Statistical Analysis of Wildfires and Future Probabilities in Washington State
Faculty Mentor
Dr. Richard Orndorff
Presentation Type
Poster
Start Date
May 2025
End Date
May 2025
Location
PUB NCR
Primary Discipline of Presentation
Geosciences
Abstract
Using data from the National Interagency Coordination Center, we determined the summary statistic burned acreage and suppression costs for Washington wildland fires between the years of 2002-2023. From this information, we then conducted a binomial analysis in the R programming language to determine the probability of numbers of years with burned acreage exceeding the historic mean out of the next 10 years. We found that the most probable number of years with burned acreage above the mean is 2 years within the next 10 years (probability of 27.9%).
We acquired air temperature data for Washington from NOAA for the same period, and we used linear regression of annual acres burned onto annual mean temperature and found that there is a positive association between increased annual air temperature and area of land burned. We concluded that 37.5% of the variability in annual acreage burned could be explained by variability in annual mean air temperature.
Recommended Citation
Colvin, Stephen; Smith, Marilyn; and Orndorff, Richard, "Statistical Analysis of Wildfires and Future Probabilities in Washington State" (2025). 2025 Symposium. 13.
https://dc.ewu.edu/srcw_2025/ps_2025/p2_2025/13
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Statistical Analysis of Wildfires and Future Probabilities in Washington State
PUB NCR
Using data from the National Interagency Coordination Center, we determined the summary statistic burned acreage and suppression costs for Washington wildland fires between the years of 2002-2023. From this information, we then conducted a binomial analysis in the R programming language to determine the probability of numbers of years with burned acreage exceeding the historic mean out of the next 10 years. We found that the most probable number of years with burned acreage above the mean is 2 years within the next 10 years (probability of 27.9%).
We acquired air temperature data for Washington from NOAA for the same period, and we used linear regression of annual acres burned onto annual mean temperature and found that there is a positive association between increased annual air temperature and area of land burned. We concluded that 37.5% of the variability in annual acreage burned could be explained by variability in annual mean air temperature.
