Accounting for false positives improves estimates of occupancy from key informant interviews

摘要: Aim Much research in conservation biogeography is fundamentally dependent on obtaining reliable data species distributions across space and time. Such are now increasingly being generated using various types of public surveys. These often integrated with occupancy models to evaluate distributional patterns, range dynamics status multiple at broad spatio-temporal scales. Occupancy have traditionally corrected for imperfect detection due false negatives while implicitly assuming that positives do not occur. However, survey also prone false-positive errors, which when unaccounted can cause bias estimates. We test whether a dataset collected from surveys lead overestimation site estimators simultaneously account false-negative errors improve estimates. Location Western Ghats, India. Methods We fit large-scale key informant interview 30 large vertebrates. tested their performance against standard only negatives. Results Standard correct tended overestimate errors. accounted had greater support [lower Akaike's information criterion (AIC)] and, consistent predictions, systematically lower estimates than models. Furthermore, accounting improved the accuracy despite added complexity statistical estimator. Main conclusions Integrating modelling approaches powerful tool informing management. many if most cases, it will be important explicitly ensure reliability obtained datasets such as interviews, volunteer surveys, citizen science programmes, historical archives acoustic