ArcGMT:
A suite of tools for conversion
between Arc/INFO®
and Generic Mapping Tools (GMT)

^[1] Department of Geosciences, Oregon State University, Corvallis, OR 97331 USA
^[2] Institute of Geological & Nuclear Sciences Ltd., P.O. Box 30-368, Lower Hutt, NEW ZEALAND
^[3] College of Oceanic and Atmospheric Sciences, Oregon State University, Corvallis, OR 97331 USA

ABSTRACT

ArcGMT is a suite of tools for converting Generic Mapping Tools (GMT) grids to a format readable by the older Arc/INFO® and ArcView® geographic information system (GIS) packages, and conversely, for creating GMT grids and files from Arc/INFO formats. GMT has arisen as an unofficial standard among geoscientists for producing maps, particularly from marine bathymetry, sidescan sonar imagery, gravity, and magnetics data. Many geoscientists have also turned to the mapping, database management and spatial analytical capabilities of Arc/INFO and ArcView. Subsequently, a need has arisen, particularly in the marine geology and geophysics community, for a conversion interface between these two environments. This initial version of ArcGMT is designed for UNIX-based workstations running Arc/INFO version 7.0 or higher, GMT version 3.0 or higher, and ArcView version 3.0 with the Spatial Analyst extension. ArcGMT consists of a series of UNIX shell scripts, Arc Macro Language (AML) files, and Avenue scripts, bundled in an easy-to-use menu interface for use within Arc/INFO or as an extension to ArcView.

Keywords: data conversion; geographic information system (GIS); marine geology and geophysics; physical geography

Note added in proof, 10/9/00: The current release of MBSystem, now includes "mbm_grd2arc" for ArcView users, which converts GMT grids to ASCII grids that Arc/INFO can read.

INTRODUCTION

In recent years scores of geoscientists have turned to the geographic information system (GIS) for advanced mapping & managing of their data, as well as for integration with other types of environmental data (Wadge, 1992a; Wadge, 1992b; Walker and others, 1996). This is particularly true in the marine geology and geophysics community (Bobbitt and others, 1997; Hatcher, Maher and Orange, 1997; Fox, Bobbitt and Wright, 1996; Wright, 1996; Wright and Goodchild, in press) where scientists also rely heavily on the Generic Mapping Tools (GMT) package. GMT has arisen as an unofficial standard for producing maps of marine bathymetry, sidescan sonar imagery, gravity, and magnetics. On the GIS front, many have turned to Arc/INFO® and its smaller desktop cousin, ArcView®, both of the Environmental Systems Research Institute. Both may now be considered the world's leading vector GIS packages (GIS World, 1996) and have been adopted in full by agencies such as the USGS and NOAA, as well as university labs and classrooms throughout the Western hemisphere. A need has risen, once again, particularly in the marine geology and geophysics community, for an interface between these heavily used software packages. GMT is often used to process and create base maps but provides little means for the incorporation of detailed attributes to the locations that are mapped out. Arc/INFO and ArcView, as GIS packages, provide a combination of mapping and relational database capabilities that exceeds GMT. However, the base maps that are more readily created in GMT (in concert with the MB-System package of the Lamont-Doherty Earth Observatory, which processes raw multibeam bathymetry and high resolution sidescan) provide a crucial base layer for many applications that are desirable for the GIS. An example is the fine-scale mapping of hydrothermal vent locations, animal communities, and lava flows at seafloor-spreading centers, a task that is only ~5% complete on a global scale (e.g., Embley and others, 1995; Haymon and others, 1991). Recently, with the increased popularity of GIS, shipboard data acquisition environments have been set up to import data from submersibles or remotely-operated vehicles directly into a GIS (Bobbitt, 1996; Bobbitt, Fox and Wright, 1996; Wright, 1994; Wright, 1996), some in real time (Hatcher, Maher and Orange, 1997). This has greatly facilitated the integration of geologic observations with vent chemistry and biology. However, bathymetry or sidescan sonar imagery, which allow for the interpretation of larger geologic structures that put these fine-scale observations in regional context are most often collected, processed and archived in GMT grids. This usually occurs on separate expeditions preceding the submersible or vehicle dives. It is crucial to be able to add these data as base themes or layers to the existing GIS databases. Conversely, it may be necessary to export features mapped in the GIS for inclusion in elaborate GMT maps and plots.

