Assessing spider species richness and composition in Mediterranean cork oak forests

TitleAssessing spider species richness and composition in Mediterranean cork oak forests
Publication TypeJournal Article
Year of Publication2008
AuthorsCardoso, P., Gaspar C., Pereira L. C., Silva I., Henriques S. S., da Silva R. R., & Sousa P.
JournalActa Oecologica
Volume33
Issue1
Pagination114 - 127
Date Published2008///
Keywordsaraneae, Arrábida, Biodiversity assessment, iberian peninsula, methodology, Portugal, Quercus suber, Richness estimators, Semi-quantitative sampling, Stop-rules
Abstract

Semi-quantitative sampling protocols have been proposed as the most cost-effective and comprehensive way of sampling spiders in many regions of the world. In the present study, a balanced sampling design with the same number of samples per day, time of day, collector and method, was used to assess the species richness and composition of a Quercus suber woodland in Central Portugal. A total of 475 samples, each corresponding to one hour of effective fieldwork, were taken. One hundred sixty eight species were captured, of which 150 were recorded inside a delimited one-hectare plot; this number corresponds to around 90% of the estimated species richness. We tested the effect of applying different sampling approaches (sampling day, time of day, collector experience and method) on species richness, abundance, and composition. Most sampling approaches were found to influence the species measures, of which method, time of day and the respective interaction had the strongest influence. The data indicated that fauna depletion of the sampled area possibly occurred and that the inventory was reaching a plateau by the end of the sampling process. We advocate the use of the Chao estimators as best for intensive protocols limited in space and time and the use of the asymptotic properties of the Michaelis–Menten curve as a stopping or reliability rule, as it allows the investigator to know when a close-to-complete inventory has been obtained and when reliable non-parametric estimators have been achieved.

URLhttp://linkinghub.elsevier.com/retrieve/pii/S1146609X07001178