Spatial distribution, life history and estimates of survivorship in a fourteen‐species assemblage of larval dragonflies (Odonata: Anisoptera)

Abstract

SUMMARY. 1. Spatial and temporal changes in larval densities were used to infer patterns of habitat use and survivorship in a fourteen‐species assemblage of dragonflies (Odonata: Anisoptera) in a small fishless pond. The density of all species combined peaked at >1000 m−2 in late summer, Most species (e.g. Libellula spp.) were restricted to shallow, nearshore habitats (<1.0m in depth), but a few (e.g. Epitheca spp.) also used deeper areas of the pond. Only Perithemis tenera was most abundant in deep habitats.2. Because many species exhibited temporal shifts in their use of habitats, it was necessary to estimate survival from changes in population size, calculated as the product of density and habitat area, summed across habitats. In most species, periods of high mortality in autumn and spring were separated by 3–4 months of negligible mortality in winter. Survivorship was linear only in the two species that completed all of larval development in summer (Sympetrum vicinum and Pantala flavescens). Average survival rates for these two species (−0.0049 and −0.0079 log density d−1) were similar to those in previous studies (Lawton, 1970; Benke & Benke, 1975).3. Survivorship in many species was confounded by other life history phenomena such as (i) mixed voltinism, (ii) overlapping migrant and resident cohorts, and (iii) asynchronous development within species. Asynchrony made it difficult to estimate initial and final population sizes, hence total larval survivorship. However, based on emergence data, only 0.4–3% of larvae survived after peak abundance. None of this mortality can be ascribed to vertebrate predation, and only a little to overwintering stress and starvation. Thus, predation by invertebrates might play a major role in the regulation of these populations.