A.E.S. Bulletin No. 483 August, 1989
LAND USE TRANSITIONS IN DELAWARE, 1974-1984 John Mackenzie Department of Food & Resource Economics Agricultural Experiment Station University of Delaware Newark, DE 19717Abstract This bulletin develops and analyzes 21-category land-use change matrices for each county and for Delaware as a whole for the period 1974 to 1984. Mapped photo-interpretations of 1984 land uses and 1974-1984 land-use changes were digitized to determine aggregate acreages in each transition category. The major transitions are analyzed with respect to existing land-use policies and other published use transition statistics. Introduction The rapid pace of urbanization in Delaware is readily apparent to even the most casual observer of the state's landscape. In New Castle County, development in the Wilmington area has stretched the existing highway and water supply infrastructures to or beyond capacity. Heavy residential development along the Route 40 corridor reflects overflow development pressure from the Newark-Wilmington corridor. New Castle County has seen significant increases in acreage committed to commercial, industrial, transportation and other infrastructural uses. In Kent County, population and housing demands in the greater Dover area have expanded rapidly. The state is constructing a major highway bypass to alleviate severe traffic congestion on Route 13, the state's major north- south artery. In Sussex County, the explosion in demand for coastal recreation has fueled a boom in second home construction along the Atlantic coast and around the Inland Bays. Development has generated environmental problems in the coastal zone. Leaching sewage has degraded water quality in the Inland Bays, and increased groundwater withdrawals are causing significant saltwater intrusion into the principal coastal aquifer. To the west, the growth of the poultry industry has motivated significant conversions of land from forest to cropland to supply the industry with feedgrains. Despite the fact that these development pressures are so obvious, the aggregate land-use changes Delaware has undergone have not really been analyzed in a comprehensive way. This bulletin develops a simple accounting framework for quantifying land use transitions and analyzes some of the major land use changes which occurred in Delaware between 1974 and 1984. It is important to distinguish between gross changes and net changes in land use categories. For example, it is obvious that significant amounts of cropland have been converted to residential uses; however it is not so obvious that some land is also being converted from other uses to cropland. Farmland conversion has been an important policy focus, since both the long- run health of Delaware's agriculture and a range of more general environmental amenities depend on the maintenance of an adequate farmland base. The public tends to focus on the highly visible gross losses of farmland to urbanization; we should be looking at net losses. This analysis shows that the net loss of Delaware farmland between 1974 and 1984 was significantly smaller than the gross loss. Technical Background The Agricultural Lands Preservation Section of the Delaware Department of Agriculture has compiled a large data base to support its Land Evaluation and Site Assessment (LESA) program. In 1984, the Department contracted with the Earth Satellite Corporation of Chevy Chase, Maryland, to analyze aerial photographs of Delaware taken in 1974 and 1984 and to construct land use change maps as well as 1984 land-use/land-cover maps for each of Delaware's three counties. The main body of data presented in this pamphlet is calculated from these maps, copies of which the Department provided to the author. The 1984 land-use/land-cover maps provide a comprehensive categorization of all land in each county. The categories used in this analysis are similar to USGS's standard classification system (Anderson, 1976; USGS 1992): LULC Categories 1974-84 Delaware LULC Mapping Code(s) USGS categories(s) Undifferentiated Urban 10 = 16, 17 Residential 11, 12, 13, 14 = 11 Commercial/Services/Inst. 15 = 12 Industrial 16 = 13 Transp./Comm./Utilities 17 = 14 Recreation 18, 19 = 16, 17 (in mixed/other urban) Agriculture (confined feeding) 21 = 23 Agriculture (other) 22 = 21, 22, 24 Brushland 30 = 32, 33 (shrub-brush/mixed rangeland) Forest (decid.,conif.,mixed) 41, 