OOS 11-1 - Integrating field and remote sensing applications for characterizing tree loss in a rapidly densifying urban area

The convergence of residential infill for housing and urban development policies that marginalize urban tree canopy is leading to a national reduction of green spaces across the US. Yet, as trees and green spaces are increasingly concretized and opportunities for greening is reduced, researchers and managers have few tools for describing changes in urban green space. To address this paucity of information, we examine the strengths and limitations of using three distinct datasets -- LiDAR (1m), NAIP (60cm), and NDVI (30m) -- to describe changes in urban tree canopy over a five year period (2014 - 2019) in the Portland (OR) region. Using a visual assessment of over 400 sites, we describe differences in six quantitative data quality parameters. The results suggest that both high resolution datasets are comparable in describing presence and absence of individual trees, yet vary significantly in other parameters, such as technical skill, monetary cost, and computational power. We conclude by offering a qualitative evaluation and set of recommendations for those attempting to use different sources of data to characterize changes in urban tree canopy and green spaces.