Soil Mechanical Properties at the Apollo 14 Site

Home , Apollo 14, Apollo 4, Cone (crater), Fra Mauro formation

Journalof GeophysicalResearch

VOLUME 77 OCTOBER 10, 1972 NUMBER 29

SoilMechanical Properties atthe Apollo 14 Site

JAMESK. MITCHELL;• LESLIE G. BROMWELL,2 W. DAVIDCARRIER III, s NICHOLASC. COSTES,4 AND RONALD 1•. SCOTT5

The Apollo14 lunarlanding provided a greateramount of informationon the mechanical properties of the lunar soil than previous missionsbecause of the greater area around the landing site that was explored and because a simple penetrometer device, a special soil mechanicsti•ench, and the modularized equipment transporter (Met) provided data of a type not previouslyavailable. The characteristicsof the soil at shallow depthsvaried more than anticipatedin both lateraJand vertical directions.While blowing dust causedless visibility impairment during landing than on previous missions,analysis shows that eroded particleswere distributedover a large area around the final touchdownpoint. Measurements on core-tubesamples and the resultsof transportertrack anaJysesindicate that the average density of the soil in the Fra Mauro region is in the range of 1.45 to 1.60 g/cm3. The soil strength appearsto be higher in the vicinity of the site of the Apollo 14 lunar surface experiments package,and trench data suggestthat strength increaseswith depth. Lower-bound estimatesof soil cohesiongive values of 0.03 to 0.10 kN/m 2, which are lower than values of 0.35 to 0.70 kN/m 2 estimatedfor soils encounteredin previousmissions. The in situ modulus of elasticity, deducedfrom the measuredseismic-wave velocity, is compatible with that to be expectedfor a terrestrial silty fine sand in the lunar gravitational field.

Studiesof the soil (regolith) at the Apollo 14 formulate, verify, or modify theories of lunar site have been made (1) to obtain data on the history and lunar processes;(3) to develop compositional,textural, and mechanicalproper- information that may aid in the interpretation ties of lunar' soils and the variations of these of data obtained from other surface activities or propertieswith depthand locationat and among experiments (e.g., lunar field geology, passive Apollo landing sites; (2) to use these data to and active seismic experiments); and (4) to developlunar-surface models to aid in the solu- tion of engineeringproblems associated with • Departmentof Civii Engineering,University of California, Berkeley, California 94720. future lunar exploration. The in situ character- 2 Department of Civil Engineering, Massachu- istics of the unconsolidated lunar-surface ma- setts Ins•.i.tute of Technology, Cambridge, Massa- terials can provide an invaluable record of the chusetts 02139. past influencesof time, stress,and environment s GeophysicsBranch, Planetary and Earth Sci- on •he moon.Of particularimportance are par- encesDivision, Manned SpacecraftCenter, NASA, Houston, Texas 77058. ticle size, particle shape,particle-size distribu- 4George C. Marshall Space Flight Center, tion, density,strength, and compressibility,and Huntsville, Alabama 35812. the variationsin these propertiesboth region- 5 Division of Engineering and Applied Science, ally and locally. California Institute of Technology, Pasadena, California 91109. To date it hasbeen necessary to rely heavily on observationaldata, suchas are providedby Copyright ¸ 1972 by the American GeophysicalUnion. photography,astronaut commentary,and ex-

