Counting the Same Populations of Large Tropical Mammals from the Ground and from the Air

Abstract

The population of feral buffalo occupying 614 km2 of sedgelands and woodlands at Kapalga in the Northern Territory was counted from a fixed-wing aircraft, a helicopter and a vehicle on the ground over six consecutive days in August 1981. Aerial counts on sedgelands were 1.7-2.0 times greater than counts from ground level because of the greater lateral visibility; whereas counts from ground level of wooded habitats, much the commonest situation, were 1.7-1.8 times greater than aerial counts because ofthe reduced vertical visibility for the latter. The area of all habitats combined which would have to be sampled for a given precision differed little between the three techniques, but in wooded situations alone counting from the ground was twice as precise. Aerial counts repeated over consecutive days had a coefficient of variation of c. 10% for all habitats combined but c. 20% for wooded habitats alone. Counting from fixed-wing aircraft has great advantage in costs and logistics for extensive broad-scale surveys. Counting from a helicopter is more expensive but has advantages in a less extensive survey covering habitats in which visibility differs widely. Counting from the ground over small areas is cheap and has a particular advantage when detailed demographic data is required. The overall density of the introduced feral buffalo at Kapalga is 13 per km2 which is equivalent to a biomass of c. 5200 kg km-2. Such a biomass is high compared with wild indigenous tropical populations in Asia, at the lower end of the African range, and low by comparison with cattle in slightly drier habitats in Asia.