Get reliable science

Mapping the distribution and abundance of invasive plants is a high priority, but establishing cost-effective and practical techniques at appropriate scales remains elusive. Mesquite is a highly invasive shrub that cannot currently be reliably distinguished from other plant species using remote sensing technologies, at least not at accuracies necessary for mapping mesquite at very low densities. This paper describes and tests an alternative method. A visual, aerial technique was used to map a large mesquite (Leguminoseae: Prosopis spp.) population in Australia; 216 654 ha was surveyed in 18.5-ha grid cells to include the entire population. The objective was to test the ability of this technique to detect and map mesquite at very low densities for surveillance and to assist in prioritizing management effort and, where mesquite was well established, to categorize mesquite into broad canopy cover classes for change detection and to identify habitat associations. The survey technique was very effective at detecting isolated mesquite plants (< 0.6% canopy cover across a grid cell), which is considerably better than existing remote sensing technologies. Detection of low-density mesquite was particularly important, as most occupied grid cells (55%) had isolated mesquite, and their management may offer the best return on investment. The technique was also competitive cost wise (0.39 USD per hectare) and required relatively little expertise. Grid cells with moderate (20%-50%) to dense (. 50%) canopy covers were almost all restricted to a 32 500-ha area on the floodplain delta of the Fortescue River, where the original introductions occurred. Cover class estimates appeared to be well calibrated between observers within a survey; however, they were poorly calibrated between independently conducted surveys, suggesting that further methodological refinement is necessary if this technique is to be reliable for change detection.  The Rangeland Ecology & Management archives are made available by the Society for Range Management and the University of Arizona Libraries. Contact lbry-journals@email.arizona.edu for further information. Migrated from OJS platform August 2020