Sub-parcel scale analysis is needed to capture socially-driven canopy cover change in Baltimore, MD

Urban tree canopy (UTC) cover is rarely distributed equitably across social groups, space, and time. Over the past 20 years, research on the social, spatial, and temporal dynamics of UTC has grown considerably as municipalities adopt ambitious tree canopy cover goals. Yet less is known about how these three dimensions of tree canopy intersect. This paper brings these research areas together by examining i) which sets of social variables--population density, socioeconomic status, or lifestyle--are associated with UTC cover on residential lands, ii) how those relationships vary from front to back yard, and iii) how those relationships are associated with tree canopy cover changes in Baltimore, MD from 2013 to 2018, to more wholistically understand UTC. Socially, population density and social stratification predict tree canopy cover on residential lands, but not as well as lifestyle and life stage factors. More detailed and finer-grain social categories perform best. Spatially, models that explicitly separate front and backyards fit the data better than all-residential statistical models. Ignoring the front yard vs back yard distinction may hinder future theory development, limit the generalizability of empirical research findings, and prevent managers from realizing their canopy goals. Temporally, UTC across residential yards had a positive, though not significant, change likely from the relatively short period (5 y) considered. A fruitful next step could be to model how much planting, maintenance, and loss minimization is needed to achieve the city’s 40 % canopy goal with various scenarios for mortality, longevity, and removal over several timesteps.