After reading my first post, you might be wondering, what do artificial intelligence (AI) and data science (DS) have to do with reversing global warming? I’ll start to answer that question in this post with some examples. And of course future posts will include many more examples, and go into more depth.
Sectors
Many emissions analyses divide the world into economic sectors. The EPA publishes data on global emissions by sector based on analysis by the IPCC, and summarized in the figure below. Let’s examine some Drawdown Data Science examples for two large sectors, noting that the pie chart below is based on 2010 emissions.
First the energy sector has a huge impact on emissions. As you can see in the pie chart, the two slices for energy account for the largest proportion of greenhouse gas emissions. How can we use data science ideas here? Energy management by utilities is a complex topic that I won’t dive into right now, but suffice it to say that utilities need to make decisions about starting and stopping power stations, among other choices. If they have insufficient information it can lead to emergency energy purchases from neighboring utilities or to firing up a fossil fuel power station unnecessarily. Predicting energy consumption and generation can help with these decisions. Both types of predictions can be aided by machine learning, as illustrated by this Kaggle competition for predicting solar energy, this recent review of consumption prediction, and this DrivenData competition to forecast energy consumption.
A second, less obvious sector with many opportunities is the agriculture industry, which together with forestry and other land use accounts for 24% of the emissions pie. DS application examples here include tools that support decision making by farmers using sensor data and forecasting, automated irrigation and disease monitoring, and analyzing seed data to predict performance.
If you’d like to explore emissions visualizations by sector and many other lenses, Our World in Data offers a page on CO₂ and other Greenhouse Gas Emissions.
Organizations
There are an increasing number of companies and organizations doing interesting things with AI or DS applications to that might help reverse global warming. I highlight a handful next. I have no affiliation with these companies but have been watching the space for the past few years.
Advanced Microgrid Solutions offers an AI energy markets trading platform. They mention deep learning and probabilistic forecasting on their web site, both of which are DS techniques. They spoke at AWS re:invent about their approach to “optimize market participation of clean energy assets in Australia’s National Energy Market.” I couldn’t find this on their web page, but in their talk, they presented a mission statement that explicitly mentions their leadership role in the transformation to a clean energy economy.
At least two companies focus on intelligent building management and efficiency improvements. First, BuildingIQ has a platform that “…is built on the five pillars of data capture analysis, advanced modeling, measurement & verification, closed loop predictive control, and expert human analysis.” They mention visualization and machine learning on their home page. Second, Verdigris uses AI to analyze sensor data to detect equipment faults. Revisiting our sector pie chart above, buildings account for 6% of global greenhouse gas emissions. More carefully managing building’s energy use has potential to reduce their emissions, so I was excited to find these companies solving important problems.
On the government side, the National Renewable Energy Laboratory, or NREL is a federal lab dedicated to “research, development, commercialization, and deployment of renewable energy and energy efficiency technologies.” This page points to some of their work involving data and data science.
Finally, Microsoft has launched AI for Earth, providing grants, education, and investment for advancing environmental sustainability.
And Much More!
There are many other areas we’ll touch on. One domain is human health: an example is deciding the status of a water pump from data related to other pumps in the area. Second, even ridesharing can help reduce auto emissions, and algorithms for matching riders with drivers can include forecasting, planning, and other AI techniques. Finally, I’ll periodically include examples from the field of data visualization, such as this one from Ed Hawkins which shows the historically observed changes in temperature from 1850 onward.
What examples have you heard about?
References
