Climate Change Model Identifies Global Warming’s Impact on Electricity Usage and Demands  

March 13, 2020 by Luke James

Global warming alone could push Chicago to generate 12% more electricity for each person each summer month by 2030.

According to projections from a new climate change model developed by Purdue University researchers, U.S. midwest cities could easily face widespread power shortages if this percentage is lower. This would mean that the city would need to implement stringent measures to avoid rolling blackouts. 

This 12% estimated increase is much larger than previous projections, too, owing to the fact that it now accounts for how consumers use both electricity and water at the same time, such as when a dishwasher is used or when water is heated. 

The researchers’ model also ponders a wider range of climate features that may affect mixed-use—for example, humidity—which makes it even more accurate.


Power Lines City

The climate change model factored in the use of both electricity and water since consumers utilize both at the same time when running electrically-powered household appliances. 


Creating a New Model for Climate Change

The Purdue University team—led by Roshanak Nateghi, a Purdie assistant professor—partnered with Rohini Kumar, a postdoctoral researcher at the Helmholtz Centre for Environmental Research in Leipzig, Germany, to develop the model. 

Originally published in the journal Climatic Change, the collaborative research effort applied their model to five other cities in the U.S. Midwest: Cleveland, Columbus, Indianapolis, Madison, and Minneapolis. 

Overall, the model projected that the U.S. Midwest will be using 19% more electricity and 7% more water by 2030—and this is only accounting for the summer months. And although the model does not yet account for technology, technological advances (such as the growth of EVs), and population growth, neither do the models that utility companies currently use. Nateghi’s model does, however, account for other variables that current models do not, such as wind speed, relative humidity, and large-scale climate phenomena.  

Adding these variables makes Nateghi’s model far more representative of future climate change scenarios. 


Why Chicago and the U.S. Midwest?

The Purdue researchers used Chicago as an example because it is a very windy city, and wind speed is an important factor when estimating concurrent electricity and water use. In other parts of the country, such as in the southwest, extremes of other factors—temperature—may play a larger role. 

On the whole, the U.S. Midwest was chosen for the model’s application because this part of the country exhibits more distinct seasonal weather and thus provides a better opportunity for the model to be thoroughly tested. It can be applied to any region in the U.S. and indeed the wider world, however. 


Manhattan during a power outage.

Lower Manhattan during a power outage. 


The Model’s Predictions

To make its predictions, Nateghi’s model uses artificial intelligence. To help it learn, the model was fed years’ worth of data from the U.S. Midwest’s utilities and weather services. It was then trained to predict changes in electricity and water consumption for given climate change scenarios.

Climate scientists have predicted that global warming could cause temperature increases by 1.5 degrees Celsius by 2030 and 2.0 degrees Celsius by 2050—these scenarios are what Nateghi’s team plugged into the model. 

According to the new model, Chicago will see an increase in electricity and water use by 12% and 4% respectively (the best-case scenario) if temperatures increase by 1.5 degrees Celsius. If, however, temperatures increase by 2.0 degrees Celsius, the model predicts that, as a worst-case scenario, there will be a 20% and 6% increase in use. On average, the entire U.S. Midwest will see a 10%-20% increase in electricity use and a 2%-5% increase in water use. 

This new model could be crucial to improve our understanding of how electricity and water usage together could influence future climatic conditions. It will potentially enable researchers to adapt to possible changes and develop future-proof solutions and processes.