New methods to locate historic fossil fuel infrastructures: How computational techniques can illuminate 19th-century urban environmental hazards

Keywords: environment, fossil fuel, coal, inequality, urban, computational geography, machine learning
Session Type: Virtual Paper Abstract
Day: Friday
Session Start / End Time: 2/25/2022 12:40 PM (Eastern Time (US & Canada)) - 2/25/2022 02:00 PM (Eastern Time (US & Canada))
Room: Virtual 18

Abstract

From the mid-19th to the mid-20th century, many homes, institutions, and factories in the United States, Canada, Australia and much of Europe relied on gas produced from coal to light lamps and kitchen stoves. The manufactured gas industry sparked a revolution in urban form and function, while at the same time producing vast quantities of hazardous waste with minimal regulation. Today, the remains of former coal gas production, distribution and disposal sites number in the tens of thousands in the United States alone, mostly hidden by decades of urban change and industrial churning.

While many of these unidentified sites likely contain significant levels of highly toxic and biologically persistent contamination, locating them remains a significant challenge. This paper presents a new method to identify manufactured gas production, storage, and distribution infrastructure in bulk by applying feature extraction and machine learning techniques to digitized historic Sanborn fire insurance maps. Our approach, which relies on a two-part neural network to classify candidate map regions, increases the rate of site identification 20-fold compared to unaided visual coding.