The role of the Conservation Reserve Program in controlling rural ...

Journal of Rural Studies 17 (2001) 323}332

The role of the Conservation Reserve Program in controlling rural residential development Jerry Johnson *, Bruce Maxwell Department of Political Science, Wilson Hall, Montana State University, Bozeman, MT 59717, USA Land Resources and Environmental Sciences, Montana State University, MT, USA

Abstract Rural population growth in the form of residential development frequently results in the loss of agricultural productive land as well as loss of adjacent open space that often characterizes rural communities. A land-use prediction model was used to determine what in#uence the USDA Conservation Reserve Program (CRP) may have on urban sprawl and rural community sustainability. The model demonstrated that the projected mean rural residential growth rate was almost half the growth rate with CRP as compared to without CRP in the local land management mix. In addition, ecosystem integrity on the land surrounding a rural community was sharply increased with the introduction of CRP. However, community economics and subsequent social character of the community may have been signi"cantly impacted by CRP. In order to partially mitigate CRP-induced community impacts we propose future CRP guidelines support the establishment of within-production "eld scale ecological refuges. These refuges would satisfy the conservation requirements of the program, return a level of traditional agricultural production to the land management mix, and provide the adjacent community with aesthetic and recreational amenities that are frequently associated with modern rural economies. 2001 Elsevier Science Ltd. All rights reserved.

The transition of agricultural land from production to rural housing is a functional marketplace. On the demand side, a seemingly unlimited number of willing buyers exist for rural lands. On the other side, a ready supply of farmers and ranchers are willing sellers of agricultural land. It follows then that in addition to regulation on the demand side of the market, an essential issue for those concerned with rural sprawl must be to consider policy that constrains supply of agricultural land entering the market for rural subdivision. This paper shows how a US Federal land-use policy * speci"cally, the CRP, can play a role in slowing and even preventing sprawl on the rural countryside in regions where population is growing (e.g. the mountain valleys of the west). Using a land-use prediction model designed for a mixed agricultural/residential landscape in southwest Montana, CRP is shown to preserve and protect the vital ecological functions sustained by the preservation of agricultural land as well as curtailing the supply of land for rural residential development. With CRP as part of the land management mosaic, the area was projected to * Corresponding author. Tel.: #1-406-994-5164; fax: #1-406-9946692. E-mail address: [email protected] (J. Johnson).

have an average residential land-use growth rate of almost half that of areas without CRP enrollment.

1. Introduction Communities throughout the western United States are experiencing a population surge that may surpass the impact of the "rst western migrations. Nationally, more land has been developed for housing between 1992 and 1997, at a rate of over three million acres/yr, than the entire decade before 1992 (Stevens, 1999). In the Mountain states of the western US new housing grew at 20.6% between 1990 and 1998, that was double the national average of 10% (US Bureau of the Census, 1999). Seven of the 10 fastest-growing states in the nation lie along the axis of the Rocky Mountains; 67% of counties in the region continue to grow at rates faster than the national average (US Bureau of the Census, 1995). Much of the growth is located in agriculturally productive river valleys found throughout the Rockies. Population growth in the agricultural countryside poses two particular areas of anxiety to the region. First, as the demand for homesites consumes productive farmland, rural development brings with it a host of

0743-0167/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 7 4 3 - 0 1 6 7 ( 0 0 ) 0 0 0 5 3 - X

