Portal Home || Research || Ecology Group || Social Science Group || Remote Sensing Group || Integrative Activities || Data || Maps || IFIRE Publications

Extreme Fires Portal: Social Science Working Group

Defining Extreme Wildfires from Social Science Research

The social science team completed a case study in year 3 consisting of in-depth interviews to validate and improve our understanding of the factors that affect people’s perceptions of what makes a wildfire ‘extreme.’ The 2012 Dahl fire near Roundup, MT was ignited by lightning and burned 22,045 acres from June 26 to July 3 in the mixed ponderosa pine sage grass prairie terrain. Initial estimates from the National Interagency Fire Center were that there were over $1.6 million in damages, including 73 homes and 150 other structures. The social science team conducted 51 in-person interviews in the summer of 2013 with residents, land and fire managers, emergency personnel, community leaders, real estate agents, insurance agents, and other stakeholders affected by the fire.


  • Interviewees consistently noted that climatic conditions (very dry, hot, and windy) led to a very fast moving and intense fire and steep, rugged terrain made it even more difficult for residents and fire fighters to protect their property and control the fire.
  • Although the duration of the fire was not substantial relative to other fires investigated in this research, the erratic fire behavior and perception of the lack of ability to control it led to the fire being perceived as an atypical event. This confirms our earlier findings that biophysical and fire characteristics contribute to the perceptions of fire extremity from the social perspective.
  • Access to personal property was a key source of conflict as fire managers prohibited people from re-entering their property during the fire. This conflict exacerbated the social impacts of the fire.
  • As other case studies have highlighted, efforts immediately following the fire to organize future mitigation efforts on private property (e.g., defensible space measures) appear to have been relatively short-term as there does not seem to be sustained efforts to reduce homeowner risk to future wildfires on a community-wide scale.

Evaluation of the characteristics and trajectories for extreme wildland fire events in changing physical and political climates – The ‘25 Fires’ Project

A multilevel modeling approach was implemented to survey people across 25 fires (representing a range of duration, acres burned, and fire severity measures) in the northwest U.S. that occurred in 2011 or 2012. This approach revealed which themes and factors relate to how people experience a wildfire (e.g., people’s perceptions of impacts and attitudes towards fire management), revealing strong consistency across various fires and contexts.

Initial Results:

  • Analysis of the questionnaire data demonstrate that while fires and the populations impacted by them may differ, there are many consistent relationships between wildfire impacts and resident well-being. This indicates that there is the potential for more uniform indicators for impact to people and broader social systems. The table below shows results from linear regression analysis highlighting statistically significant predictors that explained 48% of the variance in negative impacts to people’s well-being from the wildfire. 
  • Multi-level modelling suggested that many of the significant predictors had varying intercepts but that the slopes were not significantly different across the fires between the predictor variables and the dependent variable.  For the most part, the relationship between the variables was relatively stable across fires, however, we must also recognize and account for the difference in social and ecological contexts that influence fires and their impacts. This means understanding how local histories, development patterns, and social relationships surrounding fire in a given place, might affect the vulnerabilities that people have to fire.

Ongoing: Throughout the summer and fall of 2014, we will continue conducting statistical analysis by testing multilevel and linear regression models to predict perceptions of the extremity of a fire, as well as perceptions of landscape recovery, using the predictor variables mentioned above. We will also explore psychometric properties of some of our measures in more detail, which will contribute to future wildfire and other natural disaster research.

Assessing social perceptions of landscape recovery
The interviews in 2013 revealed that an individual’s relationship with the landscape and their understanding of fire as a natural part of the ecology of the area was important to how people reacted to the landscape impacts from the Dahl fire. These initial findings encouraged us to conduct follow-up interviews in the summer of 2014 to further investigate people’s perceptions of post-wildfire landscape recovery, a combination of Landsat satellite imagery, vegetation indices (NBR and NDVI), and high resolution imagery will be used to identify areas demonstrating different trends in post-fire vegetation recovery. 

Results: Aesthetic impacts to the landscape are important to perceptions of the severity of wildfire impacts. Interviews about the Dahl fire confirmed this finding and led to a more in-depth and on-going research project that integrates biophysical and social sciences to better understand how people perceive landscape recovery after wildfire events. The goal is to improve management and communication responses before, during, and after a wildfire in efforts to reduce the negative impacts from an often drastically changed landscape.

Ongoing: These datasets will be presented to interviewees in 2014 to elicit greater understandings of the spatial and temporal aspects of how people perceive landscape recovery after ‘extreme’ fire events, how their connections to the landscape and their mental models of fire as part of the ecosystem affect their perceptions, and how their perceptions influence their attitudes towards wildfire management.