Less input same output: simplified approach for population size assessment in Lepidoptera

With the aim of creating a simplified sampling scheme that would retain the accuracy of standard mark–release–recapture (MRR) sampling, but at a greatly reduced cost, we analysed 23 capture–recapture data sets from spatially closed populations of six Lepidoptera species according to the constrained Cormack–Jolly–Seber models. Subsequently the relationships between the estimates of population parameters were investigated in order to develop a regression equation that would enable us to calculate seasonal population size without sampling the population throughout the entire flight period. The proportion of individuals flying at peak population was highly variable (CV=0.39), but the variation decreased considerably (CV=0.14) after different life span and flight period length were accounted for. Over 90% of the variance of this proportion was explained by the life span:flight period length ratio. Simulations of hypothetical sampling schemes proved that schemes covering the second and third quarter of the flight period performed much better than those restricted to the second quarter only. The accuracy of seasonal population size estimated with the regression equation developed was comparable for intensive schemes (daily sampling) and non-intensive ones (sampling once in 2 or 3 days). We propose a simplified method of surveying butterfly populations that should be based on checking the presence of flying adults at the beginning and end of the flight period to assess its length, and MRR sampling covering its middle part, with intervals between capture days corresponding to the average life span of investigated butterflies.