42, 43 = 41, 42, 43, 61(forested wetland) Wetland 50 = 62 (non-forested wetland only) Water 61, 62 = 51, 52, 53, 54 Beach/Barren 71, 72 = 72, 73, 76, 77 Other Open/Urban 80 = 17 The 1974-1984 land use change maps show both 1974 and 1984 use categorizations for acreages which underwent use changes during the period. Unmarked land on these maps did not undergo a use change. Both maps employed a scale of one inch per mile (1:63,360). It should be stressed that these maps reflect interpretations of aerial photographs. The use categories employed by Earth Satellite's interpreters have proved reliable when photo interpretations were verified against "ground truth" data in other studies, but there is generally some minor residual interpretation error, particularly when different people are involved in the interpretation process. Procedure The following accounting of Delaware's land use transitions employs a singular matrix form. Consider two alternative land uses denoted i and j, where aij represents the acreage which converted from use i in 1974 to use j in 1984, aji represents acreage converted from use j to use i during the same period, and aii and ajj represent acreages remaining in uses i and j respectively during the period. The matrix formed from these elements is aii aji aij ajj where the diagonal elements represent acreages in each use not undergoing any use change. Unidirectional development, where aij > 0 and aji = 0, implies a triangular matrix. The ordering of the land use codes employed here reflects the general directionality of development in Delaware, so that the matrices presented below do tend toward triangular. The area of each parcel undergoing a land use change was measured on the land use change maps by digitizer, and the 1974 and 1984 land uses for each parcel were recorded. The parcel areas were summed and converted to acre measures for each use transition to obtain the 420 off-diagonal elements aij of each county's land use transition matrix. The acreage converted from each 1974 land use i to each 1984 use j is presented in Appendix Tables 1.1 (Delaware), 1.2 (New Castle County), 1.3 (Kent County), and 1.4 (Sussex County). Blank spaces indicate zero acreage in a given transition. The 21 diagonal elements aii, ajj, ... are not shown in these tables. The row sums for each 1974 use represent the gross loss of acreage from that use (TOTLOSS). The column sums for each 1984 use represent the gross gain in acreage to that use (TOTGAIN). The net change in acreage in each use is the difference between the gross gain and the gross loss (NETCHG). This analysis neglects likely minor changes in total county land areas attributable to shore erosion or deposition. It assumes that the total area of each county remained constant over the 1974-1984 period: the sum of each gross loss column equals the sum of the corresponding gross gain column, and the sum of each net change column is zero. Appendix Table 1.1 (all of Delaware) was constructed by matrix addition of all but the last columns of Appendix Tables 1.2, 1.3 and 1.4. All discrepancies (0.1 acres and less) between Appendix Table 1.1 and the element sums of the three county tables are due to rounding error. The amount of land in each 1984 use was measured on the 1984 land- use/land-cover maps by digitizer. Individual parcel areas (or parcel sections for complex areas) were measured and summed by use to obtain total acreages in each land use category in 1984. Grand totals of acreages across all uses were compared against Census statistics on total land areas for each county (State and County Data Book). In the case of New Castle County, water acreage in the Delaware River was excluded from the analysis. Discrepancies between the Census acreage statistics and the sums from the digitized data were less than two percent, implying a linear measurement error of less than one percent. Each county's acreage figures for each land use category were slightly re-scaled so that the total acreages would correspond exactly to the published data. Since data on acreages in each 1974 use were lacking, these data were calculated back from the 1984 acreage measures (N84ACRES), where N74ACRES = N84ACRES - NETCHG. Finally, the net percent change in acreage in each land use category was calculated from a 1974 base (PCTCHG). Appendix Tables 2.1 through 2.4 show complete acreage transition matrices, including