Assistant Professor of Environmental Studies Heather Kropp recently published a paper in the journal Environmental Management. “Historical changes in tree and impervious surface cover following urban renewal in a small postindustrial city,” discusses a study in which she used artificial intelligence and historical aerial imagery of Utica, N.Y., to track changes in certain types of land cover.
In her study, Kropp applied a convolutional neural network image segmentation approach to historical aerial imagery in order to delineate changes in Utica’s land cover in 1957, 1974, and 2017. She said this model predicted tree, pavement, and building cover in each year with overall accuracy ranging from 82% to 87%.
Among her findings was that from 1957 to 2017, tree cover declined in many areas and impervious surface cover increased. She found that most tree cover gains occurred in less populated, natural areas, while the greatest declines in tree coverage were seen in many residential areas following the start of urban renewal efforts in 1957.
Kropp said that “a better understanding of the legacies of historical policies and neighborhood-scale changes in land cover can assist in highlighting priorities for urban forest management and justice-oriented urban forestry approaches to urban tree planting.”
