Selected current and past research

Note: This page is currently being updated

Land Use Modeling and Change Analysis Research

1. Integrated Land-Use, Transportation and Environmental Modeling: Complex Systems Approaches and Advanced Policy Applications. 2007-2011. unded by the UVM Transportation Center/ US Department of Transportation. Click for a flowchart of the project. See http://www.uvm.edu/envnr/countymodel/ for more details

2. Dynamic Transportation and Land Use Modeling. US Department of Transportation, Federal Highway Administration. 2006-2009. Click for final report. See http://www.uvm.edu/envnr/countymodel/ for more details.

3. Analysis of the drivers of urban growth and second home development in the Northern Forest Region of Vermont. Northeastern States Research Cooperative. 2009-2011. This project proposes to simulate future land use change stemming from two distinct processes: 1)primary residential development in urban and suburban areas and 2)second or vacation home development. The study area includes nine counties in northern Vermont. Based on these results the project will characterize expected future forest fragmentation and habitat connectivity conditions. The outcome of a baseline simulation (business as usual scenario) will be compared against the results from simulations under alternative land use policy scenarios. Future land use patterns will be simulated using one modeling approach for urban/suburban residential development and another for second/vacation home development.

Projects Contributing to the Baltimore Ecosystem Study:

Development of Advanced Spatial Analyses for Examining the Relationships among Social Groups, Land Managmeent and Vegetation Structure. 2006-2009. Funded by a research joint venture agreement with the USDA Forest Service.

A Longitudinal Analysis of the Social Dynamics of Environmental Equity in Baltimore. 2006-2009. Funded by the National Science Foundation (Multi-institution PI Christopher Boone, Arizona State University)

Application of the Forest Stewardship Program’s Spatial Analysis Project to Urban Areas Integrated with the Forest Opportunity Spectrum. 2006-2008. Funded by the USDA Forest Service.

Feedbacks between Complex Ecological and Social Models: Urban Landscape Structure, Nitrogen Flux, Vegetation Management, and Adoption of Design Scenarios. 2005-2007. Funded by the National Science Foundation (Multi-institution PI Steward Pickett, Institute for Ecosystem Studies)

These grants fund research which constitute part of my role as a Co-PI for the Baltimore Ecosystem Study, a long-term ecological research (LTER) project of the National Science Foundation. Over the last year, we have collected, processed, cleaned and organized a large amount of geo-spatial data at the UVM Spatial Analysis Lab (SAL), as part of BES. These projects contribute to both BES and to the Urban Ecology Collaborative (UEC), an affiliated project that looks at urban ecological patterns and processes across several East Coast cities. Among the topics these projects look at are the socio-economic and biophysical predictors of urban private and public vegetation, methods for differentiating planed vegetation from plants established through natural succession, looking at the impacts of parks on property values and how that effect is conditioned by crime, looking at the impacts of vegetation and landscaping on property values, looking at

Object-oriented land use/land cover classification research

Functional Classification of Land Use across the Urban-Rural Gradient to Support Watershed Planning in the Northern Forest. 2006-2009. Northeastern States Research Cooperative. The purpose of this project is to develop a socially and ecologically meaningful high-resolution LULC classification system for the Northern Forest region of Vermont that can be used to predict several important biogeochemical indicators which in turn affect water quality. This classification system was be designed to characterize the unique and complex mix of human development and forest cover that is prevalent in northern Vermont, from traditional urban centers like Burlington, to low density exurbs dominated by heavy forest cover.

Development of an object-oriented framework for classifying and inventorying human-dominated forest ecosystems. 2005-2008. MacIntire-Stennis Federal Research Funds.

Throughout the Eastern United States forest cover has been returning as agricultural lands have been abandoned. While estimated aggregate figures of forest cover give great room for optimism these aggregate numbers are misleading in that they do not reflect the highly fragmented condition of this forest matrix. As development becomes increasingly diffuse and dispersed, the distinction between “suburb” and “forest” becomes less clear in this region. However, commonly used land cover classifications do not provide an adequate means for classifying land with mixed human and natural components. This project proposes development of a framework for classifying and inventorying Eastern forestland based on level of anthropogenic perturbation and fragmentation using high-resolution remote sensing data and multi-scale object-oriented (OO) classification system. It implements this framework in the contrasting settings of Northeastern Vermont and the Gwynn’s Fall watershed in suburban Baltimore, both of which have their own unique mixes of forest cover and suburban or exurban development. Because certain “perturbing” features only become evident as the scale becomes finer, this framework is scale variant. Using eCognition software, a knowledge base of classification rules and algorithms will be developed which can then be applied to other areas in the East. Final aggregate figures of forest cover under this classification will be compared for the study region to commonly used figures based on the Anderson land use classification.