diagonal elements aii, ajj, ... representing 1974 acreages remaining in the same use over the 1974-1984 period. These matrices provide a complete two-period accounting of Delaware's land base. Appendix Tables 3.1 through 3.4 show percent changes from 1974 acreages. Each element shows the percent of the acreage in 1984 use j which was converted from 1974 use i, with the column sums corresponding to total percent changes in acreage in each use category since 1974. ("ERR" notations in the percent change matrices arise from division by zero, reflecting zero acreage in a 1974 use category.) Analysis The land use transition matrices offer insights into the processes of urbanization and related land-use changes in Delaware. The concentration of acreage figures below the matrix diagonals indicates the generally uni- directional nature of development. For example, compared with the conversion of land from agriculture or forest to urban uses, the reverse flow of acreage from urban uses to forest or agriculture is small. Comparisons of net changes in each land use category are summarized in Table 1 below. ____________________________________________________________________________ Table 1: Net Acreage Transitions, 1974--1984, by Land Use Category, County and Statewide USE USECODE NEW CASTLE KENT SUSSEX DELAWARE RESIDENTIAL--single fam/duplex 11 4492.2 6836.5 4539.5 15868.2 RESIDENTIAL--multi-unit 12 505.0 145.9 215.0 866.0 RESIDENTIAL--mobile home 13 36.5 234.2 1629.4 1900.1 RESIDENTIAL--other 14 39.7 71.7 230.4 341.8 COMMERCIAL 15 740.5 1278.1 591.4 2609.9 INDUSTRIAL 16 900.5 279.7 532.5 1712.6 UTILITY/TRANS/COMMUNICATION 17 259.2 78.7 -26.9 311.1 RECREATION--water related 18 0.0 0.0 0.0 0.0 RECREATION--other 19 397.4 -1.3 300.8 696.9 AGRICULTURE--confined feeding 21 0.0 177.3 7.0 184.3 AGRICULTURE--other 22 5489.3 -6441.6 4801.9 -7129.0 BRUSHLAND 30 1461.8 -602.9 11276.2 12135.0 FOREST--deciduous 41 -970.2 -1164.2 -5414.4 -7548.8 FOREST--coniferous 42 -132.5 -180.5 -11467.5 -11780.5 FOREST--mixed 43 -105.0 -695.7 -6765.5 -7566.1 WETLAND 50 -1502.1 -2713.0 -166.4 -4381.4 WATER--river/canal/estuary/bay 61 0.0 6.4 0.0 6.4 WATER--pond/lake/reservoir 62 67.8 2978.6 15.4 3061.8 BARREN--beach 71 0.0 0.0 -86.4 -86.4 BARREN--other 72 -602.2 -76.2 448.0 -230.4 OTHER OPEN/URBAN 80 -99.2 -211.8 -660.5 -971.5 _____________________________________________________________________________ Residential land: During the period 1974-1984 New Castle County converted a net total of 5,074 acres into residential uses (categories 11 through 14). Of this, 88.5 percent was converted to the single-family/duplex residential category. About 59 percent of the land in New Castle County which was converted into residential uses came from agriculture; another 24 percent was converted from deciduous forest. Kent County converted a net total of 7,335 acres into residential uses. Of this, about 84 percent came from agriculture; about 11 percent came from forest categories. Sussex County converted a net total of 6,614 acres into residential uses. Of this, 57 percent came from agriculture and 24 percent came from forest categories. The fact that New Castle County converted less land into residential use than either Kent County or Sussex County reflects New Castle County's significantly lower rate of population growth, and higher housing density (housing units per acre) compared to Kent and Sussex Counties. As Table 2 indicates, Delaware has undergone a southward population shift. Although New Castle County has by far the largest population, its estimated population increase between 1974 and 1984 was only 7,400 people, versus estimated increases of 11,360 people in Kent County and 16,400 people in Sussex County over the same decade (Delaware Population Consortium). The 1974-1984 period shows relative lulls in population growth in all three counties; these growth rates have accelerated markedly since 1984. Between 1980 and 1990 population is projected to increase by 42,185 people in New Castle County, by 14,131 in Kent County, and by 22,846 in Sussex County. _________________________________________________________________________ Table 2: Delaware Population Growth and Population Projections, by County and Statewide -Percent Changes- 1970 1974 1980 1984 1990 70-80 74-84 80-90 New Castle 385,856 400,200 398,115 407,600 440,300 3.2 1.8 10.6 Kent 81,892 89,800 98,219 101,160 112,350 19.9 