U.S. Forest Service
20180309
RAVG Thematic Percent Change in Canopy Cover (CC-5)
raster dataset
<https://data.fs.usda.gov/geodata/rastergateway/ravg/index.php >
The USDA Forest Service Rapid Assessment of Vegetation Condition after Wildfire (RAVG) program produces geospatial data and maps of post-fire vegetation condition using standardized change detection methods based on Landsat or similar multispectral satellite imagery. RAVG data products characterize vegetation condition within a fire perimeter, and include estimates of percent change in basal area (BA), percent change in canopy cover (CC), and a standardized composite burn index (CBI). Standard thematic products include 7-class percent change in basal area (BA-7), 5-class percent change in canopy cover (CC-5), and 4-class CBI (CBI-4). Contingent upon the availability of suitable imagery, RAVG products are prepared for all wildland fires reported within the conterminous United States (CONUS) that include at least 1000 acres of forested National Forest System (NFS) land (500 acres for Regions 8 and 9 as of 2016). Data for individual fires are typically made available within 45 days after fire containment ("initial assessments"). Late-season fires, however, may be deferred until the following spring or summer ("extended assessments"). National mosaics of each thematic product are prepared annually. The associated burned area perimeters are available via the Enterprise Data Warehouse (EDW, see https://data.fs.usda.gov/geodata/edw/datasets.php). Assessment type (initial or extended assessment) for each fire is included as an attribute in the perimeter dataset.
RAVG data are produced to assist in post-fire vegetation management planning. They are intended to enhance decision-making capabilities and reduce planning and implementation costs associated with post-fire vegetation management. The primary benefit is cost-effective, efficient, and precise identification of potential areas of resource concern following wildfire. RAVG complements the Burned Area Emergency Response (BAER) Imagery Support program--which provides information integral to determining fire effects on soils--by providing information about fire effects on existing vegetation. RAVG analysis provides a first approximation of areas that may require reforestation treatments after a fire in order to re-establish forest cover and restore associated ecosystem services. This initial approximation may be followed by site-specific diagnosis and development of a silvicultural prescription to more precisely identify reforestation needs.
2013
2017
Ground condition
Published layers are considered complete with the possible exception of the most recent year. However, a layer may be updated if additional fires are assessed for the given year or if data for constituent fires are modified. Layers for additional annual national mosaics will be added as they become available. Annual mosaics for the most recent year may be expected early in the next calendar year. If any extended assessments are to be included, an updated mosaic will be provided near the middle of the year. Mosaics for 2007 to 2012 will be added at a future date as well.
-126.882
-72.763
50.019
29.174
ISO 19115 Topic Category
Biota
none
RAVG
Burn Severity
Vegetation Condition
Wildland Fire
Wildfire
Landsat
Normalized Burn Ratio
Differenced Normalized Burn Ratio
Relative Differenced Normalized Burn Ratio
NBR
dNBR
RNBR
Basal Area
Canopy Cover
Composite Burn Index
CBI
none
United States
CONUS
None
There are no restrictions on use, except for reasonable and proper acknowledgement of information sources. The USDA Forest Service makes no warranty, expressed or implied, including the warranties of merchantability and fitness for a particular purpose, nor assumes any legal liability or responsibility for the accuracy, reliability, completeness or utility of these geospatial data, or for the improper or incorrect use of these geospatial data. These geospatial data and related maps or graphics are not legal documents and are not intended to be used as such. The data and maps may not be used to determine title, ownership, legal descriptions or boundaries, legal jurisdiction, or restrictions that may be in place on either public or private land. Natural hazards may or may not be depicted on the data and maps, and land users should exercise due caution. The data are dynamic and may change over time. The user is responsible to verify the limitations of the geospatial data and to use the data accordingly.
U.S. Forest Service
physical
Washington
DC
20250
(801) 975-3800
rapid_assessment_post_fire@fs.fed.us
Questions or comments may be submitted by phone, by email or via the RAVG website: <https://www.fs.fed.us/postfirevegcondition/contact/index.php>
None
Unclassified
None
Microsoft Windows 7 Enterprise, Service Pack 1; ERDAS Imagine 14.00.0100, Build 745; Esri ArcCatalog 10.2.2.3552
No tests for logical consistency have been performed on these data. Each layer is a mosaic of numerous individual raster datasets, which were produced using a standardized process; however, some variability is expected due to the many factors that affect the underlying imagery as a function of time, location, and natural environment.
