Premium
Advances in aerial survey methods for macropods in New South Wales and Queensland
Author(s) -
Finch Neal,
Pople Anthony,
McLeod Steven R.,
Wallace Glenn
Publication year - 2021
Publication title -
ecological management and restoration
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.472
H-Index - 42
eISSN - 1442-8903
pISSN - 1442-7001
DOI - 10.1111/emr.12486
Subject(s) - transect , distance sampling , aerial survey , geography , arid , macropus , sampling (signal processing) , ecology , abundance (ecology) , forestry , cartography , biology , marsupial , filter (signal processing) , computer science , computer vision
Summary This paper describes advances made to aerial surveys of macropods in New South Wales and Queensland. Regular aerial surveys of macropods across large areas of Australia’s arid and semi‐arid rangelands were established as a response to concerns for the species’ conservation. They became integral to the harvest management of the four macropod species – Eastern Grey Kangaroo ( Macropus giganteus ), Western Grey Kangaroo ( M . fuliginosus ), Red Kangaroo ( Osphranter rufus ) and Common Wallaroo ( O . robustus ). Harvest quotas are set as a proportion of population size, requiring absolute estimates of abundance. Counts were made in strip transects from fixed‐wing aircraft, but required correction for animals that are missed. In New South Wales and Queensland, counts are now made using line transect sampling, providing survey‐specific correction. Helicopters are used in more rugged and timbered areas and where Common Wallaroo can be more frequently found and harvested. Mark‐recapture distance sampling is used in more open country that can be surveyed using fixed‐wing aircraft and improves the accuracy of density estimates by accounting for animals missed on the transect line. Macropod populations are likely to be surveyed in the future by airborne imagery and by drones, but the technology needs further development. Spatial models could usefully provide small‐scale density estimates and address non‐random sampling.