Refining sampling protocols for inventorying invertebrate biodiversity: Influence of drift-fence length and pitfall trap diameter on spiders

The limited resources available to inventory biodiversity and conduct ecological monitoring requires efficient protocols for sampling with pitfall traps. Here we consider adding different length drift-fences to pitfall traps on spiders. Four different fencing treatments (no fence, or fence lengths of 2, 4 and 6 m) were evaluated in combination with three trap diameters (4.3, 7.0 and 11.1 cm). Three-way ANOVAs revealed no significant interaction effects between any combinations of fencing treatments, trap size or the spatial positioning of transects within the study site along which traps were arranged. Post-hoc tests showed fences significantly increased the abundance of individuals and richness of spider families, and species collected. Traps with 6 m fences were significantly higher in all of these variables than traps with 2 m fences. ANOSIMs revealed taxonomic composition differed significantly between fenced and unfenced traps at familial, and specific ranks. Among fenced traps, taxonomic composition was influenced primarily by trap diameter rather than fence length. ANOSIMs showed significant differences in taxonomic composition between each trap diameter for fenced traps. An optimal combination of fencing treatment and trap diameter was determined by constructing smoothed species accumulation curves for increasing numbers of traps. Four criteria were considered: equivalent numbers of traps, standardized cumulative trap circumference, standardized cumulative fence length (fenced traps only) and standardized cumulative handling time. For the same number of traps, 11.1 cm traps with 4 and 6 m fences collected the most species. At a standardized trap circumference, long fences were best, with all trap sizes catching similar numbers of species. When fence length was standardized, 11.1 cm traps with 2 or 4 m fences collected the most species. At a standardized handling time all traps caught very similar numbers of species, although most 11.1 cm diameter traps collected more species than other trap sizes and those with 4 m fences were most efficient. Given the similar performance of fenced and unfenced traps for standardized handling time, we outline reasons why unfenced traps may be best.