12.7 14.4 Sussex 80,356 87,400 98,004 103,800 120,850 22.0 18.8 23.3 Delaware 548,104 577,400 594,338 612,600 673,500 8.4 6.1 13.3 Sources: Delaware Population Consortium, 1989; 1980 Census of Population. _________________________________________________________________________ Urbanization implies growth of interrelated land uses: for example, commercial expansion generally depends on growth in local population and income; conversely, population growth supports new commercial, residential and infrastructural development. Since residential development accounts for most of the acreage transitions to urban uses, it is helpful to compare the scale of other land use transitions to residential conversions. Table 3 shows the proportions of a "typical" acre converted to residential use in each of the four residential use categories, and shows the net urban and rural acreage conversions which were associated with this typical one acre of residential development over the 1974-1984 period, for each county and statewide. While there are some sizeable variations between counties in some of the associated acreage conversion figures, these figures provide some idea of the interdependencies between residential development and other urban land uses vis-a-vis the rural land base. ___________________________________________________________________________ Table 3: Acreage Transitions Associated with a 1-Acre Conversion to Residential Use, by County and Use Category, 197--1984 USE USECODE NEW CASTLE KENT SUSSEX DELAWARE (COMPOSITION OF NEW RESIDENTIAL DEVELOPMENT) RESIDENTIAL--single fam/duplex 11 0.89 0.94 0.69 0.84 RESIDENTIAL--multi-unit 12 0.10 0.02 0.03 0.05 RESIDENTIAL--mobile home 13 0.01 0.03 0.25 0.10 RESIDENTIAL--other 14 0.01 0.01 0.03 0.02 ALL RESIDENTIAL CATEGORIES 1.00 1.00 1.00 1.00 (ASSOCIATED ACREAGE CONVERSIONS) COMMERCIAL 15 0.15 0.18 0.09 0.14 INDUSTRIAL 16 0.18 0.04 0.08 0.09 UTILITY/TRANS/COMMUNICATION 17 0.05 0.01 0.00 0.02 RECREATION--water related 18 0.00 0.00 0.00 0.00 RECREATION--other 19 0.08 0.00 0.05 0.04 AGRICULTURE--confined feeding 21 0.00 0.02 0.00 0.01 AGRICULTURE--other 22 -1.08 -0.88 0.73 -0.38 BRUSHLAND 30 0.29 -0.08 1.70 0.64 FOREST--deciduous 41 -0.19 -0.16 -0.82 -0.40 FOREST--coniferous 42 -0.03 -0.02 -1.73 -0.62 FOREST--mixed 43 -0.02 -0.10 -1.02 -0.40 WETLAND 50 -0.30 -0.37 -0.03 -0.23 WATER--river/canal/estuary/bay 61 0.00 0.00 0.00 0.00 WATER--pond/lake/reservoir 62 0.01 0.41 0.00 0.16 BARREN--beach 71 0.00 0.00 -0.01 0.00 BARREN--other 72 -0.12 -0.01 0.07 -0.01 OTHER OPEN/URBAN 80 -0.02 -0.03 -0.10 -0.05 -1.00 -1.00 -1.00 -1.00 ___________________________________________________________________________ Other Urban Categories: The pattern of net increases of acreage in commercial uses (category 15) in each county closely parallels the net increases in residential acreage (categories 11 through 14). On average, statewide, one acre is transferred to commercial use for every 7.27 acres converted to residential use. New Castle County shows the largest net acreage increases in industrial and infrastructure uses (categories 16 and 17). Kent County shows the smallest increase in acreage in industrial use. On average, statewide, one acre is transferred to industrial use for every 11.1 acres converted to residential use. New Castle County saw considerably more acreage converted to infrastructural use (utility, transportation and communication, category 17) than either Kent or Sussex County. New Castle County converted about one acre of land to infrastructure for every twenty acres converted to housing; corresponding conversions per residential acre developed in Kent and Sussex counties were much smaller. This difference suggests that New Castle County may be experiencing diseconomies of scale in providing and maintaining infrastructure in the face of high-density urbanization. Furthermore, infrastructural needs are not always tied to population growth: the City of Wilmington has experienced a population decline, but has also experienced significant infrastructural growth to support its growing industrial and commercial base. Agriculture: Between 1974 and 1984 Delaware had a net loss of 6,945 acres of farmland (categories 21 and 22), which represents a 1.2 percent decline in the total 1974 farmland base of 595,901 acres. While 20,403 acres of farmland were converted from agriculture to other uses (mostly urban), 13,458 