AD-A269 92r%

ER2 91993.1

Terrain Characterizationfor Trafficability

Sall A. hoopJune 1993

a4

0101 AAppwdA * I%,;

AbstractTerrain material characterization is needed to predict off-rood vehicle perfor-mance, trafficability, and deformation (compaction and rutiing) resulting iromvehicle passage. This type of information is used by agricultural engineers,foresters, military engineers, the auto and tire industry, and a.iyone else con-cemed with off-road, unpaved, or winter mobility. This report appraises thestate-of-the-art of terrain (or substrate) characterization techniques for vehicletraction studies. It concentrates on field measurement of strength-related prop-erties for soil, snow, muskeg, and vegetation, but also discusses how thesecompare with laboratory measurements and the importance of other terrainfeatures (slopes, drainage, and obstacles).

Cover: Clockwise starling from upper left: Portable shear annulus (CRREL);liquid water content measurement in snow using a Denoth DielectricMeter (Instilut fAr Experimentolphysik, Universituf lnnsbruck, Austria);bevameter mounted on a Polecat (photo compliments of the KeewenawResearch Center, Houghton, Mich.); AARI penetrometer in Antarctica(compliments of the Arctic and Antarctic Research Institute, St. Peters-burg, Russia); dynamic cone penetrometer (Waterways ExperimentStation, Vicksburg, Miss.); and Clegg impact soil tester (complimentsof Lafayette Instruments, Lafayette, Ind.),

For conversion of SI metric units to U.S/British customary units of measurementconsult ASTM Standard E380, Standard Practice for Use of the InternationalSystem of Units (SI), published by the American Society for Testing andMat"rials, 1916 Race St., Philadelphia, Pa. 19103.

CRREL Report 93-6 aUS Army Corpsof EngineersCold Regions Research &Engineering Laboratory

Terrain Characterizationfor TrafficabilitySally A. Shoop June 1993

Prepared for 93-22512OFFICE OF THE CHIEF OF ENGJNFFRS

Approved for public release; distribution is unlimited.

PREFACE

This report was prepared by Sally A. Shoop, Research Civil Engineer, of the Applied Rcsearch Branch, Experimental Engineering Division, U.S. Army Cold Regions Research andEngineering Laboratory (CRREL). The project was funded by DA Project 4A762784AT42,Design, 'C,,, .. ation, and Operations 7 echnoiogy for Cola Region.s Trask CS. Work Unit007, Off-Road Mobility in Thawing Soils.

The report was originally written to be included in an American Society of AgriculturalEngineers monograph on traction mechanics, Advances in Soil Dynamics, edited by SverkerPersson.

The author expresses her gratitude to Leonard Della-Moretta (Ret. J, U.S. Forest Service,Technology and Development Center, San Dimas, California; Dr. Ronald Liston of USACRREL; and Warren Grabau (Ret.), Newell Murphy, and Richard Gillespie of the U.S. ArmyWaterways Experiment Station, Vicksburg, Mississippi, for their thorough technical reviews:Maria Bergstad for style and grammar edit; and William Bates for drafting the figures.

The contents of this report are not to be used for advertising or promotional purposes.Citation of brand names does not constitut .n official endorsement or approval of the use ofsuch commercial products.

Wt31DTIC

A..Aes Ion ,I

AA.Ditii

CONTENTSPage

Preface ........................................................... ii

Introd uctio n ............................................................................................... ........S o il ......................................................................... .......................... . . . . . . . . . . . . . .

Penetration resistance ......................................................................................... 1P late sinkage ................................................................................................. ..... 3In situ shear tests .................................... ................................................... .... 4Devices reproducing wheel motion .................................. 7Comparison of test methods .................................................................................... 8A vailab ility ............................................................................................................ 9

Organic terrain and vegetation ........................................ 9S no w cov er ................................................................................................................... 15

Unique aspects of snow .......................................... 16Snow strength indices ........................................... 17

Other factors affecting mobility .................................................................................... 19L iterature cited ........................................................................................................ 20Bibliography ................................................................................ ................... 22A bstract ............................................................................................................... . . ... 25

ILLUSTRATIONS

Figure1. Hand-held cone penetrometer ....................................................................... 22. Components of a bevameter ................................................................................ 43. Shear vane device ................................................................................................ 44 . V ane-co ne ............................................................................................................ 55. Cohron sheargraph ................................................... 56 . Shear annulus ............................................................................. ..... . ........ 67. Portable shear annulus ....................................................................................... 68. Grouser shear plate ......................................................................................... 79. In situ direct shear apparatus ................................................................................ 7

10. W heel arc test rig ................................................................................................. 711, Use of an instrumented wheel to measure tire/terrain interface forces and

tire/terrain strength parameters ................................... 812. Yield envelopes on silty sand for different test methods .................................... 913. Typical organic terrain profile ............................................................ .......... 1014. Vehicle performance on seven common muskeg terrains ................................... I 115. Relative effectiveness of vehicle design parameters on different muskeg

terrains ......................................................................................... . . . . . .... 1216, The muskeg fluke ............................................. 1217. Instruments used to measure shear resistance and tensile strength of the

vegetation mat ............................................... 13

iii

Figure Page18. Basis of trafficability index and brightness temperature by trafficability class ... 1419. Seasonal influence of different ice forms on mobility ..................... 1520. Metamorphosis of snow crystals with temperature and time ............................... 1621. Snow com paction gauge .................................................................................. 1722. R am m sonde penetrom eter .................... ...................................................... 1723. Snow hardness characterization using manual techniques and hardness gauge .. 18

TABLES

TableI. Comparison of cohesion and friction angle measurements................................. 92. Structural classification of vegetal cover cf muskeg....................................... 103. Characteristics of seven common muskeg terrains ............................................. 114. Tearing resistance of muskeg measured with the muskeg fluke ............... 125. Breaking lengths and yield/rupture ratios from tearing resistance tests on

forest so ils ................................................................................................... .. 136. Muskeg topographic classifications .............................................................. 147. Bearing strength of frozen peat .......................................................................... 158. CTI snow compaction gauge values ..................................................................... 17

iv

Terrain Characterization for Trafficability

SALLY A. SH-OOP

INTRODUCTION surface material is a well-behaved continuum (perhapsa bold assumption). A good review of various predictive

This rp•-prt appraises the state-of-the-art for charac- models, along with their theoretical and experimentalterizing terrain material (or substrate) for off-road vehi- basis, is given in Plackett (1985). Each model may re-cle traction or trafficability studies. It was originally quire different material properties as input, and al-,,written for inclusion in Traction Mechanics (Persson, in though many different methods of material characteri-preparation), a monograph in the Advances in Soil Dy- zation exist, none is universally adequate. The same isnamics series of the American Society of Agricultural true of the predictive models. It is of the utmost impor-Engineers. Therefore, although I concentrate on soil tance that the strength characterization technique satis-strength characterization, which is of primary impor- fy the need for the information and be suitable for thetance to agricultural engineers, I also include the unique terrain material in question.aspects of other surfaces such as snow and organic ter-rain, of particular interest to military and forest engi- SOILneers and others dealing with operation of off-roadvehicles. The emphasis is on field measurements, with The fundamental parameters commonly used to de-brief mention of their comparison to laboratory tech- scribe soil for engineering or agricultural purposes areniques. soil type, structure, grain size distribution, Atterberg

Terrain includes the material that comprises the limits, moisture content, and density. These and otherterrain (soil, snow, vegetation) as well as the geometry physical properties of soils, as well as how they influ-of the terrain surface (topography). The ability of the ence soil strength, are fully described in Chancellorterraintosupportandprovidetractionforvehicleopera- (1993). The strength of soil depends on these basiction is called trafficability. In trafficability studies, the physical properties. Measuring soil strength in the fieldemphasis is on the interaction between the vehicle and rather than the laboratory has the advantage of testingthe surface material, whereas mobility considers the en- the soil in its natural state. It is also generally less ex-tire effects of the terrain, including obstacles and topog- pensive and less time-consuming. Although carefullyraphy, on vehicle operation. This report focuses on the controlled laboratory tests may be more exact theoret-terrain material properties that influence trafficability ically, they are impractical for a quick assessment ofand includes a brief discussion ofothereffects to be con- field terrain strength.sidered for off-road mobility, such as terrain features(slopes, obstacles, drainage) and climatic effects on the Penetration resistanceterrain environment (changes in moisture, freeze-thaw). The field cf traction mechanics has a keen interest in

A means of characterizing the surface material is developing an easy and accurate field tool for terrainneeded to predict off-road vehicle performance, traffi- characterization for vehicle traction studies. One of thecability, and soil deformation (compaction and rutting) most popular tools, which the U.S. Army uses exten-that results from vehicle passage. Predictive models sively, is the hand-held cone penetrometer (Fig. 1)calculate the forces developed between the wheels or described in ASAE standard S313.2 (ASAE 1985),tracks and the terrain surface and generally assume the SAE Standard J939 (SAE 1967). and U.S. Army Tech-

nical Manual 5-330 on Soils Trafficability (U.S. Army The ratio of the remolded Cl to the original Cl is called1968). The hand-held cone penetrometer is a simple a remoldin2 index (RI). The product of the CI and the RIinstrument designed to give a quick and easily obtained is called the rating cone index (RCI) and is a measure ofindex of soil strength. The standard WES (Waterways the soil response to repetitive loads, such as multipleExperiment Station) cone penetrometer consists of a vehicle passes.proving ring, or some other force recording device, and A vehicle cone index (VCI is obtained usine vehiclea choice of two sizes of 300 cones. The large cone has weight, dimensions, engine, and transmission factors ina 323-mm 2 (0.5-in.2) base area (15.9-mm-diameter a series of equations and graphs detailed in U.S. Armyshaft) and is used with soft soils and sands. For harder TM 5-330 (U.S. Army 1968). The VCI is representativesoils and soils with fines, a smaller cone, 130-mm2 (0.2- of the minimum RCI required for 50 passes of the vehi-in. 2 ) base area with a narrow shaft, is used. cle. A comparison of the VCI and the soil RCI will result

