Flawed meta-analysis of biodiversity effects of forest management.

It appears that the negative effect of forest management on biodiversity has become an axiom. Whether the negative effect, however, is a fact based on solid empirical evidence is not self-evident. Most of the studies that address the issue suffer from a lack of geographic extent and taxonomic narrowness. Therefore, a synthesis drawing together results from the individual studies is direly needed. In their recent paper, Paillet et al. (2010) rise to this challenge and present a formal pan-European metaanalysis of data from 49 papers representing 120 individual comparisons across 10 taxonomic groups. Their synthesis has the potential to be a landmark paper in ecological research, but also to affect pan-European forest policies and conservation prioritizations. In any metaanalysis, selection of studies to be included is critical for the conclusions to be reliable, but in such a potentially high-profile contribution as the synthesis by Paillet et al., a particularly high level of scrutiny of the data is called for. Here we draw attention to four major shortcomings in Paillet et al. that undermine the conclusions of their meta-analysis.