Lichens are often small and inconspicuous and therefore difficult to detect. Nevertheless, detection and subsequent identification is necessary to be certain about a species’ occurrence and to derive metrics like frequency, population density, and distribution range. Failure to detect a species that is present can lead to underestimation of those metrics unless the detection error is accounted for. Despite the common use of inventory data for distribution maps and assessments of species’ conservation status, no standardized large-scale lichen inventory has so far incorporated detection probability into any analyses as a means to correct for detection errors in the field. Methods to do exactly that have, however, been developed and used increasingly in amphibian, reptile, bird, and plant surveys: Occupancy models estimate both the detection probability and the occupancy rate (i.e. frequency) of a species, using information from sites that were inventoried multiple times. In many lichen inventories, a subset of the sites was inventoried repeatedly to assess reproducibility although those data were never used to specifically estimate detection probabilities. We are using a dataset with 641 plots of 500 m2 collected as a representative sample of the distribution and frequency of epiphytic lichen species in Switzerland. Our goal is to explore the extent and the variation in detection errors. Preliminary results will be discussed with a focus on potential implications for future lichen inventories.
Complete tool for submission, evaluation and publication of proceedings for scientific events.
