Mailing address: % South African Museum, P. O. Box 61, Cape Town, 8000 South Africa.
ESTIMATING NATURAL MORTALITY RATE IN REPRODUCTIVELY ACTIVE FEMALE SOUTHERN RIGHT WHALES, EUBALAENA AUSTRALIS
Article first published online: 26 AUG 2006
Marine Mammal Science
Volume 14, Issue 4, pages 738–749, October 1998
How to Cite
Best, P. B. and Kishino, H. (1998), ESTIMATING NATURAL MORTALITY RATE IN REPRODUCTIVELY ACTIVE FEMALE SOUTHERN RIGHT WHALES, EUBALAENA AUSTRALIS. Marine Mammal Science, 14: 738–749. doi: 10.1111/j.1748-7692.1998.tb00759.x
- Issue published online: 26 AUG 2006
- Article first published online: 26 AUG 2006
- Received: 3 March 1997 Accepted: 4 September 1997
- southern right whale;
- Eubalaena australis;
- South Africa;
- mortality rate
Cow-calf pairs of southern right whales on the South African coast have been photographed in aerial surveys in October each year since 1979. In this paper 469 resightings of 177 individually identified cows photographed in the first six years of surveys have been analyzed in two ways to produce estimates of natural mortality rate. Both methods assume that all females calve either two, three, or four years after their previous calf. In Method A there is assumed to be no systematic trend with time in the probability of a female being photographed on each calving occasion. Natural logarithms of the numbers photographed 2-4, 5-7, 8-10, etc. yr after being first seen are regressed against time, the slope of which provides an estimate of natural mortality rate of 0.0255 ± 0.0071. The intercept value for this regression provides an estimate of the detection probability, or 0.769 ± 0.011. In Method B it is assumed that if a female has not been photographed for at least nine years then it is dead. Annual mortality estimates are obtained from the proportional reduction in the numbers of females known to be alive at each three-year interval after being first photographed, up to at least 6-8 yr from the present. Method B produces an estimate of natural mortality of 0.0260 ± 0.0190 (corrected to 0.0227 ± 0.0192 using the average detection probability). Both methods may be subject to various biases that tend to inflate estimates of natural mortality.