Estimating tag loss of the Atlantic Horseshoe crab, Limulus polyphemus, using a multi-state model

Background/Question/Methods

The Atlantic Horseshoe crab, Limulus polyphemus,is a valuable resource along the Mid-Atlantic coast which has, in recent years, experienced new management paradigms due to increased concern about this species role in the environment. While current management actions are underway, many acknowledge the need for improved and updated parameter estimates to reduce the uncertainty within the management models. Specifically, updated and improved estimates of demographic parameters such as adult crab survival in the regional population of interest, Delaware Bay, could greatly enhance these models and improve management decisions.  There is however, some concern that difficulties in tag resighting or complete loss of tags could be occurring. As apparent from the assumptions of a Jolly-Seber model, loss of tags can result in a biased estimate and underestimate a survival rate. Given that uncertainty, as a first step towards estimating an unbiased estimate of adult survival, we first took steps to estimate the rate of tag loss.  Using data from a double tag mark-resight study conducted in Delaware Bay and Program MARK, we designed a multi-state model to allow for the estimation of mortality of each tag separately and simultaneously.  

---

Results/Conclusions

The added complexity of a multistate model, specifically the Psi parameter, allows for the detail that is necessary to tease apart tag specific loss rates.  A set if a piori models were chosen including the covariates of sex, stratum, time, and age. These were chosen to test the hypotheses that different tags or behavioral and physical differences in sex may affect detection or the life span of a tag. Our top model suggests that Males are 2.15 times as likely to be detected (.50-1.05 95% CI) and yet .79 times as likely to lose their main tag (-.53-.08 95% CI).   However, this model carried a weight of only .232 and 96% of the model weights lay across seven models. Due to the substantial model uncertainty, the estimates were determined using a model average. Model averaged estimates for Males range from .187 (SE=.055, CI=.102 to .319) in the first year to .016 (SE=.018, CI=.002 to .139) in the eighth year, meaning roughly an average of 7.5% are lost.  For Females the model averaged estimates range from .208 (SE=.057, CI=.117 to .342) in the first year to .018 (SE=.024, CI=.001 to .206), meaning roughly an average of 8.5% are lost.