324

J. Johnson, B. Maxwell / Journal of Rural Studies 17 (2001) 323}332

detrimental impacts to ecological functions including fragmentation of wildlife habitat at many trophic levels; interruption to the water and soil nutrient cycles; and disturbs wild"re burn patterns (Gersh, 1996). Others (Lassila, 1999) suggest that residential development of the land may threaten the capability of the region to sustain productive ecosystem functions that would enable residents to make a living on the land. The second concern stemming from rural residential growth is the threat to the community integrity of the region, thereby, posing a threat to local economies. Rural population growth in the west is di!erent from past growth with respect to who lives on the land. In the past, migration to small towns was for the jobs in the extractive industries indigenous in the west. Timber harvest, ranching, farming, mining, and oil exploration attracted large numbers of workers and families during the 1940s through the 1970s. During the latter 1980s and into the 1990s, a great deal of rural in-migration was propelled by buyers who are attracted by a mix of amenities acting as pull factors (Decker and Crompton, 1993). These include scenic beauty, recreation, opportunity for small business investment, the local small town environment, and personal safety (i.e. #eeing high-crime urban centers). Many recent arrivals are retired or semi-retired or are able to make a living aided by a modern telecommunications network (Beyers et al., 1995). As a result, they are not "xed to the traditional urban infrastructure for employment. The fear is that unplanned settlement on rural lands may eventually destroy the very qualities that attract in-migrants and the jobs that follow. Coupled with a new demographic of the rural settlement of the west is the demographic makeup of the current farm and ranch population. Faced with declining agricultural prices and their children's inability or reluctance to remain in agriculture (Stauber et al., 1995), owners of farms and ranches often sell the land to developers or farm and ranch brokers. In rapidly growing counties, the potential wealth from selling to developers serves as a lucrative retirement for farmers and ranchers. Many of the new buyers of rural land will not continue the agrarian tradition or will shift from intensive agricultural production to a less utilitarian form of land management such as `hobby ranchinga or owning the land for its recreational potential. In many cases though developers purchase agricultural land, subdivide it and build `ranchettea properties of 1}5 acres and, thereby, permanently remove land from agricultural production. While a market-based economic model might suggest that greater e$ciencies are attained by moving the land from production agriculture to rural housing, there is an abundance of evidence that current landowners as well as residents of rural communities increasingly value the cultural and scenic attributes provided by a productive agricultural landscape (American Farmland Trust, 1998; Gersh, 1995; Riebsame et al., 1996).

2. Background * the conservation reserve program The Conservation Reserve Program was initiated as part of the Food Security Act (FSA) of 1985. It allowed farmers to enroll some erodable or otherwise ecologically sensitive croplands into a conservation land bank in return for annual payments in 10-year increments. By the early 1990s, more than three quarters of the authorized 18.2 million ha (45 million acres) had been enrolled in the program (USDA, 2000). As more land was bid into the program, however, controversy arose regarding the effects of CRP on rural communities economic well-being. Changes observed in small agricultural towns included a drop in agricultural employment and associated economic activity as well as changing consumer spending habits. Very simply, as less land was cultivated, less labor was needed for farming, less crop was stored in local granaries, and fewer implements were replaced or maintained. In many cases, small businesses that directly supported agriculture failed and as they did so other `main streeta businesses su!ered. The Department of Agriculture foresaw these potential community impacts and sought to minimize them by limiting the amount of land enrolled in CRP to not exceed 25% in any single county. However, in Three Forks and other sparsely populated counties, some farm and ranch supply businesses did fail. On the other hand, in the towns where the vitality of the retail sector was not locally driven * usually larger adjacent towns, regional retail trade centers have emerged and their business prospered (Henderson et al., 1992). At the same time, farmers were somewhat better o! economically as their farm incomes stabilized due to guaranteed CRP payments. The negative community aspects of CRP may somewhat overstated and many of the problems attributed to CRP may have originated earlier because of the changing nature of agricultural business practices and the number and distribution of farm operations (Saltiel, 1994). In any case, he reports that in Montana, CRP is popular among farmers and ranchers alike. So popular that by 1994 Montana ranked fourth in the nation in CRP enrollment. By 1999 1,643,400 acres were enrolled in CRP in the state (USDA, 2000). In economic terms, some (Daniels, 1988; Parks and Schorr, 1997) argue that CRP is an overly expensive program with a relatively short duration (signups are for 10-year increments) and other more e!ective programs exist to achieve similar results. We discuss those considerations below.

3. Study area description The town of Three Forks, Montana lies at the headwaters of the Missouri River in Gallatin County. The valley in which the town is sited is de"ned by the


Fig. 1. Map of the Three Forks study area.

Madison and Gallatin Rivers to the east and to the west * the Je!erson River (Fig. 1). The town site was a major encampment for the returning Lewis and Clark expedition in 1805. It is an ideal community for the investigation of rural change as it makes a transition from an economy based primarily on natural resources to one that is increasingly regionalized. Historically, the Three Forks area economy was based on railroad employment and agriculture (dry-land and irrigated wheat) and some mining-related activity. Today, the mining industry is economically stagnant. While some manufacturing (cement, talc products) takes place, no large-scale mining investment is currently underway. A small lumber mill is located nearby and the town has a small, and shrinking, retail sector. Following the citynegotiated transformation of past railroad right-of-way land into a housing subdivision in the early 1990s, employment in home construction increased from previous levels. In this and other subdivisions, the homes were purchased as soon as they became available. Older homes in the region are being acquired and refurbished. Between 1992 and 1996, Three Forks housing prices increased by 25% due in large part because of in-migration pressures from Bozeman. In recent years, a vertically integrated bakery has been in operation. This has provided a market for locally produced wheat and provided some value-added production in local agriculture. A

