Wildfire hazard is not constant across the landscape. Wildfires are driven by weather, topography and fuels. Where these elements come into alignment, (example: when a dry wind blows up a steep, sunny, brush-covered slope) fire behavior is the most dangerous. Conversely, if lightning starts a fire in a lone tree atop of a barren mountain on a rainy day, it is not likely to spread rapidly downhill. Living safely with wildfire requires that we understand the dynamics of wildfire.

At Deer Creek Resources, we couple our real-world knowledge of wildfire with modern computer modeling and mapping technologies. We use maps, graphics, aerial and ground photography, and interviews with knowledgeable locals to study, learn, and develop stratgies and tactics to reduce the wildfire hazard to people and property. This section describes some of the techniques that we use, and projects that we have done.

Using computers to model wildfire behavior

Our staff  are experts in the development and application of the FARSITE and FLAMMAP fire behavior models. These programs use fuels, weather, and topographic information to create graphical portrayals of potential wildfire spread patterns, rates of spread, and burn intensities. The resulting maps are very useful for communicating big-picture wildfire hazard issues to a lay audience.

Some applications of Fire Behavior Modeling Include:

  • Identifying high wildfire hazard areas
  • Planning tactically useful fuels management projects
  • Prioritizing areas for fuels treatment
  • Facilitating sensible land use planning
  • Designing prescribed burning projects
  • Planning wildfire response

Fire modeling analyzes the factors that contribute to fire spread and intensity, and can be used to highlight high-hazard areas. Perhaps the greatest utility in fire behavior modeling is that it provides an excellent way for fire behavior experts to express their understanding of wildland fire behavior in easily understood graphic forms.

Fire Spread Modeling/Hazard Assessment for Industrial Timberlands
Click to view full image

Modeling is part art, part science

Fire behavior modeling draws upon all facets of what we know about fire. It uses observations of past fire behavior, historic weather data, and empirical observations from laboratory experiments. For example, much of what we know about the fire intensity of different types of surface fuels comes from people carefully sorting, weighing, and stacking different sizes of pine needles, sticks, and logs, and then burning them and recording the results. One researcher burned different types of ornamental shrubs like Oleander, Juniper, and Rosemary in a laboratory burn-chamber to develop data on potential fire behavior in common Wildland Urban Interface fuel types.

Based on historic fire behavior

Modelers use the accumulated science and a lifetime of personal wildfire experience to make inferences about what types of fire behavior are possible for a given area. For example, the map below shows the actual observed spread during the first day of the Poe Fire, which burned in 2001 in Northern California’s Feather River Canyon. Fire behavior modelers working in similar areas would use these spread rates to calibrate their models.

Click to view full image

Working at many scales

Fire managment challanges occur at a variety of scales. Starting with the big picture, or the landscape-scale, we may want to look for places a fire could start, get large enough to escape control, and then threaten a community or “asset-at-risk”. At the more local scale, we may be concerned about how to protect a community from a large fire approaching from the wildlands.

The graphics below highlight examples from a 2004 fire hazard assessment conducted for Plumas County. These tools and images below were developed by DCR staff, and our partners at Wildland-Rx, as part of a project that set priorities for hazarous fuels reduction projects for all of the private lands around communities in Plumas County. A major objective of this project was to create a library of reference photos showing potential fire behavior for each major forest type in the county. For more information on this project, visit the Plumas County Firesafe Council’s Website.

Identifying high-risk areas

Fire risk is defined as the likelihood that a fire will start in a given location. Some of the best tools for ranking risk are found in historic fire reports. The map below shows all human-caused fire ignitions in the area around Quincy, California between 1970 and 1996 (thanks to the US Forest Service).Click to view full image

Modeling Crown Fire Potential Around Quincy, CA

Fire hazard describes the likelihood that an area will burn with a dangerous intensity. The map below shows crown fire potential in the area around Quincy, CA. Crown fire potential is an important metric because it describes the fire’s resistance to control.Surface fires are often easy to suppress with local resources. Torching fires are more difficult to suppress as each torching tree can cause many spot fires in nearby areas. There is very little that firefighters can do to stop a crown fire other than to get out of the way and wait until weather conditions moderate, or for the to fire run out of fuel. In the following image, Green areas are predicted to experience low-intensity surface fire,Yellow areas will have scattered torching of the overstory trees, and Red areas will experience running (active) crown fire – where fire will leap from tree to tree.

