Threat to high market value agricultural lands from urban encroachment: a national and regional perspective.
The urban-rural fringes of America's metropolitan areas continue to experience rapid population growth. As a result, the replacement of agricultural land with urban land is a pervasive problem on the urban-rural fringe. The public response to farmland conversions is usually local in nature, resulting in local and county government initiatives to preserve farmland as a valuable scenic amenity, protect farmers rights to farm, or promote local and self sufficient farm economies.(1) Local governments are not usually aware or immediately concerned with the impact of farmland conversions on the national distribution of agricultural land. However, the aggregate of farmland conversions occurring at the local level ultimately affects the quantity and quality of the national cropland base. Once farmland is converted to urban land it is permanently removed from the national inventory of agricultural lands, and the converted land is often replaced by land that is less productive for agriculture.
Metropolitan agriculture may seem insignificant when one considers the geographic extent of the nation's largest cities relative to the nation's remaining cropland base. Nevertheless, over half of America's agricultural production, as measured in terms of market value of agricultural products sold, comes from within or adjacent to the nation's metropolitan areas. The presence of so much agriculture in the shadow of metropolitan America has raised concerns among farmland preservationists about the sustainability of the nation's agriculture.(2) A recent national study of 135 counties, that grew rapidly between 1970 and 1980, determined that 790,000 acres of cropland were converted to urban land uses during the 1970s.(3) Higher national estimates of farmland conversion are reported in the Soil Conservation Service's (SCS) National Resource Inventories (NRIs), which indicate that about 900,000 acres are converted annually.(4) At the same time that cropland is decreasing in metropolitan fringe areas, cropland is increasing on marginal lands in many of the nonmetropolitan counties of the arid west.(5) These trends suggest that cropland increases in the arid west are related to the loss of high value agricultural land surrounding metropolitan areas.
Our knowledge of the national extent of the farmland conversion problem is limited by the availability of land conversion data. Recent attempts to provide answers have made use of agricultural and population censuses,(6) while others have utilized large scale aerial photography in scattered test sites.(7) Both national agricultural censuses and high resolution satellite imagery are employed in this study to examine the potential threat of urban encroachment onto farmland. It is suggested that the correlation of census variables, such as market value of agricultural products sold and population growth statistics, are not appropriate surrogates of national farmland conversion rates. Remote sensing data provide accurate assessments of the spatial extent of the conversion of farmland to urban land, but high processing costs limit the use of these data to regional scale analyses. This article considers the relative strengths of national and regional level data for understanding the scope of the agricultural land conversion process, and argues that in the absence of a national land-use policy framework, population growth will continue to threaten metropolitan agriculture.
The need to protect farmland from urban encroachment has been widely debated in the literature. Platt has attributed the lack of resolve over the key issues of the debate to the unreliability of land conversion data and to the inability of debate participants to clearly articulate their objectives.(8) Data from the National Resources Inventories (NRIs) of the United States Department of Agriculture (USDA) have provided much of the factual basis for the farmland conversion debate. Variability in collection procedures as well as changes made in definitions over time have raised suspicions about the reliability of the NRI data.(9) Nevertheless, opponents of farmland protection are quick to cite cropland acreage increases since 1959, as reported in the NRIs, in response to farmland protectionist objectives.
The concern over the conversion of farmland to urban land uses resurfaced recently with the publication of a map by the American Farmland Trust (AFT). The AFT map highlights vast areas of the U.S. where high value farmland is identified as being at high risk of conversion.(10) Data from the 1987 U.S. Census of Agriculture and the 1980 and 1990 decennial censuses of population were used by AFT to examine the relationship between high value farmland and population growth. In addition to concluding that urban sprawl was devouring prime and unique farmland, the AFT suggested that farmland protection policies be designed to encourage compact urban development and to keep farmland dedicated to long-term agricultural use.
