Detecting the effects of fishing on seabed community diversity: Importance of scale and sample size

Large-scale anthropogenic disturbances of terrestrial and marine environments, such as logging and fishing, are generally assumed to have negative effects on species diversity, but this supposition is often not supported by empirical observations. I investigated the importance of the extent of area sampled to the observed outcome of comparisons of the diversity of seabed assemblages in different areas of the seabed that experience either low or high levels of fishing disturbance. Using a finite data set within each disturbance regime, I pooled samples of the benthic communities at random. Thus, although individual sample size increased with each additional level of pooled data, the number of samples decreased accordingly. Detecting the effects of disturbance on species diversity was strongly scale-dependent. Despite increased replication at smaller scales, disturbance effects were more apparent when larger but less numerous samples were collected. The detection of disturbance effects was also affected by the choice of sampling device. Disturbance effects were apparent with pooled anchor-dredge samples but were not apparent with pooled beam-trawl samples. A more detailed examination of the beam-trawl data emphasized that a whole-community approach to the investigation of changes in diversity can miss responses in particular components of the community (e.g., decapod crustacea). The latter may be more adversely affected by disturbance than the majority of the taxa found within the benthic assemblage. Further, the diversity of some groups (e.g., echinoderms) actually increased with disturbance. Experimental designs and sampling regimes that focus on diversity at only one scale may miss important disturbance effects that occur at larger or smaller scales.