The force required to press the cone through the soil in a prediction of whether the vehicle is mobile or not inlayers is called the cone index (C0). Five to seven pene- a particular soil.trations should be performed to get a good statistical Several adaptations have been made to the basic

average and an estimate of the variability of the terrain hand-held cone penetrometer, primarily in the ftirm ofboth laterally and with depth. The cone is pressed into continuous readouts, electronic data acquisition, andthe soil at a uniform rate of approximately 30 mm/s (72 hydraulic rather than manual applied pressure (Olsenin./min), although this rate may not be achievable in 1987; Rawitz and Margolin 1991). In these advances. theharder soils. The first reading is taken when the base of proving ring is replaced with a load cell. and the depth ofthe cone is flush with the soil surface and then every 25 penetration is measured with a proximity sensor. Theor 50 mm (I or 2 in.) thereaftcr, depending on the appli- output of the device is then automatically recorded on acation. The index is reported with depth, as an average data storage module or data logger. These developments

over a range of depths or as a gradient. allow full characterization of an inhomogeneous mate-For fine-grained soils, a remolding test may also be rial in an efficient manner and at a low cost.

performed. The remolded sample is obtained by sub- A similar device is the drop cone (Godwin et al 1991jecting a 50.8-mm (2-in.) radius by 152.4-mm (6-in.) in which a 2-kg, 300 cone is dropped from a height of Iheight soil sample contained in a tube to 100 blows (for m. This has the advantage of imparting a large force onfine-grainedsoils, or25blowsforsandswithfines)with the soil without the need to transport large weights ora 1.14-kg (2.5-1b) remolding cylinder dropped from a hydraulic equipment to the field (as would be needed toheight of 0.3 m (12 in.). The cone penetrometer is then impart large forces with the standard "static" cone pen-used to measure the cone index of the remolded soil. etration). Tests at several field sites indicate linear rela-

Handle

Proving Ring

Dial Gauge

18-in. Staff 18-in, Extension Rod

300 Cone

Figure 1. Hand-held cone penetrometer (after ASAE 1985. SAE 19/67).

2

tionships between the drop cone penetration and soil The cone penetrometer is very useful for determin-moisture, vane shear strength indices, and mean wheel ing go/no-go scenarios based on a large database ofrut depth, enabling prediction of soil and crop damage known vehicles. Problems may beencountered, howev-from driving machinery in the field (Godwin et al. er, in extrapolating results to predict performance of1991). Another impact device, the Clegg Impact Soil new or different vehicles. One of the best applicationsTester, is used to assess the condition of low-volume of the cone, because of its sensitivity, is tor spatial char-unsurfaced roads (Mathur and Coghlan 1987) and has acterization of the terrain. For example, Hadas andalso been effective at monitoring soil strength recovery Shmulewich (1990) use a spectral Nialysis of cone dataafter spring thaw for assessing trafficability (Alkire and to determine the spatial arrangement of soil clods.Winters 1986). The variation of impact measurements Ohmiya and Masui (1988) have taken this type ofwithin a site is less than for other hand-held tools analysisonestepfurther.usingthree-dimensionalgraphi-because of the larger soil volume incorporated in the cal representation of the cone data to aid in visualizaili,,test. For this same reason, some researchers have found of the spatial variation of soil strength. The penetrom-that the drop cone is not sensitive enough and prefer the eter has also been very successful i n agricultural studies

static penetrometer. as an indicator of plant growth or root penetration (Tay-

The penetration resistance measured by cone pene- lor et al. 1966. Bowen 1976).trometers is determined by acombination of soil strength Although the value of the cone penetrometer de-properties: shear, compression, tension, and soil/metal pends on the type of study, it is no doubt the most uni-friction. To use mobility prediction techniques that rely versally used and widely accepted index of soil strengthon the more fundamental soil properties, Rohani and for vehicle mobility studies.Baladi (1981) developed relationships between the coneindex and the shear strength and stiffness of the soil. Un- Plate sinkagefortunately, these relationships work only for homoge- The plate sinkage test is used to determine the

neous, frictional soils. Using the theory they developed, pressure-sinkage relationship (or flotation characteris-a cone index can be calculated knowing the cohesion, tics) of the soil. The plate sinkage test performed forfriction angle, and stiffness of the soil, however, the mobility studies differs from that commonly used ininverse procedure is more difficult because of the num- civil engineering for determining bearing capacity. Forberofunknowns.A solution to this inverseproblem was mobility studies, the area of the plate should be largeproposed by Hettiaratchi and Liang (H987) for a drop enough to simulate the contact patch of the tire. (Theindenter (cone) test. By carefully choosing the geomc- same plate is also used to predict mobility of tracks, ittry of the indenter and the type of tests performed, is not the total area of a track but rather simulates thesolutions to a mathematical model of the soil indenter contact area of the track pads supporting the peak load.)can be achieved based on cavity expansion theory. The Sometimes a range of plate shapes and sizes are used.solutions are presented in the form of nomograms The plate penetration equipment can be mounted eitherrelating indentation to soil strength. on a portable test rig or on an off-road vehicle where it

A series of controlled experiments to determine the is possible to generate large normal loads. Repetitiverelationship between penetration and soil strength was loading is used to provide information on terrain re-performed byMulqueenetal.(i97j.Tici. ci . s por;e :' multiple p'(cg-es

are that the relative proportions of the different strengths The plate sinkage test, along with ia shear annulus(shear, compression, and tension) reflected in the cone measurement, is used in a well known terrain character-readings vary with moisture content and the cone be- ization apparatus called a bevameter (for Bekker valuecomes insensitive to shear strength as the moisture con- meter), shown in Figure 2. With a bevameter. platetent increases. In addition, whiie performing the experi- penetration is used to measure bearing capacity. and thements they noted that soil compacted ahead of the cone shear annulus (d;.;cusse, later) is used to determine theeffectively changed the shape of the cone and that the shearing characteristics of the soil. Hence, both the nor-cone shaft sometimes interfered with the readings. mal and shear loading of a vehicle are simulated. These

Similarly, several researchers have studied the ef- strength parameters (c, ., and K representing shearingfectsof soil physical properties on cone resistance (Col- behavior and n, kc. and k0 representing sinkageeare thenlins 1971, Voorhees and Walker 1977, Wells and Tresu- used in Bekker's analytical model for predicting vehiclewan 1977, Ayers and Perumpral 1982, etc.). A good performance (Bekker 1969). Karafiath and Nowatzkireview of the factors affecting the penetration resis- (1978), however, argue that the fundamental assump-tance-water content, bulk density, root density, soil tions behind the test method and analysisare not entire-structure, penetration rate, and soil type-is given in ly consistent with the tractive behavior of a wheel orPerumnpral (1987). track. Won- (1989) gies bevameter results on a range

3

Loading Cylinders F Amplifier

Torque Motor

Amplifier Tque and

p, "L Angular MotionSensors Pressure

-- P1- • iz P lrb2

b1 b2

.' TPenetration PlatesShear Annulus

Figure 2. Components of a bevarneter (after Bekker 1969).

of terrains including different mineral soils, organic Shear vaneterrain (muskeg), and snow. Two vehicle-mounted be- The shear vane device is a simple tool designed tovameters are described in Wong (1989) and Alger measure the shear strength of clays (Fig. 3). The vanes(1988). are typically about 70 mm in diameter and 100 mm in

height but may vary in size depending on the purpose ofIn situ shear tests the instrument. The instrument is pressed into the soil

While the penetration techniques mentioned above and then rotated, and the shear strength of the soil isrelate to vehicle sinkage and motion resistance, mea- reflected in the torque needed to rotate the device as thesurements of the shear strength of soil give information soil fails in shear in a cylindrical shape around the vanemore indicative of tractive performance. Several field circumference. Since there is no way to change the loadmethods for assessing the shear strength of soil are normal to the shear plane, the shear vane is not suitablesummarized briefly below, for frictional soils, but it is handy in silts and clays.

Although there is an ASTM standard fbr laboratory

Rod

-,----- Horizontal Sheared Surface

_• 1I • ,,•Blade

Vertical Sheared Surface

Horizontal Sheared Surface

Figure 3. Shear vane dlevice (after Kogure et al. 1988).

4

Roller Bearing

Handle

Proving Ring Indicator

FTorque Ring Handle

Torque Reacing Disk

Roller Bearing T

Helical Spring

Torque Ring 1iPenetration Rod --

Cone

Vane

FiVure 4. Vane-cone (after Yong and Youssef 1978).

tests using a miniature shear vane (ASTM Standard D4648-87), no standard exists for the field technique.

Vane-cone HandleCombining the penetration resistance measurement

of the cone penetrometer with the shear strength of theshear vane, a vane-cone penetrometer (Fig. 4) wasproposed by Yong et al. (1975). The idea is that the com-pression and flotation behavior as well as soil shear re-sistance can be evaluated with one simple and easy-to-use device. The vane-cone is pressed into the soil and Bearingsthen, at a specified depth, is rotated while the depth isheld constant. Vehicle mobility prediction equationsbased on the parameters given by the results of vane- Rcone measurements are presented by Yong and Youssef RecDrum(1978). In a soil test bin study on a soft clay, they found Recording -

that predictions based on the vane-cone were favorable, Penbut additional studies and acceptance of this combina-tion device are yet to come. •---- Spring

Cohron sheargraphOther types of she:ar devices apply a shearing force

ulong the surface of the soil. Of this category of instru-ments, the Cohron sheargraph is the most compact andeasy to use. It is hand operated by placing the shear headon the soil using the desired normal load and then • ....... Shearapplying a shearing force by rotating the device. Both Headthe normal and shear forces are recorded on the drum Figure 5. Colitron she'tairraph (a.10r Kara-graph attached to the instrument (Fig. 5). Although the fiith and Nowatrki 1978).

• • .. , i I I I I I I

N Annular shear ring SAnnular shear tests were proposed by B (kker I 9 9)and are a part of the bevameter technique of assessing

T soil strength for mooility prediction (Fig. 2 ). To assessshear strength, an annular plate is placed on the soil withan applied normal load and rotated at a constant rate.The annular plates can have either a metal or rubbersurface as well as grousers (Fig. 6). The test is per-formed at a rz age of normal loads to determine theCoulomb shear strength parameters correspondin, to

the soil/metal or soil/rubber shear. Stafford and Tanner(1982) suggest that more than six different normal loads

N be used during the field procedure to obtain significantresults.

T The shear annulus is commonly mounted on an off-

road vehicle as part of a bevameter. as described inWong (1989) and Alger (1988). However, a smaller andsimpler set-up, which can be operated by one person(Fig. 7), is described by Stafford and Tan'ner 1982).

