Priority effects, taxonomic resolution, and the prediction of variable patterns of colonisation of algae in littoral rock pools.

This study focuses on succession of macroalgae in littoral rock pools on the west coast of Italy. Previous studies in this system indicated that either canopy algae or turf-forming algae may dominate late in succession. Priority effects and non-hierarchical interactions have been proposed as possible explanations for these patterns. From previous knowledge on the timing of reproduction and recruitment of the two groups of algae and their interactions, I predicted that: (1) canopy algae would dominate patches of substratum cleared during their main period of recruitment (between April and July); (2) the turf-forming algae, although initially present, would be replaced by canopies in these patches; (3) turf-forming algae would characterise both the early stages of colonisation and the mature assemblage in patches cleared before or after the main period of recruitment of canopy algae, and (4) succession would be more consistent in space (i.e. canalised) in the presence of canopy algae than when the turf-forming plants achieve dominance. These predictions were tested in a multifactorial experiment where patches of substratum were cleared in three different periods (before, during and after the main period of recruitment of canopy algae), on three dates within each period and in two replicate pools in each date. Univariate and multivariate analyses indicated that variability at early stages of colonisation dictated much of the subsequent dynamics in this system. Predictions 1-3 were supported by the results, but only at a gross level of taxonomic resolution. Patterns of colonisation of individual species of turf-forming algae were unpredictable due to large small-scale spatial and temporal variation in abundance. Prediction 4 was not supported by the results. This study indicated that knowledge of the life-histories and ecology of individual populations is crucial to increase the accuracy and precision of ecological models that attempt to predict succession in variable systems.