New Model Uses AI to Make Predictions on Climate Change Impact

Global warming alone is sufficient to make Chicago produce 12% more electricity for each person each summer month, by 2030.

Cities such as Chicago could experience blackouts by 2030 if they don’t prepare to meet projected increases in electricity and water use due to climate change, a new study reports. Image Credit: Stock photo/Pexels.

According to projections from a model developed by researchers at Purdue University, if that percentage is less than that, the city may probably face a power shortage that may need stringent measures to prevent rolling blackouts.

That projected increase is larger when compared to the earlier projections because it considers the way consumers utilize water and electricity simultaneously. The new model also accounts for an extensive range of climate features that have an impact on this mixed-use, like wind speed and humidity, resulting in more precise predictions.

Consumers will utilize both water and electricity when landscaping, heating water, or running a dishwasher. Moreover, cities use water to produce electricity, and electricity for the treatment and distribution of water.

Chicago is a windy city and, hence, the speed of the wind is important when estimating the use of water and electricity. Temperatures may have a bigger role to play in the Southwest, where droughts generally take place.

Usually electricity and water utilities work in silos. But if we want to accurately capture climate-sensitive demand, we need to look at electricity and water together.

Roshanak “Roshi” Nateghi, Assistant Professor, Industrial Engineering and Environmental and Ecological Engineering, Purdue University

Nateghi’s group developed the model in association with Rohini Kumar—a postdoctoral researcher at the Helmholtz Centre for Environmental Research - UFZ based in Leipzig, Germany.

In a study, recently reported in the Climatic Change journal on March 5th, 2020, the collaborative research group applied the new model to five more cities in the U.S. Midwest—Columbus, Ohio; Cleveland; Minneapolis, Madison, Wisconsin; and Indianapolis.

On the whole, the model estimated that the U.S. Midwest will be utilizing 7% more water and 19% more electricity, and that is only during the summer season.

The scientists began their study with the Midwest because this region is known to experience distinct seasons; however, the new model can be applied to all types of regions.

While the projections made by the model do not consider the technological shifts or population growth, such as increased use of electric vehicles, a standard baseline model that is presently used by utilities to predict the impact of climate only considers the way precipitation and temperature have an impact on water and electricity use.

In the model designed by Nateghi’s laboratory, these variables and also wind speed, relative humidity, and large-scale climate phenomena—like El Niño that frequently results in more mild winters in the Midwest—are taken into consideration.

Adding in these other variables makes the model more representative of future climate-change scenarios.

Renee Obringer, Study First Author and PhD Candidate, Environmental and Ecological Engineering, Purdue University

Using artificial intelligence, the new model predicts the effect of climate change. This model serves as a learning algorithm and fed with years of information from the utilities and weather services of a region and further trained to predict the changes in the use of water and electricity under specific situations of climate change.

Such scenarios occur when the temperature of the Earth increases by 1.5 °C or 2.0 °C above its mean temperature at the time of the pre-industrial period, around 1881 to 1910.

According to climate researchers, global warming could exceed the 1.5 °C threshold by 2030 and the 2.0 °C threshold by 2055.

This implies that that the best-case scenario for Chicago is that water use and electricity use increase by 4% and by 12%, respectively if global warming crosses a threshold of 1.5 °C. However, if a threshold of 2.0 °C is reached, then the worst-case scenario would be an increase of 6% in water use and an increase of 20% in electricity use.

Kumar added that, “Such scenarios are fundamental for understanding the joint response of electricity and water uses to future changes in climatic conditions as to understand to what degree our current management and technological strategies need to adapt to the future changes.”

The scientists discovered that on average for every city examined in this research, there could be an increase of 2% to 5% in water and an increase of 10% to 20% in electricity during the summer period because of a warming climate.

A baseline model only looking at temperature and precipitation is used over and over again to develop policies. In the future, there could be significant shortages in water and electricity supply because these models have been significantly underestimating what the actual demand would be.

Renee Obringer, Study First Author and a PhD Candidate in Environmental and Ecological Engineering, Purdue University

While there is always space for more precision, the scientists believe that the accuracy level achieved by this new model should make it viable for use by city planners and utilities to establish policies that are more effective.

This model provides a much better sense of potential risk for variability and change,” Nateghi concluded.

The study was financially supported by the Purdue University Center for the Environment and the National Science Foundation (grants 1826161 and 1832688).

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.