Using the same data from the censuses of agriculture, others have drawn conclusions opposite from the AFT. Hart, in his case study of agricultural land use transitions in the New York metropolitan area, disputes the idea that the valuable agricultural farmland surrounding most metropolitan areas is prime farmland:
Its high price derives from its location near a metropolis rather than from any inherent quality of its soil, and it is a mere fraction of the total cropland base of the United States.(11)
Drawing upon Von Thunen's agricultural land use principles,(12) Hart argues that the high value of land on the urban rural fringe forces farmers to cultivate their land more intensively which in turn generates a higher volume of sales at the market place.
Researchers at the USDA's Economic Research Service (ERS), although recognizing that urban encroachment threatens increased loss of farmland, see metropolitan growth offering opportunities to more farmers.(13) Heimlich and Brooks provide examples of opportunities available to metropolitan farmers:
1. access to specialized markets; 2. access to off-farm employment; 3. higher farm equity; and 4. political support for farmland retention measures.
Although, the AFT concluded that compact urban development should be encouraged to improve the quality of farming on the urban rural fringe, Heimlich and Brooks concluded just the opposite:
By the mid-1970's, new development was occurring beyond the existing metropolitan fringe in areas with no urban centers comparable to the growth nodes of earlier decades. And unlike the earlier development pattern, which took land out of farm use, the new development patterns allowed more land to be kept in farms.(14) Today, these two opposing sentiments, compact versus dispersed settlement, have taken on a new fervor within the neotraditional planning movement.(15)
Clearly, there is a range in perspective on whether it is necessary to preserve farmland by managing urban growth. The above studies examined agricultural sales data from the census of agriculture to help support their arguments. We turn now to these data.
HIGH VALUE FARMLAND
The Census of agriculture, administered every 5 years by the U.S. Census Bureau, has served as the principal source of information on metropolitan agriculture. The ERS examined the 1974, 1978, and 1982 censuses of agriculture and found that the amount of farmland in metropolitan counties rose by almost haft between 1974 and 1982.(16) The large increase resulted from metropolitan counties that were added between 1971 and 1985. Consistent with previous studies, the ERS study concluded that metropolitan farms were more land intensive and more focused on high-value production than nonmetropolitan farms.
The U.S. Census Bureau, following guidelines from the Office of Management and Budget (OMB), redefines the nation's metropolitan counties after every decennial census. Out of the nation's 3,069 counties and equivalent units, excluding Alaska and Hawaii, 806 were defined as metropolitan following the 1990 census.(17) Although these counties only represent 26% percent of the nation's counties, they accounted for 34% of its gross agricultural sales in 1987 (Table 1). The high volume of fruit, vegetable, and nursery sales from these metropolitan counties underscores the significance of their land intensive agricultural practices.
The AFT recently identified counties of agricultural importance, measured in terms of agricultural sales, that were at risk of losing farmland to urban growth.(18) Counties were placed into two high risk groups and a third cautionary group. Counties that were at highest risk of losing valuable farmland from urban encroachment were identified as fast growing counties inside or adjacent to metropolitan areas with high value farmland [ILLUSTRATION FOR FIGURE 1a OMITTED]. Counties with high value farmland are counties that had gross agricultural sales and agricultural sales per acre of farmland exceeding the U.S. county average in 1987. Fast growing counties were defined as counties experiencing percentage increases in population, during [TABULAR DATA FOR TABLE 1 OMITTED] the 1980s, greater than the U.S. county average percentage increase of 4.1%. A final condition for inclusion dropped out 442 counties that were adjacent to metropolitan areas, but had population densities less than 25 persons per square mile. The rationale for the latter condition was that a critical amount of settlement was necessary before population growth could pose a realistic threat to farmland. The 25 persons per square mile value was chosen because this is the minimum population density in which a county can qualify as metropolitan under the official metropolitan guidelines.(19)
Only 10% (202) of the approximately 2,000 metropolitan and adjacent counties were considered in the group at highest risk of losing valuable farmland to urban encroachment. Although representing only 6% of the nation's counties, these counties accounted for almost a quarter of the total U.S. gross agricultural sales in 1987 (Table 2). The minimum population growth rate during the 1980s for these counties was 4.1%, but as a group they averaged an 18% rate of population increase. In addition to the heavy concentration of these counties in the states of California and Florida, a significant number of them surrounded the edges of many of the nation's metropolitan areas. The latter spatial distribution is important because the most important metropolitan population redistribution trend of the 1970s, 1980s, and now 1990s has been explosive population growth on the suburban periphery at the expense of the central cities of metropolitan areas.
Table 2. 1987 Agricultural Sales by Urban Encroachment Status
County Gross Sales % of U.S. Number of Type 1987 (000) Gross Sales Counties
High Risk 32,725,627 24 202 Moderate Risk 5,015,777 4 109 Nonmetropolitan High Value 20,458,811 15 228
A second group of counties with high value farmland and fast population growth were added to the list of those counties at risk of losing farmland to urban encroachment [ILLUSTRATION FOR FIGURE 1b OMITTED]. The principal difference between this list and the first is that counties with high value farmland were redefined as counties that had gross agricultural sales and sales per acre of farmland exceeding their state county averages in 1987. The national population growth criteria and minimum population density criteria were the same for each risk group.
The use of the state criteria added an additional 109 counties to the list of counties at risk of losing valuable agricultural land to urban encroachment. The purpose for including the latter group is that many counties in the states of California and Florida had exceptionally high value farmland which skewed the national average disproportionately in the positive direction, making it difficult for counties outside these two states to score high on the high value farmland criteria. The additional counties that were included under the state criteria only accounted for 4% of 1987 gross national agricultural sales, but the combination of these counties with the 202 other urban influenced counties account for 28% of the nation's total gross agricultural sales (Table 2). Urban influenced counties qualifying under the state criteria experienced an average rate of population increase of 17% during the 1980s, and like the previous map these counties were concentrated around the periphery of metropolitan areas [ILLUSTRATION FOR FIGURE 1b OMITTED].
A final map not included in the AFT analysis shows nonmetropolitan counties with gross agricultural sales and agricultural sales per acre of farmland exceeding the national and state averages in 1987 [ILLUSTRATION FOR FIGURE 1c OMITTED]. Although these 228 counties only represented 7% of the total counties in the U.S., they accounted for 15% of the nation's total gross agricultural sales in 1987 (Table 2). Far removed from the nation's metropolitan areas, the high value cropland in these counties are not threatened by urban encroachment, but are an important link in the cropland conversion debate. The U.S. Census of Agriculture reported a national decline of 10.5 million acres of cropland between 1978 and 1987.(20) Of the 311 urban influenced counties that have been identified by AFT, 77% have experienced decreases in cropland between 1978 and 1987 [ILLUSTRATION FOR FIGURE 2a OMITTED]. Meanwhile, 51% of the nonmetropolitan counties with high value farmland experienced cropland acreage increases over the same time period [ILLUSTRATION FOR FIGURE 2b OMITTED]. These two maps suggest that the large losses of high value cropland in metro areas is being offset by increases in high value cropland in nonmetropolitan counties. Moreover, a large share of the high value nonmetropolitan counties that experienced cropland increases are located in the arid west where irrigated cropland is putting pressure on diminishing groundwater supplies [ILLUSTRATION FOR FIGURE 2b OMITTED].
Although analyses of high value agricultural lands and population growth point out some disturbing trends, they provide us no information in terms of the actual amount of farmland that is being convened to urban land uses. Moreover, arguing that the correlation of high value cropland and population growth is an estimate of urban encroachments threat to agriculture, can not escape the criticism that it is location that accounts for metropolitan farming's focus on high value production. Of the 311 counties considered at highest risk by AFT, a third or more had sales in fruit, vegetable, and nursery sales exceeding the national average [ILLUSTRATION FOR FIGURE 3a-3c OMITTED].
