Geology, like other scientific disciplines, has seen a rapid increase in the routine use of sophisticated computer technology in recent years. Desktop computer workstations continue to become more powerful and affordable while software packages with tremendous capabilities continue to become easier to use and more robust in functionality. Of particular importance to geology and the earth sciences has been the development of Geographic Information Systems (GIS) technology, which combines comprehensive geospatial database management and analytical capabilities with the ability to produce, on the computer screen or as printed output, highly-accurate maps that represent underlying database information or analytical derivatives thereof. Unlike traditional hardcopy maps, which are generally printed in relatively large quantities and distributed over a long period of time regardless of new information that might come available, maps derived from digital data can be produced on demand and, therefore, a map representative of the most recent data contained in the database can be generated, thereby assuring currency of the information depicted. Further, modifications to GIS databases, such as corrections or new data entries, can be easily and quickly implemented, imparting long-term usefulness to the data set through a program of routine maintenance and update.
For the past several years, the Geological Survey of Alabama (GSA) has conducted its geologic mapping program, particularly at the 1:24,000, 7.5-minute quadrangle scale, using GIS technology for data compilation, storage, analysis, and output. During this time, we have developed techniques that greatly automate and expedite the process of map production, integrally involve the geologic mapper or mappers in all phases of the process from field data collection through GIS development to map finalization, and result in comprehensive and useful digital geologic data sets for a variety of applications, as well as traditional hardcopy maps and reports. This geologic mapping program and the development of GIS techniques for geologic mapping have been greatly facilitated by support from cooperative agreements with the United States Geological Survey (USGS) under the auspices of the STATEMAP part of the National Cooperative Geologic Mapping Program, which was authorized by the National Geologic Mapping Act of 1992. To date, three 7.5-minute quadrangle maps have been completed under this program (Osborne, 1995; 1996; Osborne and others, in review), field mapping and data collection for three quadrangles are underway at present, and funding for four additional quadrangles has been approved. The majority of these mapping projects involve structurally deformed areas in the Valley and Ridge geologic province of Alabama and, thus, the completed projects have provided excellent cases for developing our methodology in geologically complex areas. Although the techniques used continue to evolve, we feel that we have addressed many of the major issues and problems associated with geologic mapping in the GIS environment and that our methodology suits our purposes quite well. The purpose of this paper is to provide an overview of the techniques used for digital geologic mapping at GSA and to present some of our plans and goals for the future. The methodology described below has been developed using ARC/INFO, a commercial GIS software package from Environmental Systems Research, Inc. of Redlands, California and Excel, a commercial spreadsheet software package from Microsoft Corporation, Redmond, Washington. The use of these software packages and their registered trademark names in no way constitutes their endorsement by the Geological Survey of Alabama or the State of Alabama.
As with any geologic mapping program, the most fundamental and important aspect of mapping using GIS technology is the collection of accurate, detailed field data. GSA has a highly qualified, experienced team of geologic mappers that form the nucleus of our mapping program. This team works under the guidance of a state-wide geologic mapping committee, composed of government, academic, and private sector geologists, that assists in the selection of quadrangles to be mapped through a prioritization process. Selected quadrangles are then formally proposed for mapping to USGS under the STATEMAP part of the National Cooperative Geologic Mapping Program and funded proposals are implemented as mapping projects by GSA. Data are collected using standard and accepted field mapping techniques incorporating traverses, measurements, observations, sampling, and detailed written and photographic documentation. Field data are initially compiled and interpreted by the mapping team on scale-stable contact prints produced from scale-stable film positives of 7.5-minute topographic quadrangle sheets acquired from USGS.
Preparation of data for digital capture generally consists of three parts: (1) transfer of features for each map feature type onto scale-stable overlays to facilitate digitization, (2) uniquely identifying each feature of each type, and (3) entry of feature attribute data, along with unique identifiers, into a spreadsheet program. In order to minimize the possibility of introducing errors, the mapping geologist or geologists are responsible for data preparation in coordination with the GIS Specialist assigned to the project. We have found that this not only involves the geologist in the digital data capture process and encourages interaction between the mapper and the GIS Specialist, but also adds a level of quality control, in that the geologist, through familiarity with the study area and the data set, is more likely to recognize errors and mistakes than the GIS Specialist at this phase, thereby minimizing changes later in the process.
