sing GIS techniques, USGS scientists

Usin

have constructed a new map for the continental shelf north of San Francisco. Of note, the map shows diverse types of data that traditionally are not mapped together, such as sediment age, sediment thickness, faults, and sediment particle size. Each type of information is depicted in a different color (shown here in black and white), so as to be useful to a broad community of users, including physical oceanographers, commercial and recreational fishers, and geologists. Using digitized data sets in MAPGEN format, the map is

being compiled for formal publication, with partial support provided by the Office of Naval Research, Department of the Navy, which will use the map as background information for the STRESS (Sediment Transport Events on the Shelf and Slope) Program. Under the STRESS Program, researchers from the USGS, Woods Hole Oceanographic Institute, University of Washington and several other universities are examining the role of storms in resuspending and transporting sediment on the continental shelf.

The

map includes information on structure, morphology, and surficial geology. One of the strengths of the map is its color format, which permits overlaying contrasting information and no loss of legibility. The map consists of five layers: bathymetry, faults, surface geology, Holocene sediment thickness, and sediment particle size.

Bathymetric contours, digitized from a National Oceanic and Atmospheric Administration data base, are mapped at 20-meter intervals on the shelf and at 100-meter intervals beyond the shelf break (200 meters). Onshore and offshore faults are digitized from a USGS map. The surficial geologic units were mapped by the investigators by using high-resolution seismic-reflection profiling and sidescan sonar. Three main units are shown: pre-Quaternary outcrops, Pleistocene deposits, and Holocene deposits. The preQuaternary units include both deformed and uplifted sedimentary beds on the outer shelf and exposures of resistant sedimentary and granitic rocks on the inner shelf. The broad, thick suite of Pleistocene deposits formed during times of shoreline migration caused by the rise (transgression) and fall (regression) of sea level. The Holocene deposits, primarily composed of sediment that has accumulated during the past 5,000 years, are mapped by thickness as well as distribution, thus showing the position of local concentrations of sediment formed by shoreline and deltaic deposition. Superimposed on all units are contours showing average particle size (in millimeters) of the few centimeters of sea floor sediment. upper The map has many potential uses for the public and for research investigations. For example, the particle size information is useful to researchers for predicting movement of sediment during storms. The location of outcrops and the textural character of the sea floor are valuable to fishers for locating shellfish and certain types of bottom fish. Location and thickness of Holocene sediment are important for future assessments of placermineral and aggregate resources. Decisions about continental shelf use, such as locating suitable disposal sites for dredged material, need this type of definitive information about bottom sediment type and morphology. Other applications will doubtlessly emerge as use of the continental shelf increases.

GIS Technology and Its Application to Geologic Hazards

By Arthur C. Tarr

IS technology is transforming geologic hazards research in the 1990's. Complex hazards analyses once thought to be impractical or impossible are now routine. GIS technology automates the process of acquiring geologic data by replacing drafting vellum with a computer screen, a greenline with a

digital base map, and a set of drafting tools with a digitizer tablet and a mouse. Finished, full-color maps portraying results of research are produced and distributed in weeks or months instead of the years needed to produce a traditionally printed map.

Geologic research is a lengthy, deliberate process of data acquisition, analysis, peer review of results, and finally, publication of a report or map. Until the last decade, geologic mapping and map compilation employed cumbersome tools and laborious techniques. GIS has changed that. In the USGS, GIS is now used widely to study and depict earthquake hazards in the San Francisco Bay area, the Pacific Northwest, and the Central United States and to study the landslide hazard in numerous areas of the United States and foreign countries.

GIS promises to radically change the methods of field geologists. Imagine a geologist in the field in 1996 mapping fault breaks resulting from a recent large earthquake. The geologist begins by entering field data into a lightweight, battery-powered computer. Simultaneously, geographic coordinates of the geologist's position are computed by a Global Positioning System (GPS) unit and recorded for each field observation. The GPS unit receives radio transmissions from a constellation of navigation satellites and is small enough to be stashed in a backpack. Back at the office, the geologist transfers the images and the data from the portable computer to a workstation where the geographic coordinates and other data are converted into a standard map projection in a GIS-compatible format. The new field data and selected video frames are then displayed together on the highresolution screen of the workstation where the data can be edited and interpreted.