Here we present ArcGMT, a suite of tools for converting GMT grids to a format readable by Arc/INFO and ArcView. ArcGMT will also create GMT grids and files from Arc/INFO format. We begin by providing brief overviews of GMT, Arc/INFO and ArcView, followed by a functional description of ArcGMT.

GMT

GMT is a public-domain software package for data manipulation and generation of high-quality maps and scientific illustrations. The initial version was released in October of 1991 (Wessel and Smith, 1991) and re-released in a much-improved version 3 in 1995 (Wessel and Smith, 1995). An electronic survey (Wessel and Smith, unpublished data, 1994), as well as an assessment of ftp traffic and submitted registration forms, reveals that more than 5000 users worldwide frequently use GMT in their work. Within the marine geology and geophysics community, GMT is currently in use by scientists on all continents, aboard research vessels in the University National Oceanographic Laboratory System fleet, and even aboard U.S. Navy submarines and aircraft on scientific missions.

GMT runs on almost all computer platforms that are capable of running UNIX&tm;, from Cray supercomputers to workstations to PCs and Macintoshes. Although it is capable of filtering and gridding 1- and 2-dimensional data sets, GMT is best known and used for its advanced plotting and cartographic capabilities (Fig. 1)

Figure 1. Example of the advanced cartographic capabilities of Generic Mapping Tools (GMT). Composite figure showing 3-D mesh of Hawaiian topography from ETOPO 5 bathymetry on top of contour map of the Hawaiian geoid. After Wessel and Smith (1991). Click on the image for an expanded view

ARC/INFO® AND ARCVIEW®

With estimated sales of more than $200 million (Daratech Inc., 1994), the Environmental Systems Research Institute is the world's largest GIS company. It provides standalone software in product categories ranging from desktop GIS to high-end, advanced analytical GIS to software development tool kits. The company's flagship product, ARC/INFO, is the most commonly used GIS in a variety of market and scientific research segments, most notably petroleum and mining, groundwater mapping, geologic mapping, and marine geology. In addition, ArcView, the company's new desktop GIS, has been tremendously successful, and provides an easier-to-use, cost-effective addition to ARC/INFO. In addition to the aforementioned agencies, Arc/INFO and ArcView have been adopted by 40 of the top 50 oil companies in the world (Petroleum Intelligence Weekly, 1995), is site licensed on over 100 institutions of higher education in the U.S. (M. Phoenix, ESRI, pers. comm., 1997), and is in use by companies such as Sea Beam Instruments, the world leader in the deep ocean seafloor mapping sonar systems.

GIS packages such as Arc/INFO and ArcView provide not only an efficient means for mapping and plotting data but also for archiving, manipulating and analyzing data in relation to other layers or themes of observations. Three characteristics that distinguish GIS from mapping packages such as GMT are:

1. provision of linkages within a relational database management environment between the locations of points, lines, areas, and/or grids and detailed descriptive information about them (attributes), along with associated metadata.
2. provision of algorithms for the spatial analysis of data (e.g., spatial autocorrelation, advection and dispersion modeling, pattern recognition and analysis, buffering, spatial interpolation, overlay, etc.).
3. "spatial intelligence" or the ability to not only change map projections but to subsequently compute distances, identify nearest neighbors, calculate regions of proximity around existing features and update their attributes accordingly.

FUNCTIONAL DESCRIPTION OF ARCGMT

ArcGMT allows users to take advantage of the capabilities of both GMT and Arc/INFO or ArcView by providing a means for data conversion between the two environments. This initial version is suited only for UNIX-based workstations that are also running Arc/INFO version 7.0 or higher, GMT version 3.0 or higher, and ArcView version 3.0 with the Spatial Analyst extension. ArcGMT consists of a series of UNIX shell scripts, Arc Macro Language (AML) files, and Avenue scripts. AML and Avenue are the languages used in Arc/INFO and ArcView, respectively, to program and tailor those environments. There are two options for use of the code, depending on which GIS package will be primarily in use, Arc/INFO or ArcView.

