Current Research Initiatives:

Remote Sensing and GIS Approaches to Fire Fuel Modeling in the Southeastern US
An Early Warning System for Hemlock Woolly Adelgid in the Southern Appalachians
Habitat Change Analysis Using Landsat TM
Improved Urban Forest Management via GIS

Ongoing & Completed Research Projects:

National Park Service GIS Technical Support Center
Geospatial Data Development for Selected National Parks
Spatial Modeling for Park and Recreation Planning and Management
GIS Technologies for Natural Resource Management
Instructional Technology for GIS and Environmental Science Learning

Remote Sensing and GIS Approaches to Fire Fuel Modeling in the Southeastern US

Issue: The President’s National Fire Plan is concerned that fuel loads are reaching hazardous levels that can lead to widespread catastrophic wildfires in forest ecosystems and the forest/urban interface. Land managers need tools to enable them to classify, estimate, and monitor fuel loading, and to predict wildfire risk and behavior based on inputs of fuel, weather, and topography for a specific location.

Response: Working with personnel from the Forest Service, the National Park Service, and NC Division of Forest Resources we developed protocols for mapping fire fuel loads using digital aerial photography, satellite remote sensing, vegetation maps, and field data. We found that actual fuel loads were, in many places, considerably different than standard loadings given in fire behavior models, but loadings were highly correlated to vegetation alliances and stand history.

Impact: Three major impacts have already been realized from this research: 1) working with the National Park Service we have developed digital fire fuel maps (both complacent and available) for the high fire risk parks in the Eastern US. These maps form the basis of the park fire management planning, 2) in cooperation with the NC DFR we have constructed a completely new set of high resolution road and water source maps that are the now reference for pre-suppression fire planning state wide. These digital maps have now also been adopted by the NC Department of Agriculture for bio-terrorism planning, and 3) the fuel load mapping approach investigated in this study is currently being extended to the Alligator Wildlife Refuge and Dare County Bombing ranges to aid the US Fish and Wildlife Service and the Department of Defense in fire protection planning in this environmentally sensitive area.

An Early Warning System for Hemlock Woolly Adelgid in the Southern Appalachians

Issue: Hemlock forests throughout eastern North America are currently threatened by the hemlock woolly adelgid (HWA). If the pest continues to spread at its current rate, it will rapidly invade the remainder of the southern Appalachian region. HWA attack is almost always fatal and has already virtually eliminated the Hemlock population in the Shenandoah National Park. HWA could potentially entirely remove these ecologically important trees from the whole region if preventive measures are not developed abnd deployed in a timely manner. Currently HWA management is based almost entirely on biological control. This expensive procedure requires both careful delineation of potential HWA infestations and a clear understanding of the environmental conditions under which the control populations will succeed. Efficient remote sensing and GIS procedures for HWA infestation mapping and epidemiological modeling are essential if any hope of stemming the impact is to be realized.

Response: We developed a new approach to data fusion that allowed the combined use a multiple remote sensing data sources (Landsat TM , ASTER, etc.) and combined this with the development of a suite of decision rules involving topography, geology, and other factors to predict HWA infestation with reasonable accuracy. Using this discriminate approach, we described a new hemlock mapping protocol that can be scaled to the entire region. We then developed a predictive model of HWA risk that can be used to prioritize control measures.

Impact: Results of our current studies have produced a detailed GIS data base of the spatio-temporal distribution of HWA in the Great Smoky Mountains National Park combined with corresponding environmental conditions (e.g., elevation, surrounding vegetation and land use, aspect, soils, etc.). From this we have derived a predictive spread model that outlines the likely annual extent of HWA in the park thereby providing the managers with a basis to direct their preventive measures. We have been approached by the Forest Service to develop a program to extend these methods to other invasive pests that have potential catastrophic impacts (Sudden Oak Death, Emerald Ash Borer, Southern Pine Beatle, etc.) over large geographic areas.

Habitat Change Analysis Using Landsat TM

Issue: The North Carolina Sandhills Conservation Partnership (NCSCP) is developing a conservation strategy for the red-cockaded woodpecker, the longleaf pine forests, and the other ecosystems and biota of the Sandhills in North Carolina. The plan emphasizes the conservation, restoration, and management of key land parcels within the mosaic of existing land uses embedded in the Sandhills longleaf pine ecosystem.

Response: Regional (multispectral) image data were used to map vegetation changes significant to wildlife management objectives, and site specific (hyperspectral) image data were used in discriminating pine forest types, which provide key habitat for wildlife species of concern.

Impact: These data sets have provided NCSCP with a means of identifying lands which provide forested wildlife habitat corridors between, and habitat buffers surrounding, existing public and private conservation lands. They are also used to identify areas that can be targeted for habitat restoration for the benefit of the endangered red-cockaded woodpecker. Studies are continuing in cooperation with the Army Corps of Engineers Research Lab (CERL) to characterize stand composition and structure using hyperspectral imagery and lidar in these areas of existing or potential habitat.

