Data analysis

Using data to make cities greener

As politicians stall when it comes to dealing with climate change on a national level, local data-based projects are trying to reduce carbon emissions on their own doorsteps

The take-away

  • In Denmark, researchers are using data-analysis tools to optimise energy consumption. The goal is, for example, to have people charge their electric cars on windy or sunny days.
  • In Paris, a scaled-down model of the real world is testing nanosensors that could help cities gather more climate-relevant data.

What can you do when politicians don’t make it a national priority to cut greenhouse-gas emissions? For some cities, the answer is to tackle climate change from the ground up.

Platforms like the C40 climate group, a data-driven organisation working with 96 cities around the world to meet the Paris Agreement’s climate goals, are helping municipalities take serious steps. Many are going beyond what their countries have pledged. Oslo, for example, plans to cut its CO₂ emissions by 95% by 2030 (vs. 1990 levels), though Norway as a whole has pledged to cut by only 40% – and in reality is on track to reach a mere 7%.

Swiss researcher Claudia Binder from the École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland, says that when it comes to sustainability the key is to “think globally, act locally”. While local action alone will not stop climate change, there are plenty of ways in which cities and neighbourhoods can become more sustainable.

One example is a model developed by researchers, also at EPFL, that looks at a group of buildings as a whole to design the best energy system. The work showed that optimising energy efficiency at this level is more cost-effective than doing it for each building individually.

“Heat, cold and wind all have an influence on energy needs, and buildings have an impact on each other, too,” says Dasun Perera, a researcher at EPFL’s Solar Energy and Building Physics Laboratory. The model, which combines a building simulator with an urban climate model to find the optimal energy system for the whole system, can apply to a whole neighbourhood – or even an entire city.

 

Electricity clusters

Per Sieverts Nielsen, a systems analysis researcher at the Technical University of Denmark, Copenhagen, hopes to use smart-metre data to help utility companies persuade customers to use energy more efficiently.

Nielsen and his colleagues are using data from smart metres in the Danish city of Esbjerg to cluster people into groups based on how they use their electricity. The idea is that utility companies could target campaigns toward people who, for example, have an electric car or a swimming pool with a heat pump.

For Denmark, where a lot of energy is generated through renewable sources like wind, optimising energy use is not as simple as using less electricity. Counterintuitively, it might be better for someone to use more energy overall – but to charge their electric car only on a blustery day. “We want to use electricity when the wind is blowing,” says Nielsen. “Customers may increase their electricity consumption by 5%, but they can reduce CO2 emissions simply because they use the power when the wind is blowing or the sun is shining,” he added.

Another project looked at the energy needed to run swimming pools in summer houses in two Danish towns. The results are still preliminary, but they suggest that carbon emissions could be reduced by around 30% if people heated their pools only when lots of electricity is being generated from wind.

Simulation of a city

Optimising the efficiency of entire new neighbourhoods or repurposing data from smart metres is one thing. But to take local, data-based action on climate change further requires, well, more data. “There is a lack of research and harmonized data to compare relevant indicators,” says Binder. Scientists hoping to make cities more sustainable are turning to nanosensors that are sensitive, cheap and energy efficient.

A project called Sense-City, based in Paris, is now testing such sensors in a scaled-down model of the real world. The 20-by-20-metre chamber, which has been operational since 2018, was built to test lab-developed sensors in an environment that’s closer to reality. Smaller than life-size, it’s still representative of a real neighbourhood with houses, roads, cars and buildings. It can even imitate the weather with simulated sunshine and rain.

“You want to verify that in a real situation your sensor is going to work,” says Bérengère Lebental, a researcher at IFSTTAR in France. An air pollution sensor, for example, might have been developed to measure nitrogen dioxide (NO2), but when it enters a real-life situation it’s going to encounter more than that. “The problem is that when you get into real life you not only have NO2, you also have ozone and particulate matter – so it’s much more complex,” she explains.

Lebental says that environmental nanosensors could be on the market within two years. While other researchers are investigating possible uses of nanosensors in monitoring air pollution and greenhouse-gas emissions, she is currently working on a project called Proteus developing nanosensors for better water monitoring.

More traditional sensors are already in use in other local climate projects. At Newcastle University, UK, sensors across the city are gathering data on 50 parameters from solar radiation to wind direction. The Urban Observatory project even allows communities to request new sensors to tackle specific challenges, such as flooding or air quality.

Claudia Binder (director of the Laboratory for Human-Environment Relations in Urban Systems, EPFL), Per Sieverts Nielsen (senior systems analysis researcher, DTU), Bérengère Lebental (researcher at IFSTTAR), EPFL, Statista, Climate Action Tracker, World Economic Forum, Oslo Climate Budget