ArcGMT with Arc/INFO

Within the Arc/INFO environment ArcGMT is operated through a simple pull-down menu from which the user may select to documentation to read or choose from three types of conversion:

1. GMT grids are converted to Arc/INFO grids by writing the binary GMT to an ASCII file, and then reformatting the file to space-delimited row-major order and with header information needed by Arc/INFO, including cell size, number of rows and columns, and coordinates of the grid's lower-left corner. The file is then piped to the Arc/INFO ASCIIGRID command for conversion to an Arc/INFO grid.

2. Arc/INFO grids are converted to GMT grids by first changing the grid (a raster data structure) to a vector point coverage. A coverage is Arc/INFO's primary method for storing point, line and areal geographic features. Structurally, the coverage is a directory containing binary files that store the locational coordinates ("Arc") and ASCII database files that store the locations' attributes and topology ("INFO"). Arc/INFO coverages and grids can also be read by ArcView. The initial change from a grid to a point coverage is necessary because each cell in an Arc/INFO grid is given a value (e.g., depth or elevation) corresponding to an x-y location on the surface of the earth, but location is inherent in the storage structure and not explicitly defined as an attribute. A point coverage derived from a grid, however, explicitly define z-coordinates as an attribute.

3. Arc/INFO coverages may also be converted to text files in a format readable by GMT scripts. This would be appropriate for features digitized in Arc/INFO, such as the outlines of lava flows or fault lineations, that would need to be included in a GMT plotting script. ArcGMT prompts the user for an existing Arc/INFO line or polygon coverage, converts it to a text file and then formats it for GMT. Points may just as easily be mapped by GMT with little or no conversion.

ArcGMT with ArcView

ArcView was introduced in 1994 as a scaled-down, desktop GIS, with a graphical user interface not available in Arc/INFO, thereby allowing the user to visualize, explore, and query data much more easily. With a learning curve of 1-5 days (versus 6-12 months for Arc/INFO) many geoscientists have turned to ArcView as their primary introduction to the world of GIS. Recent revisions to ArcView have included more powerful spatial analytical capabilities, although Arc/INFO remains the full-service, industry-standard GIS toolkit. ArcGMT allows users to make the same conversions described above within the ArcView environment. This is facilitated by implementing remote procedure calls (RPCs) between Arc/INFO and ArcView. There are several forms of RPC, but both ArcView and ARC/INFO use the Open Network Computing standard. RPCs allow client applications to initiate procedures in a server application across a network. UNIX RPC servers are uniquely identified by a hostname, a server id, and a version number. An RPC client application uses this information to establish the RPC connection. In ArcGMT, ArcView makes client requests to an Arc/INFO server in the form of Avenue statements. These statements call the ArcGMT conversion AMLs and shell scripts and then allows the user to import the results into the ArcView environment. Before using ArcGMT within ArcView, the user must start Arc/INFO as a server with a simple "&type [iacopen]" directive and then start ArcView.

Using ArcGMT with ArcView requires the ArcGMT extension ("arcgmt.avx"). Extensions are add-on programs to ArcView (written in Avenue) that provide additional functionality. When an extension is loaded, ArcView's graphical user interface changes to reflect this added functionality in the form of additional menu items, buttons, and/or tools (ArcView tools are distinguished from ArcView buttons in that the user must go beyond a mouse click and perform an action with the cursor in order to get a result). The ArcGMT extension adds the following functionality to the ArcView interface:

• Buttons to perform the three conversions described above, as well as clip out subsets of grids to create new grids, and create hillshaded themes from grids (Fig. 2A).

• A tool to create an empty grid based on a rectangle drawn by the user and the user's specification of row and column numbers (Fig. 2B).

• A menu item (under the "Theme" heading in when a View is active) to convert an ArcView shapefile (the native ArcView data format) to an Arc/INFO text file.

