Environmental and Ecological Statistics 3, 311-327 (1996) A combination line transect and capture-recapture sampling model for multiple observers in aerial surveys RUSSELL ALPiZAR-JARA and KENNETH H. POLLOCK Biomathematics Graduate Program, Department of Statistics, North Carolina State University, Box 8203, Raleigh, NC 27695-8203, USA Received November 1995. Revised August 1996 We present a robust sampling methodology to estimate population size using line transect and capture-recapture procedures for aerial surveys. Aerial surveys usually underestimate population density due to animals being missed. A combination of capture-recapture and line transect sampling methods with multiple observers allows violation of the assumption that all animals on the centreline are sighted from the air. We illustrate our method with an example of inanimate objects which shows evidence of failure of the assumption that all objects on the centreline have probability 1 of being detected. A simulation study is implemented to evaluate the performance of three variations of the Lincoln-Petersen estimator: the overall estimator, the stratified esti- mator, and the general stratified estimator based on the combined likelihood proposed in this paper. The stra- tified Lincoln-Petersen estimator based on the combined likelihood is found to be generally superior to the other estimators. Keywords: detection function, program DISTANCE, program SURVIV, visibility bias, wildlife population density estimation 1. Introduction Traditionally, capture-recapture methods and line transect sampling methods have often been used to estimate the density of animal populations. For detailed examples and practical applications of capture-recapture analysis with closed population models see Otis et al. (1978), White et al. (1982), and Pollock et al. (1990). For a detailed discussion of the line transect theory, exposition of the design, and several comprehensive examples of its application see Burnham et al. (1980), Seber (1982), and Buckland et al. (1993). Better estimators for the density of animal populations can be derived by combining line transect theory and capture-recapture theory when there are multiple observers. A classic example is aerial surveys, where the assumption that all animals on the line are sighted (i.e. g(0) = 1) is generally unrealistic. If there are two independent observers (e.g. in the same plane), then it is possible to simply apply the Lincoln-Petersen method characterizing which animals have been seen by the first, second or both observers acting independently and each having different sighting probabilities (Seber, 1982, p. 457). This method ignores the distances from the line. An estimator derived from a generalized line transect, capture-recapture model is expected to be more efficient and robust than a simple Lincoln-Petersen estimator which does not use the distance data, and this is the approach taken by Alpizar-Jara (1994) in his master's thesis.This paper is based on that work. 1352-8505 © 1996Chapman& Hall