Figure 6. Shear annulus. A drawback to the technique is that the failure of thesoil beneath the shear ring can occur on a plane oblique

has beeu around tot many years and is still to the piane of dte annulus ring, so the true normal andin use (Flores 1990), it has not become widely accepted. shear stress values along the failure plan are unknownIts light weight and small size make it a handy field (Liston 1973). The development of the oblique failureinstrument, but at the same time the readings are less planes,howevercanbeavoidedbyplacingasurchareconsistent because of the small area sampled and the onthesoil aroundand inside the annular ring (Karafiathsensitivity to operator error. However, Patin (1972) and Nowatzki 1978).found the Cohron sheargraph to be more consistent thansimilar techniques (using a spline shear device) and Grouser shear p/are

stated that the sheargraph "yielded measurements that A similar concept is the shear plate where a plate orwere somewhat more indicative of the tire performance grouser is placed on the soil surface with a range ofthan those obtained with the cone penetrometer." normal loads and the plate is then sheared across the soil

20 cm

A - Annulus and shield in ground F - Shear angle transducerB - Loading lever G Instrumentation and recorderC - Torque transducer H - Lift handlesD - Mechanism for withdrawing shields I . Support legsE - Reduction gear box and manual drive J - Transport wheel

Figure 7. Portable shear anndaus (after Stqffiord and Tanner 1982).

6

ditions (moisture, density, and texture) remain closer tothe actual field conditions, as compared with gatheringand removing a sample to be tested in the laboratory.

N/ FSample preparatiop consists of carefully cutting an

,,1,0 X "undisturbed" sample or excavating the soil so that theZ / 7shear box can be placed around me soil in situ (Fig. 9).

As in the laboratory, the soil is sheared at several applied

Figure 8. Grouser shear plate, normal loads. Because of the time-consuming nature ofsample preparation, however, usually too few samples

(Fig 8.). However, rather than a rotational shear, the aretestedsoagoodsamplingofthenaterialis enerallyplate moves across the soil in a linear mode. not obtained.

Both the annular shear ring and the shear plate may Devices reproducing wheel motionneed to be mounted on a heavy portable stand or avehicle to achieve the necessary v' *ical load. In addi- Wheel-shaped devicestion, the shear plate may need high horizontal forces. A To characterize the tractive capacity of the soilportable test rig that includes a translational shear accuratel), the test device should more closely simulategrouser as well as plate sinkage and cone index capabil- the motion of a wheel (or track). Thus. a new kind of testities is described in Upadhyaya et al. (1990). rig, where the device acting against the terrain is shaped

like a segment of a wheel and acts like a wheel slippingIn situ direct shear over the soil surfaice, was proposed by Wasterlund

The in situ direct shear test essentially duplicates the (1990). The simulated wheel is made of a rubber surfacedirect shear test commonly performed in the laboratory. over rigid steel and moves in an arc across the soil. asBy performing the test in the field, the pretest soil con- shown in Figure 10. This apparatus has been used to

N. Normal Load

PROFILE VIEW

Figure 9. In situ direct shear apparatus.

1. Frame2. Adapter tool holder3. Axle spindle4. Lever for load application5. Hydraulic cylinder for loading6. Wheel segment7. Wheel spike8, Hydraulic cylinder for horizontal force9. Horizontal position indicllnr

10. Transducer for horizontal force11. Controls and setting values for hydraulic pressures and flow

Figure 10. Wheel are test rig (after Wasterhund 1990).

7

Vertical

Proximity

L ock-Oit~u GaugeHub

i k L lCFSpeedindicatorPins

Figure) 1. Use ofan instrnmented wheel to measure (top) / . Transversetire/terrain interface forces and (bottom) tire/terrainstrength parameters (after Shoop 1989, 1992). Longitudinal

T9

On, N/Area

characterize the strength of the forest floor to avoid ter- tion of the mechanics of the tire/soil interface but forrain damage from forestry operations. firm soils the internal angle of friction is also dependent

on slip velocity.Instrumented vehicle wheels

A vehicle with instrumented wheels can also be used Comparison of test methodsto assess the strength of the tire-soil system (Shoop Several studies have been conducted comparing the1989, 1992). Triaxial load cells mounted on the wheel results of various strength meaurement techniquesaxles measure the forces at the tirelsoil interface as re- (Patin 1972, Johnson et al. 1987, Kogure et al. 1988,corded through the response of the axle (Fig. I I {topD). Shoop 1989). Okello (1991 ) stroneglv recommends theDuring a traction test, the measured longitudinal force use of in situ techniques oer laboratory techniques butis equivalent to the net traction at the wheel. Gross trac- emphasizes that the size of the device (retferrirTI to p late

tion (Tg) applied to the soil surface is then estimated by sinkage) should be comparable to the size of the Iu- orsubtracting the motion resistance, and the applied trac- track elements. As expected. there is disagreement be-tive (longitudinal) and vertical forces are converted to tween the shear strength values obtained from the differ-stresses by dividing by the tire contact area. Traction ent test instruments, primarily because ofthe magnitudetests areperformed at a range o applied normal stresses and direction of the applied stress and the rate of de-by changing the tire contact area using different infla- formation.tion pressures. The terrain-tire shear parameters are One of the most comprehensive situdies of shear testcalculated using a Mohr-Coulomb approach (Figure techniques was published by Stafford and Tanner (1982).1I [bottom ]); these mobility terrain parameters are used They compared six shearing techniques on six differentto characterize the soil for mobility purposes and to pre- soils, with the results summarized in Table I Althoughdict the mobility of other vehicles on the same soil results vary with soil type, the vane shear consistentlyconditions, yields the highest values of cohesion, except when used

Komandi ( 1990) alsocalculated soil parameters from on remolded soils. Similarly, when comparing vanevehicle slip-pull curves obtained in the field. He con- shear, direct shear, and triaxial tests. Ko!ure et al.eludes that the Mohr-Coulomb theory is a valid descrip- (1988) report the vane shear to yield the highest values

I P •

Table 1. Comparison of cohesion and friction angle measure-ments (after Stafford and Tanner 1982).

DirectTorsional shear shear Triartial test Shear

Soil Box Annulus Box Undrained Drained vane

a. Cohesion (kPa)I 28.5 39.3 14.8 6.5 17.4 43.32 36.3 41.9 54.3 11.1 14.4 54.13 81.9 88.6 50.1 33.5 41.9 92.84 33.0 36.1 19.9 11.3 15.5 50.35 40.4 62.3 28.2 10.3 21.1 38.56 1.6 4.2 6.2 4.9 3.1 5.5

b. Friction angle (deg.)I 28.5 33.2 33.2 25.4 29.7 -2 20.2 13.1 21-4 17.3 26.4 -3 38.1 22.9 24.0 11.3 11.3 -4 29.0 30.8 34.4 16.9 21.9 -5 20.5 21.3 8.3 6.0 2.8 -6 14.8 8.8 31.6 31.8 33.3

Soils I= sandy clay loam 4 = peat2 = clay 5 = remolded clay3 = clay with stone 6 = remolded sand

of shearstrength and thedirect shearthe lowest. Kogure kamp, Agrisearch Equipment in The Netherlandsalso studied theeffects of sample orientation foreachof (through Sauze Technical Products Corp. of Platts-the tests and found the results from the vane shear to be burgh, N.Y.). The cone penetrometer is also availableindependent of orientation, In studies that have includ- through the U.S. Army supply system. The Clegg lm-ed the (Cohron) sheargraph, summarized in Johnson et pact Hammer is available from Lafayette Instrumental. (1987), the sheargraph was found to yield the highest Company. The in situ direct shear is usually a modifica-values of cohesion but not necessarily the highest fric- tion of laboratory direct shear instruments. and the sheartion angle. Shoop (1989) compared shear annulus, di- plate, shear annulus, and plate sinkage equipment arerect shear, and triaxial tests with terrain strength values generally made to specifications. The wheel simulationcalculated from traction tests (on silty sand) and found test rig was custom-built at the Swedish University ofthat the undrained triaxial tests most closely compared Agricultural Sciences in Garpenberg, Sweden. lnstru-with the failure envelope calculated from the forces at mented wheels and vehicles are custom built by Hodgesthe vehicle tire/soil interface (Fig. 12). Transportation of Carson City, Nev.; Testing Services

and In-strumentation of Westfield Center, Ohio; Data-Availability Motive, Inc., of Reno, Nev.; and the Cold Regions Re-

Of the techniques discussed, the sheargraph, shear search and Engineering Laboratory of Hanover. N.H.vane, and cone penetrometer are commercially avail-able through Soiltest, Inc., of Evanston, Ill., or Eijkel- ORGANIC TERRAIN AND VEGETATION

150 Organic terrain, also called muskeg, is a term used toI Idescribe terrain comprising a surface layer of vegeta-

tion with a subsurface layer of peat or fossilized plantL_ debris. It includes terrains such as peat bogs. swamps.

:2s. 100 -0 "(- tundra, and forest floor. The surface of this terrain is

\ -e\ composed of a living organic mat of mosses, sedgzes.8 .and/or grasses, either with or without tree and shrub0 50 growth. Underneath this vegetative mat is a mixture of

Shea Anns partially decomposed and di sintegirated organic materi-Sal called peat or muck. To be classified as muskeg, the

I .1 peat must be over 450 mm thick when undrained or 30W0 100 200 mm when drained and have an ash content less than 801(C-

Normal Stress (kPa) (Radforth and Brawner 1977). The typical stratigraphy

Figure 12. Yield envelopes on silty sandfor diffrrent test of organic terrain is sketched in Figure 13.methods (after Shoop 1989). As a rule, peat or muck is highly compressible

9

Surface Vegetation

Mat of Live Roots

Peat Moss

SIFigure 13. Typical organic lerrain (tnuskeý):.:i profile (after Yong 1985).

Inorganic Soil(silty -sand - clay)

compared with most mineral soils; it is characterized by yielding the nine pure vegetative coverage classes givenits very high watercontent and its extremely low bearing in Table 2. No species identification is necessary; onlycapacity (MacFarlane 1958). To a great extent, the the qualities of the vegetation are needed, making thistrafficability of muskeg depends on the strength of the classification system suitable for use by engineers orvegetative mat overlying the soft peat or muck below, scientists unskilled in plant identification. Since the

and vehicle mobility depends on the success of the classes usually occur in combinations, the terrain isvehicle to utilize the strength of the mat effectively designated by combinations of two or three of thesewithout tearing or breakage. class letter designations, starting with the most promi-

A classification system for muskeg was proposed by nent class. Seven common inuskeg classifications areRadforth (1952) and compiled into a field guide by described by Radforth and Evel (1959) in Table 3.MacFarlane (1958). The classification scheme is based The descriptions and classification of the variouson the vegetation, the contained peat/muck, the underly- types of muskeg offer qualitative indications of theing mineral soil, andthe topography. This was integrated engineering properties of the terrain, particularly withwith the system of the British Mires Research Group respect to vehicle mobility. For the muskeg classifica-

Table 2. Structural classification of vegetal cover of muskeg (peatland) (integration of British and Canadian systems,from MacFarlane 1958, 1969).