Figure 2. (Top) Buttons added to the standard ArcView button bar as part of the ArcGMT extension to ArcView. From left to right: "Bex (the bathymetry dog)" converts GMT grids to Arc/INFO grids; "black and white globe" button converts Arc/INFO coverages to GMT-formatted text files; "color globe" button converts Arc/INFO grids to GMT grids; "scissors" button extracts a portion of an existing grid after the user makes a selection with the standard ArcView rectangle tool; "contour" button creates a hillshaded theme from a grid after the user opens that theme's legend editor. (Bottom) A tool added to the standard ArcView tool bar as part of the ArcGMT extension to ArcView. The tool creates an empty grid based on a rectangle drawn by the user and the user's specification of row and column numbers.

ArcGMT also comes with an optional ArcGMT project file ("arcgmt.apr") that automatically opens ArcView with the extension already loaded.

RESULTS AND CONCLUSION

Figure 3 shows the result of a GMT to Arc/INFO grid conversion in ArcGMT. The original grid was of Sea Beam 2000 bathymetry revealing a seamount in the SW Pacific discovered in May of 1996 (Wright, Bloomer and Boomerang Leg 8 Shipboard Scientific Party, 1996). Grid cells were not square due to the gridding algorithm used when the grid was created with MB-System but ArcGMT was still able to make a perfect conversion. Figure 4 shows a GMT map of Sea Beam bathymetry overlain with fault lineations that were exported from an Arc/INFO line coverage with ArcGMT. Figure 5 shows the results of an Arc/INFO to GMT grid conversion in ArcGMT. The original grid was a 1-degree (3-by-3-arc-second data spacing), 1:250,000 USGS Digital Elevation Model (DEM) of the Big Island of Hawaii downloaded from the USGS Eros Data Center.

Figure 3. Results of a GMT to Arc/INFO grid conversion in ArcGMT. (Top) GMT contour map of a GMT grid of Sea Beam 2000 bathymetry from the SW Pacific, input to ArcGMT for conversion. Data are from Wright, Bloomer and Boomerang Leg 8 Shipboard Scientific Party (1996). Stars on map show where rocks were dredged from atop this seamount. (Bottom) Screen dump from an ArcGMT session, showing the same grid after conversion to Arc/INFO format. Grid is displayed in default grayscale within the GRID module of Arc/INFO. Click on the image for an expanded view

Figure 4. GMT contour map of a GMT grid of Sea Beam bathymetry from the NE Pacific, overlain with fault lineations that were exported from an Arc/INFO line coverage by ArcGMT. Data are from the RIDGE Multibeam Synthesis web site. Click on the image for an expanded view

Figure 5. Results of an Arc/INFO to GMT grid conversion in ArcGMT. (Top) A 1-degree, 1:250,000 USGS DEM in Arc/INFO format, displayed in default grayscale within the GRID module of Arc/INFO. Data are from the USGS Eros Data Center web site. (Middle) TIN construction of the grid using the LATTICETIN command with analytic hillshading in Arc/INFO before final conversion to GMT format. This option may be used in ArcGMT if desired. It is not the default. (Bottom) Contour map created in GMT of the resulting GMT grid (100 m contours, annotated every 500 m). Click on the image for an expanded view

A few cautions apply to Arc/INFO to GMT grid conversions in ArcGMT. The conversion is designed primarily for elevation or bathymetry data in units of positive m. If the user tries to convert another kind of grid (e.g., gravity in mgals), the z value may be off by 1-2 orders of magnitude, which can easily be remedied with "grdmath" in GMT. USGS DEMs may be in x-y units of decimal seconds, which are unacceptable units for GMT. Before converting, the user should always doublecheck the characteristics of the grid with the DESCRIBE command in Arc/INFO. The grid bounds output by DESCRIBE should be noted by the user. It may be necessary to project the units of the grid to decimal degrees and/or to make a datum transformation. In the case of the USGS DEM of Hawaii, the original WGS72 datum was changed to match a database stored in NAD83 with the following subcommands in Arc/INFO PROJECT: output, projection geographic, units dd, datum nar_d three, parameters, end. Such a three-parameter transformation may result in errors or uncertainty on the order of ~50 m.

Future improvements to ArcGMT include additional Avenue scripting to automatically read the hostname of the Arc/INFO server for the ArcView extension "arcgmt.avx" and allowing the user to interactively test projections, datum transformations, and various gridding intervals before converting grids from Arc to GMT.

