Heat derate in the Inland Empire and Valley by month
Hot panels make less power, and in Southern California the loss peaks in late summer: in August at six of the seven sites we modeled, in July in Lancaster. In our model, a common panel with a -0.36%/°C coefficient lost 6.0% of its yearly output to heat in Santa Monica and 7.4% in Fontana, against what it would make with its cells at 25 °C. In August the loss reached 11.1% in Valencia and 10.9% in Lancaster; in a Lancaster December it was only 0.9%. A panel with a -0.25%/°C coefficient cut the Lancaster loss from 7.1% to 4.8% over the year and from 10.9% to 7.5% in August.
In this guide
Heat loss by month#
Swipe to see all columns →
| Month | Santa Monica | West Covina | Van Nuys | Valencia | Lancaster | Riverside | Fontana |
|---|---|---|---|---|---|---|---|
| January | 3.4% | 4.0% | 2.8% | 2.6% | 1.7% | 4.1% | 4.0% |
| February | 4.0% | 4.7% | 3.9% | 3.4% | 3.1% | 4.8% | 4.7% |
| March | 5.1% | 5.8% | 4.0% | 5.4% | 5.1% | 4.9% | 5.4% |
| April | 5.9% | 6.5% | 5.0% | 5.2% | 6.2% | 6.6% | 6.6% |
| May | 5.5% | 7.0% | 6.0% | 8.2% | 7.9% | 7.5% | 7.7% |
| June | 6.6% | 7.7% | 8.0% | 9.2% | 9.7% | 8.0% | 8.8% |
| July | 7.8% | 9.2% | 8.6% | 10.0% | 10.9% | 9.8% | 9.6% |
| August | 8.3% | 9.9% | 9.3% | 11.1% | 10.9% | 10.2% | 10.1% |
| September | 7.2% | 9.4% | 8.8% | 9.4% | 9.7% | 9.2% | 9.3% |
| October | 6.7% | 8.0% | 7.9% | 7.9% | 7.2% | 9.1% | 9.2% |
| November | 5.2% | 5.9% | 4.9% | 5.2% | 4.0% | 5.7% | 5.7% |
| December | 2.9% | 2.5% | 1.8% | 2.0% | 0.9% | 2.6% | 2.4% |
| Year | 6.0% | 7.1% | 6.4% | 7.2% | 7.1% | 7.3% | 7.4% |
Share of the panels' DC output lost to heat, net of the cool hours when cells run below 25 °C, by month, for Qcells Q.PEAK DUO BLK ML-G10 410 (-0.36%/°C) facing south at 20° on a roof: the CEC panel model at the modeled cell temperature against the same panel with its cells held at 25 °C, under the same sunlight. Typical-year weather (NSRDB) at Santa Monica 90401, West Covina 91790, Van Nuys 91406, Valencia 91355, Lancaster 93535, Riverside 92503 and Fontana 92335. Source: Cali-Graf engine.
What the months say#
- Late summer is the worst. August had the largest loss at every site but Lancaster, where July was slightly worse. - The coast loses least. Santa Monica's cool air kept its yearly loss to 6.0% and its August loss to 8.3%. - Valleys and the Inland Empire are alike. West Covina, Valencia, Riverside and Fontana all lost 7.1% to 7.4% a year; Van Nuys, at 6.4%, was closer to the coast in this weather year. - The desert swings the most. Lancaster's cells reached 77 °C in the model, but its cold winters cut the loss to 0.9% in December.
What it means for the design#
- A better coefficient cuts the heat loss by about a third everywhere. The better panel cut the yearly heat loss in Lancaster by 31%; it lost 1.9% less of its yearly output in Santa Monica and 2.3% less in Fontana, so the gain is a little larger where it is hot. - Model the heat by month. A yearly derate hides the summer: Lancaster's August loss was 1.5 times its yearly figure.
Data limits#
The loss is computed on the panels' DC output against a 25 °C reference under the same sunlight; it does not include the inverter, wiring or soiling. One typical weather year per site; heat waves make single months worse. All figures use the roof-mount cell temperature model. Two panels only; other panels fall between and around them according to their coefficients. The newer Q.PEAK DUO BLK ML-G10+ 410 is listed by the CEC with a better coefficient than the ML-G10 410 modeled here.
How much do solar panels lose in the heat in Southern California?
In our model, a common panel lost 6.0% of its yearly output to heat on the coast and up to 7.4% inland, and up to 11.1% in August.
Which month has the biggest heat loss?
August at six of the seven sites we modeled; in Lancaster, July by a small margin.
Does a better temperature coefficient help in the Inland Empire?
Yes: in Fontana a -0.25%/°C panel lost 5.1% a year against 7.4% for a -0.36%/°C panel.
Methodology#
The Cali-Graf engine takes the sunlight reaching each panel and its cell temperature hour by hour from PySAM's PVWatts model (roof mount) on NSRDB typical-year weather, then computes the panel's DC power with the CEC single-diode model from the panel's CEC parameters, once at the modeled cell temperature and once at 25 °C. The monthly loss is one minus the ratio of the two monthly sums. Rounded half up.
Sources#
- California Energy Commission — Solar Equipment Lists, PV modules. https://solarequipment.energy.ca.gov/
- NLR — PySAM, the Python package for the System Advisor Model. https://nrel-pysam.readthedocs.io/
- NLR — National Solar Radiation Database (NSRDB), typical meteorological year. https://nsrdb.nlr.gov/
How to cite this page#
Cali-Graf. “Heat derate in the Inland Empire and Valley by month.” cali-graf.com, updated Nov 19, 2026. https://cali-graf.com/guides/heat-derate-in-the-inland-empire-and-valley-by/