Comparison of methods for estimating wild turkey poult survival

Abstract Eastern wild turkey ( Meleagris gallopavo silvestris ) population growth is driven by variation in reproduction, and poult survival is a crucial vital rate. Traditional poult survival monitoring uses flush counts of radio‐tagged females, but brood flocking (when broods and females merge into larger flocks) and mark retention are potential problems with this method and can produce biased estimates. Alternatively, radio‐tagging poults to estimate survival offers a potentially less‐biased method if tagging and tracking poults can be conducted without adverse effects on the poults. We compared poult survival estimates derived from flush counts and from radio‐tagged poults over a 56‐day monitoring period during 2018–2022 in south‐central Tennessee, USA. We radio‐tagged 183 poults and monitored 85 broods via 2‐week flush counts and observed 572 individual poults during flush counts over the 56‐day monitoring period. We used known‐fate survival models for the telemetry data, Lukacs survival models for the flush count data, and a modified known‐fate model to estimate whether radio‐tagging poults affected their survival. Two‐week survival estimates from the flush count of broods, and radio‐tagged poults were similar (flush count: day 0–28: 0.25, 95% CI: 0.20–0.31, day 29–56: 0.56, 95% CI: 0.36–0.76; radio‐tagged: day 0–28: 0.30, 95% CI: 0.23–0.37, day 29–56: 0.68, 95% CI: 0.49–0.87). Using known‐fate survival models, radio‐tagged poults had similar survival rates to non‐tagged poults within the brood (day 0–28: Ŝ Radio‐tagged = 0.15, 95% CI: 0.10–0.20 v. Ŝ Non‐tagged = 0.20, 95% CI: 0.16–0.23), which suggests the radio‐tags did not directly affect survival. However, survival of individuals that were trapped was lower than survival of broods that were not trapped during the first 14 days. Trapping methods may need to be adjusted to limit incidental capture effects when choosing to use radio‐tagged poults. Conducting flush counts every 14 days until day 56 of the brood‐rearing period likely is the most cost‐effective method for estimating poult survival. However, monitoring individual radio‐tagged poults may yield additional information, such as cause‐specific mortality, that may better inform management.