Perhaps, the most problematic issue of these analyses is their reliance on the county as the unit of analysis. Because county boundaries are arbitrary, varying in size across the nation, the extent of urban encroachment is often exaggerated. The Census Bureau's metropolitan area concept was designed to illustrate the areal extent of a city's influence and the county was chosen as its basic building block. The Census Bureau's urbanized area concept, however, reflects the actual built up area of a metropolis. The metropolitan area of Greeley Colorado, an area with high value agricultural land and high population growth, illustrates the physical discrepancy between the metropolitan area and urbanized area definitions [ILLUSTRATION FOR FIGURE 4 OMITTED]. The loss of farmland due to urban growth is largely contained to the area around Greeley's urbanized area, yet the AFT has extended its influence to the entire county.
High resolution satellite imagery offers an improved method for acquiring more accurate information on urban growth's threat to agriculture. Commenting on the quality of the SCS's NRI data, Platt concluded that satellite imagery could vastly improve our knowledge of farmland conversion rates:
Even in 1982, cropland acreage totals were not derived from satellite imagery or other standardized remote sensing methodology. Instead such data were obtained from the estimates of individual agricultural extension agents in most of the nation's 3,041 counties and equivalent units.(21)
The lack of any systematic use of satellite imagery for detecting land use change on the urban rural fringe was further underscored by Theitz et al., who suggested that our ability to access, process, and integrate satellite imagery with other data is still being developed.(22) The following case study provides conversion rates for high value agricultural land in an urban-rural fringe area of Chicago.
Northern Illinois Case Study
Land uses on the urban-rural fringe of the Chicago metropolitan region have been in constant flux for the past 40 years. As the city of Chicago's population and industry have decentralized to the city's suburbs, the amount of cropland that has been converted to urban land uses has increased substantially. The spatial pattern of these cropland conversions have also become more complex with the maturation of "edge cities" on the outskirts of the Chicago metropolitan area.(23) Generally defined as large suburban employment centers, Chicago's edge cities (Schaumburg area, O'Hare Airport area, Illinois Research and Development Corridor, and the Lake Shore Corridor area) have attracted a disproportionate share of the region's new industrial development because of the locational advantages that their sites offer. With the reconfiguration of journey to work patterns around Chicago's emerging edge cities, new residential development is being scattered over a wider range of territory.
The best place in which to detect cropland conversions, caused by urban encroachment is along the edge of a census defined urbanized area.(24) The United States Census Bureau, has delineated the metropolitan region of Chicago since the 1950 Census.(25) Today, that region, referred to as a Consolidated Metropolitan Statistical Area (CMSA), consists of 8 urbanized areas and 13 counties extending over the three states of Illinois, Indiana, and Wisconsin. The area selected for the current study is located on the western edge of the 1980 Chicago-Northwestern Indiana Urbanized Area and selected eastern edges of the 1980 urbanized areas of Aurora and Elgin. The Aurora and Elgin urbanized areas to the west include towns that originated as small industrial centers on the Fox River in the early nineteenth century, which in some respects grew independent of the Chicago urban complex to the east. Today, the cropland in between these three urbanized areas is rapidly being consumed as the areas continue to coalesce. The total area selected for our conversion study consists of 73 square miles of land in western DuPage County and eastern Kane County that was classified as nonurban in 1980.
The cropland between these multiple urbanized areas is highly productive. Yields of corn and beans on the deep mollisols are among the highest in the world. Out of an Illinois state total of 102 counties, DuPage county and Kane county were ranked number one and eight, respectively, in terms of the value of agricultural products sold. The cropland, however, is rapidly being replaced by urban development. Mapping the census tracts that experienced a population increase of 1,000 or more persons during the 1980s reveals a concentric arc of rapid growth that surrounds the city of Chicago and its inner-most suburbs [ILLUSTRATION FOR FIGURE 5 OMITTED]. During the 1980s, the populations of DuPage County and Kane County experienced percentage increases of 19% and 14%, respectively.
