Predation intensity in estuarine soft bottoms: between-habitat comparisons and experimental artifacts

Abstract

The aim of this study was to determine whether experimental artifacts have the potential to bias comparisons of relative predation intensity among different soft-sediment estuarine habitats. A field experiment was used to test (1) whether predation intensity on the infaunal clam Mercenana mer- cenarja (L ) differed between shallow-subtldal and intertidal sand flats, (2) whether predation on clams differed between intertidal sand flats and vegetated habitats (the e d g e of Spartina alternjflora (Loisel) salt marshes) of similar elevation, a n d (3) whether the outcome of these comparisons was influenced by the experimental method used. For mobile prey, some restraint on prey movement is typically required to compare predation intensity among different habitats. This is most often achieved by tethering prey. Tethering and other experimental interventions may induce different experimental artifacts In esti- mates of p r e d a t ~ o n intensity In d~fferent habitats, which would invalidate between-habitat comparisons (Peterson & Black 1994: Mar Ecol Prog Ser 111:289-297). 1 used colnblnations of 2 types of prey restraint, tethers and buried fences, to test for possible interactive effects of habitat and prey restraint on prey mortality. Tethered or untethered clams were deployed within fenced or unfenced field plots within each habitat type, in a factorial design, and were exposed to the natural assemblage of predators for 1 wk. W i t h ~ n each habitat type, both tethering and fencing increased the proportion of clams recov- ered live after I wk compared to the treatment in which no prey restraints were used. Slynificantly more live clams were recovered in the intertidal sand flats than In the subtidal sand bottoms wlthin field plots enclosed with buried fences, suggesting lower predation intensity in the Intertidal than in suhtidal sand flats. In contrast, there was no significant difference in proportions of live clams between subtidal sand bottoms and intertidal sand flats within unfenced plots. These results were obtained with both tethered and untethered clams Significantly more clams were recovered live in intertidal sand flats than in intertidal salt marshes of similar elevation in all treatment combinations, suggesting lower predation intensity in unvegetated than in vegetated habltats Thls result was unexpected because habitat structural con~plexity is known to decrease predation rates in other shallow marlne habitats, such as rocky intertidal habitats and seagrass beds. Results of this field experiment indicate that prey restraints can bias not only estimates of predation intensity within a single habitat type but can also bias between-habitat comparisons of predation intensity, probably by causing experimental artifacts of dif- ferent magnitudes in each habltat. The actual magnitude of experimental artifacts in each habitat type can only be estimated with direct observations of predator-prey Interactions in the field.