Timezone: »

Identifying causes of Pyrocumulonimbus (PyroCb)
Emiliano Diaz · Kenza Tazi · Ashwin Braude · Daniel Okoh · Kara Lamb · Duncan Watson-Parris · Paula Harder · Nis Meinert

A first causal discovery analysis from observational data of pyroCb (storm clouds generated from extreme wildfires) is presented. Invariant Causal Prediction was used to develop tools to understand the causal drivers of pyroCb formation. This includes a conditional independence test for testing $Y \indep E|X$ for binary variable $Y$ and multivariate, continuous variables $X$ and $E$ and a greedy-ICP search algorithm that relies on fewer conditional independence tests to obtain a smaller more manageable set of causal predictors. With these tools we identified a subset of seven causal predictors which are plausible when contrasted with domain knowledge: surface sensible heat flux, relative humidity at 850hPa, a component of wind at 250 hPa, 13.3 \textmu m thermal emissions, convective available potential energy and altitude.

#### Author Information

##### Kenza Tazi (University of Cambridge)

I'm a PhD candidate at the University of Cambridge on the 'AI for Environmental Risk' doctoral program. I'm jointly based at the Department of Engineering and the British Antarctic Survey. My research focuses on improving predictions of precipitation over mountainous areas using probabilistic machine learning methods, in particular Gaussian Processes. Through this project and others, I'm interested in communicating scientific results to policy makers in order to make evidence-based decisions in the face of climate change.

##### Kara Lamb (Columbia University)

I received my Ph.D. in physics from the University of Chicago with a focus on cirrus cloud microphysics. Following my Ph.D. I was a research scientist at University of Colorado's Cooperative Institute for Research in the Environmental Sciences in NOAA's Earth Systems Research Lab, where I worked on several NASA/NOAA aircraft campaigns to study atmospheric aerosols. Last summer I participated in NASA's Frontier Development Lab on the GNSS prediction team. I am currently working at Columbia University studying how hybrid-physics ML approaches can be used to improve our understanding of aerosol and cloud microphysical processes for predictions of future climate.