This dataset is intended to include burn severity data for all fires that meet standard RAVG criteria for the given years; namely, wildfires reported within CONUS that include at least 1000 acres of forested National Forest System land (at least 500 acres for Regions 8 and 9 as of 2016). Exceptions (omissions) may include fires for which suitable imagery was not available in a timely manner (i.e., within 45 days after fire containment for initial assessments).
The target scale of this dataset is 1:24,000. The horizontal accuracy of this dataset is determined by that of the moderate-resolution multi-spectral imagery from which it is derived.
RAVG burn severity data are derived from multi-spectral imagery. A Relative Differenced Normalized Burn Ratio (RdNBR) image, which portrays the variation of burn severity within a burn area, is derived from a pair of multi-spectral images: a post-fire scene and a corresponding pre-fire scene. Whenever possible, the RAVG project uses Landsat images that have been geometrically rectified, terrain corrected, and converted to top-of-atmosphere (at-satellite) reflectance. The following process was developed for Landsat imagery but applies to other multi-spectral satellite imagery that includes a near-infrared (NIR) band and a short-wave infrared (SWIR) band (approximately 2.2 micrometers).
The Normalized Burn Ratio (NBR) is computed for each image using the NIR and SWIR bands:
NBR = (NIR - SWIR) / (NIR + SWIR)
The Differenced NBR (dNBR) is computed as the change in NBR, scaled by 1000 (for representation as integer values):
dNBR = (PreNBR - PostNBR)*1000
The Relative dNBR (RdNBR) includes an additional scale factor designed to reduce the correlation with pre-fire vegetation density. It also includes an offset calculated from the dNBR in an unburned area in the vicinity of the fire having vegetation cover similar to that of the burned area. The offset is intended to account for seasonal or interannual differences between the two images.
RdNBR = (dNBR - offset)/sqrt(abs(PreNBR/1000))
RdNBR is the index used for vegetation burn severity assessments in the RAVG program. Higher RdNBR values are correlated with more severe burns.
Three burn severity indices form the complement of RAVG products: Composite Burn Index (CBI), percent change in basal area (BA), and percent change in canopy cover (CC). Values are estimated from RdNBR via regression models, which were developed from field data and Landsat imagery, both acquired approximately one year post-fire. The field data were collected from a number of fires that burned in the Sierra Nevada and Klamath Mountains in California between 1999 and 2006. RdNBR values were derived from post-fire imagery acquired at approximately the same time, and seasonally matching imagery from before each fire.
The CBI is a composite linear combination of fire effects experienced at a site and encompasses all vegetation strata from the forest floor to the upper canopy. The composite value ranges from 0 to 3. Thematic CBI-based severity classes were "calibrated" from field data collected about one year post-fire, meaning that field data collected on a given fire or fires that occurred in similar vegetation types were used to classify burn severity as unchanged, low, moderate or high severity. CBI values of 0.1, 1.25, and 2.25 were selected as the breakpoints between the successive classes, which are defined as follows:
Unchanged: Areas indistinguishable from pre-fire conditions. This does not necessarily indicate that the area was unburned.
Low: Areas of surface fire with little change in cover and little mortality of the structurally dominant vegetation.
Moderate: Areas between low and high severity, with a mixture of effects on the structurally dominant vegetation.
High: Areas where the dominant vegetation has high to complete mortality.
Percent change in basal area was calibrated through regression analysis of the RdNBR index to basal area change as measured in the same field plots where the CBI data were collected. For this measure, a tree with no green foliage one year post-fire was defined as being dead. Percent change in canopy cover was likewise calibrated through regression analysis of the RdNBR index to measured or calculated tree canopy cover change. Pre-fire canopy cover was determined through photo interpretation of pre-fire digital orthophotos or by way of the Forest Vegetation Simulator (FVS) (https://www.fs.fed.us/fmsc/fvs/).
