Field Trials of Line Transect Methods Applied to Estimation of Desert Tortoise Abundance

Anderson, David R.; Burnham, Kenneth P.; Lubow, Bruce C.; Thomas, Len; Corn, Paul Stephen; Medica, Philip A.; Marlow, Ronald W.

doi:10.2307/3803111

Cited by 80 publications

(76 citation statements)

References 21 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A way to ensure a shoulder (i.e., the shape criteria on the detection function) and hence robust estimation is to ensure adequate search effort at and close to zero distance. This can be checked during a pilot survey, and at early stage of data collection in the main survey, by examining histograms of the collected distances, and then adapting the field protocol as required (see, e.g., Anderson et al 2001). Therefore, appropriate field procedures should avoid spiked distance data, and observing spiked distance data is often an indication that an assumption might have been violated.…”

Section: Analysing Real-world Datamentioning

confidence: 99%

Estimation bias under model selection for distance sampling detection functions

2017

Self Cite

View full text Add to dashboard Cite

Many simulation studies have examined the properties of distance sampling estimators of wildlife population size. When assumptions hold, if distances are generated from a detection model and fitted using the same model, they are known to perform well. However, in practice, the true model is unknown. Therefore, standard practice includes model selection, typically using model comparison tools like Akaike Information Criterion. Here we examine the performance of standard distance sampling estimators under model selection. We compare line and point transect estimators with distances simulated from two detection functions, hazard-rate and exponential power series (EPS), over a range of sample sizes. To mimic the real-world context where the true model may not be part of the candidate set, EPS models were not included as candidates, except for the half-normal parameterization. We found median bias depended on sample size (being asymptotically unbiased) and on the form of the true detection [399][400][401][402][403][404][405][406][407][408][409][410][411][412][413][414] function: negative bias (up to 15% for line transects and 30% for point transects) when the shoulder of maximum detectability was narrow, and positive bias (up to 10% for line transects and 15% for point transects) when it was wide. Generating unbiased simulations requires careful choice of detection function or very large datasets. Practitioners should collect data that result in detection functions with a shoulder similar to a half-normal and use the monotonicity constraint. Narrow-shouldered detection functions can be avoided through good field procedures and those with wide shoulder are unlikely to occur, due to heterogeneity in detectability.123 400 Environ Ecol Stat (2017) 24:

show abstract

Section: Analysing Real-world Datamentioning

confidence: 99%

Estimation bias under model selection for distance sampling detection functions

2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Failure to detect a rare or cryptic species when it is actually present may be substantial (Gu and Swihart 2004), and for such species it is especially important to acquire accurate detection probabilities for inclusion in occupancy and population models to correct for survey bias. Unfortunately, many studies of amphibians and reptiles continue to use a count statistic or mark-recapture return rate uncorrected for detection probability (Mazarolle et al 2007 Gardner et al 1999]), as well as detection probabilities strongly affected by observer bias (Anderson et al 2001), habitat type (Gardner et al 1999), and time of year (Kéry 2002). These studies clearly illustrate the importance of adequately accounting for detectability in occupancy and population models for reptiles in order to make sound management and conservation decisions.…”

mentioning

confidence: 99%

“…Visual-encounter surveys are conducted by observers who move through a designated area and visually search for individuals of the target species, with data often expressed as individuals encountered per person-hours spent searching (Crump and Scott 1994). The effectiveness of visualencounter surveys can vary widely, however, depending on factors such as observer experience and training, habitat characteristics, and vegetation density within the survey area (e.g., Packard et al 1985, Gardner et al 1999, Anderson et al 2001.…”

mentioning

confidence: 99%

A generalized method to determine detectability of rare and cryptic species using the ornate box turtle as a model