Ecosystem Service Research

Spatial Ecosystem Service Value Transfer in Massachusetts, Washington State and California. Various project from 2003-2005 (conducted through Spatial Informatics Group, LLC); click here for our recent publication entitled: Mapping Ecosystem Services Values: Practical Challenges and Opportunities in Bridging GIS and Value Transfer in Ecological Economics 60: 435-449.

Valuing New Jersey’s Natural Capital and Ecosystem Services. 2004-2005. Funded by the State of New Jersey, Dept of Environmental Protection. (with PI: Robert Costanza and Co-PIs: Matthew Wilson and Alexey Voinov). This project is attempting to estimate the economic value of ecosystem services in the state of New Jersey. Using the “value transfer” methodology, estimated ecosystem service values are assigned to land cover types throughout the state, using an extensive database of valuation studies and GIS data. We are also generating our own ecosystem service valuation estimates specific to New Jersey for a variety of land types using a combination of ecological modeling and econometric analysis of housing data (hedonic analysis). These analyses will help us generate site-specific estimates of the ecological, recreational and amenity value of various natural land covers, and assess how those change with a variety of contextual factors.

Natural Hazards Policy Research:

I am lead editor of a book that is coming out in early 2007 from Elsevier Publishers, entitled “Living on the Edge: Economic, Institutional Management Perpsectives on Wildfire Hazard in the Urban Interface.” Click here for a chapter and contributor list.

See also my article entitled The Role of Disclosure in the flood zone: Assessing the price effects of the California Natural Hazard Disclosure Law (AB 1195). Journal of Environmental Planning and Management. 47(1): 137-162 (email me for reprints) and a policy report I wrote published by California Policy Research Center on hazard disclosure entitled An Assessment of the 1998 California Natural Hazard Disclosure Law.

Other GIS and Remote Sensing Research

Development of Functional Ecological Indicators of Suburban Sprawl for the Northeastern Forest Landscape. 2004-2006 (Morgan Grove, Co-PI). Funded by the Northeastern States Research Cooperative

Little by little, development is eating away at the once continuous Northern forests. While all urban development is by definition detrimental to the natural environment, some land must be developed to house people and provide places of employment and recreation. “Urban sprawl” is a term used to describe an inefficient pattern of development where the impacts are high relative to the services provided. Various authors have attempted to define indicators of sprawl based on its social and economic impacts, but few have attempted to define them primarily in terms of ecological impact. One of the challenges to this is that the ecological impacts of development are ecosystem dependent; what defines sprawl will be different in each context. The purpose of this project is to develop ecologically based and spatially-explicit indicators of suburban and ex-urban sprawl that are specific to the Northeastern Forests and that consider per capita impacts. Using existing empirical research, combined with extensive mapping and spatial analysis efforts, these indicators will serve to track the environmental impact of development relative to the number of people housed in a given area. Hence, while areas of dense development result in a high impact, their per capita impact is actually low, just as extremely sparse, rural development also has a minimal impact. However, in between those extremes, impact per capita increases until it crosses some threshold that we clearly recognize as being sprawl. This study will synthesize results of existing research, including original research that the PI is participating in at the University of Vermont, assessing the impacts of development on various metrics of environmental quality in Vermont. The specific metrics we will consider are habitat fragmentation, biodiversity and water quality. This information will serve as the basis of a mapping methodology which will be used to create two sprawl maps for Northern Vermont. The first is descriptive. It will show per capita impact, indicating where densities and spatial arrangements are most and least efficient. It will take into account densities, soils, topography, hydrography and natural communities. Two methods for creating this map will be compared. The first will use medium resolution publicly available data and easily replicable methods while the second will use proprietary high-resolution remotely sensed data with advanced image interpretation methods. If the two yield similar results this is an indication that these maps can be created in other locales without resorting to high cost methods. The second map will be prescriptive. It will show where, within the “sprawl zone” as defined in the first map, additional allocations of density should go, based on an exogenous growth forecast and local factors including proximity to existing urban centers, major highways and important natural areas, the density of neighboring residential developments, site suitability and surrounding land uses.

Linking Land Use Change, Stream Geomorphology, and Aquatic Biodiversity in a Hierarchical Classification Scheme. With Mary Watzin (PI) , Cully Hession, and William Keeton. US EPA Science to Achieve Results (STAR) Watershed Classification Grant. 2002-2005.

AND

Linking Land Use Change, Stream Geomorphology, and Aquatic Integrity in Changing Forested Landscapes. With Mary Watzin , Cully Hession, and William Keeton. Northeastern States Research Cooperative. 2002-2005.