In general, we have found that the compiled and interpreted working maps discussed above tend to become very cluttered with information, leading to the possibility of confusion or error during the digitization process. This is especially true in areas of complex geology. Hand-drawn and
-written information on the compilation map includes symbolized lines for geologic contacts, faults, and the axes of structural features, symbols indicating where structural orientation data were collected, geologic sections were measured, samples were taken, and contacts were exposed, and annotation, such as the names of structural features, codes for geologic units, dip values, and other notes. For digital data capture purposes, we have determined that it is desirable to separate features of different type by transferring each feature type (or subset thereof) onto a clear, scale-
stable overlay that is punch-registered to the original compilation map. Georeferencing for each overlay is accomplished by transferring the corner tics from the quadrangle to the overlay. The primary feature types on a geologic map are: (1) polygon or area features (areal extent of geologic units), (2) lines (contacts, faults, etc.), (3) points (structural data points, exposed contacts, etc.), and (4) text. It is also helpful to depict each feature on an overlay in its simplest form. For example, on the compilation map, an approximately located thrust fault would appear as a dashed line with teeth on the upper plate, whereas an approximately located geologic contact is indicated by a dashed line. For the purposes of digital data capture, these features are transferred to the overlay as simple lines. Similarly, structural data points, depicted on the map as strike and dip symbols and symbols for horizontal and vertical beds, are transferred as simple points located at the center of the symbol.
After transfer to the overlays, each feature is assigned a unique number that will later be used to link attribute information to the feature. For each overlay, features are numbered consecutively, in most cases beginning with 1. We normally begin numbering in the upper left-hand corner of the quadrangle and attempt to make the numbering scheme as easy to follow as possible, particularly in areas with dense features. This facilitates using the capability of the GIS digitizing system to automatically increment the identification number for each new feature in the feature database, thereby allowing the GIS specialist to quickly capture the data without stopping to enter unique numbers.
The last step in data preparation is the entry of feature attribute and ancillary data onto spreadsheet. We use a spreadsheet software package for data entry primarily because the majority of our staff is familiar with its use for data compilation and manipulation and, thus, no special training is required. Tables are prepared for each feature overlay and, in these tables, feature data are keyed to the above unique numbers for features. Information entered into the tables include orientation data for structural measurements (strike, dip, bearing, plunge, etc.), codes for point and line symbolization, codes for colors of geologic units, descriptive names for features (e.g., "thrust fault, approximately located"), annotation for named geologic features (e.g., "Fungo Hollow deformed zone"), geologic unit names (e.g., "Newala Limestone") and codes (e.g., "On"), notes, remarks, etc. Having this information as part of the eventual GIS database facilitates automation of many of the orientation, symbolization, and annotation requirements for hardcopy output of the geologic map, as well as provides a robust geologic database for use in various applications.
Importantly, the spreadsheet tables are also used to calculate orientation angles for geologic symbols and annotation that require orientation, such as strike and dip symbols or geologic feature names that need to appear on hardcopy output at some angle to the horizontal. In the GIS system used, all point features, including text anchored to a justification point, have a "hidden" database attribute for rotation angle ($ANGLE) and the default value for this attribute is zero. Further, negative azimuth angles indicate clockwise rotation of the point about its center. Thus, a rotation angle of "-90" indicates a clockwise rotation (toward the east) of 90 degrees. As an example, we have created a geologic symbol set in which a strike and dip symbol with zero rotation corresponds to north strike with dip to the east, whereas a symbol with an angle of -90 corresponds to east strike with dip to the south. During field data collection, the mapping team records strike and dip data in the traditional quadrant notation form, such as N35E, 20SE. To facilitate the calculation of symbol orientation, these data are entered into spreadsheet columns as follows: strike quadrant (e.g., NE), degrees from north (e.g., 35), dip direction (e.g., SE), and dip amount (e.g., 20). At this point it is possible, using the sorting functions of the spreadsheet application, to segregate measurements into groups based on the eight possible combinations of orientation (i.e., N-S strike with E dip, N-S strike with W dip, E-W strike with N dip, E-W strike with S dip, NE strike with SE dip, NE strike with NW dip, NW strike with SW dip, and NW strike with NE dip). Following this grouping, the rotation angle can be either entered explicitly for the North-South and East-West strikes (0 (N with E dip), -90 (E with S dip), -180 (N with W dip), and -270 (E with N dip)) or calculated with a formula for orientations that are oblique to the cardinal points. It is important to note that only eight entries are necessary using the "Fill/Down" spreadsheet function for each group entry, regardless of the number of data points in the each group. In this example, the formula entries for the oblique orientations are as follows: NE strike with SE dip -- rotation angle (RA) = (strike angle +180 (-1)); NE strike with NW dip -- RA = (strike angle (-1)); NW strike with SW dip -- RA = (strike angle -360); NW strike with NE dip -- RA = (strike angle -180). In our experience, the sorting and calculation process, including formula entry, takes less than five minutes. Though not absolutely necessary, as a clean-up step, we generally sort the data by unique number after the calculation process has been completed. The final step is to convert the spreadsheets to dBASE III database format files using the "Save As/DBF 3" command. The GIS can directly import dBASE III files into the INFO database, thus saving time and effort in linking the data to map features (using the unique ID) after data capture. At this point, preparation of data is complete.
