Evaluation of the taxonomic sufficiency approach for ichthyoplankton community analysis

Ichthyoplankton identification is a time-consuming task, and often larvae cannot be identified to species due to a lack of adequate early life history descriptions. As a result, ichthyoplankton assemblage data are often analyzed at the family level, which results in a loss of taxonomic resolution, or at mixed taxonomic levels (e.g. family, genus and species combined), which can lead to difficulties in interpretation of results when a single species is included in multiple taxo nomic groupings. The taxonomic sufficiency (TS) approach has been used extensively in other disciplines (e.g. benthic marine macrofauna) to address similar analytical constraints, but to date this method has not been rigorously examined for ichthyoplankton studies. In this study, an ichthyoplankton data set collected in the northern Gulf of Mexico was proportioned into 3 data subsets with varying levels of taxonomic resolution: (1) species level only; (2) species, genus and family levels; and (3) combined taxonomic levels. Comparisons were made for assemblage metrics (larval density, richness and diversity) calculated for each taxonomic subset, as well as multivariate analyses of temporal variations characterizing ichthyoplankton assemblages. Genus- and species-level similarity matrices were highly correlated, which suggests analyses at the genus level could serve as a good proxy for species when examining assemblage diversity. Multivariate results for seasonal patterns were similar among family-, genus- and species-level analyses. The common approach of analyzing ichthyoplankton assemblages at mixed taxonomic levels, however, is not as statistically rigorous as single taxonomic-level analyses.