Genetic erosion explains deviation from demographic response to disturbance and year variation in relic populations of the perennial Primula vulgaris

1. Habitat fragmentation, degradation and abiotic disturbances may strongly affect population dynamics and genetic diversity of plant populations. 2. In this study, a Life Table Response Experiment (LTRE) and stochastic simulations were used to investigate how different types of disturbance affected the fitness and population dynamics of small, fragmented populations of the long-lived perennial Primula vulgaris and how this was related to genetic diversity. Nested factors were included in the LTRE analysis in order to assess population-specific contributions to changes in population growth rate and to contrast these with contributions derived for year and type of disturbance. 3. Stochastic growth rates (lambda(s)) varied considerably among disturbance types: sites continuously disturbed throughout the year had a mean stochastic growth rate smaller than one, whereas sites receiving low or cyclic disturbance had stochastic growth rates larger than one. 4. LTRE analysis revealed that disturbance had a major impact on variation in growth rate, whereas year effects were less pronounced. However, contributions of individual populations equalled or exceeded those of the disturbance regime and mainly originated from negative or positive contributions of seedling recruitment and growth. 5. Linear regression analysis showed that these deviations of population demography from the disturbance type contribution were positively related to genetic diversity and effective population size, whereas inbreeding and isolation had no effect. Similar analyses revealed a positive effect of effective population size on lambda(s). 6. Our data indicate that disturbance has an important impact on population dynamics of this plant species, but that populations containing a higher number of reproductive adults are buffered against large fluctuations due to environmental stochasticity so that stochastic growth rates are higher. However, because population specific contributions to changes in population growth rate were as large as the contributions of disturbance itself, and because they were significantly related to population genetic diversity, our study suggests that, along with local and regional environmental variables, genetic diversity may also play a key role in determining population viability of perennial plant species.