Sullivan, J. J.; Vattiato, G.; Niebuhr, C. N.

Road-kill is increasingly recognised as an important source of mortality for wildlife, especially in densely populated urban and rural landscapes. Monitoring road-kill on fine spatial and temporal scales is necessary to better understand how the design of road networks affects road-kill. We report on a practical method for consistently and repeatedly surveying road-kill. This includes geotagging individual carcasses, reporting where on or by the road each carcass is found, categorising the age and state of each carcass, and noting whether each carcass is new or previously reported. All this can practically be done with a smartphone while on foot, biking, or as a passenger in a motor vehicle. Repeatedly mapping all carcasses, whether or not they have previously been reported, allows for estimates of carcass persistence and detection probability for both taxa and road conditions. We demonstrate the effectiveness of the method by reporting on its frequent use along one 16.2 km route of urban and rural roads in New Zealand over 12 years. Over this period, on 1,652 surveys, 4,034 new road-kill carcasses were observed, 62.4% of which were birds, 31.1% were mammals, and 3% were butterflies. Using carcass age and persistence, we estimate that the road-kill rate along this route has been at least 3,940-6,544 road-kill/100 km/year. There was considerable variation among taxa in carcass persistence and carcass position on and by the road, both of which will introduce biases into road-kill estimates if not accounted for. To better understand road-kill and estimate road-kill rates, we encourage road-kill studies to geotag all individual carcasses and track their persistence.