An improved method of censusing deer in deciduous-coniferous forests

Aerial censusing has been used to determine densities of many large mammals, including deer (Odocoileus virginianus) in agricultural areas or deciduous forests (Saugstad 1942, Morse 1946, Petrides 1953, Sanderson 1953, Berner pers. comm.). However, observability of deer from the air remains a problem in northern coniferous forests. LeResche and Rausch (1974) determined that even with the much larger and more observable moose (Alces alces) during ideal snow conditions, experienced observers counted only 68% of a known number of animals; inexperienced observers counted 43%. Caughley (1974) and Caughley et al. (1976) suggested that the best solution to the problem of observability in aerial censuses is to measure the magnitude of the biases that exist, and correct estimates accordingly. This paper describes an attempt to measure observability bias in an aerial census of deer in deciduous coniferous habitat and to produce an accurate estimate of numbers.