A new statistical model may help researchers identify fire-prone, which includes parts of California, where the Dixie Fire, pictured, this year became the second-largest wildfire in the state's history. File Photo by Peter DaSilva/UPI | License Photo
Nov. 3 (UPI) -- Researchers at the University of California, Santa Barbara said Wednesday that they have developed model designed to help identify wildfire-prone areas across the Golden State.
The new modeling approach outperformed currently used statistical models developed for certain regions of the state, accurately predicting new locations for wildfires, and the years in which they will occur, with more than 75% accuracy in some cases, the researchers said.
Based on a statistical approach known as generalized additive modeling, they were able to map annual wildfire probabilities throughout California from 1970 to 2016 by incorporating data on local climate variation, human activity and the time since the last fire event, they said.
Both local climate and human activity, such as the dryness of fuel available to burn and housing density, play key roles in determining wildfire probabilities throughout the state, the data showed.
For example, portions of the Southern California mountains such as the Angeles and Los Padres National Forests were at high risk for wildfires.
Both regions have sufficient vegetation and therefore fuel availability as well as being close to and at risk from ignitions starting in high-density housing in the Los Angeles metropolitan area, according to the researchers.
In addition, in certain environments, the amount of time since the last fire has an important influence on the potential wildfires, as do short-term climate variations involving extreme conditions, particularly in fire-prone shrublands and forests in southern California.
"This study presents a powerful tool for mapping the probability of wildfire across the state of California under a variety of historical climate regimes," the researchers wrote in an article published Wednesday by the journal PLOS ONE.
"By leveraging machine learning methods, it demonstrates the distinct ways in which local climate, human development and prior fire history each contribute to the yearly risk of wildfire over space and time," they said.
Like much of the western United States, California has been plagued by wildfires in recent years, with many communities suffering significant damage.
Many residents have lost homes, and smoke from wildfires has been linked with serious health complications, such as lung disease.
However, the factors and conditions that interact to contribute to the probability of wildfire, such as the interplay between local vegetation, precipitation and land use, are complex and vary by location and over time, according to the researchers.
With further refinements, the new modeling method could prove valuable for a variety of research and practical applications in such areas as wildfire emissions and hazard mapping for implementation of fire-resistant building codes, the researchers said.
"This study demonstrates that local climate -- through limitations posed by fuel dryness and fuel availability -- plays an important and predictable role in determining the annual probabilities of fire throughout California," the researchers wrote.
"Further, our findings emphasize the importance of incorporating human activity -- through influences on ignitions and suppression of fires -- into predictions of fire probability over space and time," they said.