Before quantifying the amount of farmland being lost to urban encroachment, we had to mask out an agricultural area on the urban-rural fringe. We used the 1980 census defined urbanized area boundaries, derived from Census TIGER files.(26) The newly defined 1990 urbanized area boundaries were not selected because of our interest in examining the land conversion process since 1980. A caveat to using the urbanized area boundary to mask out rural land is that a census defined urbanized area is not necessarily "truly bounded." Overbounding often results due to an inclusion rule in the Census Bureau's guidelines for delimiting urbanized areas which provides that as long as a high percentage of a municipality's blocks meet the minimum population threshold for inclusion then the entire municipality will be included in the urbanized area.(27) The implication here is that some of the land inside the urbanized area, especially on the urban edge, can include cropland. Although the urbanized area has some weaknesses, it represents the closest approximation of the built up portion of a metropolitan area.
Once the study area was demarcated, the urbanized area boundaries were superimposed on 1987 and 1991 SPOT satellite images of the region. These satellite images were used as the primary source for quantifying and mapping the extent of nonurban to urban land conversions. The use of 1987 and 1991 SPOT satellite images restricted our analysis to a 4 year time period, but the benefits of these high quality digital images exceeded the costs of acquiring alternative sources of land information representing a more lengthy time interval. Furthermore, the leading and most comprehensive national study of urban land conversions only dealt with a 5 year time interval.(28)
The land which was masked out by the urbanized area boundaries was then classified as urban or nonurban using the United States Geological Survey (USGS) land-use and land-cover classification as proposed by Anderson.(29) The USGS land-use and land-cover classification is composed of two primary levels with categories that range from very general to very detailed. The first level is highly generalized consisting of nine categories, one of which is urban. Detail is gained in this classification as you proceed to level II. The urban or built-up class specified in level I is composed of areas of intensive use with much of the land covered by structures. In level II these uses include several sub-categories such as residential, commercial and service, industrial, transportation and communication and utilities, commercial complexes, mixed urban or built-up land, and other urban land.
In addition to the satellite imagery, black and white aerial photographs from March 23, 1990 at a scale of 1[double prime]=2000[prime] were used to assist in the classification of land use.(30) A total of 27 photographs were required for the study area, each covering approximately four square miles. Six of these photographs were interpreted to provide information on the amount of agricultural land that existed in a typical fringe area of Chicago [ILLUSTRATION FOR FIGURE 6 OMITTED].
Once the classification scheme was established visual interpretation methods were used in acquiring land use information from the satellite images. The SPOT images consisted of four spectral bands. Three of the spectral bands had a spatial resolution of twenty meters, which were resampled to the 10 meter resolution of the panchromatic band, band 4. Bands 1, 2, and 3 record reflected green, red, and near infrared wavelengths of the electromagnetic spectrum, respectively. Similar to black and white photography, the panchromatic band covers the visible portion of the electromagnetic spectrum. An advantage of using the panchromatic band for the analysis of land use change is its 10 meter resolution. The increased spatial resolution in the panchromatic band was especially useful for classifying land use through conventional visual interpretation techniques.
A comparison of the resulting 1987 and 1991 land use classifications, illustrates that a substantial amount of rural land uses were converted to urban land uses (Table 3). Residential land use dominated the types of land use change that occurred between 1987 and 1991. Figure 7a shows a largely rural section of the study area, as it appeared in 1987. The growth of residential development spilling over the 1980 urbanized area boundary is illustrated in Figure 7b. Just under 3 square miles, or 60 percent of the total change, was residential. Commercial land use change represented 14 percent of the change over the 4 year period. The study area itself is 73 square miles, which in 1987 consisted of just over 15 square miles of urban land and 58 square miles of nonurban land. Therefore, the total amount of nonurban to urban land use change between 1987 and 1991 was approximately five square miles (3,200 acres) or about 10 percent of the study area's nonurban land.
Table 3. Rural to Urban Land Use Conversions: 1987 to 1991
Resulting Number Square Land Use of Acres Miles
Residential 1,889 2.95 Commercial 475 0.74 Industrial 56 0.09 Other Urban 764 1.20 Total Urban 3,184 4.98
The amount of rural land in the United States devoted to cropland-decreased steadily from 1958 to the late 1970s. Both the NRIs and the agricultural censuses agree on the twenty year decline in cropland acreage. NRI data indicate that the declining trend reversed between 1977 and 1982, with more recent data showing that 1.6 million acres are added to the nation's cropland base each year. The Census of Agriculture presents essentially constant figures for cropland acres in recent years with approximately 2 million acres of cropland disappearing between 1982 and 1987.
