USING ARTIFICIAL NEURAL NETWORKS TO PREDICT LIGHTNING

In this study, I built and trained Convolutional Neural Networks (CNNs) to predict cloud-to-ground (CG) lightning across the western U.S. CNN models were individually trained at each 1° x 1° grid cell and were skillful at predicting the daily occurrence of CG lightning. Interannual correlation between observed and predicted CG lightning days was also high (median r = 0.87), demonstrating that locally-trained CNNs realistically capture year-to-year variation in CG lightning activity. I then used a machine-learning visualization technique to investigate the relevance of predictor variables in each grid cell. My results showed that two thermodynamic variables - ratio of surface moist static energy to free-tropospheric saturation moist static energy, and the 700-500 hPa lapse rate - are the most relevant CG lightning predictors for 93-96% of CNNs depending on the exact method used. As lightning is not directly simulated by global climate models, these CNNs could be used to predict lightning in climate models to assess changes in future CG lightning occurrence with projected climate change. Understanding changes in CG lightning risk and consequently lightning-caused wildfire risk across the West could inform fire management, planning, and disaster preparedness.

​

Kalashnikov, D. A., Davenport, F. V., Labe, Z. M., Loikith, P. C., Abatzoglou, J. T., & Singh, D. (2023). Predicting Cloud-to-Ground Lightning in the Western United States from the Large-Scale Environment using Explainable Neural Networks. Journal of Geophysical Research: Atmospheres, 129(22), e2024JD042147.      https://doi.org/10.1029/2024JD042147​

​

​

HOW MUCH RAIN FALLS WHEN LIGHTNING STARTS WILDFIRES?

Cloud-to-ground lightning with minimal rainfall, also known as “dry lightning,” is a major wildfire ignition source in the western United States. Typically, daily-accumulated precipitation of less than 2.5 mm is used to identify dry lightning occurrence. However, there was limited knowledge of (a) the true precipitation amounts that occur with lightning-ignited wildfires, and (b) how these amounts vary across different landscapes and vegetation types. In this study, I combined wildfire, lightning, and radar precipitation data to take a deep dive into exactly how much rain falls when lightning starts wildfires across the West. My results showed that ignition precipitation amounts vary substantially across ecoprovinces and across fire types, ranging between 1.7 and 7.7 mm. This new knowledge can improve wildfire prediction and management across the different landscapes of the western US.

​

Kalashnikov, D. A., Abatzoglou, J. T., Loikith, P. C., Nauslar, N. J., Bekris, Y., & Singh, D. (2023). Lightning-Ignited Wildfires in the Western United States: Ignition Precipitation and Associated Environmental Conditions. Geophysical Research Letters, 50(16), e2023GL103785. https://doi.org/10.1029/2023GL103785

​

Interviews and news coverage:

San Francisco Chronicle, Eos.org - Editor’s Highlight, NSF Research Highlight

​

 

WHAT DRIVES CALIFORNIA DRY LIGHTNING OUTBREAKS?

Lightning occurring with less than 2.5 mm of rainfall—typically referred to as 'dry lightning'—is a major source of wildfire ignition in central and northern California. Despite being rare, dry lightning outbreaks have resulted in destructive fires in this region due to the intersection of dense, dry vegetation and a large population living adjacent to fire-prone lands. By combining lightning, precipitation, and atmospheric reanalysis data, I characterized the climatology of dry lightning and the associated meteorological conditions during the warm season (May–October) when wildfire risk is highest. My results showed that mid-tropospheric moisture and instability were consistently elevated on dry lightning days compared to background climatology. Additionally, I found that surface temperatures, lower-tropospheric dryness, and mid-tropospheric instability were all increased across the region on dry versus wet lightning days. Understanding the meteorology of dry lightning across this region can inform forecasting of possible wildfire ignitions and is relevant for assessing changes in dry lightning and wildfire risk in climate projections.

