Contrasting aerial moose population estimation methods and evaluating sightability in west‐central Alberta, Canada

ABSTRACT Population assessment is a primary component of ungulate management, but managers are continuously under pressure to reduce survey cost. Another concern in aerial surveys is accounting for undetected animals (i.e., visibility bias). Currently, a stratified random block‐survey design (hereafter, block‐surveys) is used to develop moose ( Alces alces ) population estimates in several regions of North America. In this case study, we evaluated the application of distance sampling as an alternative to block‐surveys in Alberta, Canada. We conducted distance‐sampling surveys in 2010 and 2012 and compared density estimates, precision (coeff. of variation) and flight effort (hr/100 km 2 of survey area) to block‐surveys flown in 2002, 2007, 2009, and 2012. To assess sightability bias and subsequently correct for moose missed on the transect line, we developed a predictive sightability model using 41 sightability trials with 21 radiocollared moose in 2009 and 2010. Without correcting for visibility bias on the transect line, distance sampling was more efficient in terms of flight‐hours than block‐surveys, while providing population estimates with similar or higher precision. Estimated sightability on the transect line was 67% in 2010 and 46% in 2012, which was used to re‐scale the detection functions. Considering that population estimates from block‐surveys as applied in Alberta are based on observable moose, distance sampling with a sightability correction likely provided more accurate estimates. Our results support the application of distance sampling as an alternative to block‐surveys, but we suggest further investigation of methods for correcting visibility bias on the transect line. © 2014 The Wildlife Society.