A colorful multispectral satellite image of the study area and previously mapped geologic or geophysical data also may be displayed as backdrop layers. The geologist retrieves additional field data from a colleague's computer in another State. Within the computer environment, the scientist can compose maps portraying the new data on a topographic base retrieved from a USGS data base and then plot the maps on a desktop color printer or perhaps on a large format color plotter. After revising the color maps, the geologist can make them available over a computer network to colleagues across the United States and in foreign countries.

Although some elements of GIS, such as computer analysis of spatial data and automated mapping, have been used in earthquake and landslide hazards research for decades, these elements were not originally

integrated to work together as a system. For example, the geological hazards of San Mateo County, Calif., have been analyzed and mapped since the 1970's using computerized analysis and automated mapping methods. These methods were refined during subsequent years, but progress was slow for several reasons: customized computer programs had to be written for each application; computer hardware was expensive and slow relative to today's computers; and necessary spatial data were not available in digital form. These data had to be laboriously acquired from existing source materials by digitizing paper maps.

Nevertheless, in the mid-1980's, a USGSState cooperative pilot study of a small area of the Wasatch Front in Utah demonstrated the feasibility of using GIS as an earthquake hazards assessment tool. The pilot study resulted in numerous thematic data layers portraying societal elements at risk, such as schools, fire stations, and hospitals, and geologic hazards, such as the Wasatch fault and geologic units susceptible to liquefaction. Later, GIS was employed in two different earthquake hazards study areas, one in the Pacific Northwest and one in the San Francisco Bay area; for example, GIS was used in several ground response studies following the October 17, 1989, Loma Prieta earthquake in California. By 1990 GIS was fully integrated into research plans for the New Madrid seismic zone of the central United States, and in 1991 the USGS National Landslide Information Center began incorporating GIS landslide data into a new data base, which will be available to earth scientists worldwide.

The growth of GIS in the USGS led to the establishment in 1990 of a computer graphics laboratory in Golden, Colo., to provide GIS and other scientific visualization tools to geologic hazards projects and to the creation of a Geologic Hazards Data Base (GHDB) to combine earthquake and landslide hazards data under centralized management. The GHDB currently contains more than 1 gigabyte of digital spatial data related to

geologic hazards research in the Pacific

Northwest, Central United States, Colorado, and San Francisco Bay area. Many data sets in the GHDB are accessible to USGS-sponsored researchers and collaborators over Internet, a global telecommunications network.

The figure shows how the GHDB can be used by illustrating research study results of the expected intensity (as measured on the modified Mercalli intensity scale) of a hypothetical magnitude 7.1 earthquake located near the epicenter of the 1949 southern Puget Sound earthquake. The figure demonstrates a suite of GIS functions and one technique that uses a particular algorithm to calculate circular intensity contours based on a point earthquake source (intensity is the dependent variable and has been adjusted for the effects of site geology). GIS can accommodate other algorithms that produce noncircular intensity contours (such as those from a line source instead of a point source), multiple intensity adjustments (to correct for depth to bedrock or depth to water table), or a different dependent variable (such as acceleration). Also, the technique can be generalized to produce landslide and liquefaction susceptibility

maps.

What may we expect from GIS in the next 5 to 10 years, especially as it applies to geologic hazards research? Computer hardware costs probably will continue to decrease even as the hardware becomes more powerful. Therefore, GIS software will take advantage of affordable, higher performance hardware to provide additional functions and speed to existing applications. Enormous spatial data bases, such as large-scale maps of extensive areas and high-resolution satellite images, will be distributed widely on optical media, such as CD-ROM, at low cost. Highspeed computer networks linking online data bases in the United States and in foreign countries will permit USGS scientists to access and exchange large data sets with their collaborators. In the 1990's, geologic hazards research will become increasingly efficient, comprehensive, and timely because of advances in GIS technology.

