Ithaca is a picturesque town in upstate New York famous for its waterfalls and gorges, and the world-renowned Cornell University. The area has long been hailed as a beacon of progress. Cornell’s students and scientists, for example, helped discover that water once flowed on Mars, built the first fully functioning synthetic human ear, and learned from famed astronomer Carl Sagan.
Now, Ithaca and its 32,000 residents have big plans to extend their innovative legacy by leading in urban decarbonization. The city has decided to replace its fossil fuel-burning infrastructure, like boilers and stoves, with electric-powered heat pumps and induction cookers. The work is planned for all of the city’s 5,500 buildings, with the goal of reaching carbon neutrality by 2030.
The plan is to reach 100% electrification as efficiently and economically as possible—and at the center of the project is a digital twin. The digital urban building energy model was developed by researchers from Cornell’s Environmental System Lab and the Rocky Mountain Institute. Part of the work is sponsored by Bentley Systems, the infrastructure engineering software company and a leader in digital twins and artificial intelligence (AI) for infrastructure. Bentley is also helping share lessons learned from the project with policymakers worldwide.
Insights from Ithaca could also inform how we build massive, power-hungry data centers, the backbone of the boom in AI and generative AI. The Electric Power Research Institute reported last May that data centers already accounted for a quarter of Virginia’s electric load in 2023 and “were projected to make up almost one-third of Ireland’s total electricity demand by 2026.” Hung Ming Tseng, a researcher at Cornell’s Environmental Systems Lab, says the conversation around data centers goes beyond sustainability to efficiency, resilience and security. “We’ll be thinking about these things as we scale this up to the national level,” he says.
Ithaca’s Digital Beating Heart
Ithaca Green New Deal project shows the city’s buildings color-coded from red to green, indicating different heights or heat levels. Image courtesy of ESLab, CRLab Cornell.AI is also part of Ithaca’s digital twin, which is a virtual representation of the town’s streets and buildings. The digital model uses AI and other advanced technologies to capture and render reality. Those technologies include Bentley tools that can accurately reconstruct 2D images in dazzling 3D, all the way down to the position and depth of an individual window frame. This level of detail is important because the Ithaca team aims to first retrofit the least energy-efficient buildings, like those leaking heat through their windows.
The digital twin provides a wealth of insights, such as ranking Ithaca’s properties by their cost-benefit ratio for heating electrification. This allows local planners to prioritize retrofits and deliver on their mission of supporting low-income neighborhoods through the energy transition. For example, teams can use the digital twin to analyze the entire “envelope” of individual structures, such as their age, energy efficiency and rooftop solar potential. Those details can help them find the most cost-effective next steps, which might be as simple as adding wall insulation, upgrading windows or sealing air leakages.
Beyond Ithaca, the project can serve as a model for resource-strapped municipalities around the world that often struggle to effectively manage infrastructure assets, says Dru Crawley, Bentley’s director of building performance. “It’s important to get ‘campus-scale’ twins out in the marketplace so that [municipalities] can see how they work and understand how they’ll help them achieve their goals,” he says.
Small town, big plans
Ithaca’s program began in June 2019, when it adopted the Ithaca Green New Deal resolution, setting the carbon neutrality target by 2030 and sharing the benefits with historically disadvantaged communities, including those who have been priced out of the city center. Ithaca has the advantage of partnering with experts at Cornell, which was the first university to commit to the Kyoto Protocol for carbon reduction back in 2001. Today, it operates its own renewable energy plants, including a small-scale hydroelectric plant, 15 solar projects and even a geothermal source. It also boasts the Atkinson Center for Sustainability, a purpose-built, state-of-the-art research facility.
“Ithaca’s sustainability directors have always been amazingly energetic and positive,” says Timur Dogan, director of Cornell’s Environmental Systems Lab. “They found us, and we began talking. It’s easy to say, ‘Let’s just sit it out, and the grid will eventually become cleaner.’ But we wanted to help write the template for helping a whole town or city that’s pushing to decarbonize.”
The horns of a trilemma
One obstacle for researchers was the energy trilemma, the ever-present challenge to reconcile three often conflicting priorities: sustainability, affordability and security. Tseng, the researcher at Cornell’s Environmental Systems Lab, explains that Ithaca couldn’t just scrap its gas-fired boilers, install heat pumps and flick a switch. “That would have doubled power demand, with no improvements in sustainability or affordability,” he says.
The team had limited room to maneuver on the affordability front. “The fuel price for gas in the U.S. is significantly cheaper than electricity,” says Dogan, who trained as an architect in Germany and later earned his master’s degree from Harvard University and his Ph.D. from the Massachusetts Institute of Technology.
Another challenge was finding people and budget. Just one person—Rebecca Evans, Ithaca’s director of sustainability—is responsible for simulating the energy demand of 5,500 buildings. High priority items, such as wastewater and fire protection, fiercely compete for their share of the town’s annual budget, which totals just over $106.4 million in 2025.
The digital twin has been “immensely helpful” for Ithaca, Evans says. “From strategic decarbonization planning and the deployment of renewables, to attracting financial investment and technology partners, the value of the tool is understood widely.” She says the hope for both climate and policy practitioners in the state is scaling the project.
