Alongside climate change, changes in land cover and land use have been highlighted as central drivers of change in ecosystems and the services they provide (Vitousek et al. 1997, Millennium Ecosystem Assessment 2005). These changes are largely driven by human actions, and the inherent complexity of human behaviors presents a major challenge for quantitatively predicting future change (Liu et al. 2007, Polasky et al. 2011). Scenarios provide a framework for projecting future changes when uncertainty is high: scenarios do not predict the future, but rather present collections of alternative storylines for plausible futures, which can provide insight on the range of possible outcomes (Schwartz 1991). Here, we present the development of a suite of high-resolution land-cover change scenarios for the state of New Hampshire that serve as input for process-based models, part of a U.S. National Science Foundation-funded project to evaluate temporal variability of future ecosystem services in the state (Mavrommati et al. 2016, Samal et al. 2017; Borsuk et al., unpublished manuscript).
New Hampshire has experienced large-scale land-cover change in the past, and there is a high degree of uncertainty about future land cover in the state with consequences for ecosystem services. Since the 17th century, New Hampshire has changed from primarily forest, to ~50% farmland in the mid-19th century, back to primarily forest (Foster et al. 2010). Since 1987, New Hampshire has lost forest area to development (Drummond and Loveland 2010, Foster et al. 2010, Levesque 2010, Thompson et al. 2013, Jeon et al. 2014), especially in the southeast (Johnson 2012, Sundquist 2012). Although population growth has slowed since 2000 (Johnson 2012), several projections suggest that development in the southeast will continue to expand over the subsequent several decades (Stein et al. 2009, USEPA 2009, Bierwagen et al. 2010, Stein et al. 2010, Sundquist 2012), with major implications for ecosystem services in the region. For example, three watersheds in southern New Hampshire have been ranked in the top five watersheds nationally in terms of projected degradation of water quality due to increased housing density on existing private forestland (Stein et al. 2009), and the same watersheds are classified in the 90th percentile nationally in terms of threats to forest species at risk from housing development (Stein et al. 2010). However, the state also has a long and robust history of grassroots conservation. Thirty percent (7300 km²) of the state’s land area is publicly or privately conserved (NH GRANIT 2016). This includes 3000 km² in the White Mountain National Forest, which was established in 1918 to mitigate hydrological damage to streams from intensive logging (Shands 1992, Conroy and Ober 2001). Overall, the competing land uses of development, forestry, conservation, and agriculture produce a high degree of uncertainty about the future of New Hampshire’s landscapes, and for land-cover scenarios to best inform future changes in the state, these competing uses must be addressed.
Our research builds on an extensive and growing body of literature on the use of scenarios for a variety of functions in environmental forecasting and planning (Peterson et al. 2003, Swart et al. 2004, Alcamo et al. 2006, Carpenter et al. 2006). Methods for scenario development and modeling are highly dependent on the specific objectives of a scenario effort (Mallampalli et al. 2016) and should generally meet the following goals: specific needs of the target audience (relevance), transparent and justified methods (credibility), perception of fairness (legitimacy), and provoking new thinking about uncertainty (creativity; Alcamo et al. 2006). These goals are often addressed through engagement of stakeholders, including decision makers, experts, and groups affected by scenario outcomes (Mallampalli et al. 2016). Approaches to translating available information into useful scenarios include innovative methods to involve stakeholders directly in model construction (e.g., Schmitt Olabisi et al. 2010) or to incorporate stakeholder knowledge into mathematical models (e.g., Bayesian networks; Meyer et al. 2014b), but traditional methods such as interviews and surveys remain major approaches to stakeholder engagement (Mallampalli et al. 2016). For example, key informant interviews are central to scenario development protocols recommended by the Natural Capital Project run by The Nature Conservancy and the World Wildlife Fund (Rosenthal et al. 2015). These traditional methods, alongside literature review, are well suited for collaboration with experts and can be used in combination with diverse modeling methods (Mallampalli et al. 2016). Translation of qualitative results into quantitative models can be iterative, as in the Storyline and Simulation approach in which stakeholders are asked to provide feedback on scenarios (Alcamo 2008), or, when resources are more limited or stakeholders are unable to commit to multiple workshops, a more streamlined approach can be used, drawing on published literature (Pfeifer et al. 2012).
Scenario developers must consider several components contributing to future change (e.g., in land cover): the driving forces that shape the relevant system processes; the predetermined elements that can be assumed to follow a predictable trajectory; and critical uncertainties, which are unknown, poorly understood, or can change in unpredictable ways (Schwartz 1991). However, driving forces are not always well understood, and critical uncertainties can emerge from unexpected changes to factors regarded as predetermined elements (Schwartz 1991). It is also possible for more than one set of driving forces to produce the same outcome (Moss et al. 2010). When thorough knowledge of a system is required for credibility or when modeling is used to develop a more sophisticated understanding of the system, it can be critical to explicitly model social, economic, and physical drivers (Nakicenovic et al. 2000, Wear 2011, Radeloff et al. 2012) and quantify parameter uncertainty (e.g., Alcamo et al. 2011, Verburg et al. 2013). However, excessive complexity can delay research because scenarios must be revised continually based on improved knowledge. Spatially explicit modeling of land-cover change is particularly complex, and the models are highly sensitive to underlying assumptions about driving forces; in fact, variability among models due to differing assumptions can be larger than differences among scenarios (Sohl et al. 2016). Models generally fail to predict the spatial distribution of land change at high resolution (Pontius et al. 2008), and projections are constrained by the interval of training data sets, typically 20 to 30 years (van Vliet et al. 2016), with uncertainty increasing for longer term projections (Pontius and Spencer 2005).
Our objective was to develop plausible, spatially explicit scenarios for land-cover change, population density, and impervious cover in New Hampshire from 2010 to 2100 as part of the larger NH EPSCoR Ecosystems and Society project (Gardner et al., unpublished manuscript). The scenarios served as critical input for linked terrestrial and aquatic ecosystem process models used to project changes in future environmental functions and the resulting changes in ecosystem services (Samal et al. 2017). The scenarios also served as a key source of information regarding alternate futures as part of novel deliberative multicriteria evalutions described by Mavrommati et al. (2016). Based on a combination of literature review, key informant interviews, and collaboration with subject matter experts, we produced six distinct scenarios for the future of land cover, conservation, population density, and impervious cover, and used custom models to generate decadal rasters for each of these attributes.
New Hampshire covers 24,214 km² in northern New England, between 42° 42’ and 45° 18’ N and between 70° 36’ and 72° 33’ W. The state is currently 75% forest, 5.5% developed, and 4.5% in farmland (NH GRANIT 2016; Fig. 1A). Some forest is managed for timber, especially in the northern region of the state; some is conserved for wildlife, recreation, or to protect ecosystem services; and some (particularly land held by small landowners) is subjected to a wide variety of land management practices. Most land area in New Hampshire is rural or low density (Fig. 1B), and most of the state’s population of 1.3 million lives in small towns. The three largest cities, Manchester (110,000 residents), Nashua (86,000 residents), and Concord (43,000 residents), are all located in a densely populated, central, Interstate 93 corridor in the southern portion of the state along the Merrimack River (U.S. Census Bureau 2013). The state is characterized by high rates of education (ranked seventh nationally) and family income (ranked eighth nationally), but the educated and high-income population is primarily located in the southeast (Johnson 2012). The northern portion of the state is lower income, with a rural economy dominated by manufacturing and timber industries (Johnson 2012) that are currently in decline. Farmland in the state is primarily located in the southeast and along the Connecticut River, which defines the western border of the state.
