Baltimore Ecosystem Study Institute of Ecosystem Studies

2015 BES Annual Meeting Presentation and Poster Abstracts

Tree Canopy Change and Neighborhood Stability: A Comparative Analysis of Washington, D.C. and Baltimore, MD
Chuang, Wen-Ching
Co-Authors: Boone, Christopher G., Buckley, Geoffrey L., Fragkias, Michail, Grove, J. Morgan, Locke, Dexter, Whitmer, Ali, and Zhang, Sainan

Abstract: The management of urban trees is an important sustainability priority for municipalities. Strategies to increase tree canopy, primarily on or near private residential properties, need to consider more than identifying possible planting locations and allocation of resources. An effective urban tree canopy (UTC) plan must also consider how social-ecological systems interact to influence the distribution of trees and their conservation. This study explores how neighborhood characteristics and changes in wealth over time of places impact on the distribution of UTC in the two cities, Washington, D.C. and Baltimore, MD. In this study, we use two models to explore how neighborhoods characteristics are associated with UTC. Using the Census 2000 and American Community Survey (ACS) 2013 five-year estimation, we classify Census tracts into five types: Stable impoverished, decreasing in wealth, stable above poverty, increasing wealth, and stable wealthy. Using stepwise multiple regression analysis, we identify strong predictors of UTC from variables of neighborhood types and some physical characteristics. We also conduct a Principal Component Analysis (PCA) of the socioeconomic, demographic, and housing characteristics of the two cities. Using the factor scores of the PCA, we build a spatial lag model to predict the spatial distribution of UTC cover. We found that 1) Stable wealthy neighborhoods are more likely to have higher and more consistent tree canopy cover than other types of neighborhoods; 2) Decrease or increase in income is negatively associated with UTC in Washington, D.C. but not in Baltimore. Instead, stability of income in wealthy and impoverished neighborhoods is the best predictor of UTC in Baltimore; and 3) Spatial autoregressive models that use factor scores of social, economic, demographic, and built- environment characteristics predict UTC better than the stepwise multiple regression models that use income status and built environment characteristics.