325

few other small businesses have appeared in the community. These include a "shing lodge/motel, a small cement products plant, a B & B/restaurant, and a wholesale tool supplier. While the local economy of Three Forks is not particularly healthy, the population of town and the surrounding area has undergone signi"cant growth in recent years and the region is beginning to acquire some of the social and economic characteristics of the so-called `New Westa communities (Shepard, 1993). Dependence on a regional economy, increased dependence on tourist services and most importantly, the rapid transition to a bedroom community for Bozeman, are characteristics shared with other `New Westa communities. Many residents commute via Interstate 90 to Bozeman (population: 35,000), 70 km to the east. The thriving economy there continuously provides diverse employment opportunities. In addition to the structural employment created by a regional population of over 50,000, several relatively large employers in the Bozeman area provide jobs in high technology, tourism, consumer services, retail, higher education, and light manufacturing. As more Three Forks residents "nd employment in Bozeman or spend less time working the land, they increasingly conduct shopping and entertainment activities there. This is a partial explanation why several traditional main street businesses (i.e. a retail chain, hardware, farm implement dealer) in Three Forks have closed. The overall age of Three Forks area residents is much older than the county population. In contrast to Bozeman, many Three Forks residents are retired from the railroad or agriculture and are on "xed incomes. In many respects, the Three Forks area is typical of other amenity communities in the Rockies. While some, in Colorado and other locations in Montana, for example, are adjacent to large blocks of public land (National Forest, National Parks) and others are recreation based (Jackson Hole, WY and Sun Valley, ID), there is general agreement that the combination of scenic beauty, recreation and small town atmosphere are powerful attractants to rural residents (Johnson and Rasker, 1995). Three Forks is not surrounded by large blocks of public land and only a few large parcels of land exist for industrial style agriculture. However, by virtue of its location at the con#uence of three major rivers, the landscape is naturally fragmented by rivers, hills, and large riparian zones. The fragmentation is compounded by interstate highway and other transportation infrastructure (i.e. railway, country roads), urban and rural housing, and commercial development (talc factory and airport), and a local state park. This geographically diverse landscape requires that most agricultural activity is itself highly disjointed * both how it is spatially located and in terms of production; that is, hay and grain production is mixed with beef cattle operations. However the

326


diversity of the landscape, the large riparian zones, and agricultural mix creates an attractive location for human settlement. Finally, the local microclimate is dry and windy and is comprised of locations of highly erodable soils thereby enabling much of the local agricultural land to qualify for CRP enrollment. Like virtually every other rural community in Montana Three Forks has only marginal control over local land use. Regional land-use planning is almost absent and where it does exist the citizen membership on the planning commission possess little authority to prevent development. Only in 1994 did Montana pass the Subdivision and Platting Act [Mont. Code Ann. A76-3-103(3)] which requires a minimum lot size subdivision in rural countryside to be 160 acres, down from 20 acres. Prior to the passage of the Act however nearly every agricultural landowner platted his land for 20-acre development, thereby, rendering the new regulation ine!ective for land-use planning. Zoning, as a regulatory method to compel speci"c land uses in speci"c locations, does not exist in the Three Forks area.

4. Land-use classi5cation and prediction Based on community visioning meetings held over the course of a year it was determined that a paramount need for rural communities such as Three Forks was suite of tools that would enhance their capacity to understand the nature of growth and change in the community. We determined that a tool that modeled land use into the near future would be a particularly useful for e!ective land management for the community (Maxwell et al., 2000). Using a Geographical Information System (GIS) platform, we began the study of land-use change with an inventory of land use in the area. Our land-use classi"cation was conducted in a 30 km study region on a 2.5 ha grid system using aerial photos at available intervals between 1964 and 1998. The study region was divided into cells 2.5ha in size and each was assigned a land-use designation. The observed land-use information was placed in the GIS system as independent layers of information so they could be used as independent variables in the prediction system. Other data layers were constructed to be included in our eventual model of land-use change and included natural geographical features, manmade infrastructure as well as personal interview data. The full complement of data layers as well as the years for which the data was constructed and used in the model are shown in Table 1. The complexity of the model is beyond the scope of this paper and would confuse rather than elucidate the point of CRP as a mitigating factor in rural sprawl. A complete discussion of the model can be found in Hill and Aspinal (Maxwell et al., 2000. Spatial Information For Land Use Management. Gordon and Breach, New York).

