Accommodating Unmodeled Heterogeneity in Double‐Observer Distance Sampling Surveys

Abstract

Mark-recapture models applied to double-observer distance sampling data neglect the information on relative detectability of objects contained in the distribution of observed distances. A difference between the observed distribution and that predicted by the mark-recapture model is symptomatic of a failure of the assumption of zero correlation between detection probabilities implicit in the mark-recapture model. We develop a mark-recapture-based model that uses the observed distribution to relax this assumption to zero correlation at only one distance. We demonstrate its usefulness in coping with unmodeled heterogeneity using data from an aerial survey of crabeater seals in the Antarctic.

The effect of sea-ice loss on beluga whales (Delphinapterus leucas) in West Greenland

“…Detection probability was estimated using the "point independence" method with independent observer configuration (Laake & Borchers 2004;Borchers et al 2006), implemented in DISTANCE 5.0 (Thomas et al 2006). This method involved estimating a multiple covariate distance sampling model (mcds model) for combined platform detections assuming certain detection at distance zero, and a mark-recapture distance sampling model (mrds model) for each of the independent observers using a logistic detection function model.…”

Estimating the migratory stopover abundance of phalaropes in the outer Bay of Fundy, Canada

“…In the absence of mark-recapture methods, it is difficult to gauge whether either scenario was severe enough to cause genuine nonuniformity in our data (Borchers et al 2006). We speculate that a mixture was at play and a potential outcome is that our estimators for detection probability are positively biased.…”

Model-Based Distance Sampling

“…Further, p 1 (y, z) is the unconditional probability that observer 1 detects an animal at distance y and covariates z, while p 1|2 (y, z) is the probability of detection, conditional on the animal having been detected by observer 2, with equivalent expressions for observer 2. Models for these probabilities are proposed by Laake and Borchers (2004), Borchers et al (2006) and Buckland et al (2010).…”