Improved Urban Forest Management via GIS

Issue: As NC and the Southeast have become more urbanized the need for better understanding of the Urban Forest environment has become a major issue. Local governments are spending very significant portions of the budgets trying to maintain healthy street tree and green space ecosystems with very little information on tree inventory, management costs, and environmental impact. In addition, most of these same governments have invested heavily in Geographic Information Systems primarily to assist with their tax parcel management and emergency dispatch systems. Therefore a very promising opportunity to develop and deploy detailed urban tree inventory and modeling systems exists with the potential to dramatically impact the immediate course of urban forest planning and management.

Response: NC State University has developed a prototype urban forest GIS program based on high-resolution digital orthophotography, integrated GPS and PDA technologies, customized GIS software, and wireless Internet access. This system has been adopted by the university and emulated by several other local institutions and governments and is currently being refined. We are testing its universality by running training courses for multiple government and private sector groups and supporting its deployment via our Park and Recreation Extension program.

Impact: In addition to the municipal government urban forest institutions using the current program, the integrated software design has enticed multiple agencies on the campus to participate in its development. Police, Environmental Management, Transportation, Planning, and Engineering are now employing the prototype and adding their own data layers and analysis tools. This has led to a model design of an integrated web based GIS for use in high-resolution urban vegetation management. The system was also expanded by the Science Education faculty to the allow use in the NC middle school environmental science curriculum. Over 100 K-12 teachers have now been trained in its use and lesson plans and science activities continue to be developed.

National Park Service GIS Technical Support Center

This project involves developing protocols for GIS application to National Park management problems. We construct solutions, train personnel, manage data, and assist with overall GIS planning for both individual parks and for the NPS as a whole.

The Nature Conservancy and a network of Natural Heritage Programs are developing a hierarchical vegetation classification system. NC State provided the first attempt at implementing a subset of this system. Shiloh National Military Park was one of the National Park Service units selected for application of the classification system.

Geospatial Data Development for Selected National Parks

This is a project to develop digital vegetation databases and related products for eight national parks in the Northeast Region and to create digital orthophoto mosaics for ten parks in the National Capital Region.  North Carolina State University's lead role in this project consists of everything from helping the NPS in contracting for the aerial photography to the final deliverables as well conducting accuracy assessments and producing FGDC compliant metadata. North Carolina State University has developed a methodology that allows us to create digital orthophoto mosaics from scanned aerial photography. This scanned photography is then viewed onscreen in 3d stereo and is used to delineate and classify vegetation types.

Spatial Modeling for Park and Recreation Planning and Management

There are a number of investigations underway within this project. These include urban vegetation analysis, greenway location studies, park master planning technologies, and regional environmental assessment techniques. Recent projects have involved urban ecology assessments, state park master planning, mid-Atlantic forest health inventories, and site analysis for rural park development.

The Triangle Greenways Council collected spatial and attribute information regarding trails and greenways in the Triangle area of NC. This image depicts the location of existing and planned greenways and trails. The information was digitized into a GIS and mapped by NC State. When completed, the map will aid planners in linking existing and planned trails to form a network.

GIS Technologies for Natural Resource Management

This project has centered on the design and development of GIS tools for natural resource management. New interfaces for specific applications are being developed to insure ease of use of GIS for natural resource managers. Examples include data preparation and display tools for water quality modeling, timber management, and coastal erosion as in the following example.

The potential erosion impacts of the jetty (terminal groin) at Oregon Inlet has been an on-going project. The groin protects the Bonner Bridge (the transportation link to the southern barrier islands of North Carolina) and threatens the Pea Island Wildlife Refuge. GIS is used to delineate the position of the shoreline and measure changes in shoreline position over time.

Instructional Technology for GIS and Environmental Science Learning

We are developing college undergraduate materials to facilitate GIS instruction via the Internet. This program is aimed at bringing GIS capability (both software and data) to every desktop in the university community. A potentially nation-leading effort is underway linking our GIS capability to the NC State University Libraries spatial data collection and the NC State Computing Services. In addition, we have an on going program with the North Carolina Department of Environment, Health, and Natural Resources to also bring GIS to the study of environmental sciences in K-12 classrooms.

Student-Directed, Information-Rich (SDIR) Undergraduate Education Project, 1994-96: The goal of this project was to develop teaching methodologies for GIS and other natural resources subjects that facilitate the shifting of in-class instruction from a teacher-centered mode (i.e. lecture) to a student-driven model in which the instructor is primarily a mentor and reviewer. This information-rich classroom environment is supported through multimedia computers connected to the campus communications network. The project was a joint effort combining the resources of the NC State Libraries, the College of Forest Resources, and the NC State Computing Center.