Table 1 GIS layers of information used for predicting land use change over time near Three Forks, MT Data type

1965 1979 1984 1990 1994 1995 1998

Land use Roads Streams Distance from streams Distance from roads Nearest neighbor Soil Landowner behavior

The designation `nearest neighbora is land-use type of the cell(s) adjacent to other cells. In the study area, a few land-owners control large portions of the landscape surrounding the community. We conducted in-depth interviews with area landowners to investigate directly why they made land-use change decisions that we had observed on the aerial photos. We were able to categorize their behavior into six general categories. The land-use change behavior categories could then be associated with each owner's land or portion of land that "t a particular biogeographic classi"cation. For example, some owners would be willing to place their riparian zones in conservation easements, but retain upland dry areas for crops based on maximizing net economic returns.

The data layers were used to generate a (2#n)-dimensional probability transition matrix (multiple layers of information through time and space) for land use. The approach used to predict land-use/cover change concentrates on identifying and incorporating independent driving variables directly into transition probabilities in landscapes dominated by private ownership. The model is based on the e!ects of both past land-use change as well as the additional data layers depicted in Table 1. This method assumes that future land-use patterns are driven by local patterns of land-use change and that local changes strongly in#uence nearby change. This method di!ers from a regression approach in several ways. First, the model makes the prediction of future land-use change based on the probability of change from surrounding cells. Second, whereas regression approaches tend to drop variables that do not contribute signi"cantly to prediction on a large portion of the map, a probability approach allows a full compliment of the independent driving variables to re#ect the potential land-use change (Berry et al., 1996). Finally, regression approaches tend to model change based on proximity to geographical features (roads, water) and predict a gradient of change that decreased with distance from these features (Theobald, 1998). Thus, the likelihood of development decreases with increasing distance from urban areas. Yet, observation of rural growth suggests that accessibility is perhaps not the driving factor of homes in the rural countryside. Rather, future development patterns are more likely a function of the near or distant views available to potential homeowners, characteristics


of the community or isolation from others (Maxwell et al., 2000). Computation of the transition matrix is based on the assumption that over a selected observed time period land use/cover in a selected part of the landscape can remain in the same land-use/cover category or change to one of the other possible categories. The model allows the selection of one or more data layers to be used to calculate a transition matrix. A minimum of 2 years of data layers must be selected * a primary layer and a response layer; these will be used to produce the transition matrix. The matrix represents the probability that a particular cell will change to another land-use value or stay the same. For example, Fig. 2 depicts the hypothetical transition probabilities for a cell currently designated as native range changing to other land uses. As the matrix is populated with data over the entire grid area, a map can be generated that shows future

327

land-use change scenarios based on the probability of change derived from past land-use change and the e!ects of geographical and socioeconomic factors. Accuracy of the model was enhanced by attempting to predict a known mix of land use. Various combinations of the known land-use pro"le for the years 1965}1979 in combination with the other driving factors were tested and yielded varying prediction accuracy. An aerial land-use map from observed 1998 land cover was used as the calibration. Fig. 3 illustrates the enhanced predictive capacity of the model to predict land use in 1998 given various combinations of data layers. The best prediction (93% accurate) of the observed land use in 1998 was produced by the combination of nearest neighbor, landowner attitude, and distance to stream. The model was then designed to be run forward in time to the year 2025 using the algorithm derived from the calibration. These results were utilized in the analysis

Fig. 2. Conceptual framework for calculating the multidimensional transitional matrix using land-use change over time and additional driving variables.

Fig. 3. Enhanced predictive capacity with varying combinations of driving variables for predicting the numbers of cells changing to residential designation for observed year 1998.

328


discussed below on the e!ects of CRP and rural residential development.