acres were converted from other uses (mostly forest and brushland) into agriculture. New Castle County experienced a 5.17 percent net decline in farmland, losing 5,489 acres; Kent County experienced a 3.09 percent net decline, losing 6,264 acres of farmland; Sussex County, with the largest 1974 farmland base, experienced a 1.71 percent net increase in farmland, gaining 4,809 acres. While cropland supplies the largest share of the total acreage converted to urban uses in Delaware, and is perceived by the casual observer of the Delaware landscape to be dwindling rapidly, these losses are small relative to the size of the underlying farmland base, and do not themselves constitute a threat to the long-term viability of Delaware's agriculture. On the other hand, maintaining the open-space values of farmland is a more serious concern. The agricultural land which is converted to development tends to have the highest open-space value, because of its proximity to roads and other development. A review of statistics from the 1978, 1982 and 1987 Censuses of Agriculture indicates large additions to the state's cropland base between 1974 and 1978, followed by steady declines since 1978. The overall resilience of Delaware's farmland base is attributable to several factors. First, the 1970's saw a dramatic increase in local demand for feedgrains to support the growing poultry industry. This has generated a locational premium in grain prices received. Second, preferential taxation of farmland under the Farmland Assessment Act of 1968 (through which farmland enrolled under the Act is taxed according to current use value rather than market value; see Glendenning, Cole and Mackenzie, 1988) serves as a disincentive to development of farmland; enrolled farmland is subject to a rollback tax if it is subsequently developed. Third, the concentration of development pressures above the Chesapeake and Delaware Canal, around Dover and along the Sussex coastline, leaves most of the state's farmland relatively unthreatened by development. Finally, developers tend to maintain speculative land-holdings in agriculture (enrolling it in the Farmland Assessment Program to reduce taxes, and renting it to local farmers) until development is propitious. Over 78 percent of the cropland in New Castle County is currently rented to, rather than owned by, farm operators; 60 percent of Kent County's cropland and 58 percent of Sussex County's cropland are rented. In 1982, less than 55 percent of Delaware's cropland was rented; in 1987, about 62 percent was rented (1987 Census of Agriculture). Brushland and Forest: In contrast to the resilience of the state's cropland, Delaware has seen significant net losses of acreage from forest uses (categories 41, 42 and 43), mostly to agriculture and brushland (category 30). Brushland increased by 12,135 acres (a 48.9 percent increase) between 1974 and 1984, reflecting significant clear-cutting of forest and some abandonment of agricultural land. Brushland tends to be a transitional category; much of the acreage converted out of brushland between 1974 and 1984 regenerated into forest or was cleared for agriculture. Because of its forest cover, a considerable amount of non-coastal wetland is included in the forest categories. Deciduous forest acreage underwent a net decline of 7,549 acres (3.6 percent of the 1974 base), with most of the decline occurring in Sussex county. About half of this acreage was converted to brushland, 25 percent to agriculture, and 20 percent to residential use. Coniferous forest acreage underwent a net decline of 11,781 acres (15.4 percent of the 1974 base). Virtually all of this decline occurred in Sussex County, with about 70 percent of this acreage converted to brushland and 25 percent converted to agriculture. Mixed deciduous-coniferous forest acreage underwent a decline of 7,566 acres (7.1 percent of the 1974 base). Again, virtually all of this decline occurred in Sussex County, with about two-thirds of the converted acreage going to brushland and 30 percent going to agriculture. The large loss in coniferous forest continues a declining trend that has been observed for several decades (Ferguson and Mayer, 1974). There are two reasons for this decline. First, most of Delaware's forest land naturally tends to mixed pine-hardwood forest, and pure pine stands generally require some control of hardwood growth. Second, pulpwood