Data science has been used extensively in building climate models, downscaling climate models to regions, monitoring and evaluating the accuracy of climate models through paleoclimate data as well as developing methods to mitigate the effects of climate change and develop alternative markets.
Today’s post will look at some of the more straightforward uses of data, machine learning and spatial statistics in monitoring carbon emissions as well as building alternative market systems.
The first of course is monitoring and measuring carbon emissions and emissions of other gases that contribute to the changing climate. Our first example comes from Europe.
Researchers in Europe created a tool to map the 177 regions in 27 countries of the EU and the carbon footprint associated with them. They used a database (EXIOBASE 2.3 multiregional input-output database) with detailed information about the world economy in 2007 and built a model that looked at the different factors impacting carbon emissions in different regions. Their results showed that there is significant spatial variability , income and local industries matter and carbon emissions due to outsourcing production are significant contributors.
This is a study that is fairly typical of how data science is used to solve clean tech problems – databases with large amounts of data are used, pipelines are built to extract the data, analytical models are created from the cleaned data to test different relationships and the results are presented visually to make decision making easier for the final consumer.
A second sector is looking at how models can be built that can account for carbon movement and carbon trading. Here, we take a look at what’s happening in China as well as how models built by researchers in Europe could improve the market efficiency in China.
China set up seven regional carbon trading trials in 2013 where carbon allowances totaling 120 million tonnes are traded between more than 2000 firms in various sectors. The country now plans to take these trials and develop a nationwide carbon trading market in 2017 with 3-5 billion tonnes of carbon allowances every year – a market that is going to be the largest one in the world. The market will be restricted to firms in eight sectors initially – petrochemicals, chemicals, building materials, steel, ferrous metals, paper-making, power-generation and aviation.
According to the National Development and Reform Commission’s report in January 2016, over 7000 firms are likely to participate in the market – firms that together contribute to over half of China’s current emissions.
One of the interesting aspects of making environmental markets like the one China’s proposing work is understanding how to model and account for emissions from all these different sectors and products. A study from Norway looks at how global supply chains can be modeled and mapped to track air pollution impacts around the world. The scientists combined remote sensing data with lifecycle models called “multi-region input output models” or MRIOs to build spatial maps of how economic activity drives environmental impacts around the world. As seen in the maps generated in this study, there are areas of high environmental impacts – “hot spots” – that can be targeted first to mitigate environmental impacts. Some of this targeting can be done through local regulations – the rest, which depends on the behavior of actors outside the country or region, can be met through combinations of technology and trade agreements.
This is where markets like China’s carbon market become useful tools – when companies and governments can use models like the one built by scientists in their study, they can figure out where the price of carbon should be the highest or which actors would need the most allowances to meet their emission requirements.