5. Ecosystem integrity analysis Concurrent with the development of the land-use prediction system was the assessment of ecosystem integrity for the selection of land cover classi"cations used in the model. Some argue that the concept of ecosystem integrity (EI) is a concept that is de"ned, in part, by human society. Therefore, it may be said that the integrity of an ecosystem is partially a function of land use as well as the amount or type of impact society is willing to accept for that ecosystem (Regier, 1993). Historically, high productivity dominated the societal view of what was preferred in a managed ecosystem especially in agroecosystems (Heck et al., 1991) and ecosystem monitoring has been directed accordingly (Meyer et al., 1992). A more recent trend has been towards environmental and economic optimization as the reference point for land managers (Meyer et al., 1992). That reference point has been de"ned as sustainability (Doran et al., 1996) and represents a societal level of the ecosystem integrity/economic wellbeing tradeo!. In other words, acceptable sustainability assessment must strike a balance between preservation of natural ecosystem processes and at the same time not foreclose social and economic activity on the land (i.e. agriculture, recreation, housing). Land uses that maintain ecosystem as well as human system functions are said to be more sustainable than those that do not (Doran and Parkin, 1994). Our working de"nition of ecosystem integrity assumes that it is a continuous dependent variable that is referenced to an area with minimal non-natural disturbance. High ecosystem integrity is the state in which an ecosystem maintains and sustains its function at all trophic levels but would also support the continuous present level of human activity. Parcels of land were identi"ed that were found to be consistently in speci"c land-use types since 1965. Field observations in 1995 and 1996 were used to develop an ecosystem integrity (EI) index associated with each landuse type. The EI index was assigned to each cell on historic as well as predicted maps. The EI values were then accumulated for the entire study area for a given year so that a general trend in ecosystem integrity associated with di!erent land-use change scenarios could be evaluated. A major "nding of the land-use prediction analysis and the ecosystem integrity exercise was that enrollment of land into CRP status signi"cantly changed the structure of land use in the study area and enhanced the overall EI for the region. Between 1980 and 1995, a large amount of land transitioned from dryland crop (wheat) to CRP designation. As the number of grid cells in wheat production deceased, EI scores display a concomitant increase.

The study area EI stayed the same in 1996 despite the conversion of a small area of CRP to a rural subdivision. Based on our ecosystem integrity index, we "nd that the explicit goals of the Conservation Reserve Program, to withdraw ecologically marginal land out of cultivation and revitalize the local ecosystem, were met. When the CRP funding was reauthorized in the 1997, it allowed for further improvement of environmental quality with an active land management regime for quali"ed land. We would expect ecosystem integrity to be further enhanced with these reforms. The tradeo! between dryland wheat production and ecosystem integrity scores over the 1965}1996 time period is clear * CRP enrollment played an important role in enhancing and maintaining ecological integrity in the region. The results are collaborated in other ecosystem function assessments in other regions (Parks and Schorr, 1997; Ribaudo, 1989). If CRP is in fact a useful land management tool for removing marginal agricultural land from the production stream what role might it play in curbing the amount of land entering the marketplace for rural residential development? This question led us to ask how the community might have grown if the CRP enrollments had not taken place. To test the role of the CRP policy as a moderating variable in land-use change, the CRP land-use designation was removed from the mix of land uses beginning in 1984, thereby, changing the probability function utilized in the prediction system and producing a di!erent mixture of land. A second prediction scenarios was run to produce a map without CRP as a driving variable of land-use change. The outcome is presented in Fig. 4. The census of map cells classi"ed as residential following the two growth scenarios indicate that with CRP in the land-use mix the extent of rural residential development in the study area decreased. Fig. 4 illustrates diverging trend lines that represent the number of cells that could be expected to have changed from no development to residential with and without CRP in the prediction equation. With CRP in the land management mix the number of cells in residential land use was observed and predicted to have grown 14% between 1984}1995 and 23% from 2000 to 2025. Without CRP in the mix, the area was projected to have a residential land-use growth rate of 27 and 40% growth rates for the same periods.

6. Application * CRP and the role of ecological refuges This paper provides evidence that CRP can achieve both intended and unintended consequences of private land protection. Central to the original CRP mandate was enhancing ecosystem quality by removing marginally productive agricultural lands from production. In addition however, we document an additional role for


329

Fig. 4. Number of land-use cells classi"ed as residential over time both observed and predicted in the study area with and without the CRP policy in place.

the CRP initiative: that it may have a moderating e!ect on rural residential sprawl. The "nding is important for several reasons. Aside from the obvious economic bene"ts of retaining agricultural employment and earnings in a rural economy, keeping agricultural lands intact can be a desirable amenity for community residents. The strategy preserves open space and prevents piecemeal residential development. Agricultural land in CRP can bene"t the ecology, economy, and aesthetics of the community. However, tracts of agricultural land in the midst of rapid residential growth can bring negative consequences for local residents. While rural non-farm residents may enjoy the open space provided by active farming operations, some farming activities may be considered bothersome. This includes negative public reaction to the application of fertilizers, pesticides and herbicides; smells from animal production activities; slow-moving farm vehicles on commuter roads; increased incidence of air pollution from harvesting and burning; and increased demand on local water supply for irrigation. Public concern over these and other issues has triggered numerous lawsuits that tests the right to farm legislation that exist in all of the 50 states (Pfe!er and Lapping, 1994). These negative consequences are in addition to the impacts of CRP on rural communities discussed previously .

