Seasonal drying influences odonate adults and nymphs in wetlands in an urban mediterranean‐climate landscape.

Little is known about odonate ecology and phenology in warm mediterranean climates where most wetlands are intermittent. We identified variables associated with the presence, relative abundance, assemblage composition and body size of odonates in spring and summer, to understand the influence of intermittency and season. We hypothesized that size at eclosion would be smaller in seasonal wetlands in summer due to time stress caused by drying. Nymphs, exuviae and adults were sampled in spring and summer from 22 intermittent and perennial suburban wetlands in south‐western Australia. Spatial and environmental variables within and between wetlands were measured and associated with adult and nymph distributions. Exuviae were collected to quantify size at eclosion. As in temperate perennial wetlands, wetland‐scale variables (submerged, emergent and terrestrial vegetation, water temperature) were most strongly associated with assemblage composition of adults and nymphs, but landscape‐scale variables (distance to nearest patch of native vegetation, distance to nearest large lake) were also associated with adult assemblages. Two abundant dragonfly (Orthetrum caledonicum Libellulidae, Hemicordulia tau Hemicorduliidae) and one damselfly (Xanthagrion erythroneurum Coenagrionidae) species emerged at smaller body sizes in summer than spring, but from both intermittent and perennial wetlands. Consequently, declining photoperiod and warmer summer temperatures, rather than wetland drying, probably caused reduced size at eclosion. Although the influence of vegetation, temperature and photoperiod on odonate assemblages appears similar in temperate and mediterranean climates, odonate phenology differs markedly. Fitness cost of emerging at a smaller adult size may be outweighed by the increased likelihood of successfully reaching emergence in drying waterbodies in summer. More field data on size at eclosion in warm climate regions, and laboratory experiments manipulating temperature and photoperiod, are needed to confirm the generality of patterns shown here. [ABSTRACT FROM AUTHOR]