Geospatial Analysis and Modeling Section
Illinois State Geological Survey
615 East Peabody Drive
Champaign, IL 61820
Telephone: (217) 244-2510
Fax: (217) 333-2830
e-mail: stiff@celadon.isgs.uiuc.edu
New data acquisition and map production methods are being developed and implemented at the Illinois State Geological Survey (ISGS). These changes are in response to the initiation of a comprehensive 1:24,000-scale statewide geologic mapping program. In order to keep pace with rapid mapping progress, two basic data management needs were identified: the efficient acquisition and organization of high quality, accurate data, and time- and cost-effective production of 'on-demand' maps for review, presentation, and public distribution.
On-demand maps are plotted as the need arises rather than being printed in bulk. They have been fully reviewed and published in an ISGS map series. They may be updated as new geologic information becomes available. The graphics files for these maps are archived in a digital library making it easy for the information office to respond to map requests.
The primary software packages used for GIS data management, analyses, and cartographic production are Arc/Info and ArcView. This software is run on a distributed network of Unix workstations and PCs running WindowsNT (Krumm and others, 1997). Maps are plotted on Hewlett-Packard 750c plotters. Presently, GIS expertise at the ISGS is provided by a core group of ten GIS specialist/geologists, the Geospatial Analysis and Modeling Section (GAMS), plus a number of interns and student assistants.
In the past, mapped geologic data have been submitted for automation according to the methods or mapping 'style' of individual scientists. Mappers and field scientists routinely record type and distribution of geologic features and materials on United States Geological Survey (USGS) topographic maps and in notebooks in the field. Some scientists refine and transfer these raw data onto clean topographic sheets or a mylar (USGS green-line) base. Others provide field maps for digital conversion. Occasionally, reworked (some over-worked) 'pseudo maps' from drawings, images, or even xeroxed copies have been presented for conversion to a digital file. In all cases, the original data were archived at the discretion of the individual scientist
Data have been hand-digitized using GTCO digitizing boards. Data were primarily input by GAMS members or student interns and assistants. While fully capable of capturing the data in a digital format, GIS staff might be unfamiliar with 1) complex geologic concepts, 2) the geologic characteristics of a particular area, 3) the needs of a particular study, or 4) the mapping notation and style of the individual field scientist. In addition, registration accuracy and efficiency of the individual digitizing the data likely varied from one data-entry session to the next.
Data were registered to the Lambert Conformal Conic projection at quadrangle corners that were extracted from a statewide quadrangle base file or generated using a custom program that automatically creates spatially rectified coverages - digital files that will contain the point, line, or areal data of the selected quadrangle area. RMS error for acceptable registration accuracy was 0.005 or less. After maps were digitized, they were plotted and checked against the original for consistency and completeness. Lines greater than one line width from the original were edited or redigitized. Files were then manually edited or digitally processed to remove redundant arcs, add missed contacts, close polygons, and smooth the final lines. The edited files were then coded according to the unit names assigned by the geologist on a map-by-map basis.
DATA CAPTURE SOLUTIONSHistorically, the majority of GIS users at the ISGS have worked within contract supported enclaves of 'GIS professionals' that were separated from the scientific staff. This resulted in a 'disconnect' between the science of geologic interpretation and the techniques of digital data capture. However, digital awareness is on the rise. GIS software has become easier to use, hardware/software costs are decreasing and personal computers now support robust digital processing. In addition, the Internet has enticed many more people to become computer literate. GIS software and techniques are now becoming a powerful tool for many professional geologists. Increasingly, ISGS field scientists, with support provided by GIS professionals, are entering and editing their own data. This approach to automation of geologic data reduces the number of iterations through which geologic data must pass from the scientists' interpretation to the final data file. Several potential sources of error are thereby minimized.
In view of the rise in digital awareness, the ISGS scientific staff is exploring a number of data entry methods. With ArcView software on the NT platform, geologists enter digital point locations, map unit distribution, and associated field notes using either georeferenced USGS Digital Raster Graphics (DRG) or Digital Line Graph (DLG) data as a base. Tablet registration variation is reduced because data are entered in the georeferenced environment (Universal Transverse Mercator (Zones 15 or 16) or Lambert Conformal Conic) in which they will reside. The ability to zoom in to small areas allows for better control as detailed information is entered. Updates, corrections and interpretative refinements are entered by the scientist 'on the fly'. Test plots may be sent directly from ArcView to a plotter for hardcopy output with the DRG as a base. Finally, geologic units, saved either as areal or line data files, are converted to Arc/Info format and merged into the final geologic database.
The ISGS Quaternary Geology Section is exploring the potential of using MapInfo in conjunction with Arc/Info. MapInfo offers some 'user friendly' options that are not currently available in ArcView. Unfortunately, MapInfo does not support the standard projection of most existing ISGS statewide data sets -- Lambert feet.
Geologists have also been entering field notes using the Apple Newton hand-held Personal Digital Assistant. Field notes are written to the device which decrypts the handwritten entries. It "learns" to decode the users notes as frequency of use expands. The data are downloaded to a word processor each evening. A GPS link to tie these field notes to a specific location is being pursued.
