Authors: Saaruj Khadka*, Kentucky State University, Buddhi Gyawali, Kentucky State University, Tilak Shrestha, Kentucky State University, Jeremy Sandifer, Kentucky State University
Topics: Remote Sensing, Geographic Information Science and Systems, Environmental Science
Keywords: Accuracy assessment, Ground Truth, Landsat, Ownership size, Supervised Classification.
Session Type: Paper
Start / End Time: 5:00 PM / 6:40 PM
Room: Jackson, Marriott, Mezzanine Level
Presentation File: No File Uploaded
With changes in land use patterns in Kentucky, there is a need to study temporal and spatial dynamics of landscape diversity. This study focused on examining the relationship between land cover types on different land ownership size (such as small, medium, large) and prediction of ecological productivity in Scott, Morgan, and Graves Counties, KY, which represent abundant vegetation, habitats for wildlife and farmlands with varied landscape diversity. Landsat 4-5 Thematic Mapper and Landsat 8 Operational Land Imager were used to perform supervised classification, accuracy assessment and change detections between various land cover types between 2001, 2011and 2016. Ground-truth data were compiled using the National Land Cover Dataset, Cropland Data layer, and National Agriculture Imagery Program data. Diversity by ownership was obtained from parcel data received from County Tax Assessors Offices of the respective counties. The preliminary results suggest that land cover change was relatively occurred at a higher rate in small and medium landownership classes. The change detection results at the pixel level between major land cover classes suggests major conversion of agriculture and forest lands into urban lands. The ecological productivity based on the Normalized Difference Vegetation Index (NDVI) was higher among larger parcels.