Trophic effects of sponge feeding within Lake Baikal’s littoral zone. 1. In situ pumping rates

Demosponges dominate the littoral zone of Lake Baikal, Russia. During August 1993, we measured the in situ pumping activity of a globose sponge, Baikalospongia bacillifera, common at depths of 8-50 m. Excurrent flow speed, surface area, and oscular diameter were measured with fluorescein dye release, underwater video, and image processing for nine individual sponges; diel deployments of a heated thermistor flowmeter were made on another three individuals. Mean pumping rates for oscula ranged from 0.2 to 3.3 cm s-l, with instantaneous measurements as high as 25 cm s-l. Although between-sponge variability is large, oscula from an individual maintain speeds within 1 cm s-l of each other. Volume fluxes for oscula measured from all individuals range from 0.01 to 0.60 cm” s-l. Fluxes for individuals indicate that a sponge processes its volume in 17-24 s, comparable to values obtained for marine species. Oscula are not continuously active and often reduce pumping for several hours at irregular intervals. Two of three individuals monitored over a diel cycle exhibited negative correlations between the ambient current and oscular excurrent flow. Boundary layer measurements and diffusivity calculations demonstrate that sponges are capable of depleting picoplankton near the bottom. Sponges are widespread in marine and freshwater environments and can make up a substantial component of a community’s biomass. Their abundance is particularly conspicuous on coral reefs, where demosponges may outnumber scleractinian corals and alcyonarians (Wilkinson and Cheshire 1989). Sponges are efficient suspension feeders, and through their pumping activities are capable of processing vast volumes of water (Reiswig 197 la; Vogel 1974, 1977). Sponges have been noted to selectively feed on dissolved organic matter, bacteria, and specific types of phytoplankton (Reiswig 1971b; Frost 1981; Willenz et al. 1986), including procaryotic and eucaryotic picoplankton (Pile et al. 1997). Some data exist concerning their metabolic rates (Reiswig 1974; Frost 1980, 1987), and the water-pumping rates for a few species have been measured (Reiswig 197 la, 1974; Frost 1980). The extent of heterotrophic effects of sponges on the water column, however, has not been systematically considered. Although widespread, sponges are not frequently an obvious member of the benthos in freshwater environments, and the number of species living in freshwater is greatly reduced (-300 species; Frost 1991) compared to the sea (>4,700 species; Bergquist 1978). Sponges rarely represent more than a small fraction of a lake’s benthic biomass. What specifically restricts their geographic distribution is not clear, but heterogeneities in food availability, water quality, and biogeographical factors (Jewel1 1935, 1939; Frost 1991) have been implicated. Lake Baikal in south-central Siberia is an exception to this pattern. It is the oldest (~24~ lo6 yr), deepest (1,637 m), and most voluminous lake (a fifth of the world’s unfrozen freshwater) on the planet. More than 1,500 endemic species are found in the lake (Stewart 1990). Baikal’s littoral-zone benthos is dominated by photosymbiotic sponges. Three species of chlorophyte-bearing demosponges (Baikulospongia bucilliferu, Buikulospongiu inter-media, and Lubomirskiu buiculensis) dominate the photic zone and occupy a large percentage of the available rocky substrate (Pile et al. 1997). Because of this high benthic biomass, sponges seem to be the principal constituents in the lake’s nearshore ecology, and their combined heterotrophic and autotrophic behavior may have significant effects on oxygen, food, and nutrient availability. To assess heterotrophic effects, we examined the pumping performance of B. bucilliferu, one of the three prolific symbiont-bearing species within Lake Baikal’s littoral zone. We determined how much water B. bucilliferu processes and whether it pumps fluid at rates comparable to those measured for other sponge species. We also measured the diel variability of pumping and examined whether pumping is affected by ambient flow. Finally, we examined the state of the benthic boundary layer over these sponges to gauge the effectiveness of sponge community feeding in creating food-depleted layers near the bottom. Pumping rate measurements on B. bucilliferu were made at sites along the shore near the town of Lystvyanka (51”51.91’N, 104’49.88’E and 51”50.43’N, 104’49.90’E). Water depths ranged from 8 to 20 m. B. bucilliferu is identified by its smooth, globose shape, multioscular morphology, and large-diameter (>6 mm) oscula. Individual sponges (herein defined as the collective oscula contained on one sponge body) reach diameters up to 25 cm. Sponge pumping (excurrent flow velocity and volume flux) was measured through dye-release experiments conducted in situ on 16 oscula among nine individuals of B. bucilliferu. Actively pumping sponges were identified by detecting excurrent flow; if fluorescein dye released immediately above an osculum was deflected, the sponge was accepted for measurement. Fluorescein dye was then injected just below the outer pinacoderm with a syringe. The volume of dye injected per replicate was kept below 2 ml, and the rate of dye injection was minimized (-0.5 ml ss’) to avoid disturbing the sponge’s normal activity. Dye plumes issuing from the oscula were videotaped with a camera positioned so that the effluent plume traveled in a plane parallel to the video lens. This method ensured that only the vertical component of the excurrent velocity was visible. The camera-to-subject distance was minimized to increase spatial resolution. After all replicates were completed on an individual, the individual’s morphology was recorded by videotaping in three orthogonal orientations (calipers were used to establish scale in these views). Nine individuals