The United Nations Global Pulse, an innovation initiative on big data and data science, and Western Digital recently announced the winners of the Data for Climate Action Challenge (D4CA) at the Data Innovation: Generating Climate Solutions event during the United Nations climate change conference (COP23) in Bonn, Germany.
An unprecedented open innovation challenge to harness data science and big data from the private sector to fight climate change, D4CA called on innovators, scientists, and climate experts to use data to accelerate climate solutions. Access to large amounts of data – anonymized and aggregated to protect privacy – accelerates the ability to spot connections, gain insight and develop predictive algorithms that can provide more precise direction and decisions. The Data for Climate Action Challenge demonstrates what’s possible when public and private sector organizations partner for social good.
WIRED Brand Lab: Can you give us an overview of your submission?
Constantine E. Kontokosta: We built a high spatial-temporal model of greenhouse gas emissions across New York City. We wanted to understand at an hourly and hyper-local level what the near real-time emissions from buildings and transportation were across the city.
WBL: What was it about New York City that made it an ideal candidate for the study?
Kontokosta: It’s a very data rich environment. We have access to really detailed data about the energy consumption of buildings, traffic patterns, where every street tree is, and very local temperature readings. That allowed us to combine all these detailed information sources and build this detailed look at the city in a way no one has been able to do before.
WBL: How was analysis such as this being carried out previously?
Kontokosta: New York City, like many other cities, conducts greenhouse gas inventories every year. But typically, these are done annually and at the city scale – they give no real indication of what happens in certain parts of the city at certain times. So, this was really an attempt to downscale that to provide not only climate scientists, but policy analysts an understanding of where the hotspots are in time and in space across the city.
WBL: What datasets did you use?
Kontokosta: We started with detailed land use data from the city planning department. That gave us really detailed information on the 1.2 million buildings in the city. We then layered in energy consumption data – we know when they were built, how much energy they use and how efficient they are.
We mapped in all the street trees because they have an impact on energy consumption, as well as local temperatures. We modelled the energy consumption at an hourly rate and for every building, and then we combined traffic counts and supplemented that with Waze data so we can get a more granular picture of what’s going on.
The last piece of the puzzle was sentiment analysis. We did some sentiment analysis on Twitter to see if there was some relation between poor air quality and people’s statements on the social network. It was good to see if that social listening component could become a good correlator to actual data.
WBL: Did you find a correlation?
Kontokosta: We haven’t explored it enough to give you a definitive answer. In the time frame we had and the time period we used, we didn’t have time to clear out all of the bots who were automatically tweeting out information that was skewing results.
WBL: What time period did you use to build the data visualization?
Kontokosta: We built an hourly model over the entire year of 2015.
WBL: What were some of the top-level findings of this?
Kontokosta: My team has done a lot of work on building energy and transportation emissions models across the city for a number of years. We know the landscape well. But we did find emissions were higher per capita in lower-income communities and this was not unexpected, but to be able to provide hard data for that is very powerful.
This trend was largely being driven by heavily used roadways bisecting in these areas, especially in the Bronx, in addition to lower quality housing, which is very energy inefficient, causing more emissions to be produced and released.
WBL: How does this information help policymakers in the city?
Kontokosta: On the one hand, now we can be more evidence based on policy design and analysing existing policies. We can try something out and see how it impacts emissions in a given area and measure that impact with greater accuracy.
Then there’s another piece of this dynamic—which is in real-time or near real-time. A typical example would be to look at waste collection trucks on emissions at a localised level. Previously we had no real understanding of the impact they have to the local environment for the hours they are around.
WBL: Was there anything else you saw in the results that was unusual or surprising?
Kontokosta: New York City was an interesting test case because 79% of the city’s emissions came from buildings. Because public transport is so efficient, we don’t use vehicles as much and so buildings are the biggest contributors. Contributing factors would be very different for other parts of the U.S.
WBL: What happens next?
Kontokosta: We’re working with the city of New York to build an operational version of this for use at the mayor’s office to help with long-term sustainability planning but also to aid in the design of new policies which will try and nudge behaviour change. All being well, we’d love to one day take it to other cities.
All images owned and provided by Constantine Kontokosta, NYU Urban Intelligence Lab
This content is produced by WIRED Brand Lab in collaboration with Western Digital Corporation
Data Makes Possible will continue to follow the D4CA winners as they work to implement their solutions and bring real change to our world.
Get news and stories from Data Makes Possible delivered to your inbox, including when new winner interviews go live: