Population management modeling for the Puritan tiger beetle

Population management modeling uses mathematical models to forecast risk for imperiled species under alternative management scenarios. I used that approach to inform decision-making for the Puritan tiger beetle, a federally listed threatened species, on the Connecticut River. I found that density of larvae in the Cromwell-Portland metapopulation is greater in areas with predominantly fine sand. Along the lowermost 110km of the river, a cluster of beaches in Windsor, Connecticut most closely resembles the areas where there is high density. I conducted a mark-recapture study in the Cromwell-Portland metapopulation to measure movement among habitat patches. We marked approximately 600 beetles in both 1999 and 2000 with about 30% resighting rate and 4–6% having moved among patches. Encounter histories were analyzed to estimate daily survival, movement, and resighting probabilities. I then extrapolated those daily probabilities to net movement using computer simulations. Net movement rates between pairs of habitat patches separated by about 1.4km was about 3% (ind/ind/generation) in 1999 and 12% in 2000, net movement between pairs of beaches separated by 2.7km was 0.4% in 1999 and 3% in 2000. I used maximum likelihood methods to analyze count data from the Cromwell-Portland metapopulation. Among seven models that I fit to the time series, model selection criteria indicated that the data support a simple exponential model, a model with isotropic movement, or a model with density-dependence. Regardless of the model selected, the count data reveal positive population growth (r0 ≅ 0.16 for the exponential model, 0.04 for the model with movement, and 0.2 or higher for the model with density-dependence). Finally, I compared the risk of decline within metapopulations and the chance of increased regional abundance over the next thirty years under five management alternatives. The results suggested that translocating 100 Puritan tiger beetle larvae to Windsor each of the next two years is unlikely to exacerbate the risk of decline in the source metapopulation and is more likely to lead to there being three large metapopulations than improving habitat in the source metapopulation or augmenting another, smaller metapopulation.