From the outset, we wanted to develop a suite of credible, legitimate, and relevant scenarios that were divergent in land cover, population density, and impervious cover, with contrasting consequences for ecosystem function and services. Our initial ideas were informed by previous land-cover scenarios developed for the Lamprey River watershed in southeastern New Hampshire (Scholz 2011) and Massachusetts (Thompson et al. 2013, 2014), as well as contrasting development patterns across southern New Hampshire, i.e., concentrated in the coastal city of Portsmouth vs. distributed in the many small communities along the Interstate 93 corridor adjacent to the Massachusetts border within a 72–97-km driving distance of Boston, Massachusetts.
To develop the scenarios, we sought information on possible land-cover futures from six different sources: historical trends, existing plans relating to New Hampshire’s land-cover future, surveys, existing population scenarios, key informant interviews, and subject matter experts. Scenarios were developed in parallel with information gathering, in a relatively unstructured fashion, by iteratively incorporating new details and pursuing new information sources as questions emerged from internal conversations, interviews with key informants, or input from subject matter experts.
New Hampshire’s demographic growth outpaced the rest of the northeastern United States by a significant margin through the last three decades of the 20th century and into the first few years in the 21st century (Johnson 2012). Population growth since 2007 has been slower but still greater than in other states in New England. This demographic growth has not been even across the state, however. Most of the population increase has been concentrated in the southeastern and south-central portions of the state. Corresponding with population growth, the same time intervals have seen a rapid expansion in developed land area, with most development as suburban and exurban housing (Sundquist 2012, Mockrin et al. 2013). Farmland area decreased throughout the state from the late 19th century to near the end of the 20th century, with some farms abandoned and reverting to forest and others converted to development (USDA-NASS 2012). Rates of conservation have accelerated in recent decades, primarily because of the actions of land trusts and other organizations (Meyer et al. 2014a), including broad interest in protecting the agricultural land base and expanding local food production (Donahue et al. 2014, American Farmland Trust 2016; Northeast Sustainable Agriculture Working Group http://nesawg.org). Nevertheless, most protected land is in the northern half of the state and at high elevations, where development pressure is lowest (Levesque 2010).
We relied on three reports regarding the future of New Hampshire’s and New England’s land use and land cover to inform our land-cover scenarios. The 2009 New Hampshire Climate Action Plan (New Hampshire Climate Change Policy Task Force 2009, Wake et al. 2011) represents a consensus plan developed through a process that engaged > 125 stakeholders and 29 task force members (representing a broad range of sectors and interests, including the New Hampshire House and Senate, New Hampshire state agencies, municipal government, business and industry, environmental nonprofit organizations, the forestry sector, academia, and public utilities). One of ten overarching strategies was to protect New Hampshire’s natural resources to maintain the amount of carbon sequestered by avoiding net forestland conversion and protecting existing agriculture land. In 2010, Harvard Forest published Wildlands and Woodlands: A Vision for the New England Landscape (Foster et al. 2010). The vision calls for a 50-year effort to conserve 70% of New England as forest permanently free from development, and efforts to support this vision continue. Finally, a common focus across recent New England state food system plans or strategies is in expanding food production and the agricultural land base (Vermont Sustainable Jobs Fund 2013, Metropolitan Area Planning Council 2015, New Hampshire Food Alliance 2015, Maine Food Strategy 2016). A New England Food Vision builds on the desires for expanded agricultural production in the region and describes a future in which New England produces at least half of the region’s food by 2060 via an expansion of agricultural lands from 0.8 million to 2.4–2.8 million ha (Donahue et al. 2014). Food Solutions New England’s (http://www.foodsolutionsne.org) commitment to furthering this vision has been codified in a series of annual New England food system summits beginning in 2011 and by establishing a vibrant regional network.
In 2013, New Hampshire’s nine Regional Planning Commissions sponsored a survey of 2935 New Hampshire adult residents as part of a larger project (A Granite State Future, funded by U.S. Housing and Urban Development) to engage New Hampshire citizens in a public dialogue regarding the future of their communities and state (Keirns et al. 2013). Additional input was gathered during key informant interviews with leaders of the A Granite State Future project and during a separate meeting with Commissioners from the nine Regional Planning Commissions (Table 1). More than two-thirds (70%) of respondents favor keeping future development in existing developed areas, 96% view water-quality protection as a high priority, and 76% place high priority on protecting aquatic habitats. Nearly three-quarters (74%) think that policy makers should invest more money for maintaining roads, highways, and bridges, whereas only 40% supported more funds for improving the availability of public transportation.
Population projections for New Hampshire have been generated both locally (New Hampshire Regional Planning Councils 2014) and nationally (e.g., Bierwagen et al. 2010). To represent diverging futures, we adapted the New Hampshire Office of Energy and Planning projection (2014) as a low population scenario and the A2 population scenario developed by the Integrated Climate and Land Use Scenarios used by the third National Climate Assessment for a high population scenario (Bierwagen et al. 2010). We also consulted census data for the state since 1990 as the basis for a simple linear extrapolation of current rates of population growth (U.S. Census Bureau 1993, 2013).
Key informants from 12 organizations representing five sectors related to land use and land cover (environmental nonprofits, business and industry, timber interests, public sector agencies, and academics and natural resource management consultants; Table 1) were identified based on our knowledge of the state and recommendations from individuals in leadership positions and were invited to participate in facilitated group interviews. Ten of the twelve organizations who participated in the interviews represent statewide interests, and > 150 key informants participated in these interviews (Table 1). During interviews, informants were asked two questions: (1) describe a picture of what you would like New Hampshire to look like in the future, two to four decades or more from now; and (2) describe a picture of what you expect New Hampshire to look like in the future, two to four decades or more from now. Details of our methodology are provided in Appendix 1. A majority of those interviewed favored a shift toward more compact residential development through some combination of local zoning ordinances, regulation favoring “smart growth” zoning (Addison et al. 2013), and market responses to increased demand for smaller houses and high costs of construction far from existing roads, sewers, and other infrastructure. Participants diverged on whether high or low population growth was desirable, with most public sector and industry representatives regarding higher population growth as desirable, whereas some members of environmental nonprofits expressed anxiety about the likelihood of land clearing to accommodate a growing population. A few stakeholders advocated for increased food production in the state, whereas others (representing timber and environmental interests) regarded expansion of agricultural land as a threat to natural resources. There were some stakeholders across sectors who expressed a desire for better management of forest, soil, and water to protect ecosystem services, and environmental nonprofit stakeholders in particular hoped for continued strategic land conservation.
When prompted to describe the future they expected, most stakeholders described a regulatory and zoning environment similar to their perceptions of the present day or with reduced environmental protections. The main source of differences was in the anticipated rate of population growth. Some stakeholders expected rapid population growth, continuing trends from the late 20th century, which, combined with contemporary zoning, would result in most of the state filling with residential subdivisions. Others, particularly government stakeholders, expected low growth and increased demand for infrastructure to support an aging population. Still others imagined futures between these two extremes.
Throughout the process, we engaged subject matter experts on a range of topics. The members of the NH EPSCoR Ecosystem and Society project numbered > 60 individuals from academic institutions across the state. They provided input via responses to our questions similar to those posed during key informant interviews as well as during monthly team meetings and biannual all-hands meetings. Key issues relating to forest management and existing and potential future conserved land were discussed in detail with members from the New Hampshire chapter of The Nature Conservancy and the Society for the Protection of New Hampshire Forests. We engaged Brian Donahue to help us interpret the New England Food Vision projections for future agricultural lands across New Hampshire. Demographer Ken Johnson advised us on the use of the Office of Energy and Planning population projections, which extend to the year 2040, and we subsequently obtained extrapolations to 2100 from Bob Scardamalia of RLS Demographics, the creator of the Office of Energy and Planning projection. In addition, one of the authors (CM), who originally connected with the project as a participant in key informant interviews, joined the team to provide detailed data on existing residential and commercial zoning in all 221 towns and 13 cities in New Hampshire.
