9.2 Three Views of the Impact of Local and Transported Wildfire Smoke on U.S. Solar Energy Resource Availability

Wednesday, 31 January 2024: 8:45 AM
347/348 (The Baltimore Convention Center)
Kimberley Corwin, Colorado State University, Fort Collins, CO; and C. Corr-Limoges, J. Burkhardt, P. W. Stackhouse Jr., and E. V. Fischer

The U.S. plans to rapidly expand solar energy generation over the next three decades to meet climate-mitigation goals. However, recent growth in solar energy production coincides with increases in smoke emissions as wildfires burn larger areas for longer periods of time. Accurate solar forecasting models that account for wildfire smoke’s impact on radiation are essential to meet energy goals. Previous studies document large losses in photovoltaic (PV) output on smoke-impacted days but do not quantify large-scale smoke-driven changes in baseline solar resource availability. We quantify changes in direct normal (DNI) and global horizontal (GHI) irradiance associated with smoke of varying optical depths across the U.S. at different spatial and temporal scales through daily, state-level case studies; comparisons of monthly averages at the regional and national level for high and low smoke conditions; and regression analyses of regional and national data spanning 16 years (2006-2021). We leverage irradiance data from radiative transfer models and smoke, aerosol optical depth (AOD), and cloud optical depth (COD) data from satellite observations. Locally severe reductions in mean clear-sky DNI (-32-40%) and GHI (-13-14%) on a single day are possible near large wildfires where smoke is optically thick (AOD 0.43-0.56). Optically thinner clouds (COD ~3) result in similar DNI and GHI reductions, but losses due to optically thicker clouds (COD ~11) can far exceed those caused by smoke. Comparing 2019 and 2020 monthly means, we find that increases in smoke frequency are associated with decreases in clear-sky DNI and GHI, particularly for DNI and in the vicinity of large fires. Sizable losses of clear-sky DNI persist downwind of fires (e.g., -10-15% in some North Central states). Yet, changes in GHI are minimal (< 5%) in areas further from fires where AOD increases are smaller. Results of the longitudinal regression analysis support these findings, showing that clear-sky GHI remains relatively consistent regardless of smoke coverage while DNI can decline substantially with more smoke. While smoke drives locally large reductions in irradiance at fine timescales, GHI–the main resource for PV cells–exhibits little change across the U.S. on average, even in extreme fire years with extensive smoke transport.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner