• demography;
  • Exxon Valdez;
  • harlequin duck;
  • Histrionicus histrionicus;
  • known fates model;
  • oil spill;
  • radiotelemetry;
  • survival;
  • winter

ABSTRACT  In the mid- to late 1990s, nearly a decade after the 1989 Exxon Valdez oil spill, female harlequin ducks (Histrionicus histrionicus) suffered reduced winter survival in oiled areas of Prince William Sound, Alaska, USA, relative to unoiled areas. We conducted follow-up studies from winters 2000–2001 to 2002–2003 to determine whether differential survival persisted and to evaluate whether individual-level indices of oil exposure were related to survival. Using radiotelemetry, we tracked 138 female harlequin ducks from November through March over three winters. We analyzed variation in survival in relation to season, area oiling history, age class, body mass, and an index to exposure to residual oil based on cytochrome P4501A (CYP1A) induction. We determined that survival was most strongly related to season and age class, with evidence of higher survival in late winter and after hatch year (AHY) categories, respectively. We estimated cumulative winter survival for AHY females to be 0.837 (±0.064) and 0.834 (±0.065) on unoiled and oiled areas, respectively, and we estimated hatch-year female cumulative winter survival at 0.766 (±0.138) on unoiled areas and 0.758 (±0.152) on oiled areas. Despite persistence of oil in some intertidal areas and evidence of contaminant ingestion by harlequin ducks during and beyond this study, neither area nor CYP1A were strongly related to variation in survival, suggesting that direct effects of the oil spill on harlequin duck demography had largely abated by the winters 2000–2001 to 2002–2003. Our findings offer an unprecedented description of the timeline of effects of exposure to spilled oil and contribute to a body of literature that describe demographic effects of the Exxon Valdez oil spill that persisted over a much longer time than previously assumed. An appreciation for the timescale of chronic effects of oil spills, as well as potential for demographic effects related to much lower concentrations of oil than during the immediate period of acute effects following a spill, will provide wildlife managers with a basis for risk assessment and plans for mitigation when confronted with large spills or chronic pollution.