The ideal scenario might be a land management regime where the conservation and residential development e!ects of the program are left intact but the negative agricultural impacts are reduced. We propose that CRP could support a `mixeda land management whereby conservation of water and soil (as well as visual) qualities of farmland are preserved if farmers would manage the land to focus the conservation elements attributed to CRP into `ecological islandsa or refuges within productive agricultural "elds (Diamond, 1976; Henry et al., 1999). In e!ect, this would optimize the landscape diversity of a rural community in which the farm spaces would be fragmented with ecologically functional patches of native species, protected riparian zones, and even recreational trails. Some rural housing may be interspersed on the edges of the large open spaces but the land would remain primarily in agricultural production. The ecological bene"ts of such a land-use setting would be, in e!ect, concentrated to localized areas within the larger agricultural context. The management goal for CRP would become to continue ecological protection and for the landowner it would be to preserve agricultural production. The community would bene"t through the preservation of open space. The bene"ts of such a land management mix might be signi"cant. As the island is populated with, for example,

330


native birds, we might foresee less need for pesticide application. In some locations, on a large agricultural "eld, for example, water runo! may be slowed thereby mitigating the soil erosion e!ects of heavy rains. Islands would also provide scenic qualities commonly lost to rural residential development as well as large expanses of monoculture. The heretofore negative e!ects on the community economy resulting from CRP enrollments would be somewhat diminished as productive agriculture returned to the landscape and farm based employment rises, crops return to local mills and elevators and local farm implement sales are renewed. A case can be made for optimizing the landscape diversity of rural communities by reconsidering how rural housing and productive farmland can coexist. Clearly, the open spaces produced by agricultural production provide scenic and even cultural qualities valued by rural residents. For example, research conducted in Gallatin County (Montana) in 1991 found that scenic beauty was the most important location factor for business owners who chose to live in the area (Johnson and Rasker, 1995). A great deal of the land that could be considered scenic in Gallatin and other amenity-rich counties is, in fact, private agricultural land. Other important qualities are also preserved by planning for landscape diversity such as the provision to build an interconnecting network of recreational parks and trails, and even economic savings for local taxpayers (Haggerty, 1996). Some (Beck et al., 1999) have found that a mixed land-use mosaic can have positive overall economic impacts and that environmental protection can stimulate local economies. In the case of a restructured CRP, economists may "nd similar bene"ts at a smaller scale. In many counties where CRP lands are present, local agriculture is not on the same scale as found in the Midwest where large-scale industrial farming is the norm. Rather, farming in many amenity counties is similar in scale and nature to that found in the Three Forks region. Where agricultural land quali"es for CRP and farmers wish to remain in agriculture, a mixed land-use regime may allow enough residential development to keep the operation intact. One could imagine that such a land management mix would make the farm more desirable in the marketplace when the landowner chose to retire. Signi"cant amounts of CRP land are found in many high-growth counties. For example, Teton County, ID has over 12,000 acres * 4% of the total landmass of the county and a population growth rate of almost 37%. Lincoln County, CO, a county with 23% growth rate has 156,852 acres of CRP * 6% of the landmass. Certainly, other land conservation strategies exist (White, 1998). Proponents of conservation easements point to the success of that strategy for land conservation (Daniels, 1988; Wright, 1993). A conservation easement is a legal contract between a land trust, a governmental

entity, or other quali"ed organization and a willing landowner. In exchange for a tax-deductible contribution for the value of the protected land, the easement permanently limits uses of the land in order to protect its conservation values. The restrictions run permanently with the land. A conservation easement protects the land from unlimited subdivision and development while also protecting the rights of private ownership. Examples of uses generally permitted by a conservation easement: include: continued agricultural use; sale or gift of the property; or selective timber harvest. Examples of uses generally restricted by a conservation easement are: subdivision for residential development; surface mining; or the elimination of wildlife or "sheries habitat protected by the easement. The landowner continues to own the land and continues to pay taxes on the land. Unfortunately, in many cases the agricultural landowner in rural locations is not able to implement an easement because he simply does not make enough income to take full advantage of the tax deduction. Only two western states have implemented a tax credit in exchange for signing the easement. Conservation easements do work well however where the incomes of the owners of agricultural land are large in real terms. Another strategy for land conservation is outright purchase either by the Federal government (such as #ood plains and national forest private inholdings), or a conservation group (i.e. The Nature Conservancy). These solutions for the most part a!ect relatively small amounts of high-quality land. It would seem then that the 10-year planning horizon as required in the CRP would, at a minimum, place the land in short-term conservation management while other land management options are explored or the market for rural land moderates. Finally, as Pfe!er and Lapping (1994) point out, all farmland preservation strategies are enhanced if they are integrated into comprehensive land-use planning.