British Mires Class Radfarth Sy•-stenResearch Group* symbol Texture Stature Form Examle,

Trees > 5 m A Woody 4.5 m (015 t1` or over Tree form Spruce. larchTrees < 5 m B Woody 1.5-4.5 m (5-15 ft) Young or dwarfed tree or bush Spruce. larch. willou. birch

or overShrub habit, D Woody 0-6-1.5 m (2-5 ft) Tall shrub or very dwarfed tree Willow. birch. Labrador tea

500 mm to 2 mShrub habit < 500 mm E Woody Low shrub Blueberry. laurelCreep shrub < 500 mmBroad-leaved herbs G Nonwoody Up to 0.6 m 12 ft) Singly or loose association Orchid, pitcher plantSedge-graminoid habit, C Nonwoody 06-1.5 m (2-5 f0t Tall, grasslike Grasses

1-3 ma) matsb) hummocks

Sedge-graminoid habit. F Nonwoody Up to 0.6 mn (2 ft) Mats. clumps, or patches Sedges, grassc,< I m sometimes toutchine

a) matsb) hummocks

Moss habit I Nonwoody (soft Up to I W imm (4 in.) Often continuous mats. ,los-,csor velvety) sometimes in hummock.s

Lichen habit H Nonwoody Up to I W mm (4 in.) Mostly continuous mats l.ichcn,.. . ... (leathery to crisp)

*Adapted by Radforih.

NOTE: Following classification, observer states percentage of cover class within 20%.

10

Table 3. Characteristics of seven common muskeg terrains (after Radforth and Evel 1959).

Common Associatedformulae topographi.fe/atures Subsurifce peat strun'tre

AE Irregular peat, plateaus Coarse-fibrous. woodyAEH Irregular peat, plateaus, rock enclosures Woody coarse-fibrous A ith scautered wood crrauic,DFI Stream banks Woody panicles in nonwoody fine-fibrousDEI Ridges, stream banks Woody particles in nonwoody fine-librousEH Even peat plateaus, polygons Woody and nonwoody panicles in fibrousEl Ridges, mounds Woody particles in nonwoody fine-eibrousFt Hummocks, closed and open ponds, polygons. flats__ Amorphous granular. nonwoody fine-fibrous

tions given in Table 3, a corresponding range of vehicle ter, and plant identification along with trafficabilityperformance parameters (displayed graphically in Fig- measurements at two bogs in Sweden, and they foundure 14) and desirable vehicle design factors (Fig. 15) that vegetation was the major factor influencing traffi-were assessed by Radforth and Evel (1959). Other cability.engineering characteristics of each of the muskeg cover Many of the techniques used to assess trafficabilityclasses are given in MacFarlane (1969). of soils have also been used on muskeg with varying de-

The Swedish Army has also been very successful grees of success, depending on how the test evokes theusing plants as indicators to predict trafficability of strength of the vegetative mat. Since the vegetative matmuskeg. Fridstrand and Persson (1990) analyzed data overlays very soft peat, the degree of vehicle mobilityobtained using cone penetrometer, vane shear, bevame- depends on the flotation and traction provided by the

Mechanical Terrain Reference TypesFactor AE AEH BEI DFI EH El Fl

Possibility for __

Maximum Speed6 mph

High

EffectiveManeuverability

Low

Pitching High

High

Load TowingRequirement

Low

High

DisplacementLow

HighDisplacement

re: Towed Vehiclewith No Traction

Low

Figure 14. Vehicle perfourmance on seven coinnmon oniskeg ter-rains (after Radforth and Evel 1959).

II

MechaicalTerrain Reference TypesFactor JJ AE AEH BE) DFI EH El Fl

Track RunConforming to

Ground Contour

Frontal DriveSprocket

SynchronizedWinch and Track

Ground PressureLower than

1112 lb/in.2

(10 IkPa)

Figure 15. Relative effectiveness of vehicle design parameterson differenit musk-eg terrains (after Radforrli and Eve! 1959).

mat. In addition, repeated loading of the mat by multiple (U.S. Army 1959), shear vane (ThOmsoN01 1960, Irwvinpasses may pump fine-grained muck up onto the mat and Yong 1980), and bevamneter (Won!z 1%989.surface, reducing the traction through slipperiness. The Several instruments have been designed specificall%mat'sability to suppor-tvehicles is provided by the over- to measure the tensile or tear strength ofrrmuskeg_ and

alltenilestengh o th vgetation and the interlocking vegetation mats. MacFarlane ( 1969) describes a muske

stems and roots. Many of the traditional soil strength "fluke" consisting of several spikes (Fig. 16) insertedmeasurements fail to obtain an adequate measure of the into the vegetation and attached to a cable to which atensile properties of the mat and therefore inadequately load is applied. Measurementsof the mat tearine strengthcharacterize the terrain for trafficability. Even so, some using the fluke are reported in Table 4. Scholandersuccess has been reported with the cone penetrometer (1973) measured the tearing strength of several foresl

Table 4. Tearing resistance of muskegmeasured with the muskeg fluke (afterMacFarlane 1969).

A vg .sherci or,~ te

C -terfmwwlia -{b

Fl. wet heciween El mounds 2.100 ).3i41

~" ~,El mounds, E - 1 1.6501 7.339FIE .41) 0.898

El mounds E - I 1.46)7 6,S525[ I flow. wet area) 2,607 11,.863

7, F1 (ery wet). dense F 2.483 1(.1)44El mounds. denisL E 2.7SS 12.4011

IF. I s-crN dense 1,7(00 7,560- V. -_"44 El mounds. E 1 ( .911 8,59 N

t)FI IerN e 1.951) 8.674

__ ~ Fl (wry kketl 1.650) 739t~t*El miounds. F 71 2,1)5( 9.11S

% FI. F = I 21,417 1(1.751IL117humnuioets 1.71-1 7.63'

.. .... .. ~ FIE. F I 2.367 (0.28,

Figure 16. The niuskeg fluke (front Mac Farlane /969). El 11ITiIMfO~kiý 2.000)1) I 565

12

Table 5. Breaking lengths and yield/rupture ratios from tearing resis-tance tests on forest soils (from Scholander 1974).

a. Survey of mean values of Sbmeaking (stretchdistance) of different types of vegetation andsoils.

Vegetation Soil Sbrtakingrv•__Ope texture .... mn,)_

None present Sand 50-100Grass Fine sand, silt 140-220Dwarf-shrub Sand 280-330

b. Ratio between yield and rupture limit of some uniform vegetation types.

Force Extension

Soil No. of Mean Std. Mean Sid.Vegetation texture observations value error value error

Grass Silt 32 0.78 0.02 0.58 0.03Dwarf-shrub Sand 22 0.70 0.03 0.53 0.04Grass Well moldered pat. 35 0..76 0.03 0.51 0..04.9.

soils (vegetation-covered mineral soils) by inserting a surface, in the pulling direction, with the final rupturevertical plate into the vegetation mat and applying a occurring as a tensile failure of the root mat at the sidesloadbypullingtheplatewithavehicle-mountedwinch. and bottom. This is the same failure progression ob-The r-'sults show that the vegetation cover provides served by Niemi and Bayer(1970) from tear resistancethree to five times greater tearing resistance than bare tests on muskeg using the instruments shown in Figuresoil (sand), and that the tearing resistance varies only 17. Bjorkhem et al. (1975) used plate sinkage test,, tomoderately throughout the unfrozen part of the year. evaluate the effects of roots on compressive strength (orTearing resistance varies with soil conditions, but with- bearing), finding that even though the modulus valuesin the same conditions the rupture force is a constant are nearly the same, the ultimate strength of the root-function of the breaking length. This breaking length soil system was 70% greaterthan forsoils withoutroots.can be compared with the wheel slip as a percentage of In a summary report describing several years ofthe contact length to determine if the vehicle will tear research for forest operations on peat lands in Finland,the mat. Some of the measured breaking lengths and Rummukainen (1984), Saarilahti (1982), and Saarilahtiyield/rupture ratios are given in Table 5. Scholander(1973) also observes that the vegetation fails first on the

Weights For Adjusting Normal Load

0

Figure / 7. Instruments used by Niemi and Bayer (1970) to measure (a) shear resistance and (bh) tensilestrength of the vegetation mat.

13

Mean vane Surface wetness classt 268 1 I "

shear strength 1 2 3 4 5(kN/m2)* Index of trafficabiq4y 20 T

20.0- 1 3 6 8 9 .1.4

17.5-19.9 2 4 7 8 9 j15.0-17.4 3 5 7 8 9 '0 252 -12.5-14.9 4 6 7 9 90.0--12.4 5 7 7 9 9 >

•CT

Surface wetness classes: o1 244

1-Dry, boot sole dry2-Normal, boot sole wet3-Wet, water over boot sole 236 [4-Very wet, water rises on boot upper 0 2 4 6 8

5-Extremely wet, water rises over boot upper Index of Trafficability

Figure 18. Basis of trafficability index (top) and brightness temperature by trafficabili. class(bottom) (after Saarilahti 1982, Rummukainen 1984).

and Tiuri (1981) suggest that the traditional strength muskeg terrain are associated topographic features,measurements are singular values, while continuous seasonal ice forms, and ice thickness. Small-scale tor-information is more appropriate for estimating vehicle rain roughness features, such as hummocks, polygons,mobility. They warn that indicator plants may not be re- ridges, ponds, and bars, are included in the topographicliable as they are adaptable to variations in growing classifications of muskeg shown in Table 6 (MacFar-conditions and competition from other plants. Based on lane 1958), and seasonal ice forms are characterizedevaluations of peat lands for trafficability using based on their effect on mobility, as shown in Figure 19penetrometers, vane shears, and bevameters, as well as (Radforth and Evel 1959). Some of these are not largeradar techniques to assess peat depth and water content, enough or of an areal extent to stop a vehicle, but theythey note that the strength of peat is directly related to may slow vehicle progress considerably and cause wearthe moisture content and depth. A trafficability index and tear to vehicle components. Although ice formsbased on vane shear strength and surface wetness class may impede vehicle travel, frozen peat lands are substan-was developed and related to radiometer brightness tially stronger than when unfrozen. The compressivelevels, as shown in Figure 18. Thus, radio wave tech- strength of frozen peat can be 350 to 400% higher thanniquesareproposedasanalternativetothemorelimited unfrozen peat depending on the water and vegetationpoint-wise measurements (cone, vane, and bevameter) content (Rummukainen 1984). Generally, 0.2 to 0.3 mfor evaluating peat lands for vehicle operations. of frost on wet peat lands will bear most heavy equip-

Other factors commonly influencing mobility on ment (MacFarlane 1969), and less frost will bear weight

Table 6. Muskeg topographic classifications (after MacFarlane 1958).

