PEARC23 Conference Award for Best Overall Visualization

VAPOR visualization of East Troublesome Fire in 2020

Visualizing Megafires: How AI can be used to drive wildfire simulations with better predictive skill

Recent visualization work in NCAR's Computational & Information Systems Laboratory (CISL) and Research Applications Laboratory (RAL) was awarded the Best Overall Visualization award at this year's PEARC23 conference, held July 24–27 in Portland, Ore.

The visualization effort was led by Scott Pearse from CISL using the VAPOR software developed in CISL. The wildfire modeling work was conducted by Amy DeCastro and Timothy Juliano in RAL using the coupled WRF-Fire and WRF-MET models. The visualization highlights recent advances in how artificial intelligence can be used to drive wildfire simulations with better predictive skill, as described in the publication's abstract:

"The East Troublesome Wildfire was the fourth largest wildfire to date in Colorado history, igniting on October 14, 2020. Driven by low humidity and high winds, the wildfire spread to over 200,000 acres in nine days, with 87,000 of those acres being burnt in a single 24-hour period. Wildfire simulations and forecasts help decision-makers issue evacuation orders and inform response teams, but these simulations depend on accurate variable inputs to produce trustworthy results. These wildfire visualizations demonstrate new AI tools developed at the National Center for Atmospheric Research (NCAR), which are producing superior wildfire simulation outputs than have been available in the past."

You can watch the visualization here and learn more about the work in this narrated explainer video.