Skip to main content

California Freight Cleanup → Element 7

How much California solar does wildfire smoke actually erase?

About 980k MWh, or $49 M a year, in 2019. Most published studies blame irradiance reduction; we found the bigger driver is panel soiling, by 73% to 27%. Project that into the SB100 fleet and a future of climate-amplified fire seasons, and California’s solar industry is looking at roughly $340 M/yr in lost revenue by 2050.

Decision Dashboard — compare portfolios across CRF anchors and budget scales.

We needed a quantified connection between wildfire smoke and California’s solar generation base, with explicit uncertainty bounds and a link to ratepayer-benefit framing. A single-channel treatment — smoke dims sunlight, generation drops — understates the loss by ignoring ash deposition on panels, which persists until rain or a cleaning cycle resets it. We built both channels explicitly, calibrated against peer-reviewed literature. The wildfire-PV investigation closes this element.

We ran a two-channel physical model over 5,000 Monte Carlo draws against the 2019 California fleet baseline (27 GW installed; 48 TWh annual generation). Channel A encodes smoke aerosol reducing solar irradiance at the panel plane — a linear PM2.5-to-generation-loss coefficient anchored to Juliano et al. 2022 (Environ Res Lett 17:034010) and Conceicao 2018. Channel B encodes smoke-ash deposition lowering panel transmittance until rain or cleaning resets it (Conceicao 2018: 2.5% per smoke event; 21-day cleaning cycle). Both channels draw smoke frequency and PM2.5 levels from a regional smoke climatology covering five California regions.

Climate amplification connects the 2019 baseline to 2050 projections from six global climate models, producing a 1.86× mean PM2.5 multiplier by 2050. The wildfire-vs-electrification cost comparison finds that wildfire fuel management runs $143–$430M per death avoided (Di CRF) vs. $15–$23M for transport electrification. Adding the PV-preservation benefit to wildfire portfolios improves their return by less than 0.02% of portfolio cost — nothing flips. Portfolio rankings are stable across four wildfire-load scenarios to within less than 2%. A cross-validation across structurally distinct fire years documents where the linear-scaling assumption breaks down rather than suppressing those failures.

Bar chart comparing Channel A (irradiance attenuation, 254k MWh) and Channel B (panel soiling, 693k MWh) contributions to total PV loss. Channel B bar is approximately 2.7 times taller.
Figure 1. PV loss decomposed by channel, 2019 California baseline. Channel B (panel soiling, 693k MWh) is 2.7× larger than Channel A (irradiance attenuation, 254k MWh). Most published studies estimate only Channel A, understating total losses by approximately 73%.
Side-by-side bars comparing 2019 PV loss (27 GW fleet) to 2050 projected loss at constant fleet and at SB100 100 GW fleet, showing 1.86x and 6.9x increase respectively.
Figure 2. 2019 baseline vs. 2050 climate amplification. The Investigation 7-2 CMIP6 1.86× PM2.5 multiplier raises annual PV loss from ~$49M to ~$91M at constant fleet. Scaling to the SB100 100 GW target lifts total exposure to ~6.76M MWh / ~$340M per year.

~980k MWh / $49M annual PV loss, 2019 baseline

California’s 27 GW fleet lost an estimated 980,426 MWh of generation to wildfire smoke in 2019 — 2.0% of annual PV output — with a 5,000-draw Monte Carlo P5–P95 envelope of 689k1,328k MWh ($34.5M–$66.4M). The 2019 baseline represents a moderate fire year (332,722 acres burned); the loss scales with fire intensity.

Channel B (soiling) dominates 73% / 27% — most studies miss the top channel

Of the 980k MWh total, Channel A (irradiance attenuation) accounts for 254k MWh (27%) and Channel B (panel soiling from smoke-ash deposition) for 693k MWh (73%). The published literature, including the the CEC freight solicitation background document, predominantly cites irradiance-only studies. The soiling channel is larger in moderate fire years because ash deposition accumulates across the entire inter-cleaning period; it is not limited to direct-smoke-plume days.

2050 amplification: 1.86× PM2.5 × 3.7× fleet → ~$340M/yr

Investigation 7-2’s CMIP6 L3 ensemble projects a 1.86× mean PM2.5 multiplier by 2050, driven by vapor-pressure-deficit increases across six GCMs (SSP2-4.5 and SSP5-8.5). At the 2019 constant fleet that lifts annual PV loss to ~$91M. At the SB100 100 GW capacity (3.7× current), the loss reaches ~$340M/yr. This is a reliability exposure, not a revenue signal: the backup capacity needed to cover peak smoke-day PV deficits is 2,994 MW central (2,0964,042 MW envelope), at a CAISO annualized capacity cost of ~$180M/yr.

Portfolio rankings unchanged by PV co-benefit

Adding the PV-preservation co-benefit to wildfire-intervention portfolios (C_wildfire_instead: 5% fuel reduction, $1.65B; F_maximum_impact: 30%, $13.9B) improves their 10-yr NPV by $0.24M and $1.47M respectively — 0.01% of cost in both cases. The finding is unambiguous: PV preservation is a legitimate co-benefit of wildfire fuel management, but it is a secondary monetary offset, not a ranking-flip force. The Investigation 4-3 cost-per-death-avoided gap between wildfire prevention ($143–$430M) and transport electrification ($15–$23M) is not closed by adding PV.

Wildfire excess deaths dominate electrification benefits in catastrophic years

In the 2020 fire season, wildfire smoke contributed 600 excess deaths against the best-case electrification portfolio’s 164 deaths avoided. In a 2023-type year the wildfire excess drops to ~27 deaths. The decision-relevant framing is not “which intervention is cheaper?” but “which fire-year scenario is the program implicitly budgeting for?”

Validation limits documented honestly: 5 of 13 gates pass

Investigation 7-4 cross-validates four wildfire investigations across four structurally distinct fire years. 5 of 13 coherence gates pass; 8 fail. The failures are documented scope mismatches — Investigation 7-2 operates at a 2050 climate horizon that is not commensurable with within-year current-period death counts, and Investigation 7-3 uses scenario-mapped rather than linearly-scaled per-year values. None of the failures indicate cascade errors. The 5/13 rate is reported as-found rather than gamed by post-hoc criterion loosening.

Element 4 (co-benefits / disbenefits). Investigation 4-3 anchors the wildfire-vs-electrification cost-per-death comparison. Wildfire accounts for 7% of California annual-mean PM2.5 under a 2023 baseline — rising to 15–25% in catastrophic years. Whether wildfire prevention is a co-benefit or a competing program depends entirely on which fire-year scenario the portfolio evaluation adopts.

Element 6 (ratepayer burden / portfolio NPV). Investigation 7-1’s per-portfolio PV co-benefit table feeds Element 6 directly. The PV revenue preserved by wildfire interventions is a positive NPV addend — small (<$1.5M over 10 years for the maximum-impact portfolio), but the right accounting entry for a full-stack comparison.