Using Explainable Artificial Intelligence to Quantify “Climate Distinguishability” After Stratospheric Aerosol Injection

Lightning is a major source of wildfire ignition in the western United States (WUS). We build and train convolutional neural networks (CNNs) to predict the occurrence of cloud‐to‐ground (CG) lightning across the WUS during June–September from the spatial patterns of seven large‐scale meteorological variables from reanalysis (1995–2022). Individually trained CNN models at each 1° × 1° grid cell (n = 285 CNNs) show high skill at predicting CG lightning days across the WUS (median AUC = 0.8) and perform best in parts of the interior Southwest where summertime CG lightning is most common. Further, interannual correlation between observed and predicted CG lightning days is high (median r = 0.87), demonstrating that locally trained CNNs realistically capture year‐to‐year variation in CG lightning activity across the WUS. We then use layer‐wise relevance propagation (LRP) to investigate the relevance of predictor variables to successful CG lightning prediction in each grid cell. Using maximum LRP values, our results show 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 LRP variant used. As lightning is not directly simulated by global climate models, these CNNs could be used to parameterize CG 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 WUS could inform fire management, planning, and disaster preparedness.

show abstract

Future seasonal surface temperature predictability with and without ARISE-stratospheric aerosol injection-1.5

Mayer,

Barnes,

Hurrell

2024

Environ. Res.: Climate

View full text Add to dashboard Cite

To help reduce anthropogenic climate change impacts, various forms of solar radiation modification have been proposed to reduce the rate of warming. One method to intentionally reflect sunlight into space is through the introduction of reflective particles into the stratosphere, known as stratospheric aerosol injection (SAI). Previous research has shown that SAI implementation could lead to future climate impacts beyond surface temperature, including changes in ENSO variability. This response has the potential to modulate midlatitude variability and predictability through atmospheric teleconnections. Here, we explore possible differences in seasonal surface temperature predictability under a future with and without SAI implementation, using neural networks and the ARISE-SAI-1.5 simulations. We find significant future predictability changes in both boreal summer and winter under SSP2-4.5 with and without SAI. However, during boreal winter when SAI is implemented, seasonal predictability is more similar to the base climate than when SAI is not implemented, particularly in regions impacted by ENSO teleconnections.

show abstract

Using Explainable Artificial Intelligence to Quantify “Climate Distinguishability” After Stratospheric Aerosol Injection

Cited by 3 publications

References 53 publications

Stratospheric aerosol injection for controlling greenhouse gas emission

Stratospheric aerosol injection for controlling greenhouse gas emission

Predicting Cloud‐To‐Ground Lightning in the Western United States From the Large‐Scale Environment Using Explainable Neural Networks

Future seasonal surface temperature predictability with and without ARISE-stratospheric aerosol injection-1.5

Contact Info

Product

Resources

About