Aerial survey of donkey and horse populations in the Victoria River District 2006
Saalfeld, Keith
E-Publications; PublicationNT; E-Books
0000-00-00
Made available via the Publications (Legal Deposit) Act 2004 (NT).
2001 and 2006 Aerial survey. 2001 survey was published by Parks and Wildlife Commission of the Northern Territory, Palmerston (N.T.) 2006 survey was published by Department of Natural Resources, Environment and The Arts, Palmerston (N.T.)
English
Feral livestock -- Northern Territory -- Victoria River Downs; Donkeys -- Northern Territory -- Victoria River Downs; Horses -- Northern Territory -- Victoria River Downs
application/pdf.
Check within Publication or with content Publisher.
https://hdl.handle.net/10070/230511
https://hdl.handle.net/10070/670949
https://hdl.handle.net/10070/670947
was in view. Species recorded were donkeys and horses. Observer sightings were recorded individually on Hewlett Packard HP200LX palmtop computers that had been programmed as data loggers. Each palmtop was synchronised to the nearest second to UTC (Universal Time Coordinate) and each observers sightings recorded as species, number of individuals in group, UTC date and time. Each days flying was logged using a GPS and Hewlett Packard HP200LX palmtop computer programmed as a data logger to record latitude, longitude, UTC date and time every two (2) seconds. Matching observer sighting log with the GPS position log using UTC date and time allowed latitude and longitude of each sighting on each transect to be determined . Aerial counts of feral animals are negatively biased (i.e. under estimates population abundance). Scott and Saalfeld (1999) derived correction factors for donkeys and horses from an extensive index-manipulate-index experiment in Gregory National Park. The correction factors for donkeys and horses derived by Scott and Saalfeld (1999) combine both observer and environmental components (Marsh and Sinclair, 1989) in a single correction factor for each species. Given the large scale and consistency of the observer teams for the index-manipulate-index experiment of Scott and Saalfeld (1999) the correction factors derived are considered reliable as generalised survey correction factors for donkeys and horses in the VRD. These correction factors are considered to be observer independent, however, they are applicable only to observer teams consisting of a single port and single starboard observer operating under the survey parameters described above. Due to observer availability, a total of four observer teams consisting of a single starboard and single port observer were used during the survey. The correction factors for donkeys (4.43) and horses (3.08) derived by Scott and Saalfeld (1999) have been adjusted to take into account the predicted greater sightability of animals at high densities. Calculation of the estimated donkey population using the Scott and Saalfeld (1999) correction factor of 4.43 gave a population estimate that significantly exceeded the predicted estimate based on the potential increase from 1996 taking into account the known maximum rate of increase and known off-take. Reduction of the correction factor by onethird to 2.95 gave a population estimate approximating that which would be predicted based on the potential increase from 1996 taking into account the known maximum rate of increase and known off-take. This result supports the conjecture that sightability increases with density and for this survey a correction factor of 2.95 has been used for donkeys. For horses, use of the Scott and Saalfeld (1999) correction factor of 3.08 gave a population estimate that fell well below the predicted estimate based on the potential increase from 1996 taking into account the known maximum rate of increase and known off-take. Hence, 3.08 has been used as the correction factor for horses for this survey. Population size and density and their associated errors were estimated for both donkeys and horses in each survey block using the ratio method (Marsh and Sinclair, 1989). A total population size and standard error was calculated by adding block estimates and merging variances. Minimum cell (transect separation squared = 121 km-2) density distribution maps were derived for each species. Property density distribution maps were then derived for management purposes.