Skip to Main Content

Urban climate resilience bolstered with smart use of data and sensor technology

CHICAGO – In a new study published in the AGU Journal of Advances in Modeling Earth Systems, scientists at the Discovery Partners Institute, part of the University of Illinois System, have developed a new methodology to predict temperatures at street scales. Cities worldwide grapple to counteract heat-related issues, as their strategies often lack scientific rigor and their efforts are constrained by resource availability. With accurate and prompt fine-scale temperature predictions in urban areas, communities can better prepare for the effects of climate change.

According to DPI researchers, the methodology developed in this study holds immense potential for cities to strategically deploy climate change mitigation strategies – particularly tree planting – to maximize the efficacy of their investments. For example, the City of Chicago announced a historic investment in tree equity in 2021. The investment included $46 million of the $188 million in the environmental justice and climate action budget for planting and maintaining 75,000 trees over five years. In 2023, the city planted 23,000 trees toward its goal. With smart planting – using a low-cost sensor system and the modeling framework in the study – Chicago can better analyze and predict temperatures and target zones for immediate intervention.

“Cities collect a wide range of data sets that go unused in creating solutions-oriented models that can forecast environmental stressors at scales where communities live,” said Peiyuan Li, the lead author and a postdoctoral researcher at DPI. “Our study brings high-resolution land use, lidar-based urban morphology and low-cost temperature sensors along with many other datasets together and fuses them with urban climate models and machine learning models to make useful and accurate predictions of street-scale temperatures.”

The researchers said this study highlights the need for cities to add more low-cost sensor networks and to use proposed modeling frameworks to create quick and accurate temperature predictions to guide climate action and make more equitable investments.

Elements such as parks, roads, buildings, and towering skyscrapers wield considerable influence on street-level temperatures. However, their impact is notably concentrated within specific areas. Existing urban climate models often overlook these nuanced features due to the associated high computational costs, the DPI research team said. The emergence of novel, less computationally-intensive machine learning models presents a promising solution to bridge this gap, enabling a more accurate representation of the localized effects of urban features on temperature dynamics and designing strategies for reducing heat stress.

Ashish Sharma, who leads the climate and urban sustainability program at DPI with a joint appointment at Argonne National Laboratory, is a co-author of this study. “As urban climate informatics advances, we believe this study is one of the first to invest and harvest low-cost sensor networks, urban communities, and machine learning models to improve the resilience, efficiency and livability of modern cities,” Sharma said.

This study is part of the U.S. Department of Energy Urban Integrated Field Laboratory project over the Chicago region called Community Research on Climate and Urban Science, in which Sharma is leading the climate modeling efforts to work on community-driven science. Sharma and Li emphasized the study’s relevance in guiding communities and municipalities toward informed investments, exemplifying the potential of science to empower decisionmakers and communities in building more resilient and livable cities.

“The team at CROCUS aims to build upon this prototype, connecting the study with other efforts to forecast street-scale winds and air quality in the near future, as well as observational effort to validate model results at the relevant scale. This research not only marks a milestone but also serves as a catalyst, unlocking opportunities to broaden the scope of applications in related fields,” said Cristina Negri, CROCUS project director at Argonne National Laboratory.

For additional information about this study, contact Ashish Sharma at

This work is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research’s Urban Integrated Field Laboratories research activity under contract number DE-AC02- 06CH11357.