Placing butterflies on the map – testing regional geographical resolution of three stable isotopes in Sweden using the monophagus peacock Inachis io

Abstract

Listen

Translate article

Stable isotope analyses of tissues have been used to help delineate natal regions and routes of migratory animals. The foundations of such studies are isotopic gradients or differences representing geographic regions and habitat used by the organism that are retained in selected tissues. We sampled peacock butterflies Inachis io on a regional level in southern Sweden to study natural variation and the resolving power of the stable isotope method to delineate individuals from known areas on a smaller scale than has typically been used in previous studies. Hydrogen (δD), carbon (δ13C) and nitrogen (δ15N) isotopes were obtained from butterflies at seven different locations in an area of 250×250 km over three years (2002–2004). We found sufficient isotopic differences on this regional scale to delineate approximate origins. Of the three isotopes, deuterium showed good discrimination between sites, carbon isotopes showed weaker differentiation, whereas nitrogen isotopes proved unsuitable for small scale studies in this region due to high and unpredictable variation. We found there was enough variation in δD between years to prevent a general application of the technique to resolve sub‐regional variation. Substantial part of this variation was probably caused by seasonal changes in δD of precipitation. These differences produce significant variation in δD between years in animals having short and variable tissue development times, and are difficult to estimate in natural situations. We conclude that stable isotopes are potentially powerful predictors for studies of migratory butterflies in Europe. However, without good knowledge about the sampled individuals’ previous life‐history, a lot of the natural environmental variation in tissue δD cannot be controlled for. In the case of migratory species, this information is difficult to obtain, making the confidence intervals for prediction of natal areas fairly wide and probably only suitable for longer distance migration.