Limits to generality: seasonal and temporal variation in dispersal of an intertidal gastropod

Abstract

Seasonal and temporal variation can prevent the generalization of ecological models to situations other than those in which they were developed. This paper reports a study specifically designed to test the models that seasonal or temporal variation prevents the accurate prediction of dispersal of intertidal gastropods ( Bembicium auratum) in mangrove forests using models developed on rocky shores. Analysis of variance was used to compare directly the outcomes of experiments replicated several times and in different seasons. In general, the dispersal of Bembicium was very consistent through time, particularly on rocky shores. Differences in the behaviour of Bembicium inhabiting rocky shores and those inhabiting mangrove forests occurred in all runs of the dispersal experiment. The model that seasonal variations in the dispersal of Bembicium inhabiting one habitat prevented prediction of dispersal in the other was rejected. There was, however, some aseasonal temporal variation in the dispersal of Bembicium, particularly at one of the mangrove forests. This suggests that prediction of dispersal in mangrove forests will be less precise than for rocky shores unless further work is done to evaluate causes of temporal variation in that habitat. Experiments must be replicated both spatially and temporally in order to assess the generality of ecological processes, to facilitate proper comparisons with other studies and to provide a sounder basis for the prediction of ecological processes within and between habitats.