Assessing bias in demographic estimates from joint live and dead encounter models

Abstract

Joint encounter (JE) models estimate demographic rates using live recapture and dead recovery data. The extent to which limited recapture or recovery data can hinder estimation in JE models is not completely understood. Yet limited data are common in ecological research. We designed a series of simulations using Bayesian multistate JE models that spanned a large range of potential recapture probabilities (0.01–0.90) and two reported mortality probabilities (0.10, 0.19). We calculated bias by comparing estimates against known probabilities of survival, fidelity and reported mortality. We explored whether sparse data (i.e., recapture probabilities <0.02) compromised inference about survival by comparing estimates from dead recovery (DR) and JE models using an 18-year data set from a migratory bird, the lesser snow goose (Anser caerulescens caerulescens). Our simulations showed that bias in probabilities of survival, fidelity and reported mortality was relatively low across a large range of recapture probabilities, except when recapture and reported mortality probabilities were both lowest. While bias in fidelity probability was similar across all recapture probabilities, the root mean square error declined substantially with increased recapture probabilities for reported mortality probabilities of 0.10 or 0.19, as expected. In our case study, annual survival probabilities for adult female snow geese were similar whether estimated with JE or DR models, but more precise from JE models than those from DR models. Thus, our simulated and empirical data suggest acceptably minimal bias in survival, fidelity or reported mortality probabilities estimated from JE models. Even a small amount of recapture information provided adequate structure for JE models, except when reported mortality probabilities were <0.10. Thus, practitioners with limited recapture data should not be discouraged from use of JE models. We recommend that ecologists incorporate other data types as frequently as analytically possible, since precision of focal parameters is improved, and additional parameters of interest can be estimated.

Local population collapse of Ross's and lesser snow geese driven by failing recruitment and diminished philopatry

“…We estimated survival and fidelity probabilities using a multistate joint live and dead encounter model (Burnham 1993, Weegman et al 2020). We assumed that adult reported mortality probability, Ra t , and that of juveniles, Rj t , as well as capture probability, P t , were temporal random effects such that logit( Ra t ) followed a Normal distribution with mean µ [ Ra ] and standard deviation σ [ Ra ] , logit( Rj t ) followed a Normal distribution with mean µ [ Rj ] and standard deviation σ [ Rj ] , and logit( P t ) followed a Normal distribution with mean µ [ P ] and standard deviation σ [ P ] ; µ was modelled as a linear temporal trend similarly to survival and fidelity parameters.…”

“…We used individual encounter history data, in which the likelihood in year t was determined by status in the previous year as well as Sa t , Sj t , Fa t , Fj t , Ra t and Rj t , and encounter data were determined by P t . Hence, our model definition was similar to Weegman et al (2020, Boxes 1 and 2), but with an age‐specific parameterization. A multi‐state Burnham model is usually implemented by applying state transition matrices to individual capture histories (Kéry and Schaub 2012).…”

“…Therefore, we developed a novel approach specific to a multi‐state Burnham model, where capture histories were summarized in two extended m‐arrays (for adults and juveniles), with both alive and dead state transitions. Our m‐arrays were an expanded version of those used for recaptures or recoveries (Burnham et al 1987, Williams et al 2002, Weegman et al 2020). The m‐arrays covered all transitions between observable states; latent (unobservable) states were marginalized by summing over probabilities of all possible latent state combinations.…”

“…To test these hypotheses, we estimated survival, per capita production of young, emigration probability and immigrant numbers, as well as population growth rates for Ross's and snow geese in an IPM. Information for our IPM included joint live and dead encounters (Weegman et al 2020), production of young (Ross et al 2017(Ross et al , 2018 and breeding population estimates (Alisauskas et al 2012a, Wilson et al 2016. We compared demographic components between species to quantify their synchrony.…”

Missing Release Data in Capture-Mark-Recovery Analyses: Consequences for Inference

“…For example, if survival and recovery probabilities between adults and juveniles are highly correlated but juvenile release data are missing in a year, the direction and magnitude of adult probabilities could be used to help inform juvenile estimates during the year of no releases. Other analysis methods include the use of models that incorporate additional information like recapture data for the Burnham joint live-dead recovery model (Burnham 1993); however, sufficient recapture data are needed for adequate model fit and to improve estimate precision (Koons et al 2019;Weegman et al 2020). Integrated population models (Kéry and Schaub 2012) might be useful where both abundance and productivity data are available to help estimate demographic parameters when band-recovery data are missing.…”