Authors: David L. Miller*, University of California Santa Barbara, Erin B. Wetherley, University of California Santa Barbara, Dar A. Roberts, University of California Santa Barbara, Christina L. Tague, University of California Santa Barbara, Joseph P. McFadden, University of California Santa Barbara
Topics: Remote Sensing, Land Use and Land Cover Change, Urban Geography
Keywords: California, remote sensing, drought, urban, forestry, land cover change
Session Type: Virtual Paper
Presentation File: No File Uploaded
Trees and turfgrass lawns provide many ecosystem services in semi-arid cities, such as reducing temperatures and providing recreational amenities. However, it is challenging for water-limited cities to maintain green vegetation cover during extended droughts, and the effects of drought may be distributed unevenly across space and time. Here, we track changes in vegetation cover across the Los Angeles metropolitan area, using airborne remote sensing imagery acquired each May/June as drought conditions progressed from 2013 to 2018. This time period coincided with the exceptional 2012-2016 California drought. Specifically, we calculate subpixel fractions of tree, turfgrass, non-photosynthetic vegetation (e.g., senesced plant material), and non-vegetated surfaces (e.g., pavement, roofs, soil) at 18 m spatial resolution using Multiple Endmember Spectral Mixture Analysis and hyperspectral imagery from the Airborne Visible Infrared Imaging Spectrometer (AVIRIS). We found tree and turfgrass trajectories differed through time. Overall, mean tree cover declined from 20.5% in 2014 to 18.3% in 2018. Turfgrass cover was also lowest in 2018 at 3.0%, but was highest in 2017 at 3.9% after springtime rainfall. By comparison, non-photosynthetic vegetation increased from 5.7% in 2015 to 7.1% in 2018. These changes could have significant implications for water management as tree and turfgrass cover differed based on single-family residential outdoor water use. Our results also suggest that the effects of the drought on ecosystem services were unevenly distributed: for example, urban residents in high income areas had much greater green vegetation cover than lower income areas throughout the drought, despite pronounced differences in vegetation losses through time.