7. Conclusion The CRP initiative originated as part of the Food Security Act (FSA) of 1985 shows promise as a tool with which to decrease the incentives for agricultural landowners to sell agricultural land that could, in many cases, be developed for residential use. This paper demonstrates that CRP reduced the amount of land available for development. While the CRP lands are de"ned as marginally productive for agriculture, they represent the open spaces many rural residents "nd attractive. When the distribution of CRP lands and rapidly growing counties in the Rocky Mountains are considered together, the potential for using CRP as a constraint on sprawl is substantial (see: http://www.fsa. usda.gov/ DAFP/cepd/18thcrp/crp1099.htm). While CRP appears to play a key role in the agricultural community of Three


Forks, other communities may be subject to other Federal or state programs that can act as land management tools (i.e. wetlands protection policy, National Park lands, forest stewardship programs). In some jurisdictions, local policy can be designed to achieve similar goals. Most countries have similar land conservation programs that could be studied for their potential policy impacts on sprawl. Planning and policy against rural sprawl has reached a high level of national attention in the United States and other locations among politicians, environmentalists and agriculturists alike. Indeed, in 1999 American voters approved over 240 anti-sprawl local government ballot initiatives nationwide and Federal government leaders have supported the concept of `smart growtha to help cities and counties manage the e!ects of sprawl (McOmber, 1999). National environmental groups (i.e. Sierra Club) have made sprawl a major focus of their political activity. Others, such as the American Farmland Trust, work in concert with environmentalists and agricultural producers in an e!ort to safeguard farmland from development. While a great deal of attention is aimed at decreasing demand for rural land or mitigating the e!ects of sprawl, far less attention is given to decreasing the supply of farmland for development. Finally, landscape planning by cities and counties in the rural west is a fragile process that is often plagued by lack of public and political support. However, agricultural lands whether in active crop production or enrolled in the Conservation Reserve Program, play an increasingly important role as a viewshed feature in many highgrowth communities. The protection from sprawl provided by a government program such as CRP, for all practical purposes, takes some land out of the planning equation thereby reducing the potential for con#ict in shaping the future of land adjacent to the community. These private lands are, in e!ect, zoned as non-developed land as long as it is in the landowner's economic interest and CRP funding continues. It should be noted that in 1999 USDA announced that funding for CRP will increase by $100}125 million/yr, FY 2000}2002 (http:// www.fsa.usda.gov/dafp/cepd/crpinfo.htm). If this trend continues the mitigating e!ects of CRP on rural sprawl may be felt in more communities.

References American Farmland Trust, 1998. American Farmland Trust living on the edge: the costs and risks of scatter development, http://farm. "c.niu.edu/cae/scatter/e-loe.htm. Beck, R.J., Kraft, S.E., Burde, J.H., 1999. Is the conservation of land from agricultural production to a bioreserve boon or bane for economic development? Journal of Soil and Water Conservation 54, 394}401. Berry, M., Flamm, R.O., Hazen, B.C., MacIntyre, R.M., 1996. The land use change analysis system (LUCAS) for evaluating landscape

