How Much Output Do Dirty Solar Panels Lose in LA? What the First Rain Gives Back, Measured on 219 Roofs
Not much. Across 219 rooftop systems and 41 rains since 2022, the first good rain after a dry spell of three to six months lifted clear-day output by a median 3%; a rain after a few dry weeks moved it by about 1%, inside the normal day-to-day noise. Modeled over a whole year, dust cost a typical San Fernando Valley roof about 1% of its output (0.6–1.6% in 2023–2025). The exception was ash: after the January 2025 fires, the February storm lifted output by 9.6%.
Dust settles on panels between rains and the rain washes it off, so the jump in clear-day output after a rain shows what the dirt had been costing.
The jump after rain, by length of the dry spell
41 rains at four weather points, 2022–2026.
| Dry days before the rain | Rains | Raw change | Corrected for cooler weather |
|---|---|---|---|
| under 30 days | 20 | +1.1% | +0.6% |
| 30–90 days | 16 | +1.4% | +1.2% |
| 90–180 days | 5 | +5.2% | +3.2% |
| 180 days or more | 0 | — | — |
| Dry days: days since the last day with 5 mm of rain or more. A rain counts when it reaches 5 mm in a day and at least 20 systems have three clear days on each side of it. The rain that ended the long 2024 drought, in late January 2025, had too few clear days after it to measure. | |||
https://cali-energy.com/research/dirty-solar-panels-output-loss-los-angeles. Aggregated data are free to use under CC BY 4.0 with a link to this page.The pattern is what dust would produce: the longer the dry spell, the bigger the jump. But the numbers are small. Fitted across all 41 rains, clear-day output recovered about 2% for every 100 dry days, and a rain after a month or two of dry weather is hard to tell apart from ordinary day-to-day variation.
Five rains up close
San Fernando Valley weather point.
| Rain | Date | Rain, mm | Dry days before | Systems | Change, corrected |
|---|---|---|---|---|---|
| End of the 2022 dry season | September 9, 2022 | 20 | 164 | 31 | +3.2% |
| Tropical Storm Hilary | August 20, 2023 | 67 | 107 | 57 | +2.2% |
| First rain of November 2023 | November 15, 2023 | 14 | 45 | 61 | +2.8% |
| After the January 2025 fires | February 4, 2025 | 33 | 8 | 81 | +9.6% |
| End of the 2026 dry season | September 6, 2026 | 6 | 146 | 82 | +5.8% |
The storm of early February 2025 stands apart. It came only 8 days after a lighter late-January rain, yet lifted clear-day output by 9.6% — about three times a normal first rain. The Eaton and Palisades fires of January 2025 had just covered the region in ash and soot, and a light rain does not rinse that off; a heavy one does. If your panels are under ash after a fire, a rinse is worth more than after an ordinary summer.
Tropical Storm Hilary, after 107 dry days, gave back 2.2%. The end of the 2026 dry season, after 146 days, gave back 5.8% — the clearest dust signal in the record.
What dust costs over a year
| Year | Longest stretch without a 5 mm rain | Output lost to dust | For a 7 kW system (about 9,700 kWh a year) |
|---|---|---|---|
| 2023 | 107 days | 0.6% | about 50 kWh |
| 2024 | 261 days | 1.6% | about 160 kWh |
| 2025 | 144 days | 0.8% | about 80 kWh |
| Model: loss grows by 0.019% a day since the last rain of 5 mm or more, weighted by clear-sky sunshine through the year, using the actual rains of each year. | |||
Built from the measured recovery rate and the actual rain calendar, dust cost a San Fernando Valley roof about 0.6% of its output in 2023, a wet year, 1.6% in 2024, when 261 days passed without a washing rain, and 0.8% in 2025. Most of the loss falls in late summer and autumn, the end of the dry season, when the sun is still strong.
That is lower than the figure most often quoted for California. A UC San Diego study of 186 installations (Solar Energy, 2013) found panels lost 7.4% after a 145-day summer drought, a little under 0.05% a day; LA roofs recovered about 0.02% a day. The two figures measure different things: what a rain gives back on clear days, against the loss of a never-cleaned panel; part of a post-rain jump may also come from cleaner air rather than cleaner glass. More on upkeep: solar panel maintenance in California.
Should you wash your panels?
After a normal summer
Usually not worth the effort. Dust costs about 1% of a year, and the first rain takes most of it.
After a fire
Yes. Ash and soot stick, and the jump after the February 2025 storm was about three times a normal first rain.
Bird droppings, leaves
Spots that cover part of a panel can cost that panel more than a summer of dust; check the panel-by-panel view in your app.
Safely
From the ground, with a hose and water at air temperature, early or late in the day. No pressure washer, no climbing on the array.
A panel that stays low after a good rain is probably not just dirty; see why one panel produces less and the daily solar index to rule out the weather.
Methodology
- Systems
- 219 Enphase systems in our monitoring fleet with clean daily data; 114,318 clear system-days.
- Clear-day output
- Each day’s output as a share of the same roof’s clear-day norm (NASA POWER clear-sky sunshine times the roof’s 90th-percentile ratio within ±10 days of the year). Only days with ERA5 sky at 92% of clear or more.
- Rains
- Episodes of consecutive days with at least 1 mm of rain (ERA5 at the system’s nearest of four points) that reach 5 mm in a day. Dry days: days since the last day with 5 mm or more.
- Change
- Median clear-day output in the 14 days after the rain against the 14 days before, per system (at least three clear days each side), minus 0.4% per °C of cooler daily highs after the rain; then the median across systems (at least 20).
- Noise
- The same calculation around 111 dates with no rain of 1 mm or more within 14 days.
- Year model
- Loss growing by 0.019% a day since the last 5 mm rain (the fitted slope across all rains), averaged over each year weighted by clear-sky sunshine.
- Limits
- Cleaner air after rain is not separated from cleaner panels. Panel tilt, location near roads or farms and bird activity are not observed.
Sources: Cali Energy monitoring fleet (Enphase); NASA POWER; ERA5 reanalysis, ECMWF / Copernicus, via Open-Meteo; F. A. Mejia and J. Kleissl, “Soiling losses for solar photovoltaic systems in California,” Solar Energy, 2013.
Frequently asked questions
How much output do dirty solar panels lose in Los Angeles?
Measured on our roofs, about 1% of a year’s output, and a few percent at the end of a long dry summer. The first good rain after three to six dry months lifted clear-day output by a median 3%.
Does rain clean solar panels?
Yes, a good rain of 5 mm or more does most of the job. Light rain after a few weeks changes output by about 1%, which is within normal day-to-day variation.
Should I clean my solar panels in LA?
After an ordinary summer it is rarely worth it. After a fire, when ash and soot coat the panels, or when droppings or leaves cover part of a panel, a gentle rinse from the ground pays off.
Did wildfire ash reduce solar output in 2025?
Our data suggest it did: the February 2025 storm lifted clear-day output by 9.6%, about three times a normal first rain, after the January fires.
How much does dust cost a 7 kW system a year?
About 80 kWh in a typical year in the San Fernando Valley, and about 160 kWh in a year with a very long dry spell.
More research
Next steps
Output still low after a good rain?
Prepared by Cali Energy, October 15, 2026. Figures describe systems in our monitoring fleet and are not a guarantee of what any specific roof will produce; orientation, tilt, shade, equipment and weather change results. Not engineering, financial or utility advice. See our Content Disclaimer.