is Delaware's major commercial forest product, and is most profitably harvested from coniferous forest rather than mixed forest. Demand for Delaware's hardwoods is chronically weak, and hardwood stands are comparatively under-harvested. Forested land in farms enjoys preferential tax assessment under Delaware's Farmland Assessment Act of 1968. Other forest plantations may receive 30-year tax exemptions (7 Del. Code 3501-3508). There has been little interest in these latter exemptions. Since agricultural uses are given equivalent preferential tax treatment, these tax provisions do not represent any significant disincentive to conversion of forest to agriculture. Wetlands: Delaware experienced a net loss of 4,381 acres of wetland (category 30, mostly non-forested coastal wetland), a 4.5 percent decline. Since most of Delaware's inland wetland was counted as forest, net declines in this land use could not be quantified. Over 60 percent of the state's net decline in coastal wetland occurred in Kent County. These losses occurred despite passage of Delaware's two major wetland preservation initiatives. Industrial development of coastal wetlands is regulated by Delaware's Coastal Zone Act of 1971, which prohibits new heavy industry and restricts other manufacturing development within two miles of the coast. Dredging and filling of any wetland (coastal or inland) is restricted under Delaware's Wetlands Act of 1973. Net losses of wetlands would probably have been larger without these laws. Nevertheless, this record falls far short of current Federal "no net loss" wetland policy objectives (Heimlich and Vesterby). Other Rural Categories: Acreage transitions in and out of the other rural land use categories are of small magnitude. Delaware experienced no substantive change in open water acreage (category 61), and a net increase of 3,281 acres in confined water acreage (category 62, a 53 percent increase from a small acreage base). There were minor net declines in beach (category 71) and other barren acreage (category 72). Conclusions This analysis has provided a brief overview of a decade of land use transitions in Delaware, and offers some insights into the current development trends affecting Delaware's land resource base. The state's population is currently increasing by about 1.8 percent annually (Delaware Population Consortium, 1989), intensifying development pressures and urban congestion in several areas. The high visibility of this continuing urbanization leads to public misperceptions about its scale relative to the rural land base. In particular, Delaware's farmland is perceived to be seriously threatened by urbanization, since a high proportion of the acreage converted to urban uses is former farmland. In fact, agriculture is an intermediate land use on the economic spectrum: while market forces convert significant farmland acreage to urbanization on one side, they also convert brushland and forest to agriculture on the other. Farmland generally commands higher market prices than the forest and brushland it displaces, and the state's agriculture appears to have considerably better long-run economic viability than its forestry industry. The land use transition patterns in Delaware which are described here are consistent with patterns of urbanization in the U.S. generally. Frey and Hexem (1974) noted that, on a national scale, "changes in cropland and pasture acreages were barely perceptible during 1978-82." Vesterby and Heimlich (1989) summarized a more recent analysis of land use changes at about 27,000 points on paired aerial photographs of fast-growth counties (defined as counties with 1970-80 population increases of at least 25,000 people and at least 25 percent). Their findings confirm the fact that, while a high proportion of acreage converted to urban uses is removed from agriculture, these losses are very small relative to the farmland base. As Figure 3 indicates, between 1974 and 1984 Delaware lost almost four times as much forest land as farmland, and about three-fifths as much wetland as farmland. A significant portion of the lost forest acreage which was converted by clear-cutting to brushland will be replanted or will regenerate naturally; however permanent losses of forest still appear to exceed farmland losses. This analysis suggests that the losses of forest and wetland acreage, and the loss of the watershed, wildlife and other environmental amenities which they support, deserves just as serious attention from planners and policy-makers as the loss of farmland and its environmental amenities. References Delaware Population Consortium. 