Publication Abstract
Mapping of Phragmites Australis in Gulf of Mexico Wetlands Using Small UAS
Samiappan, S., Turnage, G., Hathcock, L. A., & Moorhead, R. J. (2016). Mapping of Phragmites Australis in Gulf of Mexico Wetlands Using Small UAS. 2016 Gulf of Mexico Oil Spill & Ecosystem Science Conference Proceedings. Tampa, FL: GOMRI.
Abstract
The wetland species, Phragmites australis, also refereed as Common Reed, is an invasive plant present on every continent except Antarctica. In the North America its presence is marked by native and non-native subspecies with the non-natives quickly displacing native plants. Along the Gulf Coast, Phragmites grows in impenetrable stands at heights of 15 feet and is usually the tallest grass species in wetland, estuary, and marsh ecosystems. Phragmites invasion has been shown to have adverse impacts on the local ecosystem, most notably through decreased biodiversity. Furthermore, it presents a navigation hazard to smaller boats by impairing visibility along shorelines of canals and rivers. Management efforts targeting non-native Phragmites rely heavily on accurately mapping invaded areas. Historically, mapping has been done by walking the perimeter of a stand with a GPS unit, using satellite imagery, or through aerial photography from manned aircraft. These methods are time consuming, expensive, can have inadequate resolution, and are prone to human error. To overcome these drawbacks, we employed an Unmanned Aerial System (UAS) capable of collecting geo-referenced high spatial resolution imagery. This system negates the low resolution and long update times of satellite imagery, the cost issues and pilot error associated with manned aircraft, and the hazards of on-the-ground fieldwork. We have collected over 20 square miles of imagery using an Altavian NOVA UAS platform. The images were collected at an altitude of 700 feet with ground resolution of approximately 2 inches. Images are then mosaicked to form an orthomosaic. Initial pilot experiments with texture based mapping analysis shows over 90% agreement with manual analysis.