Source code and a README file for ArcGMT are available here (RIGHT click and "Save As").

Acknowledgments -- This work is supported by the National Science Foundation under grant OCE-9521039. The paper was improved by the helpful reviews of two anonymous referees.

REFERENCES

Bobbitt, A. M. (1996) RIDGE Observatory exploration and data integration using ALVIN and geographic information systems. RIDGE Events 7(1), 10-12.

Bobbitt, A. M., Dziak, R. P., Stafford, K. M. and Fox, C. G. (1997) GIS Analysis of oceanographic remotely- sensed and field observation data. Marine Geodesy 20(2-3), 153-161.

Bobbitt, A. M., Fox, C. G. and Wright, D. J. (1996) ArcView supports deepsea research with famous submarine. Proceedings of the 16th Annual ESRI User Conference, Palm Springs, CA, Paper 082.

Daratech Inc. (1994) New GIS market study. GIM: International Journal for Surveying, Mapping, and Applied GIS 8(5), 36-37.

Embley, R. W., Chadwick, W. W., Jr., Jonasson, I. R., Butterfield, D. A. and Baker, E. T. (1995) Initial results of the rapid response to the 1993 CoAxial event: Relationships between hydrothermal and volcanic processes. Geophysical Research Letters 22(2), 143-146.

GIS World (1996) 1996 International GIS Sourcebook, GIS World, Inc., Fort Collins, Colorado, 771 pp.

Hatcher, G. A., Maher, N. M. and Orange, D. L. (1997) The customization of ArcView as a real-time tool for oceanographic research. Proceedings of the 17th Annual ESRI User Conference, San Diego, CA, Paper 676.

Haymon, R. M., Fornari, D. J., Edwards, M. H., Carbotte, S., Wright, D. and Macdonald, K. C. (1991) Hydrothermal vent distribution along the East Pacific Rise Crest (9° 09'-54'N) and its relationship to magmatic and tectonic processes on fast-spreading mid-ocean ridges. Earth and Planetary Science Letters 104, 513-534.

Fox, C. G., Bobbitt, A. M. and Wright, D. J. (1996) Integration and distribution of deepsea oceanographic data from the NE Pacific using Arc/Info and ArcView. Proceedings of the 16th Annual ESRI User Conference, Palm Springs, CA, Paper 293.

Petroleum Intelligence Weekly (1995) Top 50 oil companies. Petroleum Intelligence Weekly, December, 1995.

Wadge, G. (1992a) Geological applications of GIS. Journal of the Geological Society, London 149, 672.

Wadge, G. (1992b) GIS for geology. Terra Nova 3(1), 93- 98.

Walker, J. D., Black, R. A., Linn, J. K., Thomas, A. J., Wiseman, R. and D'Attilio, M. G. (1996) Development of geographic information systems: Oriented databases for integrated geological and geophysical applications. GSA Today 6(3), 1-7.

Wessel, P. and Smith, W. H. F. (1991) Free software helps map and display data. Eos, Transactions, American Geophysical Union 72, 441.

Wessel, P. and Smith, W. H. F. (1995) New version of the Generic Mapping Tools released. Eos, Transactions, American Geophysical Union 76(33), 329.

Wright, D. J. (1994) From Pattern to Process on the Deep Ocean Floor: A Geographic Information System Approach. Ph.D. dissertation, University of California, Santa Barbara, California, 341 pp.

Wright, D. J. (1996) Rumblings on the ocean floor: GIS supports deep-sea research. Geo Info Systems 6(1), 22-29.

Wright, D. J., Bloomer, S. H. and Boomerang Leg 8 Shipboard Scientific Party (1996) New Sea Beam 2000 bathymetry and sidescan data from the Tonga Forearc and Trench: Results from Boomerang Leg 8, an ODP site survey. Eos, Transactions of the American Geophysical Union 77(46), F326.

Wright, D. J. and Goodchild, M. F. (1997) Data from the deep: Implications for the GIS community. International Journal of Geographical Information Systems 11(6), 523-528.

Go to Dawn's Research