Click to view full image

Crown Fire Modeling Method

We used the following data items to create this map:

  • Surface Fuel Type (grass, brush or timber litter)
  • Elevation, slope, aspect
  • Historic fire weather
  • Canopy base height (height to live crown of trees)
  • Tree height, crown density, and canopy closure

The fire behavior model (FLAMMAP) looked at the surface fuel type, slope and aspect, and determined potential surface flame lengths if the area was burning under historically high-hazard fire weather. If the flame lengths from the surface fuel were high enough to catch the overstory trees on fire, then this meant there was “crown fire potential”. If there was sufficient tree crown density to carry a fire from tree to tree, the area was rated as having “active crown fire potential”.

Wildland Urban Interface Hazard Assessment

DCR staff have developed mobile mapping tools for assessment of fuels hazard in the wildland urban interface (WUI). These tools allow field mapping crews to quickly assess fire-hazard attributes for individual structures. We then analyze this data with slope, weather, and remotely-sensed fuels data to classify the risk that wildfire poses to each structure. The photos and parcel maps below show wildfire hazard in the community of Cameron Park – in the Sierra Nevada Foothills. For this project we collected information for each of 7,300 parcels to assess those parcel’s suceptability to wildfire ignition.

In the photo and map below, a fire starting at the red icon would have the potential to destroy many homes. Contributing factors to this high level of risk are: location on a steep slope, flammable construction details like overhanging decks or shake roofs, uncleared brush and flammable ornamental vegetation like juniper or rosemary directly adjacent to the structures, and poor or unsafe access for firefighters.


Map below shows area in photo above.

Green areas = Low Hazard
Yellow areas = Moderate Hazard
Orange areas = High Hazard
Red Areas = Extreme Hazard

Modeling Wildfire Hazard in the Wildland Urban Interface

Similarly to the landscape-scale assessment above, we analyzed our parcel survey data to look at places where there was potential for a fire to run uphill thru heavy fuels and with the prevailing winds. A key consideration in modeling fire behavior in the WUI is that – while the fuel loads are often very similar to undeveloped brushlands in surounding areas – the first fire suppression resources on-scene will likely be busy protecting structures and less available to actually halt the spread of the fire. What this means is that a fire that might easily be suppressed in undeveloped brushlands has greater potential for growth if structures are present.

Prioritizing Community Defense Fuels Treatments

While the scales differ in the two examples above, both of these modeling exercises allow the end user to identify high hazard areas, and to prioritize these areas for hazard mitigation – either thru fuels reduction projects, community outreach, or code enforcement.

Designing Effective Fuels Reduction Projects

Effective fuels projects increase the effectiveness of fire suppression. Toward this end, all fuels treatments should be implemented with the objective of reducing the surface fuel loading within the treated areas. Projects must reduce potential fire intensity to a level which allows firefighters to safely work in the area during extreme fire weather conditions (when most large wildfires occur). Fire behavior modeling allows us to assess what kinds of vegetation managment will be necessary to accomplish this goal for a given area.

Project design tools

After priority areas have been established, terrain modeling and satellite fuels mapping data developed to run the models are also useful in the development of fuels treatment prescriptions. In the case of the Quincy, California area examples above, the Firesafe Council wanted to estimate how many acres of each type of fuels treatment were possible around each community in Plumas County. Slope steepness, vegetation type and percent canopy cover were used along with buffers drawn off of the community boundaries to generate the map below.

Click to view full image
View data table created from this map data.

More maps from this project.