5641 5642 M•ITCHELL ET AL. aminationof returnedlunar samples,since quan- formations that are-characteristic of the Fra titative data sources are limited. Semiquanti- Mauro formation.This extensivegeological unit tative analysesare possible,however, as Costes is distributed in approximately radially sym- et al. [1969] showedfor Apollo 11 and Scott metric fashion around the Sea of Rains over et al. [1970] showedfor Apollo 12. Such analyses much of the near side of the moon and is con- are enhanced through terrestrial simulation sideredto be older than the Apollo 11 and 12 studies [Co.steset al., 1971; Mitchell et al., mare sites.It is believed,also, that someof the 1971]. , material accessible at the site has come from The resultsof the Apollo 11 and 12 missions depths of tens of kilometerswithin the original have generally confirmedthe lunar-surfacesoil lunar crust. Bright ray material from Coperni- modeldeveloped b.y Scott and Roberson[1968]; cus Crater covers much of the landing site. that is, the lunar s0il is similar to a silty fine ConeCrater, a young,blocky, crater of Coperni- sand, is generally gray-brown in color, and can age, 340 meters in diameter, penetratesthe exhibits a slight cohesion.Evidence of both soil layer east of the landing point. The Apollo compressibleand incompressibledeformation 14 traversesand a map of the major geologic has been observed. The lunar soil erodes under featuresare shownin Figure 1. the action of the exhaust of the lunar module DESCENT AND LANDXNC descentengine during lunar landing, kicks up easily under foot, and tends to adhereto most Blowing dust. The astronauts commented objects with which it comesinto contact. The that blowingdust was first observedat an alti- value (or range of values) of the in situ bulk tude of approximately33 meters and that the density of the lunar soil remainsuncertain, al- quantity of dust from that altitude downto the though Apollo 11 and 12 core-tube data and surface seemed less than had been encountered core-tube simulations [Carrier et ed., 1971] duringthe Apollo 11 and 12 landings.The crew suggesta range of 1.5 to 1.9 g/cms for the upper estimatedthe thicknessof blowing dust to be few tens of centimeters of the lunar surface. less than 15 cm; rocks were readily visible Observationsat five Surveyor landing sites throughit. The sun angleat landingwas higher and the Apollo11 and 12 landingsites indicated for the Apollo 14 missionthan it had been for relatively little variation in surface soil condi- the Apollo 12 landing. The appearanceof the tions with location.Core-tube samples from the blowing lunar-surfacematerial in motion pic- Apollo 12 missionexhibited a greatervariation tures taken during the Apollo 14 descentseems in grain-sizedistribution with depth than had qualitativelysimilar to that observedduring the been found for Apollo 11 core-tubesamples, Apollo 11 landing. Dust was first observedat however. altitudesof 24, 33, and 33 metersfor the Apollo The Apollo 14 mission provided a greater 11, 12, and 14 landings,respectively. Because amount of soil mechanics data than either of the of the effect of sun angle and spacecraftorien- previous missionsfor two reasons.The crew tation, however,the appearanceof the dust in covereda much greater distanceduring the ex- the motion picturesmay not be a reliableindi- travehicularactivity periods,and the Fra Mauro cation of the quantity of material removedfrom landing site representeda topographicallyand the surface. geologicallydifferent region of the moon. In Surface erosion. The astronauts reported addition, three featuresof particular interest to that the lunar surface gave evidence of the soil mechanicswere new to the Apollo 14 mis- greatest erosion in an area approximately 1 sion: the Apollo simple penetrometer,the soil meter southeastof the region below the engine mechanicstrench, and the modularizedequip- nozzle, where as much as 10 cm of surfacema- ment transporter (Met). Each of thesehas been terial may have been removed during the land- used to shed new light on lunar soil character- ing. A distinct erosionalpattern is visible in istics. Figure 2. Except for a disturbedarea in the left middle distance,the surface gives the appear- ArOLLO 14 LANmNG Sx• ance of having been swept by engine gasesin The Apollo 14 landingsite in the Fra Mauro the same way as on previous missions.The region of the moon containsexposed geological disturbed area may have developedas a conse- So•, AT APo•,Lo14 S• 5643

