Adding a Third Dimension to Terrain Analysis

Adding a Third Dimension to Terrain Analysis by Captain Charles R. Graham and Dr. J. Richard Jones Modern computer technology now makes it possible to prepare a new kind of terrain map that shows the lay of the land far more graphically than standard topographic maps.

The application of this technology offers an effective method of improving a field commander’s assessment of terrain and trafficability, the most important factors in the successful employment of armored fighting vehicles. This article describes how the technology works and how it was applied to map a section of terrain many American tankers know well, the maneuver area within the Fort Hood, Texas, military reservation. The article includes maps plotted using this new technology and will show how the techniques can be applied elsewhere. While the experiment described here was conducted on a large mainframe computer, theoretically the approach is applicable to smaller microcomputers, too, once appropriate software is developed. This process offers the prospect of instantly generated, highly realistic terrain maps which appear to be threedimensional. With a microcomputer and inexpensive X-Y plotter mounted in a command poet track, it might even be feasible to create maps while the unit is on the move, using terrain data stored in the computer memory.

Moreover, the method can generate maps on any scale and the computer program allows these maps to be pre pared from any elevation of view or azimuthal orientation, allowing the commander-quite literally-almost any perspective he wants on the battlefield he seeks to dominate.

The Problem Armor commanders must appreciate the lay of the land, not only to understand where their forces and the enemy’s can move, but to plan obstacles, to assure interlocking fields of fire, to take advantage of terrain masking, to plan artillery support, and to estimate where an economy+f-force will suffice.

Usually, this understanding comes from map study, but because standard topographic maps represent three Map 1. dimensions as two, considerable inkpretation and imagination are required.

Ideally, it would be better for the commander to get out on the ground himself for a personal appreciation, but this is often impossible or impractical, especially in high-speed operations.

Some on-thespot terrain information will come in from forward scouts and from word-of-mouth, the collected experience of the soldiers who have fought the same terrain previously. This happens routinely at Fort Hood, for example. US. tankers in training there have long understood that there are times of the year when Cowhouse Creek is unfordable; they know there are few good armor routes up Manning Mountain, and they know the places a tank is likely to throw a track or become mired. This terrain knowledge is passed on in the verbal tradition of tankers who have trained here for many years.

These same maneuver areas have been mapped and studied by the US.

Army Corps of Engineers, but this information, too, is seldom put to use systematically. Elevation and soils data, along with information on vegetation, is also readily available, though seldom used.

The computer allows all of these kinds of information to be stored, quantified, manipulated and presented when necessary. Often, the commander will find that when all factors are taken into account in his evaluation of terrain, trafficability may turn out to be better-and sometimes worse-than a pure map study might suggest. In the experiment described here, an area thought to be impassable to armored vehicles, on the basis of a map evaluation, turned out to have avenues of trafficability, along with some areas that really were as impassable as they seemed.

The Study Area The authors selected a section of the Fort Hood reservation that did not have a prepared trafficability map, but with well-studied soils, vegetation, and elevation data. The area is shown in Map 1, a reduction (for space reasons) of the standard 1:50,000 topographic map. (Defense Mapping Agency Topographic Center, 1976, Sheet 6446, “Killeen”). Fort Hood terrain seen in conventional 1:5O,OOO topographic map, opposite page, is shown at right in four computer-generated views, each from a different azimuth. Computer program also allows user to select angle of viewing eieva-tion. A more detalled view of the NE quadrant Is seen on DBW 11. The letters correspond to the points on these maps. Much of this area of Fort Hood, until recently, was little used by armor units for training; the areas occupied a sort of “forbidden zone” in the tankers’ minds because it was so unfamiliar. It therefore offered excellent pros-peds for a test area; the necessary terrain data was accessible as separate maps but had not been correlated. Moving a little ahead of ourselves, the results of the study are shown in Maps 2-5, which show four different views of the study area, each from a separate azimuth, offering a full 360-degree evaluation. Map 6 (See Page

11) further demonstrates the versatility of the process. This map represents an isolation of the quadrant of the best area (corresponding roughly to boundary coordinates PK 3058 to 4258 and 3050 to 4250. The section presented is enlarged four times over the scale of the four basic maps (Maps 2-5). The aspect of Map 6 is oriented on an azimuth of 315 degrees.

