Monitoring in the presence of species misidentification: the case of the Eurasian lynx in the Alps

摘要: Inferring the distribution and abundance of a species from field records must deal with false-negative false-positive errors. False-negative errors occur if present goes undetected, while are typically consequence misidentification. False-positive observations in studies rare may cause an overestimation or distort trend indices. We illustrate this issue monitoring Eurasian lynx Alps. developed three-level classification according to their reliability as inferred whether they were validated not. The first category (C1) represents ‘hard fact’ data (e.g. dead lynx); second (C2) includes confirmed tracks verified by expert); third (C3) unconfirmed any kind direct visual observation). For lynx, which is comparatively well-known Alps, we use site-occupancy modelling estimate its show that highly sensitive presence sign category: it larger based on C3 compared more reliable C1 C2 records. believe reason for fairly high frequency among This suggests many lesser-known be similarly unreliable, because mostly exclusively thus soft data. Nevertheless, such form considerable part assessments presented, example International Union Conservation Nature Red List. However, can often not discarded only information available. When inferring carnivores, especially expanding shrinking range, recommend rigorous discrimination between fully un- partly data, order identify possible methodological problems maps related