Exploratory Thermal Image Analysis: Cooney Ridge Fire

Project details
The National Center for Landscape Fire Analysis (NCLFA) is analyzing thermal imagery from the Cooney Ridge Fire of August and September, 2003, using data gathered as part of the Joint Fire Science Program project Demonstration and Integration of Systems for Fire Remote Sensing, Ground Based Fire Measurement, and Fire Modeling (Study # JFSP-03-S-01). Airborne thermal images were collected using the Pacific Southwest Research Station's FireMapper  and ground-based thermal images were collected by the Rocky Mountain Research Station Fire Science Laboratory coincident with ground (in situ) measurements during a burnout operation on the Cooney Ridge Fire near Florence, Montana. This is the first simultaneous analysis of collected airborne and ground-based thermal imagery from a wildfire. The project seeks to understand how the thermal information provided by each image type can be useful to scientists and land managers.

The research addresses questions such as: What information is being shown in a thermal image? From thermal images, can fire intensity be predicted? From fire intensity, can fire effects on fuel be known? To get to the answers, the NCLFA must first understand the science of what energy the thermal cameras "see" and how the aircraft and ground-based measurements relate to each other.

Using the air- and ground-based image data collected from Cooney Ridge, the NCLFA can start to determine why the ground-based measurements are much hotter than air-based measurements and which data are most useful for predicting a fire's energy. As depicted in the images above, these two remote sensing techniques yield quite different portrayals of fire intensity. Preliminary results show that the ground-based system consistently measures higher temperatures (heat flux) than the airborne remote sensor. Thus, in order for managers to use these temperature maps to assess fire behavior and fire effects, research must quantify these differences. The end result of the research might be an equation that will predict duff consumption or fire effects; or a near-real time look at fire behavior so that fire managers and fire behavior analysts can have accurate information to manage a fire during the course of the fire. Following a fire, this understanding of thermal image data can be used to predict fire effects, which is important for restoration management - temperature measurements can be used to predict the fire's effects on fuels, soils, and watersheds, for example.

Project development
The Joint Fire Sciences Program-sponsored Rapid Response Project aimed to combine satellite, aircraft, on-the-ground measurements, and ground-based images. Currently, fire incident managers generally use MODIS satellite platforms to get a heat signature and determine where the fire is. During some large fires, infrared flights may be used to determine fire perimeters based on heat location. Satellite and aircraft can collect heat images, while instruments on site can validate what's being captured. The Rapid Response project and its follow-up projects, such as this thermal image analysis, will put together air-captured images, ground-based images, and ground-based instrumentation measurements for the first time. These various data types and collection methods can give pre-, during-, and post-fire ground measurements coincident with pre-, during-, and post-fire air measurements.

Principal Investigator: Casey Teske
Project Partners: Colin Hardy, Bret Butler, Patrick Freeborn, Bryce Nordgren --  USFS Fire Sciences Laboratory
Phil Riggan -- Pacific Southwest Research Station