Statistical Methods for Aerial Surveys Using the Double-Count Technique to Correct Visibility Bias

In double-count surveys two observers search the sampled area for the species of interest. The presence of the two observers permits one to calculate a survey-specific correction for visibility bias. This correction factor can be negatively biased, because animals difficult to see for one observer are also difficult to see for the other. This paper investigates methods for reducing this bias. One solution is to classify, during the survey, the animals seen according to variables potentially influencing their visibility. Statistical methods are proposed to evaluate the efficiency of such a classification. Another proposal to reduce the bias of double-count estimators is to use parcel-specific correction factors. Variance estimators for double-count estimators of population totals are proposed; three variance components are identified. The proposed techniques are illustrated using data collected in a double-count deer survey carried out on Anticosti Island in the Gulf of St. Lawrence.