This study focuses on the highly disturbed coastal sage scrub habitats of San Clemente Island, California. The purpose of this study is to map shrub cover and the ability to assess shrub cover change dynamics based on classification of high spatial resolution orthoimagery and LiDAR data. A multi-temporal analysis was conducted over a nine-year period from 2010 to 2018 to map shrub cover and determine if the methods and available data are suitable for tracking shrub cover changes. Trimble eCognition and ERDAS IMAGINE software programs were used to create vegetation growth form maps for three areas of interest. Pixel-based and object-based image analysis (OBIA) approaches to the classification of the imagery were tested. Shrub cover was estimated for 10, 20 and 40 m grid sizes. The classification approach that generated the most accurate estimates of shrub cover was the OBIA method with the incorporation of a normalized difference surface model (nDSM), applied to the highest spatial resolution imagery. Overall, the fractional cover products derived from the 2015 and 2017 Near-Earth Observation System (NEOS) imagery yielded the highest accuracies. Major factors that influenced the mapping and fractional cover estimation include the date and spatial resolution of the imagery that was classified, the type of classifier used, the inputs to the classifier, and the grid size used for cover estimation. While tracking actual changes in shrub cover over time proved to be challenging, this study shows the importance of consistent mapping approaches and high-quality inputs, including high spatial resolution imagery and a nDSM.