Testing assumptions for unbiased estimation of survival of radiomarked harlequin ducks

Unbiased estimates of survival based on individuals outfitted with radiotra nsmitters require meeting the assumptions that radios do not affect surviva l, and animals for which the radio signal is lost have the same survival pr obability as those for which fate is known. In most survival studies, resea rchers have made these assumptions without testing their validity. We teste d these assumptions by comparing interannual recapture rates (and, by infer ence, survival) between radioed and unradioed adult female harlequin ducks (Histrionicus histrionicus), and for radioed females, between right-censore d birds (i.e., those for which the radio signal was lost during the telemet ry monitoring period) and birds with known fates. We found that recapture r ates of birds equipped with implanted radiotransmitters (21.6 +/- 3.0%; (x) over bar +/- SE) were similar to unradioed birds (21.7 +/- 8.6%), suggesti ng that radios did not affect survival. Recapture rates also were similar b etween right-censored (20.6 +/- 5.1%) and known-fate individuals (22.1 +/- 3.8%), suggesting that missing birds were not subject to differential morta lity. We also determined that capture and handling resulted in short-term l oss of body mass for both radioed and unradioed females and that this effec t was more pronounced for radioed birds (the difference between groups was 15.4 +/- 7.1 g). However, no difference existed in body mass after recaptur e 1 year later. Our study suggests that implanted radios are an unbiased me thod for estimating survival of harlequin ducks and likely other species un der similar circumstances.