Based on the diverse themes and information derived from these six sources (Table 2), we developed a suite of qualitative land-cover scenarios (narratives) that were subsequently translated into quantitative land cover, population density, and impervious cover, essentially applying the story and simulation (SAS) approach (Mallampalli et al. 2016). The major driving forces that emerged from this process were population growth, zoning, conservation, and the role of agriculture in the state. Our translation methodology was iterative but relatively simple, relying on the authors’ progressive synthesis of available information.
Based on the information we collected, we realized that it was crucial to capture a range of patterns of population distribution, with an upper limit on compact development defined by the restriction of new growth to redevelopment of previously developed land area (New Hampshire Climate Change Policy Task Force 2009). However, we were less clear on how the opposite extreme of dispersed population should be defined and distributed. Key informant interviews confirmed the importance of population distribution and growth as drivers of New Hampshire land cover and provided more specificity: in a high-population future, anticipated development was described as progressing northward from the southeast along major transportation corridors, whereas in the middle of the state, growth would be concentrated in a select few municipalities (Plymouth, Hanover, and Lebanon). In a low-population future, the location of growth would be defined by the county-based New Hampshire Regional Planning Councils (2014) projections.
Our original vision for the modeling process was to use land-cover change models derived from regression trees (De’ath and Fabricius 2000) and decision trees and based on 1996–2011 land-cover change data from the National Oceanic and Atmospheric Administration Coastal Change Analysis Program (NOAA C-CAP; NOAA 2014), emulating a scenario effort for neighboring Massachusetts (Thompson et al. 2011, 2016). Regression trees provide the advantage of being intuitive to interpret and modify and have been shown to perform with similar realism to more sophisticated models for long-term projections (Tayyebi et al. 2014). However, this approach presented some challenges when the driver of land development was local change in population. Models for the low-density development scenario were therefore structurally modified to first simulate allocation of population among municipalities using a model derived from county-level population projections but designed to be consistent with a northward progression along transportation corridors. This custom model incorporated detailed information on residential lot sizes zoned for each municipality, and eventually we settled on a cost-distance weighted gravity model (see details in Appendix 2). The original regression tree models also tended, in some cases, to produce artificial and unrealistically sharp boundaries between converted and unchanged lands, and after both internal discussion among the authors and informal external conversation with stakeholders and scholars in the broader EPSCoR team, we decided to avoid these artifacts by adjusting the scale at which regression trees were computed (e.g., subregions within New Hampshire instead of for the whole state) and using careful selection of input variables that would be commensurate with the scale of simulation models (e.g., avoiding distance variables that operate at large scales when simulating land-cover change within a municipality).
Information gathered during key informant interviews suggested that the future of land conservation would play an important role in determining the effects of development on ecosystem services, a perception supported by the history of land conservation in the state. However, it was not clear what quantity or distribution of conservation would be considered necessary to protect these services. Thus, we engaged subject-matter experts at The Nature Conservancy and the Society for the Protection of New Hampshire Forests, who had previously identified high-priority areas for conservation. GIS experts from each organization shared their maps of conservation plans and helped us determine the priority threshold that would be most relevant for a high-conservation future. At the other end of the spectrum, we internally debated whether a low-conservation future would have slow conservation, no conservation, or perhaps even a loss of conserved land relative to the present day. For this, we relied on local planning expertise and research (author CM), which suggested a natural political shift toward cluster zoning with land conservation as communities are built out, and this observation was incorporated into our final models. We also discovered that a linear extrapolation of conservation trends from the past two decades, which we had planned to use as an intermediate scenario between the two extremes, would in fact result in a larger area of conserved land by 2100 than was considered desirable by our conservation stakeholders. We believed this outcome was an unrealistic consequence of the unusually high rate of conservation from 1990 to the present (Meyer et al. 2014a). In this case, we decided that ending the extrapolated conservation in 2060 would give a more realistic result while providing similar simplicity to the general assumptions of a linear model.
Projecting future change in agricultural land area posed a particular challenge in the development of scenarios for land cover. Total land area and changes in land cover for this category were relatively small, limiting power for statistical analyses. In addition, the user and producer accuracy of Landsat-based land-cover data are particularly low for the pasture/hay land category (Wickham et al. 2010), which represents 75% of cleared farmland in the state (USDA-NASS 2012). Our colleagues at NH GRANIT assisted us by providing an enhanced version of the NOAA C-CAP rasters (Rubin and Justice, unpublished data). We were previously familiar with A New England Food Vision (Donahue et al. 2014), which presents two scenarios for agricultural expansion to increase food security in the region. Given the mix of perspectives provided during the key informant interviews, we decided to include two different compact development scenarios, one assuming dramatic agricultural expansion based on Donahue et al. (2014), and another with much more modest agricultural expansion, continuing recent trends. We consulted with Brian Donahue to determine what land area would be appropriate for New Hampshire based on A New England Food Vision, and he provided specific agricultural land areas for the year 2060. He also advised that we use the more modest of the two scenarios in A New England Food Vision, which represents a normative combination of increased food production alongside protection of forests. To keep this scenario simple, we assumed linear expansion in agricultural land area between the present day and the 2060 agricultural land provided in A New England Food Vision.
The final products of the translation process were custom land-cover change models developed for each scenario (see Appendix 2), which produced decadal rasters of land cover and population density. Maps of impervious cover were derived from population density maps based on a regression model relating contemporary impervious cover to population density. To summarize the change in impervious cover at a watershed level, the proportion of impervious cover was calculated within HUC10 watershed boundaries (USDA-NRCS et al. 2015). We used a simple threshold model to classify watersheds as degraded (> 30% impervious), impacted (> 10% impervious), or not impacted (< 10% impervious cover; Arnold and Gibbons 1996).
The six scenarios we developed fall on a continuum from dispersed development with a low value placed on shared ecosystem services to concentrated development with a high value placed on shared ecosystem services (Fig. 2A). The “Backyard Amenities” scenario (Backyard) is characterized by dispersed development with little regulation or effort to preserve ecosystem services combined with rapid population growth. The “Community Amenities” family of scenarios (Community) is characterized by concentrated development and a strong focus on the preservation of existing forest and agricultural lands. The Community family contains four scenarios, each informed by Foster et al. (2010). These differ along the axes of population growth (“Small Community” vs. “Large Community”) and degree of agricultural expansion (“Protection of Wildlands” [Wildlands] vs. “Promotion of Local Food” [Food]; Fig. 2B), giving four scenarios within the family: Small Community-Wildlands, Large Community-Wildlands, Small Community-Food, and Large Community-Food). For all scenarios in this family (and to be consistent with the recommendations of the New Hampshire Climate Change Policy Task Force ), all development occurs as redevelopment within existing developed areas. Between the Backyard scenario and the Community family is the “Linear Trends” scenario (Linear). To enhance the usefulness of these scenarios for social scientists and planners, each scenario is accompanied by general descriptions of associated economic, governmental, and cultural change (Appendix 3).
