Scaled Datasets through Leveraging Technology Better Represent Plant Communities Classified Within National Land Cover Database (NLCD)

Scaled Datasets through Leveraging Technology Better Represent Plant Communities Classified Within National Land Cover Database (NLCD)

Technological advances in remote sensing have allowed for access to and application of large- scale datasets of plant communities using cover classification and chlorophyll content measurement with the use of multispectral sensors and vegetation indices. The United States’ primary remote sensing platform Landsat Thematic Mapper is a series of satellites providing 30-meter resolution imagery of the Earth’s surface. Efforts have been made to best utilize Landsat derivative National Land Cover Database (NLCD) datasets and provide spatial resolution commensurate with the needs of plant ecologists and range managers. However, usage of 30-meter spatial resolution imagery and indices does not allow detection of subtle changes in features such as plant species changes or spatial patterns and often will overestimate dominate features while underestimating smaller features due to pixel size. Here we investigate the use of unmanned aerial vehicles, agriculture imaging sensors, and high precision GPS units to better scale NLCD cover classes using Juniper (Juniperus spp) encroachment on grassland ecosystems as a model for scaling. We theorized that high resolution reclassification of NLCD cover classes will change metrics such as patch size and connectivity. In a woody-encroached grassland in central Texas, we determined that lower resolution classification underestimated woody encroachment. Smaller patch sizes were detected with greater connectivity of woody patches also detected when using high-resolution enabled classification.