Dispersal is one of the key elements of species’ metapopulation dynamics and, hence, influences global conservation status. Furthermore, determining the geographic variation in magnitude and direction of dispersal throughout a species’ distribution may expand our understanding of the causes of population declines in species of conservation concern. For instance, western burrowing owl (Athene cunicularia hypugaea) populations have declined at the northern and eastern edge of their breeding distribution during the 20^th century. In the same period, large areas of thornscrub that did not support breeding owls were converted to irrigated agriculture in the southern edge of the subspecies’ breeding distribution in northwestern Mexico. These farmlands now support some of the highest breeding densities of owls. We tested the hypothesis that owls that colonized this recently created habitat originated from declining migratory populations from the northern portion of the subspecies’ range. We used stable isotopes ²H, ¹³C, and ¹⁵N in owl feathers to infer breeding dispersal patterns throughout the subspecies’ breeding range. Populations near the northern edge of the subspecies’ breeding range had immigrants that dispersed over larger distances than immigrants at low and mid latitude populations. However, agricultural populations in northwestern Mexico disrupted this latitudinal pattern, attracting owls from more distant locations. We also found immigrants originated from further distances in declining populations than increasing populations. Stable isotopes provided no evidence of contemporaneous breeding dispersal from Canadian populations to northwestern Mexico but suggest that agricultural areas in the southern edge of the subspecies’ distribution have altered the continental dispersal pattern.

We collected feathers from nestling and adult burrowing owls during the breeding seasons of 2004-2009 at 36 study locations throughout the species’ breeding range in Canada, Mexico, and the United States. We pulled breast, back, and head feathers from nestlings that were 10-40 days-old, and we pulled the third right rectrix from adult burrowing owls. We processed all our samples in the Environmental Isotope Laboratory at the University of Arizona. We estimated the non-exchangeable δ²H in burrowing owl feathers using a modification of the “comparative equilibrium” technique of Wassenaar and Hobson (2003, Isot. Environ. Healt. S. 39:211–217) by Fan and Dettman (2014, Isot. Environ Health Stud. 51(2):214-30) as described by Macías-Duarte and Conway (2015, Auk 132:25-36). We measured δ¹³C and δ¹⁵N on a continuous-flow gas-ratio mass spectrometer (Finnigan Delta PlusXL). We combusted samples with added oxygen in an elemental analyzer (Costech) coupled to the mass spectrometer. Standardization was based on acetanilide for elemental concentration, NBS-22 and USGS-24 for δ¹³C, and IAEA-N-1 and IAEA-N-2 for δ¹⁵N. Precision based on repeated internal standards was better than 0.08‰ for δ¹³C and better than 0.2‰ for δ¹⁵N. Values of δ²H, δ¹³C, and δ¹⁵N are computed for the Vienna Standard Mean Ocean Water standard, PeeDee Belemite standard, and atmospheric N₂, respectively.

File isotope_data.csv contains raw isotope ratio measurements (permil) for individual burrowing owls. File locations_coordinates.csv contains longitud and latitude of study locations. File sample_info.csv contains study location and year of collection, age and sex for each individual burrowing owl.