For the Backyard scenario, development rates were based on population projections, a database of the mean area zoned for future residential lots in each municipality, and historical trends in zoning changes in response to development pressure (C. Mitchell et al., unpublished manuscript). As each municipality is built out to 50% and then 62.5%, land area per lot is reduced to reflect a political shift toward increasingly compact cluster zoning. The population density of newly developed land increases based on lot size and average households of 2.5 individuals (New Hampshire Office of Energy and Planning 2015), and impervious cover is updated based on population density. For the Community family, developed land area does not increase, so the projected increase in population is allocated to developed land with existing population (assumed to be residential areas), and impervious cover is correspondingly increased to reflect the necessary redevelopment to accommodate increased densities. In addition, where population density decreases, impervious cover is permitted to decline based on reduced residential requirements. The Food scenarios assume a linear expansion of agricultural land area to meet the target of 3640 km² by the year 2060. Because this vision is designed to be compatible with forest protection, we keep agricultural land area constant after 2060. The Linear scenario extrapolates recent trends in development for each of four regions in the state: the southeast, southwest, central, and north (see Appendix 2).
Rasters for the year 2100 of the four land-cover scenarios (Fig. 3) and four population-density scenarios (Fig. 4) reveal a collection of highly distinct futures for the southeastern and central portions of the state, with relatively minor differences in the north. Impervious cover increases in southeastern and central watersheds for all population scenarios except Small Community (Fig. 5A). This change is particularly dramatic for the Backyard scenario; by 2100, the number of watersheds in the impacted category (impervious cover > 10%; Arnold and Gibbons 1996) quadrupled relative to the present day (Fig. 5B). Only 750 km² of new land is conserved for Backyard compared with 4000 km² for Linear and 4700 km² for the Community family (Fig. 6). The spatial distribution of land conservation is fragmented and haphazard in Linear and targeted and contiguous for Community.
The scenarios presented here reflect four main driving forces in New Hampshire: population growth, zoning, conservation, and the role of agriculture in the state. The result is one scenario with rapid dispersed development (Backyard), a suite of compact growth scenarios (Community), and one intermediate scenario (Linear). In the Backyard scenario, southeastern and central New Hampshire are almost entirely built out by 2100, a higher end estimate of what the state could look like if recent trends continue (see shorter term projections from New Hampshire Fish and Game 2005, Stein et al. 2009, 2010). Impervious cover increases, bringing 15 new HUC10 watersheds (USDA-NRCS et al. 2015) in southern New Hampshire into the “impacted” category by 2100 (Arnold and Gibbons 1996). In contrast, the future presented by the Community family of scenarios maintains all of the undeveloped land of the present day, aligned with the goals set out by the New Hampshire Climate Change Policy Task Force (2009) and reflecting information gathered in the 2013 survey (Keirns et al. 2013) and key informant interviews. The Wildlands and Food scenarios within the Community family present two contrasting futures for rural land use within the state: one similar to the present day; and one in which the landscape is transformed to support a large portion of the food demand within the region, resulting in a dramatic expansion in farmland in areas suitable for agriculture (i.e., central New Hampshire, coastal plain, and along the Connecticut River valley), a future that in some ways resembles land cover in the mid- to late 1800s. The Small Community scenarios represent little change in population or impervious cover, whereas in the Large Community scenarios, many southeastern and central population centers experience growing high-density populations while rural spaces are protected. The Small Community scenarios are similar to the Large Community scenarios but without any significant population growth, requiring little change to municipal infrastructure from the present day.
Our modeling priority was to produce maps that would resonate as plausible to New Hampshire stakeholders. To achieve this, we used an iterative SAS approach to develop and calibrate custom models based on diverse local sources of information (Table 2). Key knowledge incorporated into the scenarios included details of contemporary zoning (C. Mitchell et al., unpublished manuscript), existing conservation plans (The Nature Conservancy New Hampshire Chapter, unpublished data; Society for the Protection of New Hampshire Forests, unpublished data), the New Hampshire Climate Action Plan (New Hampshire Climate Change Policy Task Force 2009), and a normative vision for the future agricultural land area in the state (Donahue et al. 2014). For two scenarios, we also used a population projection from the state’s Office of Energy and Planning (New Hampshire Regional Planning Councils 2014). Because that projection represents a low population future and some stakeholders expected a high population future, we also used national projections from the U.S. Environmental Protection Agency’s Integrated Climate and Land-Use Scenarios (EPA ICLUS) report (Bierwagen et al. 2010). The final result was a suite of six distinctly local scenarios relevant in different ways to a diversity of local stakeholders.
We worked to limit the number of scenarios based on the longstanding recommendation that scenario development should be restricted to three or four scenarios (Schwartz 1991, Peterson et al. 2003). However, the diversity of interests among stakeholders prompted us to expand the Community scenario into a family of scenarios to better match the interests of different stakeholders. Nevertheless, we recommend that users of these scenarios identify two to four scenarios most relevant for their purposes. For example, researchers and planners interested in the relative importance of population and population density might focus on the Backyard, Large Community-Wildlands, and Small Community-Wildlands scenarios, whereas food systems professionals might focus on the capacity of New Hampshire to feed a growing population in the Backyard, Large Community-Food, and Large Community-Wildlands scenarios.
An important methodological decision in our approach was to bound the scope of our models to the specific driving variables identified by stakeholders, rather than to attempt to develop an integrated model of all aspects of the system. Thus, the model for Backyard explicitly includes zoning and population, but we did not model the economic mechanisms that would drive the spatial distribution of population or development. Likewise, for the Food scenarios, we modeled a simple trajectory consistent with Donahue et al. (2014) and, because our scenarios focused on New Hampshire and not the region as a whole, we did not assume that regional demand would necessarily differ between the two population variants. We combined statistical models, extrapolation, expert knowledge, normative vision, and zoning data to simulate each scenario but did not attempt to predict scenario differences with a single model. Our approach differs, for example, from econometric models, which assess the potential consequences of specific policy decisions on land-cover change based on current knowledge of markets (e.g., Radeloff et al. 2012), and Bayesian belief networks, which rely on the ability of expert stakeholders to characterize the interrelationships among driving variables (McCloskey et al. 2011, Meyer et al. 2014b, Carpenter et al. 2015). Although the resulting models are less suitable for mechanistic analysis of the interdependence of the factors driving governance and land-use choices than other approaches, we believe that our models satisfied our primary goal of developing divergent scenarios consistent with stakeholder intuitions. The scenarios also represent particular divergence with regard to population distribution as compared to other scenario sets (e.g., Bierwagen et al. 2010).
In part to maintain the simplicity of our models and in part as a result of the limited resources available for a small-scale scenario effort, we borrowed in places from the approach used in the development of the representative concentration pathways, which are the basis for current scenarios in climate modeling, and which describe trajectories of changing emissions without modeling the mechanisms that produce each pathway (Moss et al. 2008, 2010). Notably, we used published projections for our high (Bierwagen et al. 2010) and low (New Hampshire Regional Planning Councils 2014) population scenarios, but we consider the magnitude of change to be more important than the underlying demographic variables such as age structure, fertility, or migration. For instance, although the EPA ICLUS A2 population projection assumes high fertility rates and high domestic migration, our Backyard and Large Community scenarios could be equally useful for high-population futures produced by other mechanisms. Such growth might result, for example, if domestic populations shift in response to changing climate. New England is expected to receive slightly increased precipitation under most climate change scenarios (Hayhoe et al. 2007, Melillo et al. 2014), whereas it is generally accepted that the recent droughts in the U.S. southwest will continue (Walsh et al. 2014). Migration of even a small percentage of U.S. residents from drier areas to wetter areas over the coming decades could produce a dramatic increase in the state’s population.