Contourtype Feature Description

a Hummock Includes "tussock." has tufted top. usually vertical sides. occurringin patches, several to numerous

b Mound Rounded top. often elliptic or crescent-shaped in plane viewc Ridge Similar to mound but extended. often irregular and numerous,

vegetation often coarser on one sided Rock gravel plain Extensive exposed arease Gravel bar Eskers and old beaches (elevated)f Rock enclosure Grouped boulders overgrown with organic depositg Exposed boulder Visible boulder interrupting organic depxosith Hidden boulder Single boulder overgrown with organic depositi Peat plateau (even) Usually extensive and involving sudden elevationj Peat plateau (irregular) Often wooded. localized and much contortedk Closed pond Filled with organic debris, often with living coverageI Open pond Water rises above organic debris

m Pond or lake margin (abrupt)n Pond or lake margin (sloped)o Free polygon Forming a rimmed depressionp Joined polygon Formed by a system of banked clefts in the organic deposit

14

Ice Form Season

Jan - Mar Apr - Jun Jui - Sep Oct - Dec

Ice Knolls

Joined Polygons

InterruptedIce Sheets

Frozen RiverBanks

Ice Depletion inDisturbed Terrain

Lensing

Thin Sheeting

"Ridging

Figure 19. Seasonal influence oftdifferent iceforms on mobilitv (after

Radforth and Evel 1959).

Table 7. Bearing strength of frozen peat (after rence. These and other environmental aspects are moreRummakainen 1984, Hakkarainen 1949). thoroughly presented in Radforth and Brawner (1977).

Thickness (m) offrozen peat laver SNOW COVER

Dry top Wet lop There are a variety of techniques for characterizingpeat laver peat laver Bearing capacity

snow for vehicle mobility or tire traction testing. Snow0.10 0.05 Will bear a horse surfaces that are used for testing are either natural or

0.15-0.20 0.10 Will bear 6-t horse sled traffic groomed and vary widely in strength and texture. In0.20-0.35 o. 15-0.25 Will bear empty 4-t truck some ways, the methodology of characterizing snow is.0.3_5-0.50 _0.25 .Will bar 10-t trk traffic. similar to that for soil. Grain size, structure, metamor-

phic state, temperature, density, free water content,according to the guidelines (Table 7) provided by Hak- hardness, and strength are measured or described atkarainen (1949). each significant layer within the snow pack. Since some

Vehicle traffic can also adversely affect the vegeta- of the techniques used are also used on soil, the follow-tion and the sensitive environment typical of organic ing is a summary of the techniques or aspects unique toterrains. Plant damage causes losses in forestry opera- snow. A more extensive summary of snow character-tions, significant changes in drainage patterns, and ization techniques for mobility and snow pavements isassociated erosion. In permafrost areas, changes in the presented in Shoop and Alger (1991) and Abele (1990)vegetation cover alter its thermal characteristics, resull and classification of seasonal snow cover in Colbeck eting in thermokarsts and changes in permafrost occur- al. (1990).

15

Unique aspects of snow low (uniform) snow, a temperature measurement at 25The size and shape of the ice grains that make up a mm and at the snowlground interface is sufficient.

snowpack have a marked influence on the mechanical The texture and structure of a snow cover are contin-behavior of the snowpack as a whole. Large rounded ually changing. Becausea fallensnowflakeisinaphysi-crystals tend to roll past each other, w! small angular cally unstable form on the ground. it changes its shapecrystals tend to pack tightly together when loaded, with time and is strongly influenced by temperatureCrystal size and shape are generally documented using gradients. Typical stages of snow metamorphism area magnifying glass or hand lens and a measurement shown in Figure 20 (Colbeck 1987). Metamorphismgrid. A comprehensive guide for classification of snow affects the shape, size, and bonding of the crystals andcrystals is given in Colbeck (1986) and Colbeck et al. therefore the strength characteristics of the snow cover(1990). and how it will react when trafficked. Freeze-thaw

Because snow exists close to its melting point, the cycling, for instance, can cause ice lenses to form. mark-temperature of the snow environment is extremely edly increasing the strength of the snowpack in a veryimportant. Temperature can be measured by use of a short period of time.simple thermometer or with arrays of thermocouples or Even when temperatures are below freezing, thethermistors. The temperatures are normally measured snow mass may contain some free or liquid water and,in a profile through the thickness of the snow pack. because of the melting of the snow grains when heated.Temperature can be used to estimate the probable the free water content measurement is much different"'wetness"ofthe snow and, when coupledwith adensity from the standard soil water content measurement. Anmeasurement, can also give a very crude estimate of estimate of whether water is present can be made usingstrength. When working with snow, the air temperature visual observations such as squeezing and forming theand snow temperature should always be measured. If snow or by chemical indicators that change color \vhenthe snow is deep or if temperature gradients exist within in contact with liquid water. The quantitative measurethe snow (i.e. the air or ground temperature is •.gnifi- of liquid watei contrent was historically determinedcantly different from the snow temperature), a profile of using either freezing or melting calorimetry, whichsnow temperature measurements is required. For shal- require a good deal of time and careful effort and are

E

Time

Figure 20. Metamorphosis of snow crvstals with temperature and tine (after Colbeck /1987).

16

therefore not desirable for field use. A more recent Table 8. CTI snow compaction gaugeadvancement in liquid water measurement is a capaci- values (after SAE 1985).tance meter that is accurate and easily operated. This C-7, ,,,•,,,,c.iI,gauge consists of a plate that is placed on or in the snow Suýýce descripiot taWgc

and a small meter that is used to read out the capaci- ...

tance. By taking a reading in the air and one on the snow Steel 1oosurface, the free water content can be determined. This Ice 91 -95

Extra hard hard-pack snow h4 93method is becoming increasingly more popular since it Standard medium hard-pack snow 7) .- ••does not require any special fluids or bulky equipment Soft-pack or loose-pack snow 50 7(0

and the gauge can easily be carried in a back pack. Virgin snow - No rating: UseBoyne and Fisk (1990) compare these three methods of depth and moisture content

moisture measurement in snow (alcohol calorimetry, Waterfreezing calorimetry, and capacitance).

Snow density is measured in much the same way as Osborne (1989), and Wong (1989). The drop cone.soil density, by collecting a sample of a known volume however, is slightly different from that used on soils: itand weighing it. is sometimes referred to as a snow compaction gauge.

The snow compaction gauge shown in Figure 21. builtSnow strength indices by Smithers Scientific Services, Inc.. of Akron. Ohio, is

The methods presented below are a summary of field similar to a soil drop cone except that the cone has beenmethods used for quickly assessing the strength of a rounded. The 220-- (7.75-oz) cone is dropped i-om asnow cover. More sophisticated snow strength and heightof219 mm (8.5 in.). The penetration is a result otindex property measurements are given in a review of vertical and horizontal compaction and shear and indi-snow mechanics by Shapiro et al. (1993). cates the compaction resistance of the snoh ,over. The

penetration distance is converted to compaction num-Bevameter and drop cone bers using the standardized scale shown in Table 8.

All of the strength measurement techniques used insoils have been tried on snow with varying degrees of Rammsondesuccess. The most common of the soil strength charac- The rammsonde is similar to the cone penetrometerterization techniques that are also applied to snow are except the standard ramm cone is much larger in sizethe bevameter and the drop cone. The bevameter was and is driven into the snow using a drop hammer (Fig.adequately covered above and its use on snow is dis- 22). Generally a complete ramm set-up will have twocussed in more detail by Alger (1988), Alger and different sized hammers along with a hammer slide and

i;i

Steel 100 ZJciIce 98e -93

Extra hard-pack snow 8-Standard medium hard-pack snow 7

Soft-pack or loose-pack snow - 60-50

Figure 21- Snow compactioni gauge, also called a drop Figure 22. Ram iisondell elet roneter(atfterAhcle 1990).cone (after SAE 1985).

17

several rod extensions for use in deep snowpacks. To Manual snow hardness classificationuse the rammsonde, the cone is placed on the snow Snow hardness can also be classified mnanually.surface and the slide hammer is dropped from a mea- without the aid of gauges or instruments, as described insured height. The penetration of the cone is measured the Swedish Terrain Classification System for Forestryand the process is repeated until the ramm has penetrat- Work (Swedish Forest Operations Institute 1992) anded the entire depth of the snow pack (or to whatever CRREL Instructional Manual I (CRR-'L 1962). Thedepth is desired). This in:m,:umf nt has been most suc- hardness is tested and classified by the ease of pushingcessful in deep packs such as avalanche zones and in the a fist, ou:stretched hand, finger. pencil, or knife into theArctic and Antarctic to obtain hardness profiles through snow. The hiardness is determined along the profile of adeep layers of snow. Correlations between the ramm- snow cover' the overall hardness of the cover dependssonde and several other snow properties and strength on the percentages of each classification present. Themeasurements are presented in Abele (1990). Use of test technique and hardness classifications arc dia-datafrom arammsonde for vehicle performance predic- gramrmed in Figure 23 and are roughly correlated withtion is described in Wong (1992). values from the hardness gauge, as indicated on the

figure.