This particular sector, judged strictly on topographic map considerations, is one of the most formidable and varied terrain sectors within the overall study area, but when all values are considered, a broad area of excellent trafficability can be seen encompassing much of the Owl Creek Mountain plateau, with wide avenues of approach from the northwest, east, and south. The most northeas-terly portion could virtually be excluded from tactical planning.

Although the following technical description of the Fort Hood terrain mapping project may include mre detail than many ARMOR readers will require, the information is presented here to enable those familiar with data processing equipment to mre clearly visualize the experiment so that they can attempt to duplicate the authors’ results, if thy wish. Ed. The equipment needs for the project was minimal: a standard 1:50,000 military topographic map and maps of soil aqd vegetation types. The computer used was the Control Data Corporation Dual Cyber 170/750 system with two programs on line, the Statistical Package for the Social Sciences (SPSS) and the interactive SYMAP/-SYMW graphic software program.

Although the SPSS was originally developed to analyze social sciences data, many of the statistical fun0 tions within the package are valid for processing terrain data. For example, the package includes bivariate and multivariate routines useful in terrain analysis. Additionally, there are options for transforming data and writing subprograms. (Table 1 lists the simple computer statements that form the basis for this study.) SPSS can compute a series of data analyses with such simple statements. The example in Table 1 demonstrates this simplicity, utilizing the COMPUTE and IF commands. The application of these commands will be explained shortly.

Of the numerous graphics software packages available, the SYMAP/SYMW package is suggested for the presentation of three-dimensional terrain data. This interactive program has been used successfully to interpret data from a variety of terrain environments.

SYMAP consists of six subpackages:

A-Outline-Defines the outer boundary of the study B-Data Points-Locates the geographical coordi-area.

Four Views of the Study Area SE I P K 4 2 3 9 l

HW I P K l V 5 B l NE SW N W SE S W NE Maps 2-5. nates of the data.

C-Otolegends-Allows for legends, scales, etc. D-Barriers-Creates data interpolation barriers of E-Values-These represent the data valuea. F-Map—A series of electives relating to map size, available statistical techniques, contour intervals, interpolation radius of contours, etc. The major options of the SYMW package are the azimuth orientation and the viewing angle from which the plat is observed. (Maps 2-6, for example, are viewed from the perspective of 35 degrees elevation). The SYMW package can operate independently or interactively with a data matrix mated with SYMAP. The output is produced as a threedimensional plot. the interative mode was used during this experiment. variable strength.

T a b 1. Computer Statements Used in Study Card Column 1 Run Name Variable List Input Medium Input Format N of Cases Compute Recode Compute IF IF Recode Print Format Write Cases Read Input Data List Cases Finish Card Column 16 Fort Hood Terrain Analysis Rise, Run, Soil, Veget Card Fixed (3X, 2(F5.2, lX), 2(F1.0)) 292 Slope=(Rise/RunxlOO Slope (0.0 Thru 8.0=1) (8.1 Thru 15.0=2) (15.1 Thru 30.03) (30.1 Thru 40.0=4) (40.1 Thru 50.0=5) (50.1 Thru 60.0=6) Index=(Slope + Soil + Veget) ((Slope GT2) and (Soil GT 2)) Index =Index+l5 ((Slope GT2) and (Veget GT2)) Index =Index+l5 Index (1 Thru 4=1) (5 Thru 6=2) (7 Thru 8=3) (9 Thru 12=4) (13=5) (14 Thru Highest=6) Index (2) (19X,F1.0) Index Cases=292Nariables=lndex How the Experiment Was Set Up Setting up the computer to prepare the threedim ensional views show in Maps 26 entailed the following steps: *In order to adapt the standard 1:5O,OOO military topographic map and grid system to a format consistent with SYMAP/SYMVU, it was necessary to first overlay the map with scaled graph paper (five squares to an inch). The graph N-S and E-W grid linea were coordinated accordingly. This provided a map format from which the terrain model was developed. A standard X-Y digitized plotter could also have been used for greater efficiency. *Data points were selected and marked on the graph overlay. To ensure an adequate number of samples and randomness, we decided to use the existing military grid intersections, transposed to the overlay, as the data points. Points lying outside the reservation boundaries, in park areas, or lying wholly within a body of water were discarded. A total of 292 data points were selected. *The map coordinate points were recoded into graph coordinates and key-punched onto standard computer cards. *Using the Corps of Engineers soil and vegetation maps, six categories were established for each of these Table 2. Slope, Soil and Vegetation Categories (Categories are for designating allowances for mobility only. Ranking is from least inhibitive to most inhibitive.)