The ISGS is in the process of acquiring a wide-format, high-resolution optical scanner. In the future, field maps may be scanned and georeferenced. Geologic data may then be digitized from the rectified image, converted from raster to vector format, or used in raster format. In addition to reducing some of the variability of digitizer coordinate registration, this will result in a digital record of the original mapping and field notation. The traditional method that involves digitizing, editing, and coding data from hand-drawn originals will continue as a method to address back-logged map automation and special situations.
It is more efficient to manage and explain a single, consistent file structure than to manage and explain data developed for the particular needs of a specific project, as has been done previously. This fact became obvious in the course of data acquisition for two separate projects. The ISGS automated published geologic maps from twenty-four contiguous southern Illinois quadrangles. Mapped by fourteen different geologists with different mapping styles and methods of geologic interpretation, the data were difficult to join into a single, area-wide data set. At about the same time, ISGS personnel digitized stack unit geology for the fifteen quadrangles of Henry County (with Whiteside and Rock Island counties to be added later). These counties were mapped by a single individual. Not surprisingly, it was far easier to supervise map automation of the Henry County areas, and ultimately join them into a seamless data set.
Creating a seamless, universal data structure required development of an adjustable, standardized data model that would accomodate various types and styles of mapping. The new data structure maintains all unit names, definitions, abbreviations, and descriptive notations found on the individual maps, in a standardized format and sequence. It also includes a twelve-digit numeric code that can be parsed for basic geologic information as the data are read.
One of these extracted code segments of basic information is a four-digit geologic unit designation developed by Tom Buschbach in the 1970's and later updated by David Gross. The original coding system was based on the Pleistocene Stratigraphy of Illinois (Willman and Frye, 1970) and The Handbook of Illinois Stratigraphy (Willman and others, 1975). Age was tacitly implied by numeric sequence, younger units having smaller numbers. In the early 1980's, the ISGS Coal Section began using and customizing the four-digit coding system to automate their production of Coal Resource Maps.
Further mapping studies and evaluation of new data by other ISGS staff have reduced the efficacy of the original four-digit code. In particular, reclassification of surficial units in northeastern and central Illinois (Hansel and Johnson, 1996) and the previously mentioned bedrock mapping in southern Illinois have altered earlier stratigraphic nomenclature and/or generated new geologic units. Adding spaces to the original four-digit code provides room for change as ISGS staff continue to update and expand geologic interpretation.
Consistent numeric attribution provides coherent access to both old and new data sets. Coincidentally, it facilitates database development, map production, and data distribution. Existing geologic databases are available on-line at the Illinois Natural Resources Geospatial Data Clearinghouse (http://www.isgs.uiuc.edu/nsdihome/ISGSindex.html). The developing geologic databases will ultimately be used to produce interactive maps on the Internet. Seamless digital data from each comprehensive quadrangle mapping project may also be written to CDs that will include free viewing software such as Arc Explorer.
In the past, ISGS maps were produced by two different methods. Illinois Geologic Quadrangle (IGQ) maps were produced using traditional methods by the Publications, Graphic Arts and Photography Section. The graphics for these select maps were manually drafted and sent out for plate preparation and printing. Maps expected to have a smaller audience were produced digitally by the Geospatial Analysis and Modeling Section. Some of these digital maps became ISGS-Open File publications, others were printed in the Illinois Map series. Still others, considered poster-type maps because of their limited distribution, were never published. All geologic maps are now produced using digital methods and software - mainly some combination of Arc/Info and CorelDraw. The few that are commercially printed are prepared in digital format.
The geology of approximately fifty-four quadrangles is presently being mapped under various federal and state mapping programs. Mapped for bedrock, Quaternary and/or stack-unit geology, twenty-seven of these quadrangles have geologic data in digital format and are awaiting publication.
In 1996, the Illinois Geologic Mapping Program (IGMaP) was initiated. IGMaP is a comprehensive mapping program involving development of suites of maps through in-depth analysis of the three-dimensional geology of each quadrangle by mapping teams. Basic maps for IGMaP include: the geologic materials at land surface (surficial geology), a three-dimensional portrayal of glacial/post-glacial sediments, the uppermost bedrock geology, a three-dimensional portrayal of bedrock materials, and a digital orthophoto that has been reclassified to a single thematic layer applicable to geologic interpretation. Structural geology, geologic unit isopachs, three-dimensional models, block diagrams, geologic columns, and cross sections may be added to the basic maps. Derivative maps included in all basic map sets are: groundwater resources, flood plains and/or flood prone areas, and aquifer sensitivity. Other derivative maps may be added at the discretion of the mapping team. Examples of these derivative maps might be: parent materials, geotechnical properties, geochemical properties, aggregate resources, coal resources, oil and gas resources, and so on.
Seven quadrangles are being mapped during fiscal year '97-'98 and another seven are scheduled for fiscal '98-'99. The expectation is that seven to fifteen maps for each of fourteen quadrangles will be ready for either review, presentation, or publication on-demand within the next two years in addition to maps produced under the USGS National Cooperative Geologic Mapping Program's STATEMAP component.