Canadian hardness gaugeThe Canadian hardness gauge (and similarly, the Commercial tre traction testing <on snow

CRREL hardness gauge) measures resistance to pene- To assess the tractive performance of diffcrent tiretration with small plates designed to be carried in a pack designs on snow, the tire and automotive companiesin the field. Its major use has been in the area of ava- standardized thesnow with re2ard to the tractive perl'or-lanche prediction; it is best used in hard virgin snow. mance of a Standard Reference Test Tire (SR1T-) orThe plates are various sizes, and the size used depends Snow Monitoring Tire (SMT)_ The snow test section ison the strength of the snow cover. The plate is pushed prepared by tilling, grooming. and compacting as nec-into the snow either horizontally or vertically, depend- essary to provide a specified tractive coefficient usingingonthepurposeofthetest, and the resistance to pene- the SRTT as outlined in SAE Standard J1466 (SAEtration registers on a gauge built into the instrument 1985). If the SRTTperforms within the specified range.handle. Hardness is generally measured at each of the the snow is considered adequate for comparing thesnow layers within the cover, traction of other tires. Each tire is tested several times

Testing snow hardness Corresponding hardnessThe hardness class is based on which of the Hardness gauge readingfollowing can easily be pushed into the snow. class (g.em2)

Closed fist covered Very softwith glove

0-500

Flat extended hand f Softwith glove

Extended glove-covered I Medium hard 500-2500index finger

Pencil Hard 2500-5500

Knife - Very hard >5500

Figure 23. Snow hardness characterization using matnual techniques and hardness gatge (afterCRREL 1962, Swedish Forest Operations Institute 1992).

18

andondifferentdatesthroughoutthe winter.Onagiven cle movement and in extreme cases actually sloppingtest date, the SRTT is tested many times throughout the vehicle motion. On the other hand. vegyetation can alsoday (every third tire) to be assured that the snow is provide flotation and traction, supporting vehicles incontinually meeting the standard traction criteria (Shoop very wet and soft ground environments not otherwiceet al. 1993). trafficable, such as bogs or wet forest floor. In these

cases it is very important to limit the breakage of theOTHER FACTORS AFFECTING MOBILITY vegetative mat (as discussed in Organic Terrain and

Vegetation above). Larger vegetation presents obsta-Aside from the strength ofthe substrate, other terrain ties to vehicular movement and limits visibility. This

factors influencing vehicle mobility include vegetation, kind of impediment is often characierized by trunk andobstacles, terrain profile (micro relief), water courses, stem diameter, spacing. and branching frequency. Siln-and slopes. Any of these factors may change with time ilarly, boulders can create obstacles and/or prosidedue to natural conditions such as rainfall or snowmelt or reinforcement to otherwise weak terrain material (suchman-made conditions such as farming or construction, as wet soil).

In general, the vehicle and driver respond to those Nearlyall ofthese factors are subject tochanes withfactors that absorb energy (by increasing motion resis- tnie: daily, seasonally, annually, or over rnany year*.tance, inducing drag, reducingtraction,oractivating the Temporal changes occur in nearly all of the terrain-vehicle suspension), thus reducing or eliminating too- related factors such as soil moisture and density. planttion. Grabau* groups these terrain factors into three growth, stream flow, runoff, water deplh. stream cur-categories based on how they affect vehicle operation: rent velocity. freezing and thawing of ground sorfaces

"* those dealing with surface geometry (small- and and water bodies, and accumulation of snow and ice.large-scale surface irregularities including obsta- For these reasons, a mobility prediction scheme mustcles) take climatic data into account, In addition to natural

" those that produce drag on the vehicle (vegetation changes in terrain, human intervention in the form ofand shallow water) construction or agricultural practices can also change

"* those dealing with the supporting material or sub- terrain conditions (such as altering water drainage pat-strafe (discussed in detail earlier., terns) very quickly.

Surface geometry affects vehicle mobility at a range A good overview of these types of parameters andof scales from millimeter-sized pebbles on a road to vast how they influence mobility can be found in Koeppelchanges in the slope angle and orientation of the land. andGrabau (1987). An example of how these factors areAll of these scales can occur at the same location, such included in terrain-based mobility prediction models isasagra,,ei-covered, washboard road on a slope. One of documented in U.S. Army (1968) and Turnace andthe most important considerations in assessing the ef- Smith (1983). Currently, these types of terrain classifi-fects of surface geometry is how the amplitude and fre- cation schemes are incorporated into GIS (Geographicquency of the surface irregularities excite thc vehicle Information Systems)-based mobility predictionsuspension system. Small-scale surface roughness, such schemes (Edmark et al. 1990. Fridstrand and Perssonas gravel on a road. may do little more than cause tire 1990, U.S. Army 1992). Similar schemes of terrainnoise, but intermediate terrain roughness of tilled farm classification are used in forestry to plan forest opera-land or a washboard road may severely reduce the speed tions and costs. Examples of forestry terrain classifica-and effectiveness of the vehicle operation. affecting the tion systems are presented in Sý%cdish Forest Opera-driver and cargo to such an extent as to make the traverse tions Institute( 1992) and. from Norway, Samnset ( 1975 .intolerable. Other surface irregularities (such as streams. Becauseofthe great spatial and temporal variation inditches, large boulders, mounds, and pits) create obsta- the parameters affecting mobility, it is logical to incor-cles to vehicle passage because of incompatibility with porate a statistical representation of the variability intothe shape of the vehicle: the vehicle "hangs up" on a any mobility prediction model. This type ofprobahili,-steep bank or "bottoms out" on a protruding rock." tic approach is necessarily a current area of research inThese features can slow the motion of the vehicle, stop the advancement of mobility prediction. A probabilisticprogress entirely, or delay movement by the additional approach is needed for both the descriptive input parame-time required to avoid the obstacle. ters, such as the statistical distribution of cone penctra-

Vegetation can fall within both geometrical effects tion data(Kogure etal. 1985, Heiminig 1987 ). as well asand drag-inducing effects. Smaller vegetation causes the predictive end results such as vehicle speed madeadditional frictional resistance or drag, impeding vehi- good (Lessem et al. 1992). Therefore. while in eswcncc

the mobility model may be deterministic for a specificW.E, Grahau. personal commnunication. t992. vehicle over a specified terrain, in reality the terrain

19

input is considered as an average value representing a Colbeck, S.C. (1986) Classification of seasonal snowterrain -unit" with specified variability, and thus the cover crystals, WAater Resources Re.'earch, 22(9): 9S-operational use of the model generates a range of 70S.vehicle performance that can be expected over the Colbeck. S.C. (1987) Snow meetamorphisin and clasýsi-variable terrain. ficatiol 1 n Seasonal .Snokicovers.\ Ph vsics, Cliemir ott%

Hvydroiolv. H.G. Jones and WiJ. Orville-Thomas (cdsj.)LITEATUE CTEDDordrecht, The Netherlands: 1). Reidel Plublishin2 Co.,LITERTURE ITEDp. 1-35.

Abele, G. (1990) Snow roads aid runways. USA Cold Colbeck, S.C., E. Akitaya, R. Armstrong, [I. Cu bier,Regions Research and Engineering Laboratory, Mono- J. Lafeuille, K. Lied, D. McClung and E. Morrisgraph 90-3. (1990) International classification for seasonal snow oil

Alger, R.G. (1988) Shear and compressive strength the ground. Technical Report MP 2794, issued bV themeasurements of snow using the bevameter. In Pro- International Committee on Snow and lee, Internationalceedings of the 1st International Conference on Snow Association of Scientific Hydrology, and InternationalEngineering, Santa Barbara, Calif., Jul 'y. USA Cold Glaciolo~ical Sodiety. Boulder: University of Colorado.Reoions Research and Engineering Laboratory. Special World Data Center for Glaciology.Report 89-6. Collins, J.G. (1971) Forecasting trafficab ility of soils.Alger, R.G. and M.D. Osborne (1989) Snow charac- Technical Memo 3-33 1. Vicksburg, Miss.: USA Corpsterization field data collection results. Final report to of Engineers Waterways Experiment Station.Department of the Army, Contract No. DACA89-85-K- CRREL (19627) Instructions/oritMaking and Recordiirg0002- Snow Observations. CRREL Instruction Manual 1. Ha-Alkire, B.D. and L. Winters (1986) Soil strength nover,N.H.: USAColdRe-Lions Research and Eiiineer-recovery using a Clegg impact device. In Proceedings in- Laboratory.of the 41h International Conferenice on Cold Regions Edmark, P., K. Fridstrand and MI. Svantesson (1990)Engineering, Anchorage, Alaska, February, p. 155- Cross country trafficability assessmen, wvith a GI S, In1 66. Proceedings of the 3rd Scandinaviian Research Confe r-ASAE (1985) Soil cone penetrometer. American Soci- ence on Geographic Infiorniationi SYstems, lf elsinglir,ety of Agricultural Engineers, Standard ASAE S3 13.2. Denmark, November.Ayers, P.D. and J.V. Perumpral (1982) Moisture and Flores, J.A. ( 1990) A field method to determi ne Alip anddensity effect on cone index. Transactions ofthe ASAE, compaction in a work day. ASAE paper from ASAE25(5): 1169-1 172. conference, Columbus. Ohio, June.Bekker, MI.G. (1969) Introduction to Terrain -Vehicle Fridstrand, K. and J.F. Persson ( 1990) Franikow I iý,hetSvstemns. Ann Arbor, Mich.: University of Michigan i 6vre Norriand. Utveckling och utvardering, av framiko-Press. mli-hetsmodell (A model for cross-country mobility).Bjork'lcm, U., G. Lundeberg and J. Scholander FOA Report C 20819-2.6 (2.7). Sundbyberg. SwNeden:(1975) Root distribution and compressive strength in National Defence Research Establishment.forest soils, Root mapping and plate loading tests in Godwin, R.J., N.L. Warner and D.L.O. Smith (199 1)thinning-stage stands of Norway spruce. Swedish Roy- The development of dynamic drop-cone de\ ice for theal College of Forestry, Depts. of Forest Ecology and assessment of soil strenath and the effects of mnachinervForest Soils, Research Notes No. 22 (in Swedish with traffic. Journal o~f Agricultural Engineering Research.English summary). 48(2): 123-131.Bowen, H.D. ( 1976) Correlation of penetrometer cone Hadas,A. and 1. Shmulewich ( 1990) Spectral analysisindex with root impedance. ASAE Paper No. 76-15 16. of cone penetrometer data for detecting spatial arrange-St. Joseph, Mich.: American Society of Agricultural ment of soil clods. Soiland Tillage Research, 18:4.7--2.Engineers. Hakkarainen, A. (1949) Routa Talviieiden rakenuija-Boyne, H.S. and D.J. Fisk (1990) A laboratory com- na. Metsatehon tiedoitus 29 (in Finnish).parison of field tcchniques for measurement of the Heiming,G.W.(1987)Statistical procedurestbrevaliu-liquid water fraction of snow. USA Cold Regions Re- ation of terrain measuring data. In Proceedings oft/ic 9thsearch and Engineering Laboratory, Special Report 90- C'onference oIj the Internauional Society fur Terralin-3. VehicleS 'vs~enjs(ISTVS), Barcelona. Spahin,p. 143-151.Chancellor, W. (1993) Soil physical properties. Ad- Hettiaratchi. D.R.P. and Y. Liang (19S7) Nomograinsvances in Soil Dynamics Monograph Series, Vol. 1, for the estimation of soil strength fronm indenitationl tests.American Society of Agricultural Engineers. ASAE Journal o~f Terrainechanics, 24(3): 1718Monograph, in press. Irwin, G.J. and R.N. Yong (1990) Demionstration of a