Use of GIS Technology in Preparing WaterResources Publications

By Gregory J. Allord and

Richard W. Paulson

The

The USGS National Water Summary (NWS) program documents national assessments of the water resources of the United States in a series of USGS WaterSupply Papers. Each Water-Supply Paper presents information on hydrologic and meteorologic conditions for 1 or 2 water years (October to September) and focuses on a particular topic, such as water quality or water use. The information is presented in a series of articles and in State-by-State summaries. By using GIS technology, the USGS has speeded up the analysis, presentation, and publication of this information.

The national assessments that are presented in these NWS reports aggregate existing data from USGS and other Federal and State data bases. Because the NWS reports are intended for a nontechnical audience, the often-complex data are presented in specially designed formats for that audience. The multicolor graphics that can be produced with GIS technology are an effective method of presenting this complex data.

The State summaries provide for comparison of water-resources conditions between States because each State summary shares a common organization of text, tables, and illustrations. Manual preparation in a timely manner of the text and illustrations for each State summary rarely is possible because of the large quantity of data represented, the need to replicate common presentation formats, and the deadlines of report preparation. Consequently, the preparation of computer-based text and illustration data sets allows the information to be combined and compared in

needed formats.

Examples of small-scale, national digital data sets that have been created in support of the NWS reports include State and county boundaries; hydrography; hydrologic unit boundaries; average annual runoff; population growth and distribution; location of waste sites; water withdrawals and use by county, aquifer, and drainage basin; and areal extent of major floods and droughts in each State. Digital data sets that currently are being prepared include land use, physiographic provinces, updated population distribution, and water-quality trends and conditions at selected sites. A national digital wetlands data set at a scale of about 1:2,000,000 also is in

preparation in cooperation with the U.S. Fish and Wildlife Service, Department of the Interior.

Because most USGS hydrologic studies are local or regional in scope, and few result in the creation of nationwide, small-scale data sets on a specific subject, organizing and displaying hydrologic data in NWS reports present unique problems for the compilation, analysis, and review of the data. The topical information selected for the NWS, however, is aggregated on a regional and national basis and must be consistent from State to State and at a scale for publication as page-sized maps. Almost all of the State-summary illustrations are now prepared using GIS-based techniques.

The figure (facing page) is a black and white example of a multicolor, GIS-produced State-summary illustration. The illustration shows six time-series graphs of data from USGS stream-gaging stations that show periods of major drought, a State map that shows the locations of the stream-gaging stations, and State maps that show the areal extent of droughts. Each drought is identified by a particular color that links a highlighted period in the time-series graphs with the areal extent of the drought on a State map and with the color key in the map explanation. Additionally, if the severity of the drought fluctuated in the State, as in the case of the 1980-82 drought, two shades of the color are used on the State map to show areal extent of classes of drought severity. Time-series data from the six hydrologic stations in Tennessee were extracted from the USGS National Water Data Storage and Retrieval System data base and analyzed to identify the major droughts in the State. The areal extent of each major drought was then defined on the basis of the analysis of data from as many as 40 hydrologic stations.

Links between ARC/INFO, the GIS used by the Water Resources Division, and the traditional publishing process have been created for the preparation of illustrations, and several software applications and computer systems are used. For example, desktop publishing software on a personal computer was used to lay out the different elements of the drought illustration. In conformance with the overall design of the illustration, individually scaled graphic elements, such as the drought maps, graphs, and illustration explanation, were first put in position using Adobe Illustrator 3.0. The horizontal and vertical page position coordinates were then determined, and the graphic elements were digitized and entered into an ARC/INFO file in one of the USGS Distributed Information System (DIS) computers. This ARC/INFO map then was converted to PostScript format, a page