Our experience in scenario development exemplifies the challenges of small-scale scenario development with limited resources. Our interdisciplinary team included a mix of natural scientists and social scientists but did not include researchers with preexisting expertise in scenario development or with specialized knowledge of statistical and engagement tools that might have aided in systematic translation of stakeholder perspectives into internally consistent models (Mallampalli et al. 2016). For example, use of fuzzy cognitive maps or Bayesian networks could have produced more immediate translation from stakeholder perspectives into quantitative scenarios (Mallampalli et al. 2016). Our project would likely have been enhanced by the inclusion of additional academic disciplines on the team such as a demographer who could provide more customized population projections or an economist with knowledge of land valuation, supply chains, and distribution networks. In addition, although we endeavored to engage stakeholders from a balanced collection of sectors in the state, our reliance on snowball sampling could have introduced selection or gatekeeper bias (Bonevski et al. 2014), preventing us from hearing the perspectives of important groups not represented in our professional networks. For example, a clear area for improvement in the scenarios would have been more sophisticated modeling of the timing and spatial distribution of increased farmland area in the Food scenario, which might have been enhanced by more engagement of experts in food production or distribution or, alternatively, by the inclusion of an economist on the team.
The transparency of our process also would have been improved if we had used more systematic methods to collect data from key informant interviews, including recording, transcription, and quantitative analysis of responses. It is therefore not possible to delve further into the interviews to identify patterns among the data and the different informants, which could help to predict whether there might be identifiable triggers associated with the different futures. Finally, our timeline did not allow for iterative engagement of the stakeholders that we interviewed to invite them to critique our preliminary scenarios, as is recommended in the classic SAS approach (Alcamo et al. 2006). This method could have provided significant refinement of the details of the different population distributions and the accompanying economic, governmental-regulatory, and cultural change narratives.
Overall, our scenario development process strongly centered on the decisions and intuitions of the research team and assigned priority to the perspectives of stakeholders with close working relationships with the authors. While this can enhance collaboration between researchers and specific technical experts outside of academia (Pulver and VanDeveer 2009), a clear criticism of this approach is the bias it introduces, amplifying the voices of some stakeholders over others. On the flip side, the use of engagement methods that require minimal investment by stakeholders may have increased the breadth of perspectives included, and reliance on existing collaborative relationships may have enabled us to include more specific detail than might have been included if we insisted on a more statistical representation of stakeholders.
In addition to the use of these scenarios in the ecosystem modeling efforts that combine land-cover scenarios with high and low emissions climate scenarios (Samal et al. 2017) and an investigation of the role of imagined futures in shaping the value placed on ecosystem services (Borsuk et al., unpublished manuscript), they are also informing conversations among municipal planners. Based on a 2016 stakeholder workshop, high-resolution maps have been generated to inspire further discourse (http://ddc-landcover.sr.unh.edu/). Future research will build on these projects to further assess dynamic changes to ecosystem services in response to climate and land-cover change. An important area for refinement of the scenarios will be spatially explicit modeling of land use and land management, particularly for agriculture and managed forests.
We thank Barbara Wauchope for helping with key informant interviews, Fay Rubin and David Justice for developing the base maps of land cover and answering questions about NH GRANIT data sets, Joshua Plisinski for processing the elevation and slope maps, Robert Spoerl for conservation dates from the NH DRED database, Chris Wells and Will Abbot for providing the Society for the Protection of New Hampshire Forests conservation plans, Peter Steckler for providing The Nature Conservancy conservation plans, and Bob Scardamalia for extrapolating the New Hampshire Office of Energy and Planning population projection. We thank Jennifer Wilhelm and Brian Donahue for reviewing ideas for the agricultural expansion scenarios, David Lutz and Ross Jones for information on forest management in the White Mountain National Forest, and Shan Zuidema for ideas on modeling impervious cover. We thank Michele Dillon for friendly review and three anonymous reviewers who provided valuable critiques during the revision process. Finally, we thank our numerous stakeholders for their contributions to the development of our scenarios. This research was supported by a grant from the National Science Foundation (EPS-1101245).
Addison, C., S. Zhang, and B. Coomes. 2013. Smart growth and housing affordability: a review of regulatory mechanisms and planning practices. Journal of Planning Literature 28(3):215-257. http://dx.doi.org/10.1177/0885412212471563
Alcamo, J. 2008. The SAS approach: combining qualitative and quantitative knowledge in environmental scenarios. Pages 123-150 in J. Alcamo, editor. Environmental futures: the practice of environmental scenario analysis. Elsevier, Amsterdam, The Netherlands. http://dx.doi.org/10.1016/S1574-101X(08)00406-7
Alcamo, J., K. Kok, G. Busch, J. A. Priess, B. Eickhout, M. Rounsevell, D. S. Rothman, and M. Heistermann. 2006. Searching for the future of land: scenarios from the local to global scale. Pages 137-155 in E. F. Lambin and H. J. Geist, editors. Land-use and Land-cover change. Springer, Berlin, Germany. http://dx.doi.org/10.1007/3-540-32202-7_6
Alcamo, J., R. Schaldach, J. Koch, C. Kölking, D. Lapola, and J. Priess. 2011. Evaluation of an integrated land use change model including a scenario analysis of land use change for continental Africa. Environmental Modelling and Software 26(8):1017-1027. http://dx.doi.org/10.1016/j.envsoft.2011.03.002
American Farmland Trust. 2016. Mission and history. American Farmland Trust, Washington, D.C., USA. [online] URL: https://www.farmland.org/mission-history
Arnold, C. L. Jr., and C. J. Gibbons. 1996. Impervious surface coverage: the emergence of a key environmental indicator. Journal of the American Planning Association 62(2):243-258. http://dx.doi.org/10.1080/01944369608975688
Bierwagen, B. G., D. M. Theobald, C. R. Pyke, A. Choate, P. Groth, J. V. Thomas, and P. Morefield. 2010. National housing and impervious surface scenarios for integrated climate impact assessments. Proceedings of the National Academy of Sciences 107(49):20887-20892. http://dx.doi.org/10.1073/pnas.1002096107
Bonevski, B., M. Randell, C. Paul, K. Chapman, L. Twyman, J. Bryant, I. Brozek, and C. Hughes. 2014. Reaching the hard-to-reach: a systematic review of strategies for improving health and medical research with socially disadvantaged groups. BMC Medical Research Methodology 14:42. http://dx.doi.org/10.1186/1471-2288-14-42
Carpenter, S. R., E. M. Bennett, and G. D. Peterson. 2006. Scenarios for ecosystem services: an overview. Ecology and Society 11(1):29. http://dx.doi.org/10.5751/ES-01610-110129
Carpenter, S. R., E. G. Booth, S. Gillon, C. J. Kucharik, S. Loheide, A. S. Mase, M. Motew, J. Qiu, A. R. Rissman, J. Seifert, E. Soylu, M. Turner, and C. B. Wardropper. 2015. Plausible futures of a social-ecological system: Yahara watershed, Wisconsin, USA. Ecology and Society 20(2):10. http://dx.doi.org/10.5751/ES-07433-200210
Conroy, R. G., and R. Ober, editors. 2001. People and place: Society for the Protection of New Hampshire Forests: the first 100 years. Society for the Protection of New Hampshire Forests, Concord, New Hampshire, USA.
