Source: UNIVERSITY OF VERMONT submitted to
A MODEL FOR SIMULATING FOREST FRAGMENTATION DUE TO SUBURBAN DEVELOPMENT IN VERMONT
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
MCINTIRE-STENNIS
Reporting Frequency
Annual
Accession No.
0193809
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Oct 1, 2002
Project End Date
Sep 30, 2004
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Project Director
Troy, A.
Recipient Organization
UNIVERSITY OF VERMONT
(N/A)
BURLINGTON,VT 05405
Performing Department
SCHOOL OF NATURAL RESOURCES
Non Technical Summary
Urban sprawl contributes to the fragmentation of forest habitat and to the reduction in ecosystem function. This project models the impacts of predicted future urban growth on forest fragmentation levels in Northwestern Vermont and simulates how these outcomes change under alternative policy scenarios.
Animal Health Component
35%
Research Effort Categories
Basic
15%
Applied
35%
Developmental
50%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1230699206015%
1236040206015%
1237210206020%
6100699206010%
6106040206020%
6107210206020%
Knowledge Area
123 - Management and Sustainability of Forest Resources; 610 - Domestic Policy Analysis;

Subject Of Investigation
6040 - Real estate; 0699 - Trees, forests, and forest products, general; 7210 - Remote sensing equipment and technology;

Field Of Science
2060 - Geography;
Goals / Objectives
This project proposes implementation of a predictive land use/land cover change model for two counties in Vermont's Champlain Basin, integrated with tools for forest fragmentation analysis. The objectives of the project are to: 1) spatially quantify loss of forest habitat to urban growth in Chittenden and Addison Counties over the last ten years, 2) predict patterns of residential development under current policies, based on historical trends and models of economic behavior, 3) describe the relationship between development patterns and metrics of forest fragmentation and model future forest fragmentation based on predicted development patterns, and 4) simulate how policy alternatives would likely affect future levels of forest fragmentation. The policy alternatives modeled in this project will concentrate on hypothetical changes to Vermont's Land Use Law (Act 250) as well as changes to local zoning ordinances and development impact fees.
Project Methods
This project will analyze recent changes in land use/land cover using remotely sensed imagery. Changes will be modeled statistically as a function of economic, locational, environmental and demographic characteristics. Urban growth will then be simulated dynamically by entering these parameter estimates into the UrbanSim model (Noth et al. 2000, Waddell forthcoming) . UrbanSim is an agent-based behavioral simulation model that operates under dynamic disequilibrium and integrates an economic framework. Outcomes under alternative policy scenarios will be modeled by altering model inputs and seeing how these affect predicted development patterns. To run the model, various spatial and non-spatial data inputs are needed. These include land use in the base year, population and employment figures and forecasts, current and planned transportation networks, physical constraints to development (e.g. wetlands, floodplains, steep slopes), and current policy constraints, including land use plans, zoning/density maps, protected/conserved land maps, locations requiring development impact fees, or any other user-specified constraint. Most of these layers are currently available in some form for the study area. The second component of this project involves assessing changes in forest cover and fragmentation. Digital imagery from the two different time periods used for the historical land use change analysis will be analyzed so that a baseline level of forest fragmentation can be obtained and a relationship between certain types of development and forest fragmentation metrics can be derived. This will allow for prediction of how future development will affect those forest conditions. Towards this end, the project will use remotely sensed imagery to derive numerous metrics of forest fragmentation at a variety of scales. To quantify metrics of forest fragmentation, the Patch Analyst Extension for Arc View will be used. References: Noth, M., A. Borning, and P. Waddell. 2000. An Extensible, Modular Architecture for Simulating Urban Development, Transportation, and Environmental Impacts. UW CSE Technical Report 2000-12-01. Waddell, P. Forthcoming. UrbanSim: Modeling Urban Development for Land Use, Transportation and Environmental Planning, Journal of the American Planning Association.

Progress 10/01/02 to 09/30/04

Outputs
This project seeks to better understand the extent of conversion from forest cover to urban development in Northwestern Vermont and to assess the possibilities for predicting these conversions into the future. For this project we conducted an extensive land use/land cover change analysis for Chittenden, Addison, Franklin, Lamoille and Washington counties, using time series LANDSAT satellite imagery. This change detection looked at land use conversion in six three year increments, from 1984 to 2002 and resulted in maps showing the location of lands that were converted during these times. Because of limited validation information on building patterns, we have only been able to conduct a full accuracy assessment on the last time increment, from 1999 to 2002, which is the only time increment where high resolution aerial photos are available for the beginning and ending of the time increment. As part of this project we also created an updated 2002 land use map for Vermont, improving on the current best-available land use data, which is from 1992. Finally, as part of this project we prepared all the data inputs needed to run UrbanSim, which is a spatial urban growth simulation model that will simulate patterns of future urbanization into forested areas. UrbanSim has not yet been run, but it will be shortly as a component of two related projects (one funded by the EPA and one by the Northeastern States Research Cooperative) which are building upon this project. UrbanSim will also be used to model how the pattern of fragmentation changes under several different policy scenarios, including the construction of proposed highway projects.

Impacts
This project is of great interest to planners, politicians and researchers in Northwestern Vermont because it provides badly needed and previously unavailable information on the extent, rate and spatial pattern of urbanization, and shows how much of that is occurring on previously forested land. Moreover, it provides an updated land use map for Vermont, showing the pattern of forests, agriculture and urban development in 2002, which is ten years later than the next most recent land cover data set. Such information can help planners assess whether development is occurring in the desired pattern and locations and how it is affecting the natural resource base. It can also help them in designating areas for conservation. The simulation is of great interest to planners because it can help them choose where to allocate infrastructural investments, where to encourage growth and where to restrict it, based on the predicted pattern of future development and the predicted response under different policy or investment scenarios. While other simulations are being conducted regionally, this simulation is the most empirically based and will be important for comparison to other regional models. To date, we have worked closely with the Chittenden County Regional Planning Commission and the Metropolitan Transportation in sharing data and methods and have an agreement to share our results with them when available.

Publications

  • No publications reported this period