Research Document - 2015/046
Instantaneous availability bias correction for calculating aerial survey abundance estimates for bowhead whales (Balaena mysticetus) in the Canadian High Arctic
By Cortney A. Watt, Marianne Marcoux, Bernard Leblanc, and Steven H. Ferguson
Abstract
Twenty-five bowhead whales (Balaena mysticetus) were fitted with satellite tags near the communities of Igloolik and Pangnirtung, Nunavut in July-August 2012 and 2013 and twenty-two provided information on the time whales spent at 0-2, 0-3, 0-4, 0-6, 0-8 m depths in August. Bowhead whales diving to depths up to 8 m depths were considered possibly available to be viewed by observers during aerial surveys. To obtain an accurate estimate of bowhead abundance, for calculating total allowable catch and managing stocks, a correction for bowhead whales that are present but not visible to aerial observers is necessary. An instantaneous availability correction factor used to correct aerial surveys can be estimated from the proportion of time diving animals spend near the surface where they can be detected and identified. To provide a bias correction for the 2013 aerial survey, the proportion of time bowhead whales spent at different depths (0-2, 0-3, 0-4, 0-6, 0-8 m) was analyzed in a mixed effect model with individual whale as a random variable and period of August (early (August 1-15), mid (August 16-23), and late (August 24-31)), time of day (day or night), sex, and summering area (Prince Regent Inlet/Gulf of Boothia/Foxe Basin (PRI/GoB/FB), or Cumberland Sound) as fixed factors. To investigate other environmental factors related to bowhead diving behaviour, we also tested models with slope and depth as fixed factors.
Models that included time of day and period of August performed the best in the 0-2 m and 0-3 m bins, while models that included time of day, period of August, and area of tagging performed best in the 0-4, 0-6, and 0-8 m bins. Since aerial surveys were primarily conducted in early and mid-August, always during the day, an availability bias which excluded night, and combined both sexes for each area was calculated. Without experimentation, previous studies have relied on a correction for bowhead whales based on the 0-4 m bin. Since we are unsure of the deepest depth to which bowhead whales can be seen, we wanted to increase the variance around our estimate and calculated uncertainty in the availability bias correction based on a combination of the 0-2, 0-4, and 0-6 m depth bins since there is no evidence to suggest which depth is best. In early August, bowhead whales spent 21.6 ± 3.12 % of their time in the 0-2, 0-4, and 0-6 m combined depth bins in Cumberland Sound and 24.7 ± 5.11 % in the PRI/GoB/FB region, and thus we recommended an instantaneous availability bias correction of 4.63 (± 0.669) for Cumberland Sound and 4.05 (± 0.838) for the PRI/GoB/FB region for strata surveyed in early August 2013. In mid-August, bowhead whales spent 17.6 ± 1.65 % of their time in the 0-2, 0-4, and 0-6 m combined depth bins in Cumberland Sound and 29.1 ± 7.09 % in PRI/GoB/FB; thus, we recommend an instantaneous availability bias correction of 5.68 (± 0.533) for Cumberland Sound and 3.44 (± 0.838) for the PRI/GoB/FB region for strata surveyed in mid-August 2013. For areas surveyed outside of Cumberland Sound or the PRI/GoB/FB region in 2013 we recommend using a combined estimate for the areas based on the 0-2, 0-4, and 0-6 m combined depth bins. In early and mid-August, bowhead whales from both areas spent 24.3 ± 4.52 % and 25.1 ± 5.30 % of their time respectively in the 0-2, 0-4, and 0-6 m combined depth bins, resulting in an instantaneous availability bias correction of 4.12 (± 0.766) for early and 3.98 (± 0.840) for mid-August. Future research is necessary to determine the depth at which aerial observers can detect bowhead whales.
Accessibility Notice
This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.
- Date modified: