Determination of weather parameters used to predict fire behavior and growth

Predicting potential fire behavior has become standard for developing fire management plans for land management agencies and is used regularly during fires to determine where control actions are necessary, the efficacy of control actions, and to ensure safety to firefighters. Predictions are developed using various models such as BEHAVE, FLAMMAP, FARSITE, and RERAP. These models are used to model a point in time and place, fire movement across a landscape for several days, or probabilities of fire movement based on historic situations. Regardless of the modeling objectives and tools, weather analysis is instrumental to all fire behavior predictions.

The installation of a network of Remote Automated Weather Stations (RAWS), a central data depository (KCFAST, WIMS), and the development of the Fire Family Plus weather analysis tool, has facilitated the easy acquisition and synthesis of historic weather information throughout the country. However, there is no standard for developing or reporting the weather conditions to use for fire behavior modeling at the various temporal and spatial scales. Assessments may use weather that occurred during a particular fire that exhibited a certain fire behavior. Some assessments include weather at the xth percentile; frequently the 90th and 97th percentile. Rarely is it documented why a particular weather scenario was chosen or how it was developed. The objectives of this study are: A) Compile the methods used to determined weather used in fire behavior analysis. B) Compare the fuel moistures and wind speeds derived using these methods to determine if there is ‘significant difference'. C) Compare fire behavior characteristics modeled with the fuel moistures and wind speeds to determine if there is ‘significant difference'. D) Propose and demonstrate methods to determine weather attributes based on modeling objectives.

Methods used to determine weather for fire behavior analysis are compiled by reviewing published literature, fire behavior assessments, and surveying practicing fire behavior analysts. Using the same methods, fuel moistures and wind speeds from several weather stations are derived from Fire Family Plus. Because many fire behavior models in the United States use the Rothermel model to calculate flame length, rate of spread, and fireline intensity, these fire behavior characteristics are modeled with BEHAVE and FLAMMAP for each fuel moisture and wind scenario. Additionally, FARSITE is run to determine fire size and growth using a standard landscape and randomly selected ignition points. RERAP is run using a standard assessment line from the randomly selected ignition points. Comparisons are made based on calculated flame lengths, rates of spread, fireline intensity, fire size, and probability of fire reaching a point of concern to determine the effects of various weather selection methods. Weather during past fire events are assessed to determine conditions that lead to low, moderate and high rates of spread and fire growth. The percentile weather at which these conditions occur is determined. Standard methods for determining weather parameters for fire behavior assessments are proposed.