Assumptions:

Test vehicle is M60 family MBT, combat loaded. Weight: 56.6 tons Gnd Pressure: 11.5 psi Eng HP: 750 Grade AscenVDescent: 60% Grade Side-Slope: 30% Slope categories 0 - 8.0% - 1 8.1-15.0% - 2 15.1-30.0% - 3 30.1-40.0% - 4 40.1-5O.O0/o - 5 50.1-60.0% - 6 Vegetanon oescriptlon Grasses; 1 m or less in height; Treeslscrub does not exceeed 10Y0 Mixed coniferous/deciduous scrub; 3-3.75 cm dia; spaced 2 m or less; height less than 3 m Coniferous scrub; 6.25 cm dia; spaced 1-4 rn; height 3 m Mixed coniferous/deciduous trees: 5.5-6.25 cm dia; spaced 1-3 m; height 4.5-5m Deciduous trees; 11.25-15 cm dia; spaced 2-3.5 m; height 4.5-6.5 m Deciduous scrub; 12.5 cm dia; 0 m space; height 4 m Soils Description Speck/Purves ASSOC.

Dark gray to red brown clay w/clayey sand lenses Some rutting.

Tarrent 8. Brackett Assoc. Dark to light brown clayey cobble. Some erosional rutting.

Unnamed

Gray-Brown to brown clayey gravel and sand. Some large stones. Shallow.

Unnamed

Dark gray to black clay w/silty sand wet; subject to flooding. Denton-San Saba-Krum Assoc. Gray-brown to brown silty clay. High s hrin k-swel I.

Trinity-Houston Black-Frio Assoc. Dark gray to black clays and silty clay. Wet; subject to flooding. caegory 1 2 3 4 5 6 category 1 2 3 4 5 6 two variablea (see Table 2). The value occurring at each coordinate grid intersection was rank ordered for individual effecta on trafficability, category 1 being the best and category 6 being the worst. .Slope percentages were calculated for each point using the formula rise/run X 100 = %. To maintain simplicity, the results of these percentages were also recoded into the six categories (outlined in Table 2.) Ultimately, however, only the first three slope categories were used because percentages were not identified that exceeded the upper limit of category 3. .The variable categories were then key-punched onto computer cards and indexed by the computer. There were also two IF statements provided which took into effect the compounding effect of poor soils and impassable vegetation on slope accessibility (see Table 1). The resulting index number, a single value for each grid intersection, was then also automatically repunched into an index data set. .The two sets of 292 computer cards, one for the data point coordinates (B-data points), the other for the index values of trafficability (E-values), were combined into a single data set for the computation of the three-dimensional map with the S Y W / SyMW package. The final products, illustrated in Maps 2 through 5, represent four different azimuthal views providing a full 360-degree evaluation of the test area. Being familiar with military maps of topographic relief, the tanker can quickly and easily grasp the relationship between trafficability and mobility.

Map 6, the isolation of the NE quadrant, allows the maneuver battalions and companies a large scale map analysis of an area for which they could conceivably be held responsible. Maps 2-5 provide information for the general planning of a small scale map operation (brigade and higher).

Conclusion

The three-dimensional presentation of trafficability provided by these maps gives useful information to assist the commander. He can determine whether or not a particular avenue of approach into or out of his sector is indeed the high speed and practical route it may appear to be on his two-dimensional topographic map. He can determine which parts of his sedor require his primary focus of attention, or conversely, which areas can be simply reinforced with additional obstacles and defended by a smaller force than he might otherwise assign. After all, if the Threat’s mechanized forces can’t move through a given area, why expend a great deal of effort in defending it?

