Computational ‘time machine’ shows solar and wind on track for 2°C target but not for 1.5°C

Wind and solar power have grown faster than almost anyone predicted but projecting their future expansion remains surprisingly difficult. Researchers at Chalmers University of Technology, Sweden, have developed what they call a computational ‘time machine’ – a model that outperforms existing projection methods by using AI techniques to analyse historical growth patterns across countries. Their central projection shows that onshore wind is likely to supply around 25% of global electricity by 2050, with solar reaching about 20%. This is consistent with the 2°C target, but falls short of what is required for 1.5°C.

Predicting the future is particularly challenging for technologies like wind and solar, where rapid cost declines are offset by growing barriers such as public opposition, infrastructure constraints and policy shifts. 

‘Existing models are very good at identifying what needs to happen to reach climate targets, but they can’t tell us which developments are most likely. That is the gap we wanted to fill,’ said Jessica Jewell, a professor at Chalmers University of Technology. 

Across more than 200 countries, the researchers identified a recurring pattern in how wind and solar power grow: long periods of relatively steady expansion punctuated by sudden growth spurts often triggered by policy shifts.

‘Most models assume a smooth s-shaped growth curve, but that is not how it actually looks in the real world. Growth often comes in bursts, and if you ignore that, you can misjudge how fast technologies will expand,’ said Avi Jakhmola, PhD student at Chalmers University of Technology and first author of the paper published in Nature Energy.

So, with the goal of improving the predictions, Avi Jakhmola created a model built on 13,000 virtual worlds. In each of these worlds, solar and wind power develop in different ways – from the fastest possible expansion to the slowest – and everything in between. A machine learning algorithm was then trained on all these worlds to learn to predict global outcomes from early national trends.

‘When we apply the model to real world data, it can tell us what is the most probable outcome for the future – given what we have seen so far and given all the virtual worlds it has seen,’ said Avi Jakhmola.

By 2050, the model projects onshore wind reaching around 26% of global electricity (central range: 20 to 34%), and solar around 21% (15 to 29%). This broadly aligns with 2°C compatible pathways but falls short of what is needed for 1.5°C.

The projections also put the COP28 pledge to triple renewables capacity by 2030 in perspective. The pledge falls near the 95th percentile meaning that it would require growth rates rarely observed. 

‘The tripling of renewables pledge is not impossible, but it would require everything to go extremely well in all countries,’ said Jessica Jewell.

The researchers also tested what would actually be required if we are to reach the 1.5°C goal. 

‘If we start now, the required growth rates are demanding but not unprecedented, comparable to what the EU targets for wind with REPowerEU and what India has planned for solar power,’ said Avi Jakhmola. ‘But if we delay until 2030, the acceleration needed becomes much steeper and much more abrupt. The window for ramping up closes quickly.’

The researchers also used the model to test the reliability of its projections – by going back in time. 

‘We wanted to know if our projections will hold up ten or 20 years from now. When we fed the model only data from 2015, we found that it correctly predicts what has happened since then. This is what we mean by a ‘computational time machine’ and it gives us real confidence in the projections going forward,’ said Avi Jakhmola. 

The study points toward a broader ambition to develop scientifically rigorous methods for projecting the most likely growth paths for other low carbon technologies, not just wind and solar.

Jessica Jewell said, ‘It has long been a joke how bad technology forecasts are. But if you are a decision maker, trying to figure out how hard to push for change, you need a realistic baseline. Our study is the first step towards developing such a realistic view of the future.’

The paper 'Probabilistic projections of global wind and solar power growth based on historical national experience', has been published in Nature Energy. The researchers have also made an online visualisation tool of the results, available at the Energy Technology and Policy website. The authors are Avi Jakhmola, Jessica Jewell, Vadim Vinichenko and Aleh Cherp. The researchers are active at Chalmers University of Technology and Lund University in Sweden, University of Bergen in Norway, International Institute for Applied Systems Analysis and Central European University in Austria.