India is the third-largest contributor to global energy use and greenhouse gas emissions, and the country is rapidly urbanizing. Transitioning to low-carbon cities is critical to meeting the country’s climate goals. In a new study published on April 13, 2021, in the journal Scientific Data, a team of researchers from Princeton University and the University of Minnesota developed an energy use database covering urban areas across 640 urban districts in India. Such baseline data are essential for Indian cities to develop and evaluate low-carbon policies that align from local to state and national scales.

The researchers reported their model — All Urban areas’ Energy-use (AllUrE) — aligns with reported energy-use at the city, state, and national scale. Therefore, AllUrE-India data can be used to model urban energy transitions consistent with national energy and climate goals. 

The team combined a number of publicly available datasets, such as censuses, industrial surveys, household surveys and records of registered vehicles, and applied a new machine learning approach to estimate vehicle ownership and travel across these 640 cities. “Residential energy” captured both the population and inequality in access to energy and consumption of energy in each district, using census data for population and access and consumer survey data for inequality in consumption. “Industrial energy” determined the number of industrial manufacturing workers by sector in each city from census data and sectoral energy use intensity derived from the Annual Industrial Survey. “Commercial energy” reflects the number of workers in commercial sectors in each city. “Transportation energy” was calculated using modeled registered vehicles at the city level, which in turn is found to be highly correlated with local social-economic parameters in each district, as well as miles traveled represented for city-types (larger than 6 million or smaller than 6 million population) in India. The team chose the year 2011 because it provided the latest publicly available population census, with detailed demographics, employment, housing conditions, residential energy use structure and water sanitation infrastructure data covering both rural and urban areas in each district.

“Our energy use dataset covers multiple energy use activities — electricity, cooking fuel, transportation fuel, industrial energy use — in all districts,” said Kangkang Tong, a postdoctoral research associate in the Department of Civil and Environmental Engineering, and the study’s first author. “The energy use data are aligned well when aggregated to and compared with state totals and national totals.” Previous work on urban energy use in India was only done at the individual city level which only covered 20 or so cities, explained Tong, and these datasets from individual cities could not be used to estimate what the aggregated carbon impact of urban actions will be for the whole of India. 

The research was advised by Anu Ramaswami, the Sanjay Swani ’87 Professor of India Studies; professor of civil and environmental engineering, the Princeton Institute for International and Regional Studies (PIIRS), and the High Meadows Environmental Institute (HMEI); and the director of the M.S. Chadha Center for Global India. “We know that India is rapidly urbanizing, but urban specific energy use data in India, across many sectors — industry, households, mobility — and fuels — electricity, wood, dung cake, gasoline, diesel, kerosene — have not previously been collated in one place in a way that systematically covers all 640 districts in India and aligns with the national total,” she reiterated. 

This dataset has many applications: exploring carbon emissions patterns across cities of different types, evaluating the impact of urbanization levels on emissions in different sectors, and quantifying the collective impact of urban decarbonization strategies on national INDCs. This novel dataset development method can be translated to other developing countries with relatively sparse data to develop their all-cities’ energy use and carbon foot-printing databases for supporting low-carbon policies and tracking progress. 

Tong is now exploring the typology of cities in India to quantify the collective carbon emission from all Indian urban districts and emission patterns from different city types. The Ramaswami lab intends to develop new science and methods to enable this all-urban data set to be updated every year, instead of every decade when census data are made available. “This will be a key initiative of the Chadha Center as part of the India Sustainable Urban-Rural Systems Data project,” said Ramaswami. 

The project was conducted with co-author Ajay Nagpure, who was a postdoc in the Ramaswami lab at the University of Minnesota and is now head of the air pollution in the Sustainable Cities program at World Resources Institute in India. “Making data actionable and useful in India though partnerships on-the-ground is our ultimate goal”, says Ramaswami.

The study, ”All urban areas’ energy use data across 640 districts in India for the year 2011,” was published Apr. 13 by Scientific Data. The research was funded in part by NSF-SRN: Sustainable Healthy Cities [#1444745].