The Vincennes and Villa Grove Quadrangles were pilot studies for the new IGMaP initiative. To date, the Villa Grove Project has ten maps in preparation or review, and the Vincennes Project four. One GIS specialist participated in the compilation of data and production of maps for each quadrangle.
Several map publication procedures were defined during data development and production of these maps. Uniformity of appearance among the maps of a set was needed. A design format that could be customized was developed. Arc Macro Language programs (AMLs) for title and authorship, north arrow, scale bar, and location map modules and a general map layout were developed and placed in a general access directory. Both scientists and GIS specialists have adapted these programs to produce maps for the Vincennes and Villa Grove Quadrangles. This standardized format is also available as a CorelDraw template.
It became apparent that a vector-format base, produced in-house, was neither time-efficient nor practical. A USGS DRG was a viable alternative base for the Villa Grove Quadrangle. However, a number of limitations to the use of DRG data were encountered while working with the Vincennes Quadrangle data (Stiff, 1997), including high-relief, urban areas, supplementary contours, and poor raster coding. Vector-format DLG data was an option for some quadrangles (although the text layer is not available at the 1:24,000-scale), but not the Vincennes Quadrangle; topography, hydrology and roads were digitized from the Vincennes DRG on-line. This required a considerable investment of time and resources even without setting the text. Experiments with high-resolution scans of USGS topographic maps provided a viable raster-format base. These black and white scans may be used as a TIF images or converted to grid-format or polygon coverages depending upon plotting requirements for any particular map.
With the advent of IGQ and Illinois Map series geologic maps in digital format, higher levels of visual clarity and cartographic quality were required. Programs that produce the maps may be easily adjusted as new technology and output devices become available. Custom color and line sets were generated to produce accepted geologic symbols. New procedures to track mapping/map production progress for each quadrangle have been established.
In a perfect GIS-based cartographic world, all geologic data would be input on site by the investigating geologist who is the best and most accurate source of geologic interpretation at a particular site. These data would be imported into a single database with standardized attribute tables containing unit names, abbreviations, definitions and characteristics. Standardization would embed symbology in a database that would drive user-friendly access programs for digital map production and data delivery via the Internet. The resulting maps would be produced on 600 dpi plotters resulting in 'museum quality' output. We are working toward these goals.
Geologists using Arc/Info and ArcView have begun entering geologic data and producing some of their own maps. Standard attribute sets have been developed but the standard data structure has not been fully implemented survey-wide. There is a trade-off between field science and data accessibility. However, our work has shown that scientists can retain the freedom to map and interpret geology and simultaneously reduce the delay between data discovery and delivery. Production of 'on-demand' digital maps is becoming a viable reality -- too quickly for some, not quickly enough for others. Much remains to be worked out regarding acceptable quality.
Other members of Geospatial Analysis and Modeling -- Robert Krumm, Section Head, Curtis Abert, Chris Goldsmith, James Hester, Chris McGarry, Renee Nagy, Dan Nelson, Matt Riggs and Lisa Smith -- provide a valuable resource for Arc/Info, ArcView, AML, three-dimensional modeling, graphics software and imaging, and database management.
In particular, Curtis Abert, the Villa Grove Project's GIS Specialist, expanded the possibilities for imagery on maps, implemented new digital mapping methods, and developed the first Quadrangle Overview and Index Plate, a poster-like presentation of a quadrangle's geologic package.
Thanks also to Curtis Abert, Dan Nelson, Jon Goodwin, and Richard Berg for reviewing this paper.
Hansel, A.K. and Johnson, W.H., 1996, Wedron and Mason Groups: Lithostratigraphic Reclassification of Deposits of the Wisconsin Episode, Lake Michigan Lobe Area: Illinois State Geological Survey Bulletin 104, 106 p.
Krumm, R.J., Abert, C.C., Nelson, D.O., and Hester, J.C., 1997, Review of Digital Mapping Techniques: The Illinois Experience; in Soller, D.R., ed., Proceedings of a Workshop on Digital Mapping Techniques: Methods for Geologic Map Data Capture, Management and Publication: U.S. Geological Survey Open-File Report 97-269, p. 5-8.
Stiff, B.J., 1997, Use of Raster Imagery and Vector Data in Support of a Geologic; in Soller, D.R., ed., Proceedings of a Workshop on Digital Mapping Techniques: Methods for Geologic Map Data Capture, Management and Publication: U.S. Geological Survey Open-File Report 97-269, p. 23-26.
Willman, H.B., and Frye, J.C., 1970, Pleistocene Stratigraphy of Illinois: Illinois State Geological Survey Bulletin 94, 204 p.
Willman, H.B., Atherton, Elwood, Buschbach, T.C., Collinson, Charles, Frye, J.C., Hopkins, M.E., Lineback, J.A., and Simon, J.A., 1975, Handbook of Illinois Stratigraphy: Illinois State Geological Survey Bulletin 95, 261 p.
U.S.Department of the Interior, U.S. Geological Survey
<https://pubs.usgs.gov/openfile/of98-487/stiff.html>
Maintained by Dave Soller
Last updated 10.07.98