The following models were used for extended assessments; that is, assessments based on post-fire imagery acquired during the growing season one year after the fire:
CBI = (1/0.6124)*LN((RdNBR + 123.3)/196.8) [2016 and earlier]
CBI = (1/0.3890)*LN((RdNBR + 369.0)/421.7) [2017 and later]
BA Percent Change = 100*(SIN((RdNBR-166.5)/389))^2
CC Percent Change = 100*(SIN((RdNBR-161.0)/392.6))^2
Calibrated thresholds for initial assessments were derived through regression modeling of satellite reflectance values from imagery collected one year post-fire to imagery acquired immediately post-fire. Application of the calculated correction factor yields the following initial assessment models:
CBI = (1/0.6124)*LN((RdNBR/1.1438 + 123.3)/196.8) [2016 and earlier]
CBI = (1/0.3890)*LN((RdNBR/1.1438 + 369.0)/421.7) [2017 and later]
BA Percent Change = 100*(SIN((RdNBR/1.1438-166.5)/389))^2
CC Percent Change = 100*(SIN((RdNBR/1.1438-161.0)/392.6))^2
A thematic version of each continuous burn severity raster was created based on the following class breaks:
Four-category severity classification based on CBI:
0 = outside perimeter
1 = unchanged (0 <= CBI < 0.1)
2 = low severity (0.1 <= CBI < 1.25)
3 = moderate severity (1.25 <= CBI < 2.25)
4 = high severity (2.25 <= CBI <= 3.0)
9 = unmappable
Seven-category percent change in basal area (BA):
0 = outside perimeter
1 = 0% BA loss
2 = 0% < BA loss < 10%
3 = 10% <= BA loss < 25%
4 = 25% <= BA loss < 50%
5 = 50% <= BA loss < 75%
6 = 75% <= BA loss < 90%
7 = BA loss >= 90%
9 = unmappable
Five-category percent change in canopy cover (CC):
0 = outside perimeter
1 = 0% CC loss
2 = 0% < CC loss < 25%
3 = 25% <= CC loss < 50%
4 = 50% <= CC loss < 75%
5 = CC loss >= 75%
9 = unmappable
When a perimeter for a given fire was available from the incident management team or local unit, it was used as the initial approximation for the RAVG burned area perimeter. In many cases, supplied perimeters were edited to account for obvious discrepancies with the visible burned area (based on the post-fire imagery). If no perimeter was provided, the burned area boundary was delineated from the dNBR and post-fire reflectance imagery. Open water and areas obscured by clouds, shadows, active fire, smoke, or snow were masked and identified as "unmappable".
Note that derived burn severity products represent a snapshot in time and that the models are applied to the entire extent of each burned area without regard to vegetation type. Factors such as delayed mortality, resprouting, the presence of non-tree vegetation, and the occurrence of non-fire disturbances can contribute to errors in the burn severity estimates.
20180227
All thematic burn severity images of a given severity type (percent change in basal area, percent change in canopy cover, or composite burn index) for a given year have been mosaicked into a national burn severity raster dataset. If needed, constituent raster datasets were resampled to a common resolution (30 m) and snapped to a common grid (the 2011 National Land Cover Database (NLCD)). Note that initial and extended assessments, if any, are combined in the mosaics. Assessment type for each fire is included as an attribute in the associated burned area perimeter dataset, available via the Enterprise Data Warehouse (EDW, see https://data.fs.usda.gov/geodata/edw/datasets.php).
20180227
Raster
Pixel
Albers Conical Equal Area
29.500000
45.500000
-96.000000
23.000000
0.000000
0.000000
row and column
30.000000
30.000000
meters
North American Datum of 1983
Geodetic Reference System 80
6378137.000000
298.257222
CC5
RAVG 5-class Percent Change in Canopy Cover (CC) class
U.S. Forest Service
Value
5-class Percent Change in Canopy Cover (CC) Category
U.S. Forest Service
0
Outside of fire perimeter
Internal
1
0% CC loss
Internal
2
0% < CC loss < 25%
Internal
3
25% <= CC loss < 50%
Internal
4
50% <= CC loss < 75%
Internal
5
CC loss >= 75%
Internal
9
Unmappable
Internal
Red
Red Color Value (0-255 scale)
U.S. Forest Service
0
255
Green
Green Color Value (0-255 scale)
U.S. Forest Service
0
255
Blue
Blue Color Value (0-255 scale)
U.S. Forest Service
0
255
Count
Pixel count for burn severity category
U.S. Forest Service
0
Unlimited
USFS Chief Information Office, Enterprise Data Warehouse
physical
Washington
DC
20250
Please send an e-mail to the address below.
data@fs.fed.us
Map service and downloadable data
The U.S. Forest Service makes no warranty, express or implied, nor assumes any liability or responsibility for the accuracy, reliability, completeness, or utility of these geospatial data or for the improper or incorrect use of those data. The data are dynamic and may change over time. The user is responsible for verifying the limitations of the geospatial data and for using the data accordingly.
20180309
USFS Chief Information Office, Enterprise Data Warehouse
physical
Washington
DC
20250
Please send an e-mail to the address below.
data@fs.fed.us
FGDC Content Standards for Digital Geospatial Metadata
FGDC-STD-001-1998