Taxonomic Sufficiency (TS) has been proposed as a short-cut method to quantify changes of biological assemblages in environmental monitoring. However, issues about the efficacy of taxonomic surrogates in depicting long-term temporal patterns of marine assemblages are still scant. Here we report on the adoption of TS combined with data transformations to describe patterns of North Adriatic polychaete assemblages through 20 years. Univariate and multivariate analyses revealed large spatialetemporal variation charactering the assemblages. The efficiency to discriminate between the two research periods (1990e1994 vs. 2004e2008) was reduced when data were analyzed at family and order level leading to misinterpreting the sources of assemblage variation. Further information was lost with data transforms. Families may represent appropriate assemblages’ descriptors in long-term monitoring, but using TS coupled with data transformations could lead to hazardous loss of information. We suggest that periodical analysis at fine taxonomic level should be routinely alternated to long-term monitoring based on TS in order to check its effectiveness
