Hide and seek: extended camera-trap session lengths and autumn provide best parameters for estimating lynx densities in mountainous areas

A tool commonly used in wildlife biology is density estimation via camera-trap monitoring coupled with capture–recapture analysis. Reliable regional density estimations of animal populations are required as a basis for management decisions. However, these estimations are affected by the session design, such as the length of the monitoring session, season, and number of trap sites. This method is regularly used to monitor Eurasian lynx (Lynx lynx) which mostly occupy the forested mountain ranges in Central Europe. Here we used intensive field sampling data of a major Central European lynx population to investigate (1) the optimal monitoring session length considering the trade-off between population closure and number of recaptures for density estimates, (2) the optimal time window within the year considering the stability of density estimates, detection probability, recapture number, and reproduction, and (3) the number of trap sites and trap spacing required to achieve robust density estimates. Using two closure tests, we found that 80 days are the minimum to ensure adequate data quality. A spatially explicit capture–recapture model revealed the best monitoring period to be late summer to early winter. Based on our results, we recommend for similar management units of comparable size (~300 km2) and similar recapture numbers to sample for at least 80 days in autumn with traps spaced about every 2.5–3 km. Our results also indicated that stable density estimates could still be maintained when the sampling area is enlarged to 760 km2 with trap spacing every 5–6 km if session lengths are increased.