The art of digital twins
Creating high fidelity 3D Models for the City of Ithaca that can be used in building simulation. Image courtesy of ESLab Cornell.Cornell researchers started by honing a retrofit recipe for Ithaca that would provide the greatest opportunity for cost-effective decarbonization. The team then worked out what kind of buildings would give them the best return—or in Dogan’s words, “bang for the buck”—in reducing emissions and supporting low-income communities.
The team had plenty of technical know-how. Dogan studied urban energy modeling at MIT, while other members of the Cornell team participated in a project that examined the grid impacts of electrification from rapid adoption of electric vehicles and heat pumps. (That project was funded by Siemens, the global tech company, and the New York State Energy Research and Development Authority, or NYSERDA.)
For the Ithaca project, researchers harnessed their automated urban building energy model, which gave them a digital replica of the energy performance of the town’s buildings. The Cornell team fed hard data from maps, tax records, utility bills and other sources into this digital model. When combined with the RMI Green Upgrade Calculator, the data allowed researchers to measure the cost and environmental benefits of common decarbonization solutions for the typical American home, such as installing an air-source heat pump or rooftop solar.
Researchers also sharpened their urban digital model with geospatial data, including images obtained by LIDAR, a remote sensing method that uses laser to capture the contours of buildings, down to the all-important window seals. “That allows us to precisely reconstruct buildings,” Dogan says, noting that such details would help Ithaca accurately assess insulation requirements.
To drill down even more, Bentley tools harnessed photogrammetry techniques, a method of generating 3D structures using 2D images. “It uses machine learning that allows us to go down to understanding where a building’s windows are,” Dogan says. That would be crucial for locating and understanding every building owner’s bogeyman: radiation through windows.
Stunning visuals
Drone remote controller displays aerial view of Ithaca. The captured images were used to create high fidelity 3D models of the city that is used in building simulation. Image courtesy of Jason W. Koski.Visually stunning, the digital twin renders buildings and blocks in bright colors: yellow for commercial properties, and red for residential ones. It gives users a dynamic, bird’s eye view of Ithaca’s buildings. Users can hover at 10,000 feet for an overview of the color-coded cityscape or zoom in on individual buildings for a snapshot, including details such as tax rate, construction date and materials.
The digital twin churns out insights. For example, it has deduced that air-source heat pumps are not always economical because of Ithaca’s chilly climate and power prices. But if building owners add insulation and upgrade windows as they install the pumps, they can save $330 per month in operating expenses.
Adding rooftop solar into the mix boosts the benefits. According to calculations by the digital twin, an average home built before 1980 could save more than $12,400 over 15 years by adding solar panels along with an air-source heat pump, insulation and upgraded windows, compared to a property that continues to use gas and air conditioning. “We learned that you need to pair any electrification upgrades with something extra to really make sense,” Dogan says.
The digital twin is also making some unexpected discoveries. The Cornell team was expecting Ithaca’s commercial buildings to provide the best potential for cost-effective decarbonization. But the model indicated that some multi-family buildings in low-income neighborhoods were a better bet in terms of kilograms of carbon dioxide saved per dollar spent. “That was a nice surprise,” Tseng says.
He thinks there’s much more to come. The digital twin can identify “symbiotic relationships” between clusters of properties, allowing Ithaca to unlock economies of scale. For example, the digital model might indicate that hundreds of houses in a certain district have a similar heating load or share a geothermal well, which allows for the bulk purchasing of equipment and more efficient heat pumps, helping to lower costs.
As Ithaca moves closer to its 100% electrification goal and adds renewable capacity, the city’s load curve—or the pattern of its power consumption throughout the day—will change. This puts extra pressure on utilities that are constantly aiming to match supply with demand. But they, too, can draw on the digital twin’s insights about the rate and nature of electrification to smoothly ride the curve.
Dogan thinks closer cooperation between Ithaca and its energy suppliers will be positive. “It’s good to collaborate with the folks that have been predicting and distributing your energy for decades,” he says.
Data-driven insights also allow Ithaca to remain pragmatic in its approach. “Digging up the road and replacing gas pipes is a huge cost, so we could start [electrification] with branches that are at the end of their life,” says Tseng. Conversely, the city might discover that electrification is economically unjustifiable in certain areas. “In that case, we might just leave the gas pipe where it is,” he explains.
The Cornell researchers, who are in close conversations with the National Institute of Building Sciences, believe that the digital twin is highly scalable. “I’d love to see this applied to somewhere bigger than Ithaca,” says Crawley, Bentley’s building performance leader. “It would be great to see AI that allows building owners to instantly understand what incentive programs will help their property reach their city’s sustainability goals in the most cost-effective way.”
Through its charitable foundation, Bentley is sponsoring digital twin demonstration projects at leading research universities in Lithuania, Ireland, Scotland, Canada, India and the U.S.
Bentley’s Infrastructure Policy Advancement (IPA) team is working with the Cornell and Ithaca teams to share learnings from the Ithaca project and apply them across the country and globe. “To our knowledge, this is the first time a digital twin has been applied in this way in the United States,” says Rory Linehan, director of IPA for the U.S. “We think this has real potential to accelerate municipalities’ sustainability and inclusion goals while achieving cost savings. There is no reason this couldn’t be applied to a city even the scale of New York City.” In fact, the Cornell and IPA teams will deliver their findings of the Ithaca pilot to New York City officials at Bentley’s Transforming Infrastructure Performance Summit in New York City on May 22.