​

Kalashnikov, D. A., Abatzoglou, J. T., Nauslar, N. J., Swain, D. L., Touma, D., & Singh, D. (2022). Meteorological and geographical factors associated with dry lightning in central and northern California. Environmental Research: Climate, 1(2), 025001. https://doi.org/10.1088/2752-5295/ac84a0

​

Interviews and news coverage:

Washington Post, San Francisco Chronicle, CNN, BBC, The Weather Channel, New Scientist, Discover Magazine, The Hill, Weather Underground, Gizmodo, FOX Weather, Earth.com 

​

​

CO-OCCURRING PM2.5/OZONE AIR POLLUTION EPISODES

Wildfires and meteorological conditions influence the co-occurrence of multiple harmful air pollutants including fine particulate matter (PM2.5) and ground-level ozone. In this study, I explored the spatiotemporal characteristics of PM2.5/ozone co-occurrences and associated population exposure in the western United States (US). I found that the frequency, spatial extent, and temporal persistence of extreme PM2.5/ozone co-occurrences have increased significantly between 2001 and 2020, increasing annual population exposure to multiple harmful air pollutants by ~25 million person-days/year. Using a clustering methodology to characterize daily weather patterns, I identified significant increases in atmospheric ridging patterns conducive to widespread PM2.5/ozone co-occurrences and population exposure. I further linked the spatial extent of pollutant co-occurrences to the extent of extreme heat and wildfires. These findings suggest an increasing potential for co-occurring air pollution episodes in the western US with continued climate change, and call for urgent efforts to reduce public harm. 

​

Kalashnikov, D. A., Schnell, J. L., Abatzoglou, J. T., Swain, D. L., & Singh D. (2022). Increasing co-occurrence of fine particulate matter and ground-level ozone extremes in the western United States. Science Advances, 8(1), eabi9386. https://doi.org/10.1126/sciadv.abi9386

​

Interviews and news coverage:

NY Times, Forbes, Smithsonian Magazine, New Scientist, Popular Science, Science Daily, Climate.gov, Atmos Magazine 

​

LIGHTNING-PRODUCING METEOROLOGICAL PATTERNS IN THE WEST

Thunderstorms during warm-season months across interior portions of the western United States (WUS) pose potential hazards ranging from wildfire ignition to flash flooding. In this study, I performed a comprehensive and spatially contiguous analysis to describe the synoptic-scale meteorological patterns and a subset of thermodynamic variables associated with summertime lightning activity in the interior WUS. When analyzed across all 24,000+ grid cells across the region, composites reveal that lightning is generally associated with positive 500-hPa geopotential height anomalies located to the northeast of the location experiencing lightning. Meanwhile, negative sea level pressure anomalies are found to the northwest and collocated with local lightning days. Areas not commonly affected by the North American monsoon system including the western Great Basin and northern Rocky Mountains show higher-amplitude anomalies of geopotential height, moisture, and midtropospheric instability patterns suggesting the importance of episodic midlatitude dynamics to lightning days in such locations. Results from this observational analysis provide a foundation for projecting future lightning occurrence using climate models.

​

Kalashnikov, D. A., Loikith, P. C., Catalano, A. J., Waliser, D. E., Lee, H., & Abatzoglou, J. T. (2020). A 30-Yr Climatology of Meteorological Conditions Associated with Lightning Days in the Interior Western United States. Journal of Climate, 33(9), 3771-3785. https://doi.org/10.1175/JCLI-D-19-0564.1

​

​

​

​

* Contributed analysis as co-author

​

Singh, D., Bekris, Y., Rogers, C., Doss-Gollin, J., Coffel, E. D., & Kalashnikov, D. A. (2024). Enhanced solar and wind potential during widespread temperature extremes across the U.S. interconnected energy grids. Environmental Research Letters, 19(4), 044018. https://doi.org/10.1088/1748-9326/ad2e72

​

Swain, D. L., Abatzoglou, J. T., Kolden, C., Shive, K., *Kalashnikov, D. A., Singh, D., & Smith, E. (2023). Climate change is narrowing and shifting prescribed fire windows in western United States. Communications Earth & Environment, 4(340). https://doi.org/10.1038/s43247-023-00993-1

​

Loikith, P. C. & *Kalashnikov, D. A. (2023). Meteorological Analysis of the Pacific Northwest June 2021 Heat Wave. Monthly Weather Review, 151(5), 1303-1319. https://doi.org/10.1175/MWR-D-22-0284.1

​

Kalashnikov, D. A. “Lightning-Caused Wildfires” and “Compounded Health Effects of Fine Particulate Matter and Surface Ozone”. In: Fleishman, E., editor. (2023). Sixth Oregon Climate Assessment. Oregon Climate Change Research Institute, Oregon State University, Corvallis, Oregon. https://doi.org/10.5399/osu/1161

​

Touma, D., Stevenson, S., Swain, D. L., Singh, D., Kalashnikov, D. A., & Huang, X. (2022). Climate change increases risk of extreme rainfall following wildfire in the western United States. Science Advances, 8(13), abm0320. https://doi.org/10.1126/sciadv.abm0320

​

​

​

​

​

​

​