Demographic Viability of Populations of Silene regia in Midwestern Prairies: Relationships with Fire Management, Genetic Variation, Geographic Location, Population Size and Isolation

1 We studied the demographic viability of populations of a long-lived iteroparous prairie perennial, Silene regia, in relation to management regimes, population sizes, geographical region (Ohio and Indiana vs. Missouri and Arkansas), degree of isolation and amount of genetic variation. Demographic data were collected from 16 populations for up to 7 years. 2 This species has high survivorship, slow growth, frequent flowering and episodic seedling recruitment. Matrix projection methods were used to summarize population performance with and without recruitment. Median finite rates of increase by population varied from 0.57 to 1.82 and from 0.44 to 0.99, respectively. 3 Populations with the highest rates of increase had been burned. Six of eight populations, for which stochastic modelling predicted persistence for 1000 years, included fire in their management. None of the five populations with predicted 100-year extinction probabilities of 100% was managed for conservation or burned. An intermediate group of three populations with at least 10% probability of extinction between 100 and 1000 years was not managed, but was none the less kept open by mowing and herbicide application. 4 Analysis of composite elasticities showed that growth and fecundity terms were higher for growing (vs. declining) populations and that growth elasticity was higher in burned than unburned populations. Lack of burning shifts the elasticity spectrum from that typical of open habitat herbs (higher growth and fecundity elasticities) to values usually found for closed habitat herbs (higher survival elasticities). 5 In multivariate analyses predicting finite rates of increase (with and without recruitment), fire management and region were the strongest predictors, followed by genetic variation, population size, isolation and interactions of population size and fire, and region and fire. Populations with the highest rates of increase were burned, eastern, more genetically diverse, larger and less isolated. Discrimination of populations with different extinction risks (three classes) was related mainly to fire, genetic variation and region. 6 Most of these conclusions support conservation biology predictions that population viability will be highest in larger, less-isolated, more genetically diverse populations. However, management and geographic trends have overriding roles affecting demographic viability. Habitat fragmentation and genetic depletion have the potential to threaten residual