Land information is most useful if spatially referenced. One way of accomplishing this is to use land-ownership parcels as the unit of observations, as described in Chapter 5. This allows comparison among parcels or among aggregations of parcels. Alternatively, land information can be organized into other types of homogeneous units, which, like parcels, are observed as polygons having uniform character. Or an arbitrary grid can be imposed over the mapped data and land characteristics then attributed to each grid cell. First, we examine land-ownership parcel schemes for spatial referencing.
Among the wide range of Geographic-Information Systems (GIS) described in Section 1.4.4, the systems that attribute land data to the visual center of homogeneous polygons or grid cells are classified as recording land digitally in a discrete rather than a continuous manner. Discrete systems contain data on units of observation, say for parcels, city blocks, or homes, but the emphasis is on the units and not on how they relate to each other at their boundaries. Each unit is treated independently; the boundaries of the units are not described. On the other hand, continuous systems are digital maps that partition the land space with points, lines, and areas describing the spatial extent and juxtapositions of land parcels or other natural and cultural features that make up the landscape.
Parcel-related records organized by parcel index numbers constitute a discrete information system. The records are often processed without reference to a cadastral overlay, which locates the parcel boundaries. Discrete parcels serve as units of observation in the system. For limited purposes, discrete Land-Information Systems are effective and relatively easy to use.
Land-parcel data can be considered continuous rather than discrete when keyed
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 100
6 The Evolving Land-Information Environment Land information is most useful if spatially referenced. One way of accomplishing this is to use land-ownership parcels as the unit of observations, as described in Chapter 5. This allows comparison among parcels or among aggregations of parcels. Alternatively, land information can be organized into other types of homogeneous units, which, like parcels, are observed as polygons having uniform character. Or an arbitrary grid can be imposed over the mapped data and land characteristics then attributed to each grid cell. First, we examine land-ownership parcel schemes for spatial referencing. Among the wide range of Geographic-Information Systems (GIS) described in Section 1.4.4, the systems that attribute land data to the visual center of homogeneous polygons or grid cells are classified as recording land digitally in a discrete rather than a continuous manner. Discrete systems contain data on units of observation, say for parcels, city blocks, or homes, but the emphasis is on the units and not on how they relate to each other at their boundaries. Each unit is treated independently; the boundaries of the units are not described. On the other hand, continuous systems are digital maps that partition the land space with points, lines, and areas describing the spatial extent and juxtapositions of land parcels or other natural and cultural features that make up the landscape. Parcel-related records organized by parcel index numbers constitute a discrete information system. The records are often processed without reference to a cadastral overlay, which locates the parcel boundaries. Discrete parcels serve as units of observation in the system. For limited purposes, discrete Land-Information Systems are effective and relatively easy to use. Land-parcel data can be considered continuous rather than discrete when keyed
OCR for page 100
to a map of parcel boundaries. The map locates the data over continuous space rather than just by discrete parcels. This continuous framework enables integration of land parcels with other continuous land data. 6.1 INTEGRATION OF DATA THROUGH SPATIAL REFERENCES The geodetic framework of a cadastre provides for integration of land-ownership information with other information, if the latter is spatially registered according to the same geometric framework with some indication of positional accuracy. This ability to relate data spatially is a powerful tool in the analysis and management of land and activities. As Figure 3.1 illustrated, the overlaying of layers of data provides an ability to meet a wide range of diverse needs. Manual overlay of map products meets many needs for mapped information but depends on human integration of the composited information. Manual overlaying is limited by the content choices of the various layers made by the map designers. Users can select layers to composite for their particular application, but any variation in scale must be accomplished by photographic enlargement or reduction, and variation of map content or data categories requires remapping. In digital form, users have a choice of the themes or layers to composite and a choice as to detail and ranges in terms of the content. In addition, choice of scale is accomplished mathematically. Base mapping according to national mapping standards is central to the multipurpose cadastre concept. This enables spatial registration of data layers. Spatial registration of the map graphics would not be an issue if locations of all land data were recorded as numerical field measurements, as they are for property boundary corners (see Section 4.2.1). However, because many users will continue to use aerial photo imagery as the source of their location data, especially for natural phenomena, determining their coincidence in space will continue to depend on accurate two- or three-dimensional plotting. Adherence to base-mapping standards avoids subsequent problems of separately mapped data not correctly relating spatially. Poor-quality base mapping or attempts to collect spatial data from uncontrolled maps will create spatial registration problems. To a certain extent these problems can be anticipated and dealt with. This may mean visual and manual reconciliation of spatial inconsistencies on a single base map and then digitization of the manually reconciled spatial data. Alternatively, each data layer or theme can be digitized from whatever source maps are available, and then the data can be rotated, transformed, and scaled to fit. This may require considerable editing in terms of redigltizing of point and line data to achieve fitting. A major problem is “slivers and gaps” that occur (1) when common boundaries between polygons of a single layer are separately digitized; (2) when boundaries are common across themes, such as where a property boundary is a street right-of-way; or (3) where a political boundary is a natural feature.
