Search

The curvilinear relationship between species richness and habitat area (species-area relationship ( SAR)) is a fundamental ecological pattern. The relationship is often viewed from a long-term perspective across relatively large spatial scales, reflecting a balance between immigration and extinction dynamics. We explored whether predictions of SAR also manifest over short time periods (days) in benthic habitat patches of a dynamic floodplain river where littoral faunal assemblages are continuously assembled and disassembled with changing water levels. We examined the relationship of patch size with faunal abundance (i.e. fish and aquatic invertebrates), taxonomic richness, trophic group richness and overall assemblage composition. Strong taxa-area relationships emerged despite the relatively short experimental time period (21 days); larger patches had more taxa and trophic groups. For the smallest patches, taxonomic richness was especially sensitive to abundance of individuals; abundance of individuals was a less important predictor of taxonomic and trophic group richness for the largest patches. Despite the relatively short time frame for study within this temporally dynamic ecosystem, our findings indicate a strong SAR for fishes and macroinvertebrates inhabiting patchy habitats in the littoral zone of this tropical river. [ABSTRACT FROM AUTHOR]

Community assembly is affected by environmental filtering that restricts viable phenotypes and by species interactions that impose limits on interspecific trait similarity. Relative influences of these processes should vary according to habitat features and dispersal. Species dispersion within assemblage trait space also should vary in relation to species richness, strength of competition, and the spatiotemporal scale of analysis. We examined ecomorphological diversity of two freshwater fish families (Neotropical Cichlidae, Nearctic Centrarchidae) to test theories of local assembly from regional species pools and theories of species packing within mesohabitat patches. Cichlid and centrarchid assemblages were surveyed in four floodplain rivers (two in South America and two in North America) during low-water periods when fish densities are highest. Surveys were conducted in four mesohabitat types (submerged wood, leaf litter, rocks, sand bank) within river channels and floodplain lakes. We measured 23 morphological traits associated with locomotion and feeding. Principal components analysis was performed on the species3traits matrix, and species axis scores were used to calculate species pairwise Euclidean distances and indices of dispersion within assemblage morphospace: mean nearest-neighbor distance (indicating similarity), mean distance to centroid (assemblage morphospace size), and standard deviation of nearestneighbor distance (evenness of dispersion within assemblage morphospace). A null model was used to assess whether patterns were significantly nonrandom. When data for all mesohabitat types were combined for each river, species were significantly overdispersed and the assemblage morphospace was larger than predicted at random in every case. Analysis of assemblages within mesohabitat patches of different types revealed, in every case, significant overdispersion of species in morphospace indicative of limiting similarity. The total assemblage morphospace was greater than expected for tropical cichlids, but not for temperate centrarchids. Trends of species dispersion with assemblage morphospace in relation to species richness within mesohabitat patches were not consistent among or within river systems, possibly indicating that patches were already saturated with these perciform fishes. Interregional comparisons suggest an influence from both adaptive diversification and environmental filtering at broad spatial scales. At the scale of mesohabitat patches in lowland rivers, cichlid and centrarchid assemblages revealed patterns of trait complementarity that imply limiting similarity and strong influence of biotic interactions. [ABSTRACT FROM AUTHOR]

Abstract: The relationship between food web dynamics and hydrological connectivity in rivers should be strongly influenced by annual flood pulses that affect primary production dynamics and movement of organic matter and consumer taxa. We sampled basal production sources and fishes from connected lagoons and the main channel of a low-gradient, floodplain river within the Orinoco River Basin in Venezuela. Stable isotope analysis was used to model the contribution of four basal production sources to fishes, and to examine patterns of mean trophic position during the falling-water period of the annual flood cycle. IsoSource, a multi-source mixing model, indicated that proportional contributions from production sources to fish assemblages were similar in lagoons and the main channel. Although distributions differed, the means for trophic positions of fish assemblages as well as individual species were similar between the two habitats. These findings contradict recent food web studies conducted in temperate floodplain rivers that described significant differences in trophic positions of fishes from slackwater and floodplain versus main channel habitats. Low between-habitat trophic variation in this tropical river probably results from an extended annual flood pulse (ca. 5 mo.) that allows mixing of sestonic and allochthonous basal production sources and extensive lateral movements of fishes throughout the riverscape. [Copyright &y& Elsevier]