1987. Delaware Dataline: Population Projections, 1987 Version. Mimeo. Delaware Data Center, College of Urban Affairs, University of Delaware, Newark, DE. Delaware Population Consortium. 1989. Delaware Dataline Vol. 3, No. 1. Delaware Data Center, College of Urban Affairs, University of Delaware, Newark, DE. Ferguson, Roland H. and Carl E. Mayer. 1974. The Timber Resources of Delaware. USDA Forest Service Bulletin NE-32, Northeastern Forest Experiment Station, Forest Service, USDA, Upper Darby, PA. Frey, H. Thomas and Roger W. Hexem. 1985. Major Uses of Land in the United States: 1982. Agricultural Economic Report No. 535. Natural Resource Economics Division, Economic Research Service, USDA, Washington, DC. Frieswyk, T.S. and D.M. DiGiovanni. 1989. Forest Statistics for Delaware. Resource Bulletin NE-109, Northeast Forest Experiment Station, U.S. Forest Service, Broomall, PA. Glendenning, Roger, Gerald L. Cole and John Mackenzie. 1989. The Impacts of Delaware's Farmland Assessment Act. Agricultural Experiment Station Bulletin No. 470, University of Delaware, Newark, DE. Heimlich, Ralph E. 1989. Personal communication regarding correction of category 80 inconsistencies. Heimlich, Ralph E. and Marlow Vesterby. 1989. "Conversion of Wetlands to Urban Uses: Evidence from Southeastern Counties." Review draft. RTD/ERS/USDA, Washington, DC. Lesher, W.G. 1975. Land Use Legislation in the Northeast: Delaware. NE-90 Project Report. A.E. Res. 75-20, Dept. of Agricultural Economics, Cornell University, Ithaca, NY. Mackenzie, J. 1988. "The Effectiveness of Use-Value Assessment as a Farmland Preservation Policy." Pp. 353-360 in Quinlan, R. (ed.) Planning for the Changing Rural Landscape of New England: Blending Theory and Practice. The New England Center, Durham, NH. Tiner, R.W. 1985. Wetlands of Delaware. U.S. Fish and Wildlife Service, National Wetlands Inventory, Newton Corner, MA. U.S. Bureau of the Census. 1980 Census of Population: Number of Inhabitants-- Delaware. U.S. Department of Commerce, Washington, DC. January, 1982. U.S. Bureau of the Census. 1980 Census of Population: General Population Characteristics--Delaware. U.S. Department of Commerce, Washington, DC. June, 1982. U.S. Bureau of the Census. 1980 Census of Housing: General Housing Characteristics--Delaware. U.S. Department of Commerce, Washington, DC. July, 1982. U.S. Bureau of the Census. 1982 Census of Agriculture: Geographic Area Series--Delaware. U.S. Department of Commerce, Washington, DC. June, 1984. U.S. Bureau of the Census. 1987 Census of Agriculture: Geographic Area Series--Delaware. U.S. Department of Commerce, Washington, DC. June, 1989. U.S. Bureau of the Census. County and City Data Book, 1988. U.S. Department of Commerce, Washington, DC. Vesterby, Marlow. 1988. Land Use Change in the 1970s. ERS Standard Data Product No. 88018. RTD/ERS/USDA, Washington, DC. Vesterby, Marlow and Ralph E. Heimlich. 1989. "Land Use and Demographic Change: Results from Fast Growth Counties." Contributed paper, Population Association of America. RTD/ERS/USDA, Washington, DC. Acknowledgements Thanks to the following people for their help with various aspects of this bulletin: Kevin Donnelly and Michael McGrath of the Agricultural Lands Preservation Section, Delaware Department of Agriculture, for providing the maps on which this analysis is based, and for their initial encouragement; Terry Dunne and Ellen Whaley, summer research associates, for their patient work with the digitizer; Betsy Olmstead, microcomputer consultant in the College of Agricultural Sciences, for designing the graphics; Ralph Heimlich and Marlow Vesterby of the Land Branch, Resources and Technology Division, ERS/USDA, for the advice they provided in constructing and interpreting the land use change matrices; and Bill Lessley, Ralph Heimlich, Marlow Vesterby, Larry Libby, Paul Barkley, Doug Morris, Jerry Vaughn, Kevin Donnelly and Mike McGrath for their helpful reviews of earlier drafts of this bulletin. It is the policy of the University of Delaware that no person shall be subjected to discrimination on the grounds of race, color, religion, sex, national or ethnic origin, age, handicapped or veteran status.