We are developing and evaluating a watershed and stream reach classification system based on the relationship between land use change, river geomorphic condition, riparian habitat fragmentation, and riverine ecological condition. The goal is a geographically independent management tool that can be applied in a hierarchical approach, linking changes in land use to expected changes in geomorphology and the riparian corridor, and then linking geomorphology to aquatic ecology. Because different factors and responses might be expected at different scales, the classification system will explicitly address issues of scale. Our objectives are as follows: (1) Beginning at the finer scale, determine which geomorphic classes can be consistently related to the ecological condition of a stream reach, (2) Expanding to a coarser scale, determine what land use metrics best predict those geomorphic classes that consistently relate to ecological condition, (3) Evaluate the ability of our classification system to target sites in greatest need of watershed management and stream restoration based on current land use and geomorphology, (4) Develop and evaluate the ability of our classification system to identify sites in need of conservation based on predicted land use change and resulting effects on geomorphology and aquatic ecology, and (5) Develop a general framework for the Vermont Department of Environmental Conservation and others to use the resulting classification system as a foundation for statewide watershed protection, management, restoration, and education. To predict land use change, we are creating an urban growth simulation model for Northern Vermont, which predict development as a function of a variety of economic, spatial and environmental variables. We are experimenting with using discrete metrics of land use change (binary pixel values) as well as continuous measures that incorporate sub-pixel measures of change in impervious surface. We will also use the land use change model to conduct a policy simulation. By altering policy inputs to the model, we can see how predicted development patterns would change and how that, in turn, would influence watershed function.

Land Use Change in Northern Forests: Assessments and Recommendations for Conserving Biodiversity. With Drs. Therese Donovan and Alexei Voinov (Gund Institute for Ecological Economics). Northeastern States Research Cooperative, 2002-2005.

This project looks at the impacts of urban growth and land use change on metrics of biodiversity. It seeks to understand the connections between urban sprawl and fragmenting habitat through a coordinated field and modeling effort. The research goal is to predict how land use change will affect biodiversity across Vermont at multiple spatial and temporal scales, and, based on those results, to assess the implications for biodiversity of current and alternative land use policies. Coordinated assessments of biodiversity across Vermont are being used to build predictive models of occurrence for a variety of taxa. Meanwhile, urban growth for Vermont is being simulated under a variety of policy scenarios, including no change in current policies. For each policy scenario, the corresponding impact on biodiversity will be quantified. Finally, spatial optimization methods will be used to identify land use patterns that are optimal for conserving an array of species, subject to socio-economic constraints. The proposed research will result in a decision-making tool that informs stakeholders of how projected land use change scenarios will likely affect different levels of biodiversity. This tool will allow policies to be driven by their potential effects on biodiversity, thereby permitting proactive land use planning that maximizes conservation value in this region. My role in the project is creating the urban growth/land use change simulation and mapping the resulting fragmentation of the landscape. The model we will use is UrbanSim, an economically-based dynamic disequilibrium model that allows for prediction under a variety of policy scenarios.

Hayfields as sustainable habitat for grassland bird populations: Ecological and economic impacts of 'bird-friendly' hay. With Drs. Allan Strong (PI), Sid Bosworth and Terese Donovan. USDA-CSREES-IFAFS and USDA NRI. 2001-2006

This project looks at the effect of hayfield management practices in the Champlain Valley on songbird habitat. In particular, we are looking at whether currently utilized haying practices turn hayfields into "ecological traps" for birds. My contributions are to: (A)to look at the economic feasibility of farmers adopting bird-friendly haying practices and assess how that varies with farm type, size and quality,(B)use GIS and remote sensing to classify Champlain Valley farms by type, size, and hay productivity, (C)use GIS to determine the extent of hayfield acreage that might serve as songbird habitat across the Champlain Valley, and (D) model the regional economic impacts of switching to bird-friendly management, based on the classification of farms. We are interested in eventually looking at grassland habitat on non-agricultural, medium to large properties, and determining whether changes to mowing practices on these lands might be more feasible than on active agricultural lands. More information coming soon...

Economic and Institutional Feasibility of the Vermont Forest Ecosystem Management Demonstration Project Treatments. With Dr. Cecilia Danks. Vermont Monitoring Cooperative. 2002-2003.

This seed project looks at the social, economic and institutional feasibility of small-scale forest lot owners adopting alternative silvilcultural management systems designed to increase forest structure complexity. It also looks at local and regional markets and pricing for various traditional and non-traditional forest products, assesses variability of input costs, and attempts to determine the importance of non-market values in informing small woodlot owner cutting decisions.

Home