Crater

I illIll I I I I I 100 0 500

Fig.1. Apollo14 site map [from $wannet al., 1971]:Cc•, materials of ConeCrater; Is, smoothterrain matedhal of the Fra Mauro formation;I•r, ridgematedhal of the Fra Mauro formation;open circle, panorama station; open triangle, station without panorama. Contact: long dashedlines indicate approximate locations, and shortdashed lines indicate that loca- tionis inferredwithout local evidence. Foot o• scarp:the linebounds a smallmesa, and thesolid triangles point downslope; the short dashed lines indicate inferred location. Edge o.•hill: longdashed lines indicate approximate locations, and shortdashed lines indicate inferredlocation. The open trianglespoint downslope.Traverse routes for first and second periodsof extravehicularactivity are marked by solidlines with arrows. quenceof grazingcontact of the q-Y footpad ejected radially from the surface below the contactprobe during the landing. spacecraftat predominantlylow anglesto the In the Apollo14 descentmotion pictures, it horizontal.There is thus,during the descent,a is evident that the lunar surface remains indis- regionfrom whichsoil is beingremoved, and tinct for a number of secondsafter descent- an adjacentregion, kilometers in lateral extent, engineshutdown. This event was probably on which the ejected particles descend.Since causedby venting from the soil of the exhaust the.spacecraft traverses laterally over the sur- gas stored in the voids of the lunar material face at a decreasingaltitude, erosionin some duringthe final stagesof descent.The outflow- regionswill be followedby depositionof parti- ing gas carrieswith it fine soil particlesthat cles removed at later times' from other ar•a.s. obscure the surface. The entireregion in the vicinity of the landed Implications o[ blowing dust. The lunar spacecraft to a radius of several hundred me- soil removedby the engineexhaust gas is tersis particularlysubject to thisprocess, which' ½-

