Research
Working papers
When are D-graded neighborhoods not degraded? [PDF]
Abstract
This paper explores how geography shapes the legacy of redlining, the systemic mortgage lending bias against minority us neighborhoods. On average, redlined neighborhoods lag behind adjacent, less-discriminated areas in home values, income, and racial composition. Yet, redlined neighborhoods near parks and water fare better. To help understand convergence, we inventory waterfront renovations, apply machine learning to historical imagery to track tree canopy changes, and instrument such changes exploiting tree replacements due to geographic variation in tree plagues and susceptible species. Findings suggest that enhancing waterfronts and increasing tree canopy can mitigate the long-lasting effects of institutionalized discriminationTrees of green: constructing panels of tree canopy from aerial imagery [PDF]
Abstract
This paper develops a fully-automated workflow for constructing panels of tree canopy from high-resolution multispectral imagery with limited near-infrared (NIR) training data. The proposed workflow utilizes the tree-pixel detection algorithm developed by Yang, Wu, Praun, and Ma (2009) and Bosch (2020) on a large set of U.S. urban areas but modifies it by creating automatic ground-truth masks through various visual graphics techniques that leverage modern high-resolution NIR data. By matching colors across different imagery periods, the workflow predicts tree presence in older images without NIR data, using the recent images with NIR data. Using a subset of cities that represent the different U.S. climate regions, I quantify the effectiveness of the workflow by implementing the algorithm without pre-processing in the creation of ground-truth masks, without equalizing colors across periods, and using a universal model for all areas. The comparison shows that my workflow is the option that leads to better results in terms of accuracy, recall, and precision.Work in progress
City buzz: Face-to-Face interactions and the urban environment,with Dani Arribas-Bel, Souneil Park and Diego Puga