![2018 Graduate Student Showcase](../../assets/md5images/5b1e1babc9ce92adcada27a680795892.jpg)
Title of Submission
Degree Program
Geoscience, MS
Major Advisor Name
Nancy Glenn
Type of Submission
Scholarly Poster
Abstract
Remote sensing of dryland ecosystem vegetation is notably problematic due to the low canopy cover and fugacious growing seasons. Relatively high temporal, spatial, and spectral resolution of Sentinel-2 imagery can address these difficulties. In this study, we combined vegetation indices with robust field data and used a Random Forests ensemble learning model to impute landcover over the study area. The resulting vegetation map product will be used by land managers, and the mapping approaches will serve as a basis for future remote sensing projects using Sentinel-2 imagery and machine learning.
Funding Information
W9126G-15-2-0030