Assessing Persistence of Two Rare Darter Species Using Population Viability Analysis Models

Abstract

The objective of this study was to assess whether results predicted from population viability analysis (PVA) models resemble historical and current presence/absence data for Etheostoma boschungi and Etheostoma brevirostrum. Etheostoma boschungi is a migratory species and has two distinct, but adjacent habitats: non-breeding and breeding. Etheostoma boschungi spawn from late February to late March and are known from 30 sites in tributaries of the Tennessee River drainage of Alabama and Tennessee. Etheostoma brevirostrum prefers rocky runs and riffles with a fast current and spawn from early April to late May. Etheostoma brevirostrum is known to occur in disjunct populations located in tributaries of the Coosa River drainage and are believed to live three years. We used information on fecundity, life span, and population size to determine fertility and survival vital rates for a three-stage, pre-breeding, PVA model for each species. The age-structured model used Leslie projection matrices to calculate deterministic population growth rates (?) and relative elasticities of the vital rates. The population size for E. boschungi decreases when adult fertility rates are below 0.896. The population size for E. boschungi falls below one individual after five years when ? = 0.309 for a one-batch fecundity and after five years when ? = 0.322 for a two-batch fecundity. The E. boschungi population size increases to more than 1,000 individuals after ten years when ? = 1.158 and after nine years when ? = 1.184 for one-batch and two-batch fecundities, respectively. The population size for E. brevirostrum decreases when adult fertility rates are below 0.737, 0.929, and 0.818 for the upstream, downstream, and combined populations, respectively. Etheostoma brevirostrum population size projected for upstream, downstream, and combined stream segments (using E. coosae and E. pyrrhogaster survivals) resulted in more than 50 times the initial population size after one year. As long as ? < 1, then the population will decrease but when ? > 1, then the population increases. The elasticity analysis of the matrix indicated the fertilities made the largest relative contribution to ?, while the survivals were smaller for all sites and for both species. Despite some of the limitations in gathering data for rare species, we believe PVA models are useful for studying fishes. State agencies should focus on improving the habitat within each of the watersheds to improve survival of the juveniles to year one, in addition to monitoring the watershed more closely for possible land use or pollution impacts.