Digital Revision of the
New Jersey State
Base Map

By David E. Catts and Robin G. Fegeas

I

n a pilot effort, the USGS is updating the 1:500,000-scale State base map of New Jersey by using all-digital techniques.

This entirely digital environment stretches the limits of technology in a number of areas. The most dramatic development is map generalization-the extremely complicated process of combining many detailed maps covering the State at mostly 1:100,000 scale into just one map at the much smaller and less detailed scale of 1:500,000. This process is easy to do manually but difficult to accomplish digitally.

The production of the new New Jersey State base map is the lead project of a modernization activity that is evaluating the current state of geographic information system (GIS) technology and its integration into USGS map production processes. Although the work is based on extensive knowledge and techniques developed in earlier projects, many new technical areas are being explored. Research investigations include developing automated feature generalization routines, enhancing stream ordering algorithms for hydrographic feature identification and selection, assessing alternative automated methods of text placement, and compiling software, without significant reprogramming, for universal application to production projects.

The USGS 1:500,000-scale State base map for New Jersey is a natural selection for this project because the map was originally printed in 1978 and had been scheduled for manual revision. Also, the production location is near a GIS research laboratory, which provides ready access to expertise and the latest computer resources.

The digital data required for production of the new New Jersey State base map are assembled or created from various State and Federal sources, including available USGS 1:100,000-scale digital line graph (DLG) files, 1:250,000-scale land use-land cover political boundary files, and Geographic Names Information System data. Digital bathymetric data

Figure 1. Conversion of multiple digital line graph (DLG) sections to a single smaller scale map of continuous coverage. The creation of continuous coverage requires edge matching, paneling, and generalizing to combine area coverages at various format stages and scales.

are supplied by the USGS Exclusive Economic Zone sea floor mapping effort and are supplemented with National Ocean Survey charts. The State of New Jersey provides information on the location of State lands and the use and ownership of railroad lines. Existing State base map topographic contours are being scanned and vectorized (converted into line data), and several options are being explored to generate the urban area tint.

GIS technology allows data from these varied sources to be integrated into a data base by converting map coordinates according to a series of identified common control points. Other integration steps follow, depending on the data source. In most cases, non-DLG data are extensively edited manually (in the digital environment) to conform to the 1:100,000-scale DLG base data.

The use of the 1:100,000-scale DLG data files themselves, as well as other data at a scale of 1:250,000 or smaller, includes edge matching and paneling together of numerous files, each covering a small area, into combined areal coverages (fig. 1). Generalization, reducing the amount of map detail, is a significant aspect of manual map production. This process had to be translated into equivalent computer instructions for digital production. A digital filtering process reduces the number of features and the number of coordinates required to describe the remaining features. The resultant computer data storage for the project files is reduced by as much as 88 percent. Figure 2 shows the results of generalizing 1:100,000-scale DLG data to 1:250,000 and 1:500,000 scales.

Actual map production of the New Jersey State base map includes developing graphic symbology sets. Much of the symbology used in editing and correcting the data did not exist previously and has been developed for this and future projects. The symbology must be developed for final graphicproduction, and the negatives for map publication must be created directly on a high-resolution digital plotter. Finally, the actual conversion from the 1:500,000-scale digital base data to a set of digital graphic images for plotting requires an additional filtering process and a resolution of any symbology overprinting problems.

The digital-to-graphic techniques being developed during data preparation stages make the digital revision of the New Jersey State base map a truly ground-breaking enterprise. At present, it is still costly and labor intensive to perform this work in an entirely digital environment. The lessons learned and technology developed in this project will go far toward accomplishing cost-effective digital integration, revision, map generalization, and map production.

Using GIS to Link
Health and
Earth Science

Data Bases

By David R. Wolf

The National Center for Health Statistics (NCHS), one of the Federal Centers for Disease Control, is the principal Federal organization having legislative authority to collect statistics on many health-related issues. Currently, the USGS and the NCHS are investigating the potential for spatial association between national earth science data bases and disease mortality data bases.

GIS (geographic information systems) and statistical techniques are being used to explore the possibility of spatial relationships between the environment and human health, which may reveal patterns of association that otherwise are difficult or impossible to detect. For demonstration purposes, ultraviolet-B radiation measurements and malignant melanoma mortality data are being spatially and statistically analyzed.

USGS water-quality data also may be examined and spatially analyzed in conjunction with other U.S. public health data bases. Examination of indigenous (endemic) nephritis in Yugoslavia has led to a spatial investigation in the United States for kidney disease of similar origins. Additional disease candidates for study include certain types of cancer and other mortality and morbidity conditions occurring in the U.S. population.

Quality Map Products From Digital Data

By Thomas M. McCulloch

D

igital spatial data serve a wide range of users, many of whom use the data as base information in geographic information system (GIS) applications. The GIS provides users with a flexible and diverse set of computer tools for integrating, analyzing, and displaying different types of data.

A network of GIS research laboratories supports studies relating to GIS development, image processing, visualization, and spatial data research, collection, and exchange. The state-of-the-art software and hardware systems in the laboratories are used in cooperative projects among the USGS, Federal agencies, and State, regional, and local agencies. The projects demonstrate the use of GIS hardware and software in combining and analyzing spatial data to solve specific problems.

As GIS technology matures, more users are taking advantage of proven GIS capabilities to build digitally based systems for making standard topographic and thematic maps. Producing these graphic maps from digital data is particularly important because, as the data are revised, new maps will be needed to reflect the latest information available in the data.

The USGS produces and distributes planimetric digital spatial data in digital line graph (DLG) format. The DLG data are based on two models that represent, in computer-readable form, the information. shown on a printed map. These models are the DLG-3 model, the current standard for content, and the DLG-E model, designed to support advanced cartographic and GIS applications. The DLG-3 model is built on vectorbased data obtained primarily by converting to computer-readable form the center line of symbols from USGS quadrangle maps. The model uses basic spatial elements, topological

relationships, and descriptive attributes to digitally portray the cartographic features of published maps.

The DLG-E model is also vector based but is derived primarily from image sources, published maps, and existing DLG-3 data. The model adds a feature structure and specific nontopological relationships such as flow direction to a reorganized set of spatial elements and a broader set of descriptive attributes, including names for selected features. The DLG-E model is designed to represent a wide range of geographic spatial information, including additional features and attributes the user may define.

The feature structure for the DLG-E model is implemented through a set of rules for handling the more than 200 features shown on a standard topographic quadrangle map. To represent so broad a domain of features, a wide array of rule sets, such as extraction specifications, representation rules, symbol specifications, and product generation rules, are needed to support the effective implementation of the DLG-E data model. This new data model and the rules supporting it will enable the USGS to generate upto-date graphic products from existing digital data.

Graphic production systems are being designed to run on high-performance graphic workstations. The increased data handling capabilities and declining costs of these workstations make them an effective solution to the problem of producing graphic products from digital spatial data. GIS applications software running on a workstation can perform the data manipulations needed for cartographic symbolization and then generate a data file that can be plotted on existing largeformat, high-quality plotters. From these film plots, press plates can be made to print the maps. The combination of digital spatial data and GIS's, once limited to demonstration projects in the GIS research laboratories, has expanded to enable the use of advanced data models and structures to produce publicationquality standard and thematic maps.