Hongli Yang is a graduate student in the Geosciences department at Murray State. She will receive her MS in Geology/Earth Science in May 2017. This research was done with her mentor, Dr. Robin Zhang.
She would like to thank Mammoth Cave National Park for the support, particularly Rick Olson, Lillian Scoggins, Rick Toomey, and Shannon Trimboli. The Watershed Studies Institute at Murray State University provided partial funding for the project, along with the Provost's Office.
Up-to-date and detailed vegetation map provides critical information for habitat management. In addition, a vegetation map is necessary for the Park’s Fire Management, for classification of fuel types, and for delineation of fire management units. There have been several attempts of vegetation mapping in 1934, 1975 and 1997. Recent advancements in mapping technology and the availability of high resolution Lidar data call for a new vegetation map for the Park’s management team.
For my research, I used Landsat-8 Operational Land Imager (OLI) imagery as the main data source for analysis. Obtained from USGS between June 2015 and February 2016, Landsat-8 dataset covered Mammoth Cave National Park in 30m pixels for the multispectral band and 15m for the panchromatic band, during both the leaf-on and leaf-off seasons. I applied Landsat-8 dataset to categorize vegetation types according to their different reflectance characteristics. Utilizing both leaf-off and leaf-on images better distinguished coniferous and deciduous trees. In addition, I used a vegetation index to identify specific vegetation species. To test classification accuracy, I used the field plots for different vegetation categories. The classification results were also compared to 2011 National Land Cover Image.
"Vegetation Habitat Mapping of Mammoth Cave National Park Using Multi-date Landsat-8 Imagery and Lidar data,"
Steeplechase: Vol. 1
, Article 9.
Available at: http://digitalcommons.murraystate.edu/steeplechase/vol1/iss1/9