The census data indicate considerable movement within the data set, however. Cropland is being lost to urbanization and other land uses, and the census figures show that cropland is shifting to western states. More than one-half of the counties in the 17 western have had increases in cropland between 1982 and 1987. One question that many researchers are approaching is whether or not urban expansion onto land previously in crops is necessarily damaging for American Agriculture? Some have suggested that high market value agriculture is at risk due to urbanization, while others have suggested that market value is a function of proximity to urban markets. Furthermore, since cropland is increasing, or at the very least remaining essentially constant, there simply may not be a problem with cropland conversions.
There may be no pervasive answer. Hart's example in New York state does not apply so readily in Illinois.(31) The soils and climate in the Chicago metropolitan region produce some of the world's highest yields of corn and soybeans. These deep mollisols over gently rolling topography are not fragile. Replacing acres lost in northeastern Illinois with cropland farther west will require more land and more input. It can be argued that in some cases the replacement may represent a non-sustainable agricultural system.(32) What we have shown here is that at the national level, the data are not the most appropriate for assessing urbanization's threat to the nation's cropland base. On the otherhand, satellite imagery has provided us with a look at the rapid pace in which high value farmland is being lost to urban encroachment within an important agricultural region.
Our findings also raise a number of policy questions that are difficult to address in the absence of a national policy designed to protect farmland from urban encroachment. The authors recognize that land-use controls are enacted and enforced at the municipal or county level and that it would be politically unpopular and difficult to administer land-use controls at the national level. Although, some states, including Florida, Vermont, and Oregon, are beginning to approximate this ideal by taking back some of the land-use controls traditionally delegated to local governments.(33) There also appears to be broad support for state purchase of development rights (PDR) programs, with the strongest support being voiced by the residents of fast growing counties.(34) Despite the advances made by a select group of states, urban encroachment continues to consume prime farmland on the fringes of many metropolitan areas. In the context of the earlier discussion of the complex relationship between the loss of metropolitan agriculture and the replacement lands of the arid west, it is imperative that a national perspective be introduced into the farmland conversion debate.
Acknowledgments: The authors wish to thank the anonymous reviewers for their helpful comments on this article. The research for this study was funded in part by the American Farmland Trust and the Northeastern Illinois Planning Commission.
1. Marlow Vesterby and Ralph E. Heimlich, "Land Use and Demographic change: Results from Fast-Growth Counties," Land Economics, 67 (1991):279-291.
2. American Farmland Trust, "Farming on the Edge," The Magazine of American Farmland Trust (Summer, 1993):10-17.
3. Vesterby and Heimlich, "Land Use and Demographic change: Results from Fast-Growth Counties."
4. Rutherford H. Platt, Land Use Control: Geography, Law, and Public Policy (Englewood Cliffs: Prentice Hall, 1991).
5. John M. Harlin and Richard P. Greene, "Cropland increases In Western States," Forum of the Association of Arid Lands Studies, 9 (1993):70-75.
6. American Farmland Trust, "Farming on the Edge"; Ralph E. Heimlich and Douglas H. Brooks, Metropolitan Growth and Agriculture: Farming in the City's Shadow (Economic Research Service, USDA, Report No. 619, 1989).
7. Vesterby and Heimlich, "Land Use and Demographic Change: Results from Fast-Growth Counties."
8. Rutherford H. Platt, "The Farmland Conversion Debate: NALS and beyond," Professional Geographer, 37 (1985): 433-444.
9. Platt, Land Use Control: Geography, Law, and Public Policy.
10. American Farmland Trust, "Farming on the Edge," pp. 12-13.
11. John Fraser Hart, "The Perimetropolitan Bow Wave," Geographical Review, 81 (1991):37-51.
12. Peter Hall, Von Thunen's Isolated State (Oxford, U.K.: Pergamon, 1966).
13. Heimlich and Brooks, Metropolitan Growth and Agriculture: Farming in the City's Shadow.
14. Ibid., p. 1.
15. Lloyd W. Bookout, "Neotraditional Town Planning: A Vision for the Suburbs," Urban Land (February 1992):20-26.
16. Heimlich and Brooks, Metropolitan Growth and Agriculture: Farming in the City's Shadow.
17. U.S. Bureau of the Census, "Revised Statistical Definitions for Metropolitan Areas," OMB Bulletin No. 93-05, Dec. 28, (1992).
