The researchers from the University of Oxford, University College London, the University of Leicester, and the University of Bath have found out that coal-fired power plants, which have been the backbone of many countries’ energy systems and a major source of greenhouse gas emissions, also reduce the solar power efficiency. Their study demonstrates that atmospheric aerosol pollution generated by coal-fired power plants reduces the global solar power generation by approximately 5.8%. In 2023, this resulted in a loss of about 111 TWh of electricity which is roughly equivalent to the annual consumption of the Netherlands.
For a long time, the energy transition logic seemed fairly straightforward: the more solar power plants appear in the world, the faster fossil fuel dependence decreases. However, practice revealed that part of the effect of new solar capacity literally dissolves into the polluted air. According to the British researchers’ calculations, on average between 2017 and 2023, generation losses caused by aerosols are comparable to nearly one-third of the annual energy supply by new solar power plants across the globe.
To draw these conclusions, the scientists developed a three-step algorithm combining Sentinel-2 and PlanetScope satellite imagery, data from the OpenStreetMap crowdsourced map, and their own machine learning algorithms.
At the first stage, the researchers identified all the global potential solar power plants, a total of nearly 141,000 sites. Then, using the Segment Anything neural network, they refined the actual boundaries of solar panels, separating them from empty land areas and surrounding infrastructure. Afterward, they compared the resulting map with the NASA MERRA-2 atmospheric data, which allowed for an individual assessment of the generation losses due to clouds and air pollution for each plant.
As expected, China—the world’s largest solar energy market and, at the same time, one of the largest coal consumers—emerged as the most telling example. Nearly half of all the solar generation losses in the country are attributed to sulfate aerosols formed by sulfur dioxide emissions from coal-fired power plants. Carbon aerosols also contribute to the reduced efficiency of solar panels.
At the same time, the new methodology allowed for correction of the previous damage estimates, which often did not account for the exact location of solar power plants. According to more accurate calculations, actual losses in solar generation in China amount to about 7.7%.
The researchers also identified a key pattern: a lot of solar power plants in the country are located relatively close to coal-fired power plants, resulting in losses from aerosol pollution in some years and equivalent to more than 60% of the additional energy provided by new solar capacity.
It is a paradox but China turned out to be the only major country with steadily decreasing losses in solar generation caused by aerosols—by approximately 1.4% per year—despite a continued growth of coal-fired generation. As the authors note, this is primarily due not to closure of coal-fired power plants, but to a large-scale modernization of thermal power plants and implementation of ultra-low emission standards which made it possible to significantly reduce the volume of pollutants. By comparison, in the United States, where solar and coal-fired power plants are geographically far apart, losses in solar generation amount to about 3.1%.
The researchers have emphasized that in the coming years, coal-fired power plants are likely to preserve their important role in maintaining stability of power grids. But at the same time, they believe it is necessary to change approach to the geographical placement of solar power facilities so that to avoid an excessive concentration of solar power plants near large coal-fired power plants. Finally, the researchers believe it is important to adjust the very principles of assessing the energy transition as without accounting for generation losses caused by air pollution, solar energy contribution to reducing emissions and improving air quality may be systematically overestimated.