Refsnider

Mitchell

Streby

et al. 2011

Wildlife Society Bulletin

View full text Add to dashboard Cite

Estimates regarding population parameters are often based on data from surveys. To ensure that such estimates are as accurate as possible, it is important to know the detectability resulting from the particular survey method used. We used radiotelemetry to measure detectability of ornate box turtles (Terrapene ornata), using visual-encounter surveys in a sand prairie in northwestern Illinois, USA. We found that the overall detection probability of visual-encounter surveys was 0.03, and our high frequency of nondetection was due to a failure to detect visible turtles rather than turtles being underground or hidden in dense vegetation. Despite the substantial population density at our study site, visual-encounter surveys failed to detect box turtles on most visits, which resulted in a prohibitively high number of surveys that would be required to accurately estimate population size or to infer absence of the species from a site. Our method of using radiotelemetry to measure detection probability of a survey method could be easily applied to other small, cryptic, or rare species. However, our low detection probability and high frequency of nondetections recommend against use of visual-encounter surveys alone in estimating population parameters for ornate box turtles. © 2011 The Wildlife Society. RightsWorks produced by employees of the U.S. Government as part of their official duties are not copyrighted within the U.S. The content of this document is not copyrighted. ABSTRACT Estimates regarding population parameters are often based on data from surveys. To ensure that such estimates are as accurate as possible, it is important to know the detectability resulting from the particular survey method used. We used radiotelemetry to measure detectability of ornate box turtles (Terrapene ornata), using visual-encounter surveys in a sand prairie in northwestern Illinois, USA. We found that the overall detection probability of visual-encounter surveys was 0.03, and our high frequency of nondetection was due to a failure to detect visible turtles rather than turtles being underground or hidden in dense vegetation. Despite the substantial population density at our study site, visual-encounter surveys failed to detect box turtles on most visits, which resulted in a prohibitively high number of surveys that would be required to accurately estimate population size or to infer absence of the species from a site. Our method of using radiotelemetry to measure detection probability of a survey method could be easily applied to other small, cryptic, or rare species. However, our low detection probability and high frequency of nondetections recommend against use of visual-encounter surveys alone in estimating population parameters for ornate box turtles. ß 2011 The Wildlife Society.

show abstract

“…In practical applications of distance sampling, underestimation or overestimation of density are frequently reported to be related to assumption violations, such as undetected animals on the transect (e.g., Anderson et al, 2001) or animal movement (e.g., Southwell, 1994;Langbein et al, 1999). However, I am not aware of observed bias being attributed to measurement errors, although several authors recognize its influence in the estimation procedures (e.g., Alpizar-Jara, 1997; Chen, 1998;Buckland et al, 2001).…”

Section: Discussionmentioning

confidence: 99%

Predicting and Correcting Bias Caused by Measurement Error in Line Transect Sampling Using Multiplicative Error Models

Marques¹

2004

Biometrics

View full text Add to dashboard Cite

Line transect sampling is one of the most widely used methods for animal abundance assessment. Standard estimation methods assume certain detection on the transect, no animal movement, and no measurement errors. Failure of the assumptions can cause substantial bias. In this work, the effect of error measurement on line transect estimators is investigated. Based on considerations of the process generating the errors, a multiplicative error model is presented and a simple way of correcting estimates based on knowledge of the error distribution is proposed. Using beta models for the error distribution, the effect of errors and of the proposed correction is assessed by simulation. Adequate confidence intervals for the corrected estimates are obtained using a bootstrap variance estimate for the correction and the delta method. As noted by Chen (1998, Biometrics 54, 899-908), even unbiased estimators of the distances might lead to biased density estimators, depending on the actual error distribution. In contrast with the findings of Chen, who used an additive model, unbiased estimation of distances, given a multiplicative model, lead to overestimation of density. Some error distributions result in observed distance distributions that make efficient estimation impossible, by removing the shoulder present in the original detection function. This indicates the need to improve field methods to reduce measurement error. An application of the new methods to a real data set is presented.

show abstract

Field Trials of Line Transect Methods Applied to Estimation of Desert Tortoise Abundance

Cited by 80 publications

References 21 publications

Estimation bias under model selection for distance sampling detection functions

Estimation bias under model selection for distance sampling detection functions

A generalized method to determine detectability of rare and cryptic species using the ornate box turtle as a model

Predicting and Correcting Bias Caused by Measurement Error in Line Transect Sampling Using Multiplicative Error Models

Contact Info

Product

Resources

About