OCR for page 100
A multipurpose cadastre offers relief from most of the spatial registration problems in that all data are reconciled to a common base before or as part of the digitizing process. Nevertheless, inexactness in data-capture technologies requires care in the capture of spatial data, so as to retain the spatial registration in digital form. In sum, spatial registration provides greater flexibility in the digital data-capture process. Without spatial registration one is forced to compile separate data for the same base and then digitize or resolve discrepancies among different bases. With registration, coordinate data for different layers or themes can be related directly. One does not have to go from digital spatial data to a map product and back to a digital data base. Instead, the digital data from orthophoto map generation can be incorporated directly into the cadastre along with the coordinates from property descriptions. 6.2 APPLICATION OF GEOGRAPHIC-INFORMATION-SYSTEM CONCEPTS TO LAND INFORMATION A digital base map and a cadastral overlay are the key elements of a multipurpose cadastre, which enable it to become the basis for a powerful Geographic-Information System. An information system consists of a data base, with the necessary input, storage, retrieval, and output technologies responsive to nonroutine queries. A GIS is a special case where the data base is a digital map or consists of observations on spatially referenced features or activities, which are definable in space as points, lines, or areas. A GIS manipulates these spatial data to retrieve data for queries and analyses (Dueker, 1979). Digital maps generally are formatted as either vector or grid data. Vector data describe areal features as polygons and linear features as line segments, both composed of digitized points. Grid data partition land space into a regular lattice with location specified by address or row and column numbers. Vector-format digital maps are employed for engineering, utility, and tax map applications, while gridformat digital maps are employed for thematic mapping and resource-analysis applications. Vector data in the form of polygon encoding of coordinates capture geometric shape and location of features. Vector-format data in the form of topological facts and metric location and shape create an even more explicit digital map. The elementary objects in two-dimensional topology are points, lines of any shape, and areas. The relations among them are the incidence of areas and points separated by lines. A topological data structure enables the construction of a consistent digital map that contains relations among features, such as to select areas bounded by specific lines or lines that end at specific points (White, 1982). Topology provides capability to edit vector data and ensure logical consistency. Topologically structured vector data are essential in the creation of large digital map files.
OCR for page 100
When producing line maps from selected layers and items, vector-format data are used most often. When performing analyses that require relating data across layers, it is usually more convenient to convert from vector format to grid format. The user can specify the appropriate resolutions for the quality of the data or the resolution requirements of the analysis. This establishes the grid cell size at which the vector-described areas are resolved, so that corresponding cells can be compared to determine the interrelationship of factors, say, for example, land use by soil type and by ownership type. The resultant overlay of these three factors enables area measurements and production of maps showing, say. vacant parcels of 40 to 160 acres with class II soil that have changed hands in the last 2 years. The polygon-to-polygon comparison of one polygon set to another polygon set is computationally cumbcrsome—calculating intersections and keeping track of new polygons. Gridding or rastering the polygon sets and comparing and tabulating corresponding cells is computationally more efficient than direct overlaying of polygons. The gridding of polygons prior to overlaying avoids the problems of slivers and gaps that result from imprecision in digitizing layers separately. A similar but even more efficient technique uses horizontal scan lines that intersect the polygons to perform the overlay. 