DOCUMENT

Name:   Fuel Classification for the Southern Appalachian Mountains using Hyperspectral Image Analysis and Landscape Ecosystem Classification -- Task 4 Final Report

Date Published:  2006

Document Type:  report

Brief Description:  Excessive fuel loading and rapid urbanization are of concern throughout the southern Appalachian Mountains. Prescribed burning has been attempted only recently in these mountains and fuel models are not available. This project used a two-phase approach to model fuel loading in the region. The first phase tested an existing Landscape Ecosystem Classification Model as a basis for classification of fuels in mountainous terrain. We expected fuel classes to be correlated with the same variables used to define landscape units (geography and topography). The second phase was to determine if fuels in each class can be detected from aerial images using hyperspectral image analysis.

Complete Description:  Executive Summary: Excessive fuel loading and rapid urbanization are of concern throughout the southern Appalachian Mountains. Prescribed burning has been attempted only recently in these mountains and fuel models are not available. This project used a two-phase approach to model fuel loading in the region. The first phase tested and existing Landscape Ecosystem Classification Model as a basis for classification of fuels in mountainous terrain. We expected fuel classes to be correlated with the same variables used to define landscape units (geography and topography). Phase one used over 1,000 study plots throughout 4 states to measure fuels over a variety of topographic and disturbance conditions. Pilot studies showed that the Landscape Ecosystem Approach was difficult to use and provided little extra information than by stratifying sample plots by topographic position (southwest-facing upper and lower slopes, northwest-facing upper and lower slopes, and ridges). Therefore, each plot was described by topographic position. Disturbances were described by existing conditions and confirmation from management records. Common disturbances included fire, windthrow, southern pine beetle attack, and harvesting. Sampling required approximately 3 years to complete because of the remote locations of most sample plots. Lessons learned in Phase One: (1) Few fuel differences occurred among topographic positions for undisturbed plots, indicating that fuel accumulation is no greater on highly productive sites than on less productive sites. (2) Ericaceous shrubs were less common than expected and may be problem fuels in only limited areas. (3) Disturbance history and type played a greater role in determining fuel loads than did topographic position. (4) Disturbances were common, particularly on exposed slopes. (5) Fire tended to decrease some fuels while beetle attack, harvesting, and windthrow increased most fuels. (6) Fuel loads described by this study are directly applicable to fire planning and fire behavior modeling and can help to determine where fuel measurements for planned fire should be concentrated. Once Phase One was completed, the second phase was to determine if fuels in each class can be detected from aerial images using hyperspectral image analysis. Signatures of reflectance values were developed for important fuel categories with ground measurements. Analysis of hyperspectral images identified areas on the ground containing each category of fuel loading. Ground crews verified the accuracy of these signatures and improvements were made. Validation of the signatures was done by image analysis and ground truthing. An unexpected result of phase one was that woody fuels varied little among topographic positions, with the exception of disturbed sites and the presence of ericaceous shrubs. Because of these results, we changed our emphasis for hyperspectral imaging to the techniques for mapping ericaceous shrubs. We suggest that disturbed sites and topographic position are simple to identify with hyperspectral imaging but these features can be identified by other methods (Landsat or aerial photography) that are much less expensive, making hyperspectral imaging impractical. Also, this part of the project was limited to the study area on the Nantahala National Forest due to data limitations in the other areas. Lessons learned in Phase Two (1) The advantages of using hyperspectral imagery for mapping include improved spatial resolution (2 m) and improved spectral resolution (37 channels for a broader range of energy bands. (2) The disadvantages include; the technology is untested for these specific uses and it may be too expensive for practical forest management purposes. (3) The use of hyperspectral data and the new method of using the hyperspectral library (STIHL) with the Enhanced Spectral Angle Mapper (ESAM) to identify members of the understory plant community matches the needs of forest research and fire management. (4) The ability to detect changes in species distribution is within the capabilities of hyperspectral technology. (5) By utilizing techniques outlined in this study as well as combining others such as NDVI and other spectral indexes, it may be possible to obtain rough assessments of fuel loads without the labor and time consuming ground surveys. (6) Hyperspectral field profiles (spectra) of the plant species, created under natural conditions, have the highest likelihood of comparing favorably with the spectra extracted from remotely sensed imagery.

Citation:

Related Resources*

Resource Name:

Fuel Classification for the Southern Appalachian Mountains using Hyperspectral Image Analysis and Landscape Ecosystem Classification

Availability & Use

Distribution ®	Format ® Online Source: http://jfsp.nifc.gov/projects/01-1-4-02/01-1-4-02_final_report.pdf http://jfsp.nifc.gov/JFSP_Completed_Projects_6.htm Fees: none Instructions: 50 page .pdf available at http://jfsp.nifc.gov/JFSP_Completed_Projects_6.htm or directly at http://jfsp.nifc.gov/projects/01-1-4-02/01-1-4-02_final_report.pdf
Contact Information ®	Individual: Organization: Joint Fire Sciences Program Position: Address: Telephone: Fax: E-mail: Hours: Instructions: Information presented on this website is public information and may be distributed or copied. Use of appropriate byline/photo/image credit is requested. We strongly recommend that USDOI data be acquired directly from a USDOI server and not through other sources that may change the data in some way. While USDOI makes every effort to provide accurate and complete information, various data such as names, telephone numbers, etc. may change prior to updating. USDOI welcomes suggestions on how to improve our home page and correct errors. USDOI provides no warranty, expressed or implied, as to the accuracy, reliability or completeness of furnished data. Some of the documents on this server may contain live references (or pointers) to information created and maintained by other organizations. Please note that USDOI does not control and cannot guarantee the relevance, timeliness, or accuracy of these outside materials. For site security purposes and to ensure that this service remains available to all users, this government computer system employs software programs to monitor network traffic to identify unauthorized attempts to upload or change information, or otherwise cause damage. Unauthorized attempts to upload information or change information on this website are strictly prohibited and may be punishable under the Computer Fraud and Abuse Act of 1986 and the National Information Infrastructure Protection Act. Information may also be used for authorized law enforcement investigations. URL: http://jfsp.nifc.gov/index.html
Liability:	unknown
Use Constraints:	unknown
Access Constraints:	unknown

Spatial Domain Information

Bounding Coordinates ®
Polygon ®

Time Period of Content

Single Date:
Date Range ®

Contributors

Name/Organization	Contribution			Role	Contact Information
unknown	unknown unknown

Metadata Reference

Details ®	Created Date: 11/27/2006 Metadata Creator: Elizabeth Caldwell The Nature Conservancy ecaldwell@tnc.org Type: FRAMES document Metadata Version: 1 Security Information: Schema: http://frames.nbii.gov/metadata/documents/documents.xsd
Keyword(s):	FRAMES subject categories prescribed fire GACC regions Southern FRAMES subject categories models NBII Resource Type Management Plans and Reports none aerial photography none Appalachian Mountains none fuel loading none fuel models none LANDSAT none Landscape Ecosystem Classification none mountainous terrain none topography
Alternate ID ®

Additional Information

	For additional resources, disclaimers,  and other FRAMES contact information, please visit the FRAMES website, at http://frames.nbii.gov.
	*Please note: Depending on metadata completion, upkeep, and availability, not all related resources may be available through FRAMES.