Incorporating Covariates into Standard Line Transect Analyses

Abstract

An implicit assumption of standard line transect methodology is that detection probabilities depend solely on the perpendicular distance of detected objects to the transect line. Heterogeneity in detection probabilities is commonly minimized using stratification, but this may be precluded by small sample sizes. We develop a general methodology which allows the effects of multiple covariates to be directly incorporated into the estimation procedure using a conditional likelihood approach. Small sample size properties of estimators are examined via simulations. As an example the method is applied to eastern tropical Pacific dolphin sightings data.

Estimating abundance of killer whales in the nearshore waters of the Gulf of Alaska and Aleutian Islands using line-transect sampling

“…Both Conventional Distance Sampling (CDS) and Multiple Covariate Distance Sampling (MCDS) approaches were used (Buckland et al 2001(Buckland et al , 2004Marques and Buckland 2003). MCDS diVers from CDS because it allows for the inclusion of environmental covariates in the estimation of detection probability (P).…”

“…MCDS diVers from CDS because it allows for the inclusion of environmental covariates in the estimation of detection probability (P). Half normal and hazard rate functions were used to model P, and covariates were incorporated via the scale parameter as described by Innes et al (2002) and Marques and Buckland (2003). Models were proposed to investigate the eVects of covariates in the probability of detecting killer whale groups, but the small sample of on-eVort sightings precluded the use of more than one covariate.…”

“…Ecotype-speciWc abundance and variance estimates were obtained for each proposed model to Wt perpendicular distance data as described by Innes et al (2002) and Marques and Buckland (2003). Models were ranked according to the Akaike Information Criterion, which provides a measure of model Wt with a penalty term for the number of parameters in the model (Akaike 1973).…”

The wild life of tick-borne pathogens

“…We fitted both simple distance-sampling models without covariates on single species in each habitat, and multiple-covariate distance-sampling models, including body mass and habitat as covariates (see below) 176 . We selected the best-performing models based on AIC values as estimates of model quality 172 .…”

Monitoring marine populations and communities: methods dealing with imperfect detectability

“…cluster size or individual size), and can be either continuous or qualitative factors. Marques & Buckland (2003) proposed a methodology to incorporate covariates into the estimation of the detection probabilities, which has been implemented in DISTANCE.…”