulation of chemosynthetic communities.