At present, digital capture of geologic map features from the overlays discussed above is accomplished by manual digitization on a high-accuracy digitizing table using the GIS's digitizing system. Each overlay is attached to the table and registered to the quadrangle's corner tics extracted from GSA's master 7.5-minute Universal Transverse Mercator (UTM) coordinate system grid, which was generated in the GIS using the latitude and longitude coordinates for the corners of each 7.5-minute quadrangle with area in the State of Alabama and projecting the resulting base into the UTM coordinate system. An acceptable maximum residual mean squares (RMS) error for registration of each sheet is adopted and rigidly adhered to. If registration results in unacceptable RMS error, the overlay is re-registered until within the acceptable limit. Registration of overlays in this manner to a mathematically generated base assures highly accurate georeferencing of each. Features on each overlay are digitized consecutively by unique number. After digitization, the GIS layer for each overlay is processed to generate topology, checked for topological errors, and edited until all such errors are corrected. At this point, check plots are generated at map scale and checked against the overlays for obvious errors. After correction, additional check plots are generated at map scale on scale-stable media and checked against overlays and the original compilation map for proper registration and feature location. This process continues in an iterative fashion until all digitizing errors are corrected and the digital data exactly correspond to the overlays and original compilation map.
Upon completion of the topologically correct GIS layers for each overlay, we proceed with GIS database development. This consists of attaching the data from the spreadsheets above (saved as dBASE III files) to the GIS data sets. The dBASE III files are imported into INFO database tables and these tables are joined to the feature attribute tables using the software's table joining function on the basis of the common unique identification number for each feature. Once these tables are joined, the digital data development part of the digital geologic mapping process is essentially complete.
The digital data for a geologic map are compiled for printed output using interactive GIS map composition tools. The databases developed above contain all of the information necessary for automated symbolization, orientation, and location of geologic features and observations and text for annotation. The GIS specialist, through the use of commands to set certain parameters (map scale, page size, linesets, symbolsets, fonts, etc.) and database queries to call up desired overlay layers and features, can quickly assemble a cartographic-quality geologic map. At GSA, we have created line, symbol, and color-fill sets that incorporate standard and accepted geologic symbology and map features are drawn with these elements using codes that are included in the database. Cartographic elements, such as titles, legends, neatlines, arrows, pointers, leaders, scales, etc. are easily added to the map through an interactive, on-screen process. At the present time, we use pen, electrostatic, and ink-jet plotters for on-demand map output and have also produced scale-stable film output for use in preparation of plates for printing on an offset press.
There are several areas that are integral to the continued development and automation of digital geological (and other) mapping in the GIS environment at GSA. Our process would be greatly accelerated (and much less labor intensive) with a migration from manual digitizing to utilization of scanning technology for data capture. We presently have the software capability for raster to vector data conversion and hope to acquire a large-format, high-resolution scanner in the near future for this purpose. We would also like to begin to incorporate Global Positioning System (GPS) and digital data logging technology into our field data collection efforts. Much of the data that is presently entered in field notebooks by hand and transferred to spreadsheet tables at a later time can be directly captured in a digital format in the field along with locational data using a GPS unit with attached data logger or laptop computer. These data can be directly imported into the GIS, thus saving considerable time and duplication of effort. Finally, we would like to continue to enhance our computer and software capabilities in terms of power, speed, functionality, and storage capacity in order to be able to take full advantage of new innovations in digital mapping and GIS technology.
Osborne, W.E., 1995, Geology of the Leeds 7.5-minute quadrangle, Jefferson, Shelby, and St. Clair Counties, Alabama: Geological Survey of Alabama, Quadrangle Series 13, 22 p., 2 pls.
Osborne, W.E., 1996, Geology of the Helena 7.5-minute quadrangle, Jefferson and Shelby Counties, Alabama: Geological Survey of Alabama, Quadrangle Series 14, 21 p., 2 pls.
Osborne, W.E., Ward, W.E., II, and Irvin, G.D., in review, Geology of the Alabaster 7.5-minute quadrangle, Shelby County, Alabama: Geological Survey of Alabama, Quadrangle Series.