Model-Based Distance Sampling

Buckland, Stephen T.; Oedekoven, C. S.; Borchers, David L.

doi:10.1007/s13253-015-0220-7

Cited by 30 publications

(31 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the product of p p and p a ) which is the compliment of the temporary emigration probability (Chandler et al., ):

N_{j, k} \sim Binomial false(M_{j}, normalϕ false)

. Models for time‐ and plot‐dependent ϕ could also be considered including random effects parameterizations which account for over‐dispersion and/or temporal correlation in repeated counts (Buckland, Oedekoven, & Borchers, ). Here, we assume ϕ is constant for simplicity of presentation.…”

Section: Methodsmentioning

confidence: 99%

“…The hierarchical distance sampling (HDS) model of Royle et al. () and other standard model‐based formulations (Buckland et al., ) imply that the within‐plot distribution of individuals is uniform. In the HDS model, the observed counts follow a multinomial distribution where the vector of cell probabilities is the product of detection probability and of occurring in a particular distance class, which is specified as uniform across distance classes (adjusted for annulus area in the point transect case).…”

Section: Methodsmentioning

confidence: 99%

“…We included the linear and quadratic effects of survey date to account for variation in the probability of availability over a survey period that began prior to the arrival of a subset of individuals and then extended past peak singing (Kéry, Royle, Plattner, & Dorazio, ; Schmidt et al., ). Lastly, we included a survey‐level random effect, δ j,k , to accommodate unexplained variation in the model (Buckland et al., ; Oedekoven, Laake, & Skaug, ), in part due to the effects of weather. For the smaller American tree sparrow dataset, we fit a reduced model that included only the fixed effects of time and effort.…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Accommodating temporary emigration in spatial distance sampling models

Mizel

Schmidt

Lindberg

2017

Journal of Applied Ecology

View full text Add to dashboard Cite

Model‐based distance sampling is commonly used to understand spatial variation in the density of wildlife species. The standard approach assumes that individuals are distributed uniformly and models spatial variation in density using plot‐level effects. Thinned point process (TPP) models for surveys of unmarked populations (spatial distance sampling) better leverage the spatial information underlying individual encounters, and in the presence of within‐plot variation in density, may explain a larger proportion of the spatial variation in density. However, existing spatial distance sampling approaches are conditioned on the assumption that all individuals are present and available for sampling. Temporary emigration of individuals can therefore result in biased estimates of abundance. We extended spatial distance sampling models to accommodate temporary emigration (TPP model). Using simulations of a thinned inhomogeneous point process, we assessed the performance of the TPP model relative to the temporary emigration distance sampling (TEDS) model, which implies a uniform distribution of individuals. In addition, we compared inferences between TPP and TEDS models using data for two passerine species in Alaska. Parameter estimates from the TPP model exhibited improved coverage probability and precision relative to the TEDS model including a 26% reduction in the coefficient of variation (CV) of the population size estimate. In the applied example, the TEDS model indicated weak relationships between abundance and habitat covariates, whereas the TPP model indicated strong relationships for those same effects, suggesting that spatial distance sampling models can provide considerably stronger inference in the presence of within‐plot variation in density. In addition, the CV of the population size estimates for the two passerine species were 32% and 4% smaller under the TPP model. Synthesis and applications. We expect our extension accommodating temporary emigration will be a critical specification for spatial distance sampling models, particularly for studies assessing changes in the distribution and abundance of highly mobile species including passerines.

show abstract

“…the product of p p and p a ) which is the compliment of the temporary emigration probability (Chandler et al., ):

N_{j, k} \sim Binomial false(M_{j}, normalϕ false)

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Accommodating temporary emigration in spatial distance sampling models

Mizel

Schmidt

Lindberg

2017

Journal of Applied Ecology

View full text Add to dashboard Cite

show abstract

“…Although most distance sampling methods are not fully model-based (i.e based on a model of the sort outlined above), there is an increasing trend towards fully modelbased inference (see Buckland et al 2016) and that is what we focus on here. Johnson et al (2010) is a recent example of this approach.…”

Section: Distance Sampling As a Thinned Point Processmentioning

confidence: 99%

“…This is the form in which CDS likelihoods appear in the DS literature (see Appendix), although more commonly with binomial rather than Poisson f (n) (see Buckland et al 2016). …”

Section: Distance Sampling As a Thinned Point Processmentioning

confidence: 99%

From distance sampling to spatial capture–recapture

Borchers

Marques

2017

AStA Adv Stat Anal

View full text Add to dashboard Cite

Distance sampling and capture-recapture are the two most widely used wildlife abundance estimation methods. capture-recapture methods have only recently incorporated models for spatial distribution and there is an increasing tendency for distance sampling methods to incorporated spatial models rather than to rely on partly design-based spatial inference. In this overview we show how spatial models are central to modern distance sampling and that spatial capture-recapture models arise as an extension of distance sampling methods. Depending on the type of data recorded, they can be viewed as particular kinds of hierarchical binary regression, Poisson regression, survival or time-to-event models, with individuals' locations as latent variables and a spatial model as the latent variable distribution. Incorporation of spatial models in these two methods provides new opportunities for drawing explicitly spatial inferences. Areas of likely future development include more sophisticated spatial and spatio-temporal modelling of individuals' locations and movements, new methods for integrating spatial capture-recapture and other kinds of ecological survey data, and methods for dealing with the recapture uncertainty that often arise when "capture" consists of detection by a remote device like a camera trap or microphone.

show abstract

Using spatial distance sampling models to optimize survey effort and address violations of the design assumption

Schmidt

Deacy

2021

Ecol Sol and Evidence

View full text Add to dashboard Cite

1. Conventional distance sampling approaches rely on the design assumption (i.e. uniform distribution of individuals in relation to transects) to ensure unbiased inference to the population of interest. However, randomized design recommendations are not always followed or may be impractical to implement for some survey types, particularly in cases where transects must be placed perpendicular to the habitat gradient. Full-likelihood spatial distance sampling models provide a potential solution to violations of the design assumption by jointly modelling the detection and occurrence processes using spatially indexed habitat covariates. Through simulation and an applied example based on a survey for Dall's sheep inAlaska, USA, we used a full-likelihood distance sampling approach to investigate the potential for bias in cases where transects placed perpendicular to the habitat gradient (e.g. elevational contours) are non-randomly sampled. We also assessed the utility of spatial approaches in cases where transects are placed along linear features, such as roads or ridgelines, where habitat may be unrepresentative of the overall study area.3. Our results showed that the full-likelihood approach was generally unbiased, even in extreme scenarios where habitat was inversely related to distance from the transect. For the Dall's sheep example, our results showed that more efficient designs with reduced sampling effort in low-quality habitats are a practical solution for reducing logistical costs when the data are analysed in a spatial modelling framework.4. Together, our findings confirm and extend existing work suggesting that spatial distance sampling can be a useful solution when non-random designs are employed.Given the high cost of survey implementation in many cases, the development of valid alternatives to design-based inference will aid in the amount of information available for a variety of species. The results of our work will be useful for practitioners in assessing alternative designs relative to particular survey applications.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

show abstract

Model-Based Distance Sampling

Cited by 30 publications

References 20 publications

Accommodating temporary emigration in spatial distance sampling models

Accommodating temporary emigration in spatial distance sampling models

From distance sampling to spatial capture–recapture

Using spatial distance sampling models to optimize survey effort and address violations of the design assumption

Contact Info

Product

Resources

About