Testing 2 aerial survey techniques on deer in fenced enclosures—visual double-counts and thermal infrared sensing

Abstract

We evaluated the accuracy of 2 aerial survey techniques over 4 large enclosures (6.0–29.4 km2) where the deer (Odocoileus virginianus) population was reconstructed using hunting harvest and winter mortality data. We conducted surveys (n = 8) along equally spaced parallel lines. Six surveys using the double-count technique involved 2 independent observers located on the same side of a helicopter who simultaneously counted animals over narrow plots (60-m width). Four of these surveys yielded deer densities 64–83% of assumed densities (based on the reconstructed population). The 2 other surveys had accuracies of 37 and 46%, respectively, and were judged unreliable because the sighting probability of the front observer was < 0.40. We conducted 2 surveys with a thermal infrared sensor. One survey had the highest accuracy (89%) among all surveys while the other gave poor results (54% accuracy). We concluded that when sighting probabilities of observers exceed 0.45 of deer groups, double-count surveys provided valid estimations of densities for management purposes, although 1 deer out of 4 was missed on average. Because of closed forest canopy, thermal infrared sensing of deer along systematic survey lines was not a reliable technique.