Measurement Errors in Line Transect Surveys Where Detectability Varies with Distance and Size
- 1 September 2001
- journal article
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 57 (3) , 732-742
Abstract
Summary. When using bivariate line transect methods to estimate the biomass density of a tightly clustered biological population, it is generally assumed that both the perpendicular distance from the trackline to the cluster and the cluster size, or biomass, are measured without error. This is unlikely to be the case in practice. In this article, assuming additive mean zero errors in distance and multiplicative errors in size, we develop an estimator of density that corrects for these errors. We use the method of moments for the case of gamma cluster size, randomly placed transect lines, and the generalized exponential detection function. We derive results that show that it may not be necessary to correct for errors in distance or size when the distance and size estimates are not biased. When the size estimates are biased, the biomass density estimate has approximately the same bias as the size estimates. The work is illustrated in the context of annual aerial surveys for juvenile southern bluefin tuna in the Great Australian Bight.Keywords
This publication has 13 references indexed in Scilit:
- Measurement Errors in Line Transect Surveys1998
- Kernel Methods in Line and Point Transect SamplingBiometrics, 1998
- A combination line transect and capture-recapture sampling model for multiple observers in aerial surveysEnvironmental and Ecological Statistics, 1996
- Studying School Size Effects in Line Transect Sampling Using the Kernel Method1996
- Maximum Likelihood Estimation for the Double-Count Method with Independent ObserversJournal of Agricultural, Biological and Environmental Statistics, 1996
- Measurement Error in Nonlinear Models1995
- Simulation-Extrapolation Estimation in Parametric Measurement Error ModelsJournal of the American Statistical Association, 1994
- Optimal Rates of Convergence for Deconvolving a DensityJournal of the American Statistical Association, 1988
- Size Bias in Line Transect SamplingBiometrics, 1987
- A Family of Density Estimators for Line-Transect Sampling1978