Every year, governments and investors count new solar capacity and update their climate projections accordingly. More panels in the ground, less coal in the mix – that’s how the math works.
A new study found a systematic flaw in that accounting, driven by something traditional capacity models ignore entirely. The source sits in the air between the sun and the panels.
Researchers at the University of Oxford (Oxford) and University College London (UCL) mapped where the world’s solar panels actually sit – more than 140,000 sites in all.
The team then matched each site to atmospheric data on aerosols, the airborne particles that scatter and absorb sunlight before it reaches a panel.
In 2023, those aerosols cut global solar electricity output by 5.8 percent.
That works out to about 111 terawatt-hours of lost generation – roughly the annual output of 18 medium-sized coal plants. 
The research was led by Dr. Rui Song, a physicist at Oxford and UCL’s Mullard Space Science Laboratory.
The findings mean clean energy is delivering less than expected, noted Dr. Song.
“As coal and solar expand in parallel, emissions alter the radiation environment, directly undermining the performance of solar generation,” he said.
Here’s the harder number. Between 2017 and 2023, new solar installations added an average of 246.6 terawatt-hours to the world’s grid each year.
Aerosols on existing solar farms canceled about 74 terawatt-hours of that – nearly a third of the gain.
Earlier studies had documented the problem in heavily polluted regions. None had ever matched it to actual output, site by site, at this scale.
China generated 793.5 terawatt-hours of solar electricity in 2023 – more than 41 percent of the world’s total – and took the hardest hit.
The country lost 7.7 percent of its potential output to airborne particles.
In some years, aerosol losses erased more than 60 percent of what new solar installations added. The gain on paper kept outrunning the gain at the meter.
Sulfate aerosols were responsible for the most damage. They form in the atmosphere from sulfur dioxide – the gas coal stacks pump out when burning.
The Oxford-UCL team calculated that sulfate alone accounts for 46.2 percent of China’s aerosol-driven solar losses.
A separate analysis of pollution sources estimated that about 29 percent of those losses come from coal-fired electricity.
Coal plants release fine particles that scatter and absorb sunlight before it can reach a solar panel‘s surface. Less light in means less electricity out.
In the United States, the picture is different. American solar farms lost only 3.1 percent of their output to aerosols in 2023 – less than half China’s rate.
This can be explained by geography. Most U.S. solar capacity sits far from active coal plants.
When the researchers mapped coal capacity against solar losses region by region, they found essentially no correlation in the U.S.
China’s results came back significantly positive on the same test. Where coal stacks densest, solar output is the worst.
Here’s the twist. China sits at the top of the loss table, but it’s also the only major region where those losses are falling.
Aerosol-driven solar losses dropped by about 1.4 percent each year from 2013 to 2023.
The improvement didn’t come from closing coal plants. China’s coal-fired electricity output rose from 4,093 to 5,857 terawatt-hours over the same decade.
The cleanup came from inside the fleet through ultralow-emission retrofits: tighter scrubbers, aggressive filters, stricter pollutant limits.
A separate paper documented those upgrades in detail. Retrofits account for 91 percent of the sulfur dioxide reductions over China’s solar regions; plant closures contributed nine percent.
The researchers stitched together open-source databases, crowd-sourced records, and their own machine-learning model trained on satellite imagery to find solar farms anywhere on Earth.
Each detected site was then run through an AI model that drew clean boundaries around the panels – not the access roads or the dirt strips between rows.
That detail is what makes the new analysis work. Earlier research estimated pollution losses by assuming solar was spread evenly across regions.
The new dataset shows where panels actually sit – in China, often within 20 to 30 miles of a coal plant.
Researchers now have a number they didn’t have before. Air pollution from coal plants doesn’t just sicken nearby populations – it directly suppresses the output of the panels meant to replace them.
That 5.8 percent global drag is enough to throw off climate projections that count installed capacity alone. Cleaning up existing coal plants helps, as China’s record shows.
But the same plants kept running as grid backup will keep emitting the particles that cut into solar’s value. Leave those losses unaccounted, and the world’s clean energy progress is likely overstated.
For regulators and grid planners, the question now is where new solar gets built – and how fast the coal capacity nearby comes offline.
The study is published in the journal Nature Sustainability.
—–
Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.
Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.
—–

source