De’ath, G., and K. E. Fabricius. 2000. Classification and regression trees: a powerful yet simple technique for ecological data analysis. Ecology 81(11):3178-3192. http://dx.doi.org/10.1890/0012-9658(2000)081[3178:CARTAP]2.0.CO;2
Donahue, B., J. Burke, M. Anderson, A. Beal, T. Kelly, M. Lapping, H. Ramer, R. Libby, and L. Berlin. 2014. A New England food vision. Food Solutions New England, University of New Hampshire, Durham, New Hampshire, USA. [online] URL: http://www.foodsolutionsne.org/sites/default/files/LowResNEFV_0.pdf
Drummond, M. A., and T. R. Loveland. 2010. Land-use pressure and a transition to forest-cover loss in the eastern United States. BioScience 60(4):286-298. http://dx.doi.org/10.1525/bio.2010.60.4.7
Foster, D., B. M. Donahue, D. B. Kittredge, K. F. Lambert, M. L. Hunter, B. R. Hall, L. C. Irland, R. J. Lilieholm, D. A. Orwig, A. W. D’Amato, E. A. Colburn, J. R. Thompson, J. N. Levitt, A. M. Ellison, W. S. Keeton, J. D. Aber, C. V. Cogbill, C. T. Driscoll, T. J. Fahey, and C. M. Hart. 2010. Wildlands and woodlands: a vision for the New England landscape. Harvard University Press, Cambridge, Massachusetts, USA. [online] URL: http://www.wildlandsandwoodlands.org/sites/default/files/Wildlands%20and%20Woodlands%20New%20England.pdf
Gittell, R., and J. Orcutt. 2012. Science and technology plan: shaping New Hampshire’s economic future. University of New Hampshire, Durham, New Hampshire, USA. [online] URL: https://www.unh.edu/research/sites/www.unh.edu.research/files/docs/NH_EPSCoR/NH_Science_Technology_Plan_2011.pdf
Hayhoe, K., C. P. Wake, T. G. Huntington, L. Luo, M. D. Schwartz, J. Sheffield, E. Wood, B. Anderson, J. Bradbury, A. DeGaetano, T. J. Troy, and D. Wolfe. 2007. Past and future changes in climate and hydrological indicators in the US Northeast. Climate Dynamics 28(4):381-407. http://dx.doi.org/10.1007/s00382-006-0187-8
Jeon, S. B., P. Olofsson, and C. E. Woodcock. 2014. Land use change in New England: a reversal of the forest transition. Journal of Land Use Science 9(1):105-130. http://dx.doi.org/10.1080/1747423X.2012.754962
Johnson, K. M. 2012. New Hampshire demographic trends in the twenty-first century. Carsey School of Public Policy, University of New Hampshire, Durham, New Hampshire, USA. [online] URL: http://scholars.unh.edu/cgi/viewcontent.cgi?article=1163&context=carsey
Keirns, T. A., Z. Azem, and A. E. Smith. 2013. NH Regional Planning Commissions: a Granite State future: 2013 statewide survey. Survey Center, University of New Hampshire, Durham, New Hampshire, USA. [online] URL: http://www.granitestatefuture.org/files/1413/8023/1024/RPC_Statewide_Report_FINAL.pdf
Levesque, C. A. 2010. New Hampshire statewide forest resources assessment - 2010: important data and information about New Hampshire’s forests. New Hampshire Department of Resources and Economic Development Division of Forests and Lands, Concord, New Hampshire, USA. [online] URL: http://www.nhdfl.org/library/pdf/Planning/NH%20Statewide%20Assessment%202010%20update.pdf
Liu, J., T. Dietz, S. R. Carpenter, M. Alberti, C. Folke, E. Moran, A. N. Pell, P. Deadman, T. Kratz, J. Lubchenco, E. Ostrom, Z. Ouyang, W. Provencher, C. L. Redman, S. H. Schneider, and W. W. Taylor. 2007. Complexity of coupled human and natural systems. Science 317(5844):1513-1516. http://dx.doi.org/10.1126/science.1144004
Maine Food Strategy. 2016. The Maine food strategy framework: a tool for advancing Maine’s food system. Maine Food Strategy, Maine, USA. [online] URL: http://mainefoodstrategy.org/wp-content/uploads/2016/06/Maine-Food-Strategy-Framework_final.pdf
Mallampalli, V. R., G. Mavrommati, J. Thompson, M. Duveneck, S. Meyer, A. Ligmann-Zielinska, C. Gottschalk Druschke, K. Hychka, M. A. Kenney, K. Kok, and M. E. Borsuk. 2016. Methods for translating narrative scenarios into quantitative assessments of land use change. Environmental Modelling and Software 82:7-20. http://dx.doi.org/10.1016/j.envsoft.2016.04.011
Mavrommati, G., M. E. Borsuk, and R. B. Howarth. 2016. A novel deliberative multicriteria evaluation approach to ecosystem service valuation. Ecology and Society 22(2):39. https://doi.org/10.5751/ES-09105-220239
McCloskey, J. T., R. J. Lilieholm, and C. Cronan. 2011. Using Bayesian belief networks to identify potential compatibilities and conflicts between development and landscape conservation. Landscape and Urban Planning 101(2):190-203. http://dx.doi.org/10.1016/j.landurbplan.2011.02.011
Melillo, J. M., T. C. Richmond, and G. W. Yohe, editors. 2014. Climate change impacts in the United States: the third National Climate Assessment. U.S. Government Printing Office, Washington, D.C., USA. http://dx.doi.org/10.7930/J0Z31WJ2
Metropolitan Area Planning Council, Franklin Regional Council of Governments, Pioneer Valley Planning Commission, and Massachusetts Workforce Alliance. 2015. Massachusetts local food action plan. Massachusetts Food System Collaborative, Massachusetts, USA. [online] URL: http://mafoodsystem.org/plan/
Meyer, S. R., C. S. Cronan, R. J. Lilieholm, M. L. Johnson, and D. R. Foster. 2014a. Land conservation in northern New England: historic trends and alternative conservation futures. Biological Conservation 174:152-160. https://doi.org/10.1016/j.biocon.2014.03.016
Meyer, S. R., M. L. Johnson, R. J. Lilieholm, and C. S. Cronan. 2014b. Development of a stakeholder-driven spatial modeling framework for strategic landscape planning using Bayesian networks across two urban-rural gradients in Maine, USA. Ecological Modelling 291:42-57. https://doi.org/10.1016/j.ecolmodel.2014.06.023
Millennium Ecosystem Assessment. 2005. Ecosystems and human well-being: current state and trends. Island Press, Washington, D.C., USA. [online] URL: https://www.millenniumassessment.org/en/Condition.html
Mockrin, M. H., S. I. Stewart, V. C. Radeloff, R. B. Hammer, and K. M. Johnson. 2013. Spatial and temporal residential density patterns from 1940 to 2000 in and around the northern forest of the northeastern United States. Population and Environment 34(3):400-419. https://doi.org/10.1007/s11111-012-0165-5
Moss, R., M. Babiker, S. Brinkman, E. Calvo, T. Carter, J. Edmonds, I. Elgizouli, S. Emori, L. Erda, K. Hibbard, R. Jones, M. Kainuma, J. Kelleher, J.-F. Lamarque, M. Manning, B. Matthews, J. Meehl, L. Meyer, J. Mitchell, N. Nakicenovic, B. O’Neill, R. Pichs, K. Riahi, S. Rose, P. Runci, R. Stouffer, D. van Vuuren, J. Weyant, T. Wilbanks, J. P. van Ypersele, and M. Zurek. 2008. Towards new scenarios for analysis of emissions, climate change, impacts, and response strategies. Technical summary. Intergovernmental Panel on Climate Change, Geneva, Switzerland. [online] URL: https://www.ipcc.ch/pdf/supporting-material/expert-meeting-ts-scenarios.pdf