20

portable device for predicting vehicle tractive pertbr- 2nd Asia-Pacifi:- (onjeretwt, of15 V. j),QAo. 1hun-mance in snow. Journal of Terrainechanics, 17(4): 197- land, Decembher, p. I -1 1.206. Okello, I.A. (1991 )A rev iew \A, soilI sIieng th mneasure-Johnson, C.E., R.D. Grisso, T.A. Nichols and A.C. mient tec' iniques for prediction of terrain \ chic le perfor-Bailey ( 1987) Shear measurement foragricultural -.oils- mance. Journal nJ Agriculturail Engineerinig Re.search.a review. Transactions of the ASAE, 30(4): 935-938. 50(2): 129-156.Karariath, L.L, and E.A. Nowatzki (1978) Soil Me- Olsen, H.J. (1987) Electronic cone penetrometer forchanics for 9ff-Road Vehicle Engineering. Clausthal. tield tests. In Proceedings n/flu' 9th International (onfilr.Germany: Trans Tech Publications. ence of ISIX'S. Barcelona, Spain. p. 20--27.Koeppel, W. and W. Grabau (1987) Terrain descrip- Patin, T.R. (1972) Evaluation of'surface shear strengthtion for mobility prediction, In Proceedings of the 9thy measurements for use in laboratory mobility studiesinternational Conference of IS7VS. Barcelona. Spain, 1: WES MP MI-72-5.160- 175. Persson, S. (in preparation Advanccxs iii Soil 0%vnwnicV.Kogure, K., Y. Ohira, and H. Yamaguchi (1985) St. Joseph. Mich.: American Society ol' AgriculturalBasic study of the probabilistic approach to predict ion of Enoincers.soil trafficability-statistical characteristics of cone- Perumpral, INV. ( 1987) Cone penetrometer applica-index, Journal of Terramechanics. 22(3): 147-156. tions-a review. Transactions ol the ASA F. 30(4) 939-Kogure, K., H. Yamaguchi and Y. Ohira (1988) 944.Comparison of strength and soil thrust characteristics Plackett, C.W. (1985) A review of' force predictionamon- different soil shear tests. Journal of Terramne- methods for off-road wheels. Journal e/ Agriculturalchanics, 25(.3): 20 L-222. Engineering Research. 31: 1-29,Koniandi, G. (1990) Establishment of soil -mechanical Radforth, N.W. (1952) Su-geested classification ofparameters which determine traction on deforming soil, musket! for the engcineer. En~ern ora 350 1:Journal of Terramnechanics, 27(2): 115-124. 1199-1 210.Lessem, A.S., R.B. Ahivin and G.L. Mason (1992) Radforth, N.W. and Burwash, A.L. 11977) Trarnsp~or-Stochastic mobility forecasts for military vehicles. In tation. Chapter 1(1 in Mus'keg and thec Northern Enviton-Proceedcinigs of thieNorth Amierican ('on~ferenice ofISTVS. inent. N .W. Radforth and C.O. Brawner (eds.). Toronto:Sacramento, C'alrfornia. WES Technical Report GL,-92- University of Toronto Press.I11. Radforth, N.W. and I. Evel (1959) Musket, impedanceListon, R.A. (1973) The combined normal and tangen- factors controlling vehicle mobility. Technical Memo-tial loading of soii. Ph.D. thesis, Michigan Technical randumn No. 57. Ottawa: Canadian National ResearchUniversity, Houghton, Mich. Council.MacFarlane, I.C. (1958) Guide to afield descripuion of Rawitz, E. and M. Margolin (1991) An economnicalmuskega. Associate Committee on Soil and Snow Me- hand-held recording penetrometer. Soil and Tilloge Re-chanics, Technical Memorandum 44. Ottawa: Canadian search. 19: 67-75.National Research Council. Rohani, B. and G.Y. Baladi (1981) Correlation ofMacFarlane, I.C. (ed.) (1969) Muskeg Engineering mobility cone index with fundamental enigi neerinrz prop-Handbook. Toronto: University of Toronto Press. erties of soil. In Proceedings of the 7th InzternatitnnalMathur, T.S. and G.T. Coghlan (1987) Use of the Con frence of ISTVS, 3: 959-990, or WES Miscella-Clegg impact tester in managing and designinga a(ggtre- neous Paper SL-8 1-4. AD-A) 01409.gate-surfaced roads. In Proceedings of the 4th Interno- Rummukainen, A. (1984) Peatland properties and theirtional C'onference on Low- Volume Roads, Ithaca, N. Y.. evaluation forwood harvesting~. Final report for Han-cst-August, p. 232-236. ing on Peatlands. a research project of the Nordic Re-McKyes, E. (1985) Soil cutting and tillage. New York: search Council on Forest Operations (NSR), 1977-1983.Else vier. University of Helsinki Department of l~opoingL and U~ti-Multjueen,J.,J.V. Stafford and D.W. Tanner (1977) lization of Forest P~roducts Research Notes. No. 45,Evaluation of penetrometers for measuring soil strength. Saarilahti, MI. (1982) Studies oin the possibilities (ifJournal of Terramechanics, 14(3): 137-15 51. usin2 radar techniques in detectingi the trafficabi lity ofN~emi, E.W. and R. Bayer( (1970) Analytical predict ion peatland-.. Act-a For, Feinn.. 170 (in Finnish wvith FocI ishof vehicle mobility on muskeg. Technical Report No. summary).I1I109, I)A-20-()1 8-AMC-0989(T), Warren, Mich.: U.S. Saarilahti, M. and M. Tiuri ( 198 1) Radar techn~iques inArmy Tank Automotive Command. detecting the trafflkahility oh (rozen peatlarnk. In Pro-Ohmiya, K. and T. Masui (lS)Visualization soft- ceedings of the 7thi Internationail Confe~rence of 157V5.ware for cone resistance analysis. In Proceedings of the Calgary.v Alberta, Canadai. 111: 10 17- 1 o44-

21

SAE (1967)Off-road vehicle mcbility evaluation. SAE condensation of the Army mobility mnodel fo rosRecommended Practice Handbook Supplement. SAE country mobility mapping. U.S. Army Waierwass E's939. Warrendale, Penn.: Society of Automotive Engi- periment Station Technical Report GL-8S3 12.neers. Upadhyaya, S., D. Wulfsohn andiJ. IMehlschau (1990,)SAE (1985) Passenger car and light truck tire dynamic An instrumented device to obtain traction related param-driving traction in snow. SAE 11466 OCT85. Warren- eters. ASAE Paper No. 90-1097.dale, Penn.: Society of Automotive Engineers. U.S. Army (1959) Trafficahility ot soils. WE"-S TM 3-Saniset, 1. (1975) The accessibility of forest terrain and 240 and Suppls., p. 1-15. V icksburg. Miss.: Waterwaysits influence on forestry conditions in Norway. Reports Experiment Station.of the Norwegian Forest Research Institute, No. 32. 1. U.S. Army (1968) Planningt and design of roads. airb-aS-Scholander, J. (1973) Tear resistance of some forest es, and heliports in the theater of operations. Chapter 9,soils, rupture limit of the field and ground vegetation Soils Trafficabi/ii.V, Vol. 1H of Dept. of the Army Techni-layer. Swedish Royal College of Forestry, Department cal Manual TM 5-330 and Air Force Manual AFM 86-3.of Operational Efficiency, Research Notes No. 61 (in U.S. Army (1992) ALBE Tactical Decision Aýid (TD.A)Swedish with English summary, figures, and tables). User'st Guide, Version 2.0. Misc. Paper GI.-92-36.Shapiro, L.H., i.B. Johnson, M. Sturm and G.L. Voorhees, M.L. and P.N. Walker( l977) Tractionabil-Blaisdell (1993) Snow mechanics: Review of the state ity as a function of soil moisture. 7Transac~iions o~f heof knowledge and applications. USA Cold Regions ASAE, 20(5): 806-09.Research and Engineering Laboratory, CRREL Report Wasterlund, 1. (1990) Soil strength in forestry flea-in progress. sured with a new kind of test rig. In Proceedings of tileShoop, S.A. (1989) Thawing soil strength measure- 10th In-iernational Conferrnce ofISTV'S. Kohe, Jap~im.ments for predicting vehicle performance. North Amner- Aug-ust, p. 73-82.ican Mfeeting of 1S7'VS, Victoria, B.C., Canada. Also Wells, L.G. and 0. Tresuwan ( 1977) The response ofUSA Colo Regions Research and Engineering Labora- various soil stieng!th indices to chan-ina water content.tory, CRREL Report 92-17 (1992). ASAE Paper No. 77-1055. St. Joseph. Mich.: AmericanShoop, S.A. (1992) Precision analysis and recommend- Society of Agricultural Eng~ineers.ed test procedures for mobility measurements made Wong, J.Y. (1989) Terratnechanicsv and 0ff-Road Vehli-with an instrumented vehicle. USA Cold Regions c/es. New York: Elsevier.Research and Engineering Laboratory, Special Report Wong, J.Y. (1992) Expansion of thle terrain input base92-7. for nepean tracked vehicle performance model, NTV PM.Shoop, S.A. and R. Alger (199 1) Snow characteriza- to accept Swiss rammsonde data from deep snow. Jour-tion for traction testing:, a survey of techniques used. nai of Terraniechanics. 29(3): 341-358.First Meeting on Winter Mobilit~y, Santa Barbara, Cali- Yong, R.N. and A. Youssef (1978) Application of % ane-fornia, June. cone tests on soil for determination of trafficahility. InShoop, S.A., B. Young, R. Alger and J. Davis (1993) Proceedingsof the 6th International (onft'renceofl"STVS.Comparison of three approaches to traction evaluation Vienna, Austria, 2: 677-706.on winter surfaces. In Proceedings of the I11th Interna- Yong, R.N., A.F. Youssef and E.A. Fattah (1975)tional Conference of ISTM, Lake Tahoe, California, Vane-cone measurements for assessmient at tractivye per-Septemnber. Also a CRREL Report in preparation. formance in wheel-soil interaction. In Proceedhigs oýfStafford, J.V. and D.VWv. Tanner (1982) Field mea- the 5th International ('on ifretce o?.t IS71VS, !)cwoit.suremeiit of soil shear strength and a new design of field Michigan, 3: 769-788.shear meter. In 9th Con~ference oft/ie International SoilTillage Research Organization (lSTRO)t, Yugoslavia. BIBLIOGRAPHYSwedish Forest Operations Institute (1992) Terrainclassification system for forestry work, developed by Ala-Ilomaki, J. and M. Saarilahti (1990) Rut forma-Steffan Berg. IS'BN 91-7614-078-4. Kolding Lyntryk. tion on peat soil-Experiences from a forced-slip, whcelDenmark: Skogsarbeten/SkogForsk. tester. In Proceedings of thec /0th International ConferT-Taylor, H.M., G.M. Roberson and 3.5 Parker, Jr. ence of !STVS, Kobe. Japaun, August, 1) 457.(1966) Soil strength -root penetration relations formedi-. laladi, G.Y. (1987) Terrain evaluatio.; for off-roadum tocoarse textured soilI materials.So cine 102(I): moiiy ora fTrro'hncs 4 ý.27--I40.18-22. Burt, E.C., A.C. Bailey and R.K. Wood (I 987'ý EffectsThomson, .I.G. (1960) Vehicles in muskeg. Engineer- of soil and operational parameters on soil--tire interfaceing Journal, 43(5): 73-78. stress vectors. Journal of Terra~ntchaniev. 24(3): 235-Turnage, G.W. and j.L. Smith (1983) Adaptation and 246.

