Pattern-oriented modelling for estimating unknown pre-breeding survival rates:: The case of the Lesser Spotted Woodpecker (Picoides minor)

in population viability analysis we are often faced with a lack of knowledge of survival rates in animal populations. In particular, survival of recruits is usually hard to assess. However, data on population structure might be considered as patterns that contain valuable information to estimate missing parameters indirectly. As an example for this pattern-oriented modelling and parameterization, pre-breeding survival rate of the endangered Lesser Spotted Woodpecker (Picoides minor) was determined here using data on population structure (e.g. sex ratio) and reproductive success at the population level (e.g. nesting success). Therefore, an individual-based model was developed simulating the population dynamics for two different populations that had been empirically studied at Lake Mockeln, Sweden, and Taunus, Germany. For both populations, a small range for pre-breeding survival rates could be identified wherein all simulated patterns corresponded best to the empirical values. Prebreeding survival rate was found to be higher in the German scenario than in the Swedish and geographical variation in life-history traits is discussed as a possible reason. It is concluded that the pattern-oriented approach is a valuable method for estimating missing demographic parameters, even when using weak patterns from empirical investigations. Furthermore, it was shown that the use of multiple patterns is necessary for this purpose. (c) 2006 Elsevier Ltd. All rights reserved.