Abstract
New statistical models allow estimating the proportion of sites occupied by a species even if it is not always detected at an occupied site. These models are also useful for the analysis of monitoring data. A prerequisite for the use of this kind of models is that at least some, and possibly all, sites are visited repeatedly (at least twice). Here, I explain how these models work and discuss the underlying assumptions. As an example for illustration, I apply these models to the historic and current distribution of the adder (Vipera berus) in Switzerland.