Mitigating the conflict between pitfall-trap sampling and conservation of terrestrial subterranean communities in caves

Abstract

Abstract:Subterranean habitats are known for their rich endemic fauna and high vulnerability to disturbance. Many methods and techniques are used to sample the biodiversity of terrestrial invertebrate fauna in caves, among which pitfall trapping remains one of the most frequently used and effective ones. However, this method has turned out to be harmful to subterranean communities if applied inappropriately. Traditionally, pitfall traps have been placed in caves solely on the ground. Here we present an optimized technique of pitfall trapping to achieve a balance between sampling completeness and minimal disturbance of the fauna in the cave.Monthly we placed traps for two days in two parallel sets, a ground trap and an upper one−just below the ceiling−along the cave. In the upper set, about 10% additional species were recorded compared to the ground set. Greater species diversity in the cave was the consequence of both the increased sampling effort and the amplified heterogeneity of sampled microhabitats.In caves sampled by traditional pitfall trapping, overlooked species may be a consequence of methodological biases, leading to lower biodiversity estimates. In our research, incidencebased estimations mostly surpassed abundance-based ones and predicted 95% coverage of the species richness within about two years of sampling. The sampling used contributes at the same time to both the more effective and less invasive inventory of the subterranean fauna. Thus, it may serve as an optional sampling to achieve optimal balance between required data for biodiversity and ecological studies, and nature conservation goals.

Distributional dynamics of a specialized subterranean community oppose the classical understanding of the preferred subterranean habitats

“…; Cave Register Karst Research Institute ZRC SAZU and Speleological Association of Slovenia) (Figure ), which is in close vicinity to the Postojna cave (Šebela, ), one of the world's hotspots of subterranean fauna (Culver & Pipan, ; Culver & Sket, ; Zagmajster et al, ; Sket, ). This relatively small, horizontal cave is 122 m long, but hosts a rich and diverse community, consisting of more than 100 species, of which at least 17 can be classified as troglobionts (Kozel et al, ). This study site was chosen because of its simple morphology, linear shape, and nearly constant cross section (with a 1.5 m wide × 1 m high entrance, 2 × 2 m average cross section, and a few large sites where the ceiling reaches 5 m height), allowing us to sample the species and measure the pertinent environmental factors in a more standardized way.…”

“…The ground plan of Zguba jama, showing sampling sections (red lines) and pitfall‐trap sites (blue squares), where a pair of traps was set, one at ground level and the other set at the cave wall, just below the ceiling. Modified from Šebela () and Kozel et al ()…”

“…The fauna was recorded by visual inspection and by pitfall trapping. We applied simultaneously two sampling methods, since it is documented that complementary sampling approaches maximize the detection and completeness of diversity assessments of cave‐dwelling arthropods (Kozel et al, ; Wynne et al, ; Wynne et al, ).…”

“…Only individuals that we could not identify to the species level in the cave were taken to the laboratory for identification. In order to investigate simultaneously the fauna occurring on the ground and on walls (Kozel et al, ), we used two parallel sets of pitfall traps: a ground‐level set, and an upper set, placed just below the ceiling, along the cave wall (Figure ). Pitfall traps were baited with decomposing beef and apple juice, with a lacing of cherry and maraschino essence, and a few drops of detergent were added to reduce the surface tension.…”

“…The voucher specimens preserved in 70% ethanol are deposited at the Faculty of Natural Sciences and Mathematics, University of Maribor. For details on the traps, see Kozel et al (). Counts deriving from visual inspection and pitfall trapping were merged in the analysis (Wynne et al, ).…”