Moss, R. H., J. A. Edmonds, K. A. Hibbard, M. R. Manning, S. K. Rose, D. P. van Vuuren, T. R. Carter, S. Emori, M. Kainuma, T. Kram, G. A. Meehl, J. F. B. Mitchell, N. Nakicenovic, K. Riahi, S. J. Smith, R. J. Stouffer, A. M. Thomson, J. P. Weyant, and T. J. Wilbanks. 2010. The next generation of scenarios for climate change research and assessment. Nature 463(7282):747-756. http://dx.doi.org/10.1038/nature08823
Nakicenovic, N., J. Alcamo, G. Davis, B. de Vries, J. Fenhann, S. Gaffin, K. Gregory, A. Grübler, T. Y. Jung, T. Kram, E. L. La Rovere, L. Michaelis, S. Mori, T. Morita, W. Pepper, H. Pitcher, L. Price, K. Riahi, A. Roehrl, H.-H. Rogner, A. Sankovski, M. Schlesinger, P. Shukla, S. Smith, R. Swart, S. van Rooijen, N. Victor, and Z. Dadi. 2000. Special report on emissions scenarios. A special report of Working Group III of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK. [online] URL: https://www.ipcc.ch/pdf/special-reports/emissions_scenarios.pdf
National Oceanic and Atmospheric Administration (NOAA). 2014. C-CAP New Hampshire 1996–2010-era land cover change. Coastal Change Analysis Program Regional Land Cover, NOAA Office for Coastal Management, Charleston, South Carolina, USA. [online] URL: https://www.coast.noaa.gov/ccapftp/#/
New Hampshire Climate Change Policy Task Force. 2009. The New Hampshire climate action plan: a plan for New Hampshire’s energy, environmental and economic development future. New Hampshire Department of Environmental Services, Concord, New Hampshire, USA. [online] URL: http://des.nh.gov/organization/divisions/air/tsb/tps/climate/action_plan/nh_climate_action_plan.htm
New Hampshire Fish and Game. 2005. New Hampshire Wildlife Action Plan. New Hampshire Fish and Game, Concord, New Hampshire, USA. [online] URL: http://www.wildlife.state.nh.us/wildlife/wap.html
New Hampshire Food Alliance. 2015. Farm, fish, and food enterprise viability in New Hampshire. New Hampshire Food Alliance, Durham, New Hampshire, USA. [online] URL: http://www.nhfoodalliance.com/sites/default/files/NHFA%20VI%20Revised-Final%20Edit.pdf
New Hampshire Office of Energy and Planning. 2015. 2014 population estimates of New Hampshire cties and towns. New Hampshire Office of Energy and Planning, Concord, New Hampshire, USA. [online] URL: http://www.nh.gov/oep/data-center/documents/population-estimates-2014.pdf
New Hampshire Regional Planning Councils. 2014. County population projections, 2013: by age and sex. RLS Demographics, Rensselaerville, New York, USA. [online] URL: http://regionalplan.uvlsrpc.org/files/4613/8780/6428/Projections_FinalReport.pdf
NH GRANIT. 2016. New Hampshire’s statewide GIS clearinghouse. University of New Hampshire, Durham, New Hampshire, USA. [online] URL: http://granit.unh.edu/
Peterson, G. D., G. S. Cumming, and S. R. Carpenter. 2003. Scenario planning: a tool for conservation in an uncertain world. Conservation Biology 17(2):358-366. http://dx.doi.org/10.1046/j.1523-1739.2003.01491.x
Pfeifer, C., M. P. W. Sonneveld, and J. J. Stoorvogel. 2012. Farmers’ contribution to landscape services in the Netherlands under different rural development scenarios. Journal of Environmental Management 111:96-105. http://dx.doi.org/10.1016/j.jenvman.2012.06.019
Polasky, S., S. R. Carpenter, C. Folke, and B. Keeler. 2011. Decision-making under great uncertainty: environmental management in an era of global change. Trends in Ecology and Evolution 26(8):398-404. http://dx.doi.org/10.1016/j.tree.2011.04.007
Pontius, R. G. Jr., W. Boersma, J.-C. Castella, K. Clarke, T. de Nijs, C. Dietzel, Z. Duan, E. Fotsing, N. Goldstein, K. Kok, E. Koomen, C. D. Lippitt, W. McConnell, A. Mohd Sood, B. Pijanowski, S. Pithadia, S. Sweeney, T. N. Trung, A. T. Veldkamp, and P. H. Verburg. 2008. Comparing the input, output, and validation maps for several models of land change. Annals of Regional Science 42(1):11-37. http://dx.doi.org/10.1007/s00168-007-0138-2
Pontius, R. G., and J. Spencer. 2005. Uncertainty in extrapolations of predictive land-change models. Environment and Planning B: Planning and Design 32(2):211-230. https://doi.org/10.1068/b31152
Pulver, S., and S. D. VanDeveer. 2009. “Thinking about tomorrows”: scenarios, global environmental politics, and social science scholarship. Global Environmental Politics 9(2):1-13. http://dx.doi.org/10.1162/glep.2009.9.2.1
Radeloff, V. C., E. Nelson, A. J. Plantinga, D. J. Lewis, D. Helmers, J. J. Lawler, J. C. Withey, F. Beaudry, S. Martinuzzi, V. Butsic, E. Lonsdorf, D. White, and S. Polasky. 2012. Economic-based projections of future land use in the conterminous United States under alternative economic policy scenarios. Ecological Applications 22(3):1036-1049. http://dx.doi.org/10.1890/11-0306.1
Rosenthal, A., G. Verutes, E. McKenzie, K. K. Arkema, N. Bhagabati, L. L. Bremer, N. Olwero, and A. L. Vogl. 2015. Process matters: a framework for conducting decision-relevant assessments of ecosystem services. International Journal of Biodiversity Science, Ecosystem Services and Management 11(3):190-204. http://dx.doi.org/10.1080/21513732.2014.966149
Samal, N. R., W. Wollheim, S. Zuidema, R. Stewart, Z. Zhou, M. M. Mineau, M. Borsuk, K. H. Gardner, S. Glidden, T. Huang, D. Lutz, G. Mavrommati, A. M. Thorn, C. P. Wake, and M. Huber. 2017. A coupled terrestrial and aquatic biogeophysical model of the Upper Merrimack River watershed, New Hampshire, to inform ecosystem services evaluation and management under climate and land-cover change. Ecology and Society 22(4):18. https://doi.org/10.5751/ES-09662-220418
Schmitt Olabisi, L. K., A. R. Kapuscinski, K. A. Johnson, P. B. Reich, B. Stenquist, and K. J. Draeger. 2010. Using scenario visioning and participatory system dynamics modeling to investigate the future: lessons from Minnesota 2050. Sustainability 2(8):2686-2706. http://dx.doi.org/10.3390/su2082686
Scholz, A. M. 2011. Consequences of changing climate and land use to 100-year flooding in the Lamprey River watershed of New Hampshire. Thesis. University of New Hampshire, Durham, New Hampshire, USA. [online] URL: https://www.unh.edu/unhsc/sites/unh.edu.unhsc/files/docs/Scholz_Thesis_2011.pdf
Schwartz, P. 1991. The art of the long view: planning for the future in an uncertain world. First edition. Doubleday, New York, New York, USA.
Shands, W. E. 1992. The lands nobody wanted: the legacy of the eastern National Forests. Pages 19-44 in H. K. Steen, editor. Origins of the national forests: a centennial symposium. Forest History Society, Durham, North Carolina, USA.
