Publication Date

Fall 2022

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Meteorology and Climate Science

Advisor

Craig Clements

Subject Areas

Meteorology

Abstract

Predicting high fire danger conditions is paramount to mitigating the impacts caused by wildfires. Such warning systems as red flag warnings (RFWs) and the National Fire Danger Rating System (NFDRS) utilize atmospheric and fuel moisture properties to warn public and government entities about conditions that may lead to the ignition or rapid growth of wildfires. In this study, we use high-resolution reanalysis and wildfire growth data from 2003-2020 in California to test a variety of different variables to determine if a more viable variable combination exists that could be used to create a better warning index which would allow for a better estimate of high fire danger conditions. This is assessed by ranking combinations based on how well they organize daily fire acreage growth values within the heatmaps. It is found that turbulent kinetic energy (TKE) 50 meters above ground level presents a strong case to be used as a fire danger predictor. Sounding profiles are also created to ascertain a clearer picture of the vertical profiles typically seen on dangerous fire weather days, with the highest average daily acreage growth values seen with each combination exhibiting a dry and warm-to-hot environment. Line plots detailing daily average acreage growth rate changes indicate compounding effects of multiple variables being in extreme states at the same time as well as limiting behavior where a single variable being in a non-fire conducive state can shut-off the influence of another variable. We also find that relative humidity and sustained wind speed is the third-ranked variable combination, empirically confirming a previous mostly heuristic result.

Download

Share

COinS