6.3 EXCHANGES OF DATA BETWEEN LAND-INFORMATION SYSTEMS Implementation of a multipurpose cadastre greatly simplifies the spatial collation of data. It not only eliminates the inherent duplication of mapping among utilities and governmental jurisdictions but also facilitates assembling composites, for example, of sewer, water, power, and gas lines or of ownership, floodplain, and agricultural land. Notwithstanding these potentials, there exists in the short term a need to relate data compiled on different base maps and with inconsistent control. This might occur when land-resource data, say floodplains or agricultural land compiled on 1:24,000 U.S. Geological Survey 71/2-min quadrangle maps, are related to land-ownership data compiled at 1:1000 to 1:5000 scale. The man-made boundaries can be defined with precision without unreasonable expense. It is difficult to define most natural boundaries, as a practical matter, with the same level of precision. Care must be exercised in relating data from different sources. Table 6.1 illustrates the problems with transferring data between the different map scales normally used for cadastre and for resource thematic systems. This requires judicious choice of resolution of grid cell size so as not to lead to false accuracy assumptions or inaccurate allocations. This problem is a direct result of an order-of-magnitude difference in the scales at which the data were compiled. Resource thematic data such as soils and floodplain boundaries, are normally compiled at map scales between 1:10,000
OCR for page 100
TABLE 6.1 Problems with Exchange of Data between the Cadastre and Resource Thematic Systems Applications Cadastral Data Systems Resource Thematic Data Systems Scale 1:1000 1:10,000 Natural Boundaries (soils. floodplains) Man-made Boundaries (parcels. political jurisdictions, building locations) and 1:100,000. Transferring these already imprecise boundaries, whether by hand or by computer, to a cadastral mapping scale (1:1000 to 1:5000) implies a higher accuracy than warranted, which may create erroneous information relating to specific parcels of land. The solution to this problem lies in compiling the resource data at the scale needed for cadastral applications—an expensive approach. Otherwise, the boundaries must be drawn implying uncertainty, i.e., an imprecise, wide line. A similar problem exists in transferring exact boundary data from the cadastre for use in resource thematic systems for environmental applications. This problem has two components. One problem component is that the process of converting coordinate location from one scale or projection to another may place the transferred data in an erroneous position with respect to data already compiled on the resource thematic map. Second, the volume and detail at the cadastre scale may not be needed for resource thematic applications. Smoothing or aggregation or both of the detailed cadastre data may be warranted. A primary mode of analysis for environmental data is to overlay one factor with one or more other factors, such as soil type with land-ownership parcels. Choice of a cell size or scan-line interval is the means by which an overlay that relates inexact boundaries to exact boundaries can be accomplished. When overlaying imprecise boundaries, the analyst should select a large cell size, which will reduce the chance that a cell will be assigned to a polygon incorrectly. As a rule of thumb, the cell size or scan-line interval should be greater or equal to the width of the most imprecise line. Figure 6.1 illustrates the problem of ambiguity of cell assignment when the width of the imprecise line is greater than the width of one cell. If too many cell centers fall within the width of the line, there is uncertainty as to their correct assignment. If the need for resource analysis precedes the development of control and base mapping for cadastral purposes, the option of developing a separate and less-accurate
OCR for page 100
FIGURE 6.1 Illustration of the problem of relating small cells to a wide line. resource thematic system must be considered. Resource thematic systems at a coarse scale take less time to develop than does a detailed cadastre. Resource problems often will not await the development of the cadastre on which to build an accurate resource thematic system. Further, many of the resource thematic applications may not require the map scale/resolution irherent in the cadastre, which may yield a volume of detail that is overwhelming. Experience in urban transportation planning shows that the land-parcel data sets provided more detail than was needed for metropolitan transportation planning. The data were immediately aggregatcd to a higher level—the traffic zone level—for analysis. The sizing and location of arterial highway facilities could be performed better with data aggregated to areas at least several city blocks in size. The same may be true for most resource planning and management analysis. In the short run, the problem of exchanging data between the cadastre and other Land-Information Systems is one of spatially adjusting the data by using the computer or of recompiling the other land data on the cadastre base maps. This should be only a short-run problem, because once the base map from the cadastre is available it can be rescaled as the base for all new resource inventories, which then can be fit directly to the cadastral data. Further, the denser geodetic reference framework will also be available for new issues of map products of the U.S. Geological Survey—often used as the standard base for resource thematic mapping—and these products will be consistent with the base map of the cadastre.