DeRoock, B. and A.W. Cooper (1967) Relationship neering, 58( 10): 34-38.between propagation velocity of mechanical waves Scholander, J. (1973) Compressive strength of some

through soil and soil strength. Transactions of the ASAE, forest soils. Swedish Royal College of Forestry, Dept. of

10(4): 471-474. Operational Efficiency, No. 59 (in Swedish with EnellishDunlap, W.H., G.E. VandenBerg and J.G. Hendrick summary).(1961) Comparison of soil shear values obtained with Scholander, J. (1974) Bearing capacity of some lorestdevices of different geometrical shapes. Transactions of soils for wheeled vehicles. In Sonic Technical A'SIects

the ASAE, 9(6): 896-900. and Consequences, Swedish Royal College of Forestry.Fukue, M., T. Kobayasi and T. Nakamura (1990) Department of Operational Efficiency. Garpenher,,Terrain evaluation under water. In Proceedings of the Sweden, No. 14 (in English).10th International Conference of ISTVS, Kobe, Japan. Smith, L.A. (1978) A recording penetroineter. In Pro-

August. p. 149. ceedings of the 4th International Cont'Crence on ifecha-Hendrick, J.G. and A.C. Bailey (1982) Determining nization of Field Experiments., IAMFf'. Iowa State ('ni-components of soil-metal sliding resistance. Transac- versity, Ames. Iowa. Julv 1976, p. 336-343,and Trans-lions of the ASAE, 25(4): 845-849. actions of the ASAE, 21( 1): 12-14, 19.

Janosi, Z.J. and L.L. Karafiath (1981) Im nproved tech- Taylor, J.H. (1967) An annular shear device. Transoat-nique of field ring shear testing for soil shear strength lions of the ASAE, 10(2): 164-165, 169determination. In Proceedings of the 7th International Tong, J., Ren Luquan and Chen Bingcong t1990)

Conference of ISTVS, Calgary, Alberta. Reducing adhesion of soil by phosphoric white iron onJohnson, C.E., G. Murphy, W.G. Lovely and R.L. thebasisof bionicsprinciples. ln Procedingsofthe 01thSchafer (1972) Identifying soil dynamic parameters for International Confirence of ISTVS. Kobe. Jaacnt. AIt-soil-machine systems. Transactions of the ASAE, 15(1): gust. p. 57-63.9-13. Wilkins, D.E., W.L. Harris and J.H. Taylor (1966)Larson, W.L. and W.R. Gill 1973) Soil physical Effect of rate of displacement on shearing stress of soilsparameters for designing new tillage systems. In Pro- as measured with a torsional shearing device. ASAEceedings of the National Conservation Tillage Confer- Annual Meeting, Amherst, Mass.. June.ence, Des Moines, Iowa, p. 13-22. Williford, J.R. and L.W. Larson (1968) A study of soil

Li Guiyou, Li Qin and Li Xiapding (1990) A simple strength values included in cone index readings. ASALmethod for measuring the phi value of internal friction Paper No. 68-666.

angle in saturated paddy soil. In Proceedings ofthe 10th Woehrle, W.J. and R.J. Wozniak 1989) Tire trac-International Conference of ISTVS, Kobe, Japan, Au- tion-matching product and test technologies. SAF Pa-gust, 1: 127-134. per No. 890102. Warrendale. Penn.: Society of Automo-Qian Cong, Ren Luquan and Chen Bingcong (1990) live Engineering.Research on reducing adhesion results of soil electro- Wong, J.Y. (1980) Data processing methodolo-'y in theosmosis and its affecting factors. In Proceedings of the characterizationol (imechanicalproperties of terrain. bar-10th International Conference of ISTVS, Kobe, Japan, nal of Terratnechanics, 17(I): 13.-41.August. 1: 45-55. Wood, R.K. and E.C. Burt (1987) Thrust and motionReaves, C.A. and R.L. Schafer (1971) Soil measure- resistance from soil-tire interface stress measurements.ments related to the performance of soil-machine sys- Transactions tf the ASAE. 30(5): 1288-- 1292.tems. Presented at the Society of Automotive Engineers Youssef, A.F.A. (1984) Properties of desert silty soil inEarthmoving Industry Conference, Peoria, Ill., April, relation to vehicle mobility. In lPro-ccdid•. offthe SilPaper No. 710512. International Confi'ren'e oqf I57h'S. ( amnlr.idle. tn -Robbins, D.H. Jr., J.G. Hendrick and C.E. Johnson gland. 3:1217-1 233.(1987) Instrumentation for soil-material sliding resis- Zhang Zhaoxing. Sang Z.Z. and ;ao L.R. (19861lance research. ASAE Paper No, 87-1(011. St. Joseph, Adhesion and friction bet.cen soils and soiids. AtaMich.: American Society of Agricultural Engineers. Agroinechanica Sinica. Transactions of the ChineseRobbins, D.H. Jr., C.E. Johnson and R.L. Schafer Society of Agricultural Machin.cry, 17(l): 32-40,(1987) Modeling soil-metal sliding resistance. ASAE Zhenya, S. (1988) Research on tile micro-mechanism ofPaper No, 87-1580. soil adhesion. In Proceedings of the 2nd Asia-Pacific

Schafer, R.L., W.R. Gill and C.A. Reaves (1977) Conference of lSTVS, Bangkok. Ihailand, l)ce,'ber, p,Lubricated plows vs, sticky soils. Agricultural L'ngi- 57--63.

23

Form ApprovedrREPORT DOCUMENTATION PAGE OM~B No. 0704-0788Pubi rceporting bren for ibis clection o information isestimatdo avrgeIhupr repone nldn the i-rr ve slfciiov wc h rigi "ýc. ýK'('' "("_r Ul~!' xv .-

maitnaining the data needed, and completing and reviewing the cletoofioraonSend cm etregadnIN ourdiirt LstIrmalt or arr/ o~tac, aspc u!r 3' , t. IO ;, 0Inc uding suggestion to reducing thisburden, toWashingtonHeadquatr Sevcs D fectraefrIomlo Oprto qsana Reponsrt rjfe'ofa s

1.AEC S NY(ev ln) 2. REPORT DATE3 REPORT TYPE AND DATES COVERED

4. TITLE AND SUBTITLE 5. FUNDING NUMBERS

Terrain Characterization for Trafficability P'R: 4A70278-IATI4'2TA: C S

6. AUTHORS WU: W(7

Sally A. Shoop

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATIONREPORT NUMBER

U.S. Army Cold Regions Research and Engineering Laboratory72 Lymie Road CRREI. Report 93-6Hanover. New Hampshire 03755-1290

9. SPONSOR ING/MONITOR ING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING.MONITORiNGAGENCY REPORT NUMBER

Office of the Chief of EngineersWashington, D.C. 20314-1000

11 -SUPPLEMENTARY NOTES

12a. 015 RIBUTIONiAVAILABiLITY STATEMENT 12b. DISTRIBUTION CODE

Approved for public release: distribution is unlimited.

Available from NTIS, Springfield. Virginia 2216).

13. ABSTRACT (Maximum 200 words)

Terrain material characterization is needed to predict off-road vehicle performance. trafficability. and deformation iconi-paction and rutting) resulting from vehicle passage. This type of information is used by agricultural engincers. foresters.military engineers. the auto and tire industry, and anyone else concerned with oft-road. unlpaved. or' Winter Inioiljlv . ThiSreport appraises the state-of-the-art of terrain (or substrate) characterization techniques foi vehicle traction ,tudie", It con-centrates on field measurement of strength-related properties for soil. snow. muskeg. and vegetation, but also discusse~show these compare with laboratory measurements and the importance of other terrainl features (slope". drainage, and ob-stacles).

14. SUBJECT TERMS 15 NUIMBER OF PAG;ESMobility Soil Traction Vehicle30Musket, Strength Trafficabil ity 16. PRICE CODESnow Terrain Vegetation _____________

17. SECURITY CLASSIFICATION 18 SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20 LIMITATION Or ABST PACTOF REPORT OFITHIS PAGE OF ABSTRACT

UNCLASSIFIED UNCLASSIFIED UNCI-.ASSIFIE')

NSN 7540-01-280-5500 Standard Form 298 (Rev 2-89)flrescnbrrc by ANYI Std 239 18298 102