331

management decisions. IEEE Computational Science and Engineering 3, 24}35. Beyers, W.B., Lindahl, D.P., Hamill, E., 1995, Lone eagles and other high #iers in the Rural producer services. Paci"c Northwest Regional Economic Conference, Missoula, MT. Daniels, T.L., 1988. America's conservation reserve program: rural planning or just another subsidy?. Journal of Rural Studies 4, 405}411. Decker, J.M., Crompton, J.L., 1993. Attracting footloose companies: an investigation of the business location process. Journal of Professional Services Marketing 9, 69}94. Diamond, J., 1976. Island biogeography and conservation: strategy and limitations. Science 193, 1027}1029. Doran, J.W., Parkin, T.B., 1994. De"ning and assessing soil quality. In: Doran, J.W. (Ed.), De"ning Soil Quality for a Sustainable Environment. Soil Science Society of America Inc., Madison, WI, pp. 3}21. Doran, J.W., Sarrantonio, M., Liebig, M.A., 1996. Soil health and Sustainability. In: Sparks, D.L. (Ed.), Advances in Agronomy. Academic Press Inc., San Diego, CA, pp. 1}54. Gersh, J., 1995. The Rocky Mountain West at risk. Urban Land 54, 32}35. Gersh, J., 1996. Subdivide and conquer. Amicus Journal 18, 14}23. Haggerty, M., 1996. Costs of county and education services in Gallatin County, Montana. Montana Policy Review 6, 13}21. Heck, W.W., Cambell, C.L., Hess, G.R., 1991. EPA/ 600/4-91/013, Washington, DC. Henderson, D., Tweeten, L., Woods, M., 1992. A multicommunity approach to community impacts: the case of the conservation reserve program. Journal of the Community Development Society 23, 88}102. Henry, A.C., Hosack, D.A., Johnson, C.W., Rol, D., Bentrup, G., 1999. Conservation corridors in the United States: bene"ts and planning guidelines. Journal of Soil and Water Conservation 4, 645}650. Johnson, J.D., Rasker, R., 1995. The role of economic and quality of life values in rural business location. Journal of Rural Studies 11, 405}416. Lassila, K.D., 1999. The new suburbanites: How American plants and animals are threatened by the sprawl. The Amicus Journal 21, 16}22. Maxwell, B., Johnson, J., Montagne, C., 2000. Predicting land use change in and around a rural community. In: Hill, M., Aspinal, R. (Eds.), Spatial Information for Land Use Management. Gordon and Breach, New York (Chapter 13). McOmber, J.M., 1999. Will &smart growth' prove to be smart political topic? Seattle Times, March 1. Meyer, J.R., Cambell, C.L., Moser, T.J., Hess, G.R., Rawlings, J.O., Peck, S., Heck, W., 1992. Indicators of the ecological status of agroecosystems. In: McKenzie, D.H., Hyatt, D.E., McDonald, V.J. (Eds.), Ecological Indicators, Vol. 1. Elsevier Applied Science, London, pp. 629}658. Parks, P.J., Schorr, J.P., 1997a. Sustaining open space bene"ts in the northeast: an evaluation of the conservation reserve program. Journal of Environmental Economics and Management 32, 85}94. Pfe!er, M.J., Lapping, M.B., 1994. Farmland preservation, development rights and the theory of the growth machine: the views of planners. Journal of Rural Studies 10, 233}248. Regier, H.A., 1993. The notion of natural and cultural integrity. In: Woodley, S.K.J. (Ed.), Ecological Integrity and the Management of Ecosystems. St. Lucie Press, Ottawa, pp. 3}18. Ribaudo, O., 1989. Targeting the conservation reserve program to maximize water quality bene"ts. Land Economics 65, 320}343. Riebsame, W.E., Gosnell, W., Theobald, D.M., 1996. Land use and landscape change in the Colorado Mountains I: theory, scale and pattern. Mountain Research and Development 16, 395}405. Saltiel, J., 1994. Controversy over CRP in Montana: implications for the future. Journal of Soil and Water Conservation 49, 284}289. Shepard, J.C., 1993. The new economy of the great plains: implications for economic development. Economic Development Quarterly 7, 403}410.

332


Stauber, K.N., Hassebrock, C., Bird, E.A.R., Bultena, G.L., Hoiberg, E.O., MacCormack, H., Menanteau-Horta, D., 1995. The promise of sustainable agriculture. In: Bird, E., Bultena, G.L., Gardner, J.C. (Eds.), Planting the Future: Developing an Agriculture that Sustains Land and Community. Iowa University Press, Ames. Stevens, W.K., 1999. Sprawl quickens its attack on forests. New York Times, December 7. Theobald, D.M., 1998. Fragmentation by inholdings and exurban development. In: Knight, R.L., Smith, F.W., Buskirk, S.W., Romme, W.H., Baker, W.L. (Eds.), Forest Fragmentation in the Central Rocky Mountains. University Press of Colorado, Boulder, CO.

US Bureau of the Census, 1995. County business patterns (GENERIC). US Bureau of the Census, 1999. Mountain states show biggest increases in housing; Nevada leads the way. Census Bureau Reports, http:// www.census.gov/Press-Release/www/1999/cb99-232.html; http://www.census.gov/Press-Release/www/1999/cb99-232.html. USDA, 2000. www.fsa.usda.gov/dafp/cepd/crpinfo.htm. White, J.S., 1998. Beating plowshares into townhouses: the loss of farmland and Strategies for slowing its conversion to nonagricultural uses. Environmental Law 28, 113}143. Wright, J.B., 1993. Conservation easements: an analysis of donated development rights. Journal of the American Planning Association 59, 487}494.