Case studies: Developing remote sensing solutions for detecting wildlife over large, remote areas
The Challenge
-
Wildlife refuges are sensitive environments, requiring non-invasive monitoring strategies such as remote sensing.
-
While georeferenced imagery is useful for verifying locations and numbers of animals, accurate sampling is hindered when some animals blend into their backgrounds.
RMH Solution
-
Incorporating a thermal sensor with our multispectral and natural color cameras has been useful for deer and moose surveys in temperate climates.
-
While seabirds on islands are generally too well insulated to show up in thermal imagery, we were able to identify them with a combination of multispectral and natural color imagery.
-
Thermal coverage is not useful in desert environments, yet high resolution natural color imagery is useful for visual identification of shapes and shadows. .
Results
-
We completed the largest aerial sampling survey for wildlife ever undertaken. Working with The Conservation Society and Oyu Tolgoi, we assessed the numbers and spatial distribution of wild and domestic ungulates over a 150,000 sq km portion of the Gobi Desert in Mongolia.
-
We worked with the US Fish and Wildlife Service in the Maine Coastal Islands National Wildlife Refuge and generated matching multi-spectral and natural color orthoimagery in which each bird species could be identified and counted.