Sohl, T. L., M. C. Wimberly, V. C. Radeloff, D. M. Theobald, and B. M. Sleeter. 2016. Divergent projections of future land use in the United States arising from different models and scenarios. Ecological Modelling 337:281-297. http://dx.doi.org/10.1016/j.ecolmodel.2016.07.016
Stein, S. M., M. A. Carr, R. E. McRoberts, L. G. Mahal, and S. J. Comas. 2010. Threats to at-risk species in America’s private forests: a forests on the edge report. General Technical Report NRS-73. U.S. Department of Agriculture Forest Service, Northern Research Station, Newtown Square, Pennsylvania, USA. https://doi.org/10.2737/NRS-GTR-73
Stein, S. M., R. E. McRoberts, L. G. Mahal, M. A. Carr, R. J. Alig, S. J. Comas, D. M. Theobald, and A. Cundiff. 2009. Private forests, public benefits: increased housing density and other pressures on private forests contributions. General Technical Report PNW-GTR-795. U.S. Department of Agriculture Forest Service, Pacific Northwest Research Station, Portland, Oregon, USA. http://dx.doi.org/10.2737/PNW-GTR-795
Sundquist, D. 2012. New Hampshire’s changing landscape. Forest Notes 269:4-7. [online] URL: https://forestsociety.org/sites/default/files/fn20121.pdf
Swart, R. J., P. Raskin, and J. Robinson. 2004. The problem of the future: sustainability science and scenario analysis. Global Environmental Change 14(2):137-146. http://dx.doi.org/10.1016/j.gloenvcha.2003.10.002
Tayyebi, A., B. C. Pijanowski, M. Linderman, and C. Gratton. 2014. Comparing three global parametric and local non-parametric models to simulate land use change in diverse areas of the world. Environmental Modelling and Software 59:202-221. http://dx.doi.org/10.1016/j.envsoft.2014.05.022
Thompson, J. R., D. N. Carpenter, C. V. Cogbill, and D. R. Foster. 2013. Four centuries of change in northeastern United States forests. Plos One 8(9):e72540. http://dx.doi.org/10.1371/journal.pone.0072540
Thompson, J., K. Fallon Lambert, D. Foster, M. Blumstein, E. Broadbent, and A. Almeyda Zambrano. 2014. Changes to the land: four scenarios for the future of the Massachusetts landscape. Harvard Forest, Harvard University, Petersham, Massachusetts, USA. [online] URL: http://harvardforest.fas.harvard.edu/changes-to-the-land
Thompson, J. R., D. R. Foster, R. Scheller, and D. Kittredge. 2011. The influence of land use and climate change on forest biomass and composition in Massachusetts, USA. Ecological Applications 21(7):2425-2444. http://dx.doi.org/10.1890/10-2383.1
Thompson, J. R., K. F. Lambert, D. R. Foster, E. N. Broadbent, M. Blumstein, A. M. Almeyda Zambrano, and Y. Fan. 2016. The consequences of four land-use scenarios for forest ecosystems and the services they provide. Ecosphere 7(10):e01469. http://dx.doi.org/10.1002/ecs2.1469
U.S. Census Bureau. 1993. 1990 census of population and housing: population and housing unit counts. U.S. Census Bureau, Washington, D.C., USA. [online] URL: https://www.census.gov/population/www/censusdata/90pubs/cph-2.html
U.S. Census Bureau. 2013. Summary population and housing characteristics: 2010. Report CPH-1. U.S. Census Bureau, Washington, D.C., USA. [online] URL: https://census.gov/library/publications/2012/dec/cph-1.html
U.S. Department of Agriculture Forest Service (USDA Forest Service). 2005. Executive summary of the final environmental impact statement for the land and resource management plan: White Mountain National Forest. [online] URL: https://www.fs.usda.gov/Internet/FSE_DOCUMENTS/stelprdb5200020.pdf
U.S. Department of Agriculture National Agricultural Statistics Service (USDA-NASS). 2012. 2012 census of agriculture: New Hampshire: state and county data. U.S. Department of Agriculture National Agricultural Statistics Service, Washington, D.C., USA. [online] URL: https://www.agcensus.usda.gov/Publications/2012/Full_Report/Volume_1,_Chapter_1_State_Level/New_Hampshire/nhv1.pdf
U.S. Department of Agriculture National Resources Conservation Service (USDA-NRCS), U.S. Geological Survey, and U.S. Environmental Protection Agency. 2015. Watershed boundary dataset for HUC10, New Hampshire. U.S. Department of Agriculture National Resources Conservation Service, Washington, D.C., USA. [online] URL: http://datagateway.nrcs.usda.gov
U.S. Environmental Protection Agency (USEPA). 2009. Land-use scenarios: national-scale housing-density scenarios consistent with climate change storylines (final report). USEPA, Global Change Research Program, National Center for Environmental Assessment, Washington, D.C., USA. [online] URL: https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=203458
van Vliet, J., A. K. Bregt, D. G. Brown, H. van Delden, S. Heckbert, and P. H. Verburg. 2016. A review of current calibration and validation practices in land-change modeling. Environmental Modelling and Software 82:174-182. http://dx.doi.org/10.1016/j.envsoft.2016.04.017
Verburg, P. H., A. Tabeau, and E. Hatna. 2013. Assessing spatial uncertainties of land allocation using a scenario approach and sensitivity analysis: a study for land use in Europe. Journal of Environmental Management 127:S132-S144. http://dx.doi.org/10.1016/j.jenvman.2012.08.038
Vermont Sustainable Jobs Fund. 2013. Farm to plate strategic plan. Vermont Sustainable Jobs Fund, Montpelier, Vermont, USA. [online] URL: http://www.vtfarmtoplate.com/plan/
Vitousek, P. M., H. A. Mooney, J. Lubchenco, and J. M. Melillo. 1997. Human domination of Earth’s ecosystems. Science 277(5325):494-499. http://dx.doi.org/10.1126/science.277.5325.494
Wake, C. P., M. Frades, M. Magnusson, R. Gittell, C. Skoglund, J. Morin, and G. Hurtt. 2011. Collaborative and transparent: production of decision-relevant information for New Hampshire’s climate action plan. Northeastern Geographer 3:1-26. [online] URL: http://aagnestval.wpengine.com/wp-content/uploads/2015/09/Collaborative-Transparent.pdf
Walsh, J., D. Wuebbles, K. Hayhoe, J. Kossin, K. Kunkel, G. Stephens, P. Thorne, R. Vose, M. Wehner, J. Willis, D. Anderson, S. Doney, R. Feely, P. Hennon, V. Kharin, T. Knutson, F. Landerer, T. Lenton, J. Kennedy, and R. Somerville. 2014. Our changing climate. Pages 19-67 in J. M. Melillo, T. C. Richmond, and G. W. Yohe, editors. Climate change impacts in the United States: the third national climate assessment. U.S. Global Change Research Program, Washington, D.C., USA. http://dx.doi.org/10.7930/J0Z31WJ2
Wear, D. N. 2011. Forecasts of county-level land uses under three future scenarios: a technical document supporting the Forest Service 2010 RPA assessment. General Technical Report SRS-141. U.S. Department of Agriculture Forest Service, Southern Research Station, Ashville, North Carolina, USA. [online] URL: https://www.fs.usda.gov/treesearch/pubs/39404
Wickham, J. D., S. V. Stehman, J. A. Fry, J. H. Smith, and C. G. Homer. 2010. Thematic accuracy of the NLCD 2001 land cover for the conterminous United States. Remote Sensing of Environment 114(6):1286-1296. https://doi.org/10.1016/j.rse.2010.01.018