Data types
Data Types
• Environmental factors
o Weather
• Surface data
• Wind All physically-based, empirical, or semi-empirical models of fire behavior will require an input of wind speed and direction for estimating fire rate of spread. The location of the wind measurement or model-derived quantity with respect to the fire line may vary.
• Vertical profiles
• Because so much of atmospheric dynamics depends on the vertical structure of fields such as temperature, pressure, wind, and humidity, vertical profiles may be used to initialize the state of atmospheric models, assuming that this profile is representative of the fire environment. These may be obtained from atmospheric soundings (through the depth of the tropopause) or atmospheric profilers which emphasize the lowest levels of the atmosphere.
• 3-dimensional atmospheric state.
• Air temperature – may be used as one of many variables for initializing the three-dimensional state of a computational fluid dynamics model such as a weather model, along with air water vapor mixing ratio, air pressure, wind speed and direction. Such physically consistent states may come from another computational fluid dynamics model or integrated analysis of many sources of data.
o Terrain
• A two-dimensional database of terrain elevation at a spatial resolution appropriate for the model. Users should take care that anomalies and errors have been removed.
o Fuel
• Steady, extrinsic properties - This is complex because it is not simply the vegetation present but the part available for burning. In general, particularly kinematic model where relationships have been developed relating fire front rate of spread to fuel loads, it is meant to be the (fast burning) small-dimensioned components of a fuel complex that participate in the reactions in the fire front that carry the flaming front. More generally and particularly in dynamic models (i.e. ones where the forces of momentum created by heat release and air acceleration), it may include the larger fuel components. Reliable measurements of fuel are hard to come by, and may be based on hand measurements (which are limited by their representativeness), assignment of categorical fuel model classification (Anderson (1982), Scott and Burgan (2005), McKenzie et al. ) based on anticipated fire behavior, remotely sensed land surface data cross walked to a categorical fuel model classification system.
• Fuel data is separated by vertical arrangement - ground fuel, surface fuel, and aerial (crown) fuel.
• Transient properties. The fuel state (fuel moisture) of surface and aerial fuel may be diagnosed or predicted using information about the physical properties of the fuel and a trace of recent weather data including temperature, humidity, pressure, wind speed, precipitation, cloud cover, and solar radiation. While the smaller dead surface fuels respond over short term time scales to weather conditions, the live fuel moisture depends more on the health of the plant and therefore longer term conditions and stresses.
• Fire data
o Time & location of ignition. These are not routinely distributed in a digital form but may later be found in incident reports.
• Currently, models begin at the time of ignition, which may be a point ignition, sequence of point ignitions, line ignition, or a line ignited over a period of time (as a person walking with a drip torch). It is recognized that differing ignition patterns may influence later fire behavior, thus models may include a wide variety of options here.
o Spatial data may include a two-dimensional map of the fire extent obtained by incident teams circling the fire with a GPS, aircraft-based imagery, unmanned aerospace vehicle, and (depending on the size and intensity of the fire and the satellite characteristics) satellite-based instruments, either polar-orbiting or geostationary. Fire remote sensing data often comes from instruments within the infrared range of the spectrum, since smoke frequently obscures the visible band.
o Times series of temperature produced by thermal sensors crossed by the fire.
References
Anderson, H. E. 1982. Aids to determining fuel models for estimating fire behavior. USDA For. Serv. Gen. Tech. Rep. INT-122, 22p. lntermt. For. and Range Exp. Stn., Ogden, Utah 84401.
Scott, J. H. and R. E. Burgan. 2005. Standard fire behavior fuel models: a comprehensive set for use with Rothermel’s surface fire spread model. Gen. Tech. Rep. RMRS-GTR-153.Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 72 p.
McKenzie, D., C.L. Raymond, L.-K.B. Kellogg, R.A. Norheim, A.G. Andreu, A.C. Bayard, K. E. Kopper, E. Elman, 2007. Canadian Journal of Forest Research. 37: 2421-2437.