Structural and functional response of meiofauna rocky assemblages to sewage pollution S. Fraschetti a, * , C. Gambi b , A. Giangrande a , L. Musco a , A. Terlizzi a , R. Danovaro b a Laboratorio Zoologia e Biologia Marina, Dipartimento di Scienze e Tecnologie Biologiche ed Ambientali, Universita ` di Lecce, CoNISMa, 73100 Lecce, Italy b Dipartimento di Scienze del Mare, Facolta ` di Scienze, Universita ` Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy Abstract Despite meiofauna being one of the most popular tool for detecting the ecological effects of different sources of impact, the application of structured experimental designs to this benthic component is typically neglected, especially in subtidal rocky shores. In this study, an asymmetrical design was used to detect hard-bottom meiofaunal abundance and biodiversity response to sewage discharge. Sampling was carried out at a depth of 3–4 m by SCUBA diving, by means of a modified manual corer. Six replicate cores were collected at three sites (80–100 m apart), at each of three locations (one purportedly impacted [I] and two controls [Cs]). A total of 84,994 specimens were collected. The outfall affected meiofaunal assemblages in terms of taxon richness, by determining a significant reduction of the number of taxa in the disturbed location, and community structure, by causing functional changes in terms of a decrease of the abundance of nem- atodes and an increase of the hydrozoan component. Nematodes and syllid polychaetes showed significant lower average abundance at I. Multivariate analyses showed that both meiofaunal assemblage and syllids were significantly different at I compared with Cs. The sewage outfall also affected patterns of spatial distribution at the scale of site (100 m apart) and of replicate units (centimetres apart), both in syllids and in nematodes. Our results provide evidence that the selection of multiple controls is crucial to prevent the widespread risk of Type II error, highlighting the need of more accurate experimental designs when dealing with meiofauna. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Environmental impact assessment; Meiofauna; Nematodes; Polychaetes; Rocky subtidal 1. Introduction Patterns of distribution of populations and assemblages in most marine systems are extremely variable at several spatial and temporal scales (Levin, 1992; Wiens et al., 1993; Fraschetti et al., 2005). Natural patterns of variabil- ity, in terms of mean values of response variables and in measures of spatial variance, can be significantly modified by human activities (Underwood, 1992, 1994; Chapman et al., 1995; Benedetti-Cecchi, 2001; Balestri et al., 2004; Terlizzi et al., 2005a,b). This is becoming increasingly evi- dent from the attention to sampling design in recent studies evaluating environmental impacts. An adequate sampling design coupled with new statistical tools can provide a powerful basis for the analysis of speciesÕ responses to anthropogenic disturbance (Anderson and Thompson, 2004). However, it is still unclear whether the findings deriving for a limited number of case studies can be of gen- eral value in the assessment of environmental impacts. This lack of generality stems from the limited array of habitat types investigated, whereas several habitats and ecological systems are left unstudied. Plankton assemblages and, in general, benthic mobile fauna are less tractable from an experimental point of view than benthic sessile assem- blages, so that pseudo-replication in space and time is still common. Consequently, monitoring programs aimed at detecting impacts on populations and assemblages may 0025-326X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpolbul.2005.10.001 * Corresponding author. Tel.: +39 832 298853; fax: +39 832 298626. E-mail address: sfrasca@ilenic.unile.it (S. Fraschetti). www.elsevier.com/locate/marpolbul Marine Pollution Bulletin 52 (2006) 540–548