18. American Farmland Trust, "Farming on the Edge."
19. Office of Management and Budget, "Revised Standards for Defining Metropolitan Areas in the 1990s," Federal Register, 55 (1990):12154-12160.
20. U.S. Bureau of the Census, 1987 Census of Agriculture on CD-ROM (Washington, DC: The Bureau of the Census, producer and distributor, 1990).
21. Platt, Land Use Control: Geography, Law, and Public Policy.
22. Paul M. Treitz, Phillip J. Howarth, and Peng Gong, "Application of Satellite and GIS Technologies for Land-Cover and Land-Use Mapping at the Rural-Urban Fringe: A Case Study," Photogrammetric Engineering & Remote Sensing, 58 (1992):439-448.
23. Joel Garreau, Edge City: Life on the New Frontier (New York: Doubleday, 1988).
24. John Fraser Hart, "Urban Encroachment on Rural Areas," Geographical Review, 66 (1976):1-17; Richard P. Greene and Dennis T. Pedersen, "Monitoring Urbanized Area Expansion with the Use of SPOT and TIGER," Proceedings of the ERDAS Northern Regional User Group Meeting, (1994): 1-24.
25. Office of Management and Budget, "Revised Standards for Defining Metropolitan Areas in the 1990s."
26. U.S. Bureau of the Census, TIGER/Line Census Files: Machine-Readable Data Files (Washington, DC: The Bureau of the Census, producer and distributor, 1991).
27. U.S. Bureau of the Census, Census 80: Continuing the Factfinder Tradition (Washington, DC: Bureau of the Census, 1980).
28. Heimlich and Brooks, Metropolitan Growth and Agriculture: Farming in the City's Shadow.
29. James R. Anderson, Ernest E. Hardy, John T. Roach, and Richard E. Witmer, "A Land Use and Land Cover Classification System for Use with Remote Sensor Data," Geological Survey Professional Paper 964 (Washington: GPO, 1976).
30. Northeastern Illinois Planning Commission, Northeastern Illinois Land Use Inventory, 1990 (Chicago: Northeastern Illinois Planning Commission, producer and distributor, 1993).
31. Harlin and Greene, "Cropland Increases In Western States."
32. Hart, "The Perimetropolitan Bow Wave."
33. John M. DeGrove and Deborah A. Miness, The New Frontier For Land Policy: Planning and Growth Management in the State (Cambridge, MA: Lincoln Institute for Land Policy, 1992).
34. Jeffrey Kline and Dennis Wichelns, "Using Referendum Data to Characterize Public Support for Purchasing Development Rights to Farmland," Land Economics, 70 (1994):223-233; Hubert B. Stroud, "Controlling Growth and Development in Monroe County, Florida," Land Use Policy, 11 (1994): 17-30.
Richard P. Greene is an Assistant Professor in the Department of Geography at Northern Illinois University. His areas of interest include land-use planning, geographic information systems, and the urban environment. He is working on several projects dealing with the land conversion process on the urban-rural fringe.
John M. Harlin is Professor and Chair in the Department of Geography at Northern Illinois University. Dr. Harlin has published several articles on agricultural soil loss and groundwater depletion due to irrigation. Current research includes surface and groundwater contamination and cropland conversions in the western states.
|Printer friendly Cite/link Email Feedback|
|Author:||Greene, Richard P.; Harlin, John M.|
|Publication:||The Social Science Journal|
|Date:||Apr 1, 1995|
|Previous Article:||The occupational mobility of black males revisited: does race matter?|
|Next Article:||"Tough laws:" policymaker perceptions and commitment.|