Fire Severity and Perimeter Mapping
The NBR index is calculated for prefire and postfire images as described in section 2. Prefire and postfire images are inspected for coregistration accuracy and corrected if spatial differences are excessive and extensive (>30 m). NBR images are differenced for each fire-scene pair to generate the dNBR. A relativized dNBR (RdNBR) is also calculated based on the work of Miller and Thode (2007) to evaluate potential limitations of dNBR to characterize fire severity on low biomass sites and potentially enhance inter-fire comparability of the results at larger scales. The RdNBR data have been shown to have stronger correlations to Composite Burn Index plot data in some low biomass western ecosystems (Miller and Thode 2007), Thode 2005). Figure 1 illustrates the sequence of data layers generated.
Ecological Severity Thresholding
Processing the Landsat image data to dNBR is a straightforward series of objective calculations requiring limited analyst interaction and relying principally on automated production sequences. Subsequent to dNBR derivation, the process of developing fire severity and perimeter maps becomes much more dependent on analyst interpretation. The dNBR data are calculated as signed 16-bit with a maximum digital number (DN) range of -32767 to +32767. However, the practical range of DN values representing fire-related change and no change is typically within -2000 to +2000. Values further away from zero represent greater change as a result of both first and second order fire effects (within the fire perimeter). Negative values indicate a positive vegetation response (growth) and positive values indicate a negative vegetation response (mortality). Figure 2 illustrates a dNBR image for the Cerro Grande fire (2003), and Figure 3 depicts the associated data range. The analyst evaluates the dNBR data range and determines where significant thresholds exist in the data to discriminate between severity classes. Interpretations are conducted on the dNBR data aided by raw prefire and postfire imagery, plot data, and analyst experience with fire behavior and effects in a given ecological setting. Composite Burn Index (CBI) data (Key and Benson 2006) have been the most commonly collected ground-based data to estimate postfire effects. Correlations between CBI and dNBR have been used to demonstrate the sensitivity of dNBR to postfire effects and to establish numerical thresholds in dNBR data that discriminate severity categories (Cocke and others 2005, Key 2005). Where CBI and similar plot data have been collected, and plot-dNBR relationships published, analysts will guide their interpretations based on these relationships. Limited extrapolation of plot-based thresholds beyond their geographic bounds but within ecologically similar conditions will be examined.
Thresholding dNBR data into thematic class values results in an intuitive map depicting a manageable number of ecologically significant classes (typically 4 to 7 class values). Within this project, the thematic raster data will characterize severity in 5 discreet classes—unburned/unchanged, low severity, moderate severity, high severity, and increased postfire response. These classes will serve as a means to easily summarize severity acres across broad scales and provide a coarse look at effect gradients within fires. However, there are uncertainties in this approach stemming from analyst subjectivity and limited or no plot data to guide threshold selection. Large-scale analysis may best be conducted on the continuous dNBR data, which provide the greatest range of data quantifying postfire change. Although not a direct measure of fire severity, dNBR data have been shown to correlate to field-based estimations of fire severity (Hudak 2006, Key 2005).
Ecological significance is issue dependent, and one set of thresholds cannot be expected to apply equally well to all analysis objectives and management issues. Other severity classifications such as described by Stephens and Ruth (2005) may be used as the basis for thresholding but must be considered for the appropriateness of their application to dNBR data. Fire severity classifications that are based on fire effects not readily discernible on Landsat data, (e.g., subsurface biomass combustion or soil chemistry changes) should not be applied to these data.
dNBR Partitioning
In addition to ecological thresholds as a means of discriminating severity classes, dNBR will be arithmetically partitioned into discreet classes to facilitate objective and flexible pattern and trend analysis. Arithmetic partitioning is not intended to provide information on the ecological severity of fires at large spatial scales or limited temporal extents. Methods for partitioning dNBR have yet to be determined, and the algorithm(s) and subsequent grain of partitioning will depend on a given technique’s ability to reveal meaningful patterns in fire severity over time. Gmellin and Brewer (2002) used a simple equal interval calculation to establish objective burn severity classes between observed unburned and high-severity conditions in the Northern region of the Forest Service. Brewer and others (2005) used the same approach in a methods comparison study that concluded dNBR to be the most effective approach of those evaluated for mapping fire severity. The relative ease and quickness of arithmetically partitioning dNBR data will allow for rapid evaluation of meaningful spatial and temporal scales in the context of fire severity trends. Moreover, dNBR data can be efficiently analyzed and classified to suit the fire severity information needs of a specific management issue.
Perimeter Delineation
Fire perimeters will be generated by on-screen interpretation and delineation of dNBR images. Analysts will digitize perimeters around dNBR values reflecting fire-induced change. To ensure consistency and high spatial precision, digitization will be performed at on-screen display scales between 1:24,000 and 1:50,000. Incident perimeters, where available, will be used in an ancillary fashion to inform the analyst. This can be particularly useful in identifying unburned islands within a perimeter or isolated, disjunct spots outside the main perimeter. Due to limited and variable availability, as well as inconsistent spatial precision, incident perimeters were not considered appropriate as a source for MTBS project perimeters.
Encyclopedia ID: p3607