The S3 Operations officer can quickly assess the different route combinations to insure his commander’s References Anderson, F. E., Black, L, Watling, L. E./, Mook, W., and Mayer, L M:, 1981. “A Temporal and Spatial Study of Mudflat Erosion and Deposition” Journal of Sedimentary Petrology, V. 51, pp. 729736. Department of the Army, 1968. “Planning and Design of Roads, Airbases, and Heliports in the Theater of Operations: Technical Manual 5-330, U.S. Government Printing Office, Washington, D.C. Dougenik, J. A., and Sheehan, D. E., 1975. SYMAP Users Reference Manual, Laboratory for Computer Graphics and Spatial Analysis, Harvard University.

Huckabee, John W., Jr., Thompson, David R, Wyrick, Jim C., and Pav-lot, E. G., 1977. Soil Survey of Bell County, Texas, Soil Conservation Service, U.S. Department of Agriculture. Jones, J. R., in press. “Computer Applications in Coastal Geomorphology” in Applied Geomorphology, P. J. Fleisher and J. Costa (Eds.), Springer-Verlag, Berlin.

Jones, J. R., and Cameron, B., 1977. “Application of Computer Mapping Programs for the Interpretation of Ancient and Modem Sedimentary Environments” Bulletin, Association of American Petroleum Geologists, v. 61. p . 801.

Nie, N. H., Hull, C. H.,’Jenkins, J. G., Steinbrenner K., and Bent, D. H., 1975. SPSS, Statistical Package for the Social dciences, New York, Maraw-Hill.

US. Arfny Engineer Topographic Laboratory, 1977. Fort Hood Terrain Analysis Map Series. force arrives, and arrives in force. He can analyze the battlefield for usable avenues which may appear on his map to be uneconomical. The maps also allow everyone the opportunity to see, three-dimensionally, the advantages and disadvantages, the opportunities and the risks, offered by any given area before the first tank gets mired or stopped. Indeed, it offers this opportunity before anyone has had the opportunity to physically reconnoiter the battlefield.

The advent of the personal-type computer in tactical units offers even greater possibilities. Not only could the information be available in a visually expressive way, but the maneuver battalion will have the capability to generate a three-dimensional index map on demand, in any scale, for any purpose. The necessary data for an area can be collected (ideally from physical field sampling, but also from a variety of maps, civilian agencies, aerial photos, or other intelligence sources), compiled, correlated and stored before it is needed, before the first battle ever begins.

Future Questions There am, admittedly, some unanswered questions which require further study and refinement. The personal-type computer will prove adaptable to this method of trafficability analysis; all that is required is the development of an appropriate software package. The major unresolved question is: does the map provide an accurate analytical tool? Physical testing of the model was not possible due to a lack of access to the appropriate vehicles. There are, however, two divisions of mechanized forces at Fort Hood that could provide the answer. The point here is that it is possible. The technology is available, the data are easily obtainable, and the information provided appears to be applicable. One of the responsibilities in fighting outnumbered and winning is being smarter than our enemy. This particular mapping system provides some badly needed graphic interpretation to terrain accessibility and defense. CAPTAIN CHARLES R.

GRAHAM, USAR, enlisted in 1971 and was commissioned in Armor through Officer Candidate School in 1979. He has served as platoon leader, company executive officer and battalion adjutant with the 3d Bn, 67th Armor, 2d Armored Division, Fort Hood, TX. He was selected for the Degree talion SVAdjutant course, he is currently attending the Armor Advanced Course.

J. RICHARD JONES received his doctorate in 1977 from Boston University. He is currently an assistant professor in the Department of Geo-grapy at the University of Texas in Austin. His research interests include the quantitative analysis and interpretation of terrain data from coastal and subtropical environments.

Completion program in i982 and attended the University of Texas at Austin. A graduate of the Armor Officer Basic Course. JOMC. and the Bat- L !