Yu-Lung Hsieh, Karl Eduard Linsenmair

• Measuring and estimating biodiversity patterns is a fundamental task of the scientist working to support conservation and informmanagement decisions.Most biodiversity studies in temperate regions were often carried out over a very short period of time (e.g., a single season) and it is often—at least tacitly—assumed that these short-termfindings are representative of long-termgeneral patterns.However, should the studied biodiversity pattern in fact contain significant temporal dynamics, perhaps leading to contradictory conclusions. Here, weMeasuring and estimating biodiversity patterns is a fundamental task of the scientist working to support conservation and informmanagement decisions.Most biodiversity studies in temperate regions were often carried out over a very short period of time (e.g., a single season) and it is often—at least tacitly—assumed that these short-termfindings are representative of long-termgeneral patterns.However, should the studied biodiversity pattern in fact contain significant temporal dynamics, perhaps leading to contradictory conclusions. Here, we studied the seasonal diversity dynamics of arboreal spider communities dwelling in 216 European beeches (Fagus sylvatica L.) to assess the spider community composition in the following seasons: two cold seasons (I:November 2005–January 2006; II: February–April) and two warm seasons (III: May–July; IV: August–October). We show that the usually measured diversity of the warmseason community (IV: 58 estimated species) alone did not deliver a reliable image of the overall diversity present in these trees, and therefore, we recommend it should not be used for sampling protocols aimed at providing a full picture of a forest’s biodiversity in the temperate zones. In particular, when the additional samplings of other seasons (I, II, III) were included, the estimated species richness nearly doubled (108). Community I possessed the lowest diversity and evenness due to the harsh winter conditions: this community was comprised of one dominant species together with several species low in abundance. Similarity was lowest (38.6%) between seasonal communities I and III, indicating a significant species turnover due to recolonization, so that community III had the highest diversity. Finally, using nonparametric estimators, we found that further sampling in late winter (February–April) is most needed to complete our inventory. Our study clearly demonstrates that seasonal dynamics of communities should be taken into account when studying biodiversity patterns of spiders, and probably forest arthropods in general.…