Fig. 2. Area below the descent-engine nozzle showing erosional features caused by the exhaust gas. The --Y footpad can be seen in the distance (AS14-66-9262). may have some implicationsin the analysesof ters (200 feet) south of station G at its point of the soil and rock samplescollected. The special closestapproach,' and at this point its altitude environmental sample obtained from material abovethe lunar surfacewas 55 meters (180 feet). in the bottom of the trench dug at station G Station G is slightly south of due east of the (Figure 1) may be used as an example.It is landing site at a distanceof about 230 meters likely that this soil sample included granular (750 feet). In the descentmovie, the first signs fragmentsboth from below the surfaceand at of blowing dust are visible as the spacecraft the surface, since material fell into the trench passedover North Crater. Consequently,a small as it was being excavated. amount of erosiontook place at station G as the During descent to its landed position, the spacecraftpassed by during descent.This ero- lunar module followed a track approximately sion is probably not significantto the analysis W22øN, going slightly south of the center of of the specialenvironmental sample. However, North Crater. The spacecraftwas about 60 me- the amount of material removed from the sur- SOIL AT APOLLO 14 SITE 5645 face increasesgreatly as the spacecraft de- Station G is about 230 meters from the scends,and major quantities are eroded from landedposition of the Apollo 14 lunar module. the landingsite. If it is assumedthat the Apollo 12 and Apollo The concentrationof particlesarriving at sta- 14 vehicleseroded identical quantitiesof lunar tion G and originatingfrom the landingsite can soil in the final stagesof touchdown and that be estimated by comparisonwith the observa- the emitted particle cloud expandsspherically, tions of the Apollo 12 mission.The Apollo 12 the density of the particle cloud at station G lunar module landed 155 meters (510 feet) would be (155/230) 8 -- 0.3 of that at the Sur- from the Surveyor 3 spacecraft,which at the veyor 3 location.Consequently, it would appear time had been on the lunar surface 31 months. that each squarecentimeter of surfaceat right Detailed study of the Surveyor 3 camera re- anglesto the unobstructedline joining station G vealed a distinct shadowpattern on the paint, to the landing site would receiveof the order oi and this pattern was shown [Ja#e, 1971] to 10• impacts of particles, a few microns or more arise from a lunar soil sand blasting. It was in diameter, ejected from the landing site. To demonstrated,moreover, that the sand-blasting reach station G, the particle velocitieswould particles came from the Apollo 12 landing site needto be of the order of 100 m/sec or greater, rather than from a sequenceof points along its dependingon their ejectionangle. landing track. The particles must have had a Footpad-surfaceinteraction. 'The -- Y (Fig- velocity greater than about 70 meters per sec- ure 2) and --Z footpadspenetrated the surface ond with a shallow angle trajectory to have only to a depth of 2 to 4 cm, whereasthe +Y reachedthe Surveyor spacecraftand must have and +Z footpadspenetrated to a depth of 1'5 arrived at a fairly high concentrationto have to 20 cm. The +Y footpad penetration mech- achieved the sharpnessof shadow effect ob- anism is clearly visible in Figure 3; the foot- served. The abrasion appears to be uniform, pad contacted and plowed into the rim of a and there is no indicationof individual impacts. 2-meter-diameter crater. The motion of the foot- Therefore, the surface or surface coating has pad through the soil caused a buildup of a been struck by so many particles that their mound of soil on the north side of the pad. The impact areas overlap. It will be assumedthat +Z footpad also landed on the rim of a small the majority of particles reaching Surveyor crater, and its appearanceand penetrationwere were of micrometer size or larger and that the similar to the appearanceand penetration of average diameter of each impact might be of the + Y footpad. the order of 10 •m. If it is further assumedthat The responseof the soil to the landing (which the area of impacts just saturatesthe surface occurred with little or no shock-absorber strok- (conservative), it appears that each square ing) and the appearanceof the soil in the foot- centimeter of the abraded area was subjected pad photographssuggest that the mechanical to impact by about 106 particles. One of us properties are similar to the mechanicalproper- (Scott) has examinedthe Surveyor 3 surface ties of the lunar material on which the Apollo sampler (also exposedto blowing dust) in de- 11 and 12 lunar moduleslanded. The penetra- tail at about 100 magnification,at which he tion of the +Z and +Y footpads causedthe shouldcertainly have been able to see any im- lunar module to tilt 1ø to 1.5ø in the westerly pact marks in the size range of order 100 •m, and northerly directions. Consequently,at the but there were none. Therefore, it can be tenta- landing site, the strike of the lunar surface tively concludedthat each squarecentimeter of slope is approximatelyW16øN, and the dip is the Surveyor camera saw at least 10• particle approximately5.5 ø in the direction N16øF,. impacts from the material eroded by the de- SOIL CONDITIONS AT TIlE LANDING SiTE scentengine. As a check,it is foundthat thesenumbers cor- The behavior of the surfacesoil was in many respondto removal of the lunar soil to a depth respectssimilar to that observedduring earlier of 3 or 4 cm over a diameter of 5 meters from missions.The color was gray-brown and ap- the lunar surfacebelow the descentengine noz- pearedto changewith sun angle.The soil could zle. This is compatiblewith astronaut observa- be kicked up easily during walking, but would tions. also compressunder foot. Footprints ranged in 5646 MITCHELL ET AL.

.. -.. ---..... :"•..?•::.

.... .

-,'•-..;;':"?:.•. '•": r"..======

...:. ' -'-•:::'...'.i" ;:."%." ;?":'.-;•:-::-

......

.-..::•::.... •::-:..::,!{;•... ,' .::•;::..- ....

ß,,,--.½--.-.?:...,..•.:: :-:,;•.;;•:?;:...... : : -'-"*' :•' '"}' .%.,';"•....':.:. '*$<.' ...... :'"'::::::1.4;;!!,- ..... --

...,..,:,

Fig. 3. The '-I'-Y footpad embedded in the lunar soil. The gold foil surrounding the landing , leg is probably the protective wrapping on the transporter (AS14-66-9234). depth from 1 to 2 cm on level groundto 10 cm with blocky ejecta [Swarm et al., 1971] lead to on the rims of fresh, small craters. The trans- estimates of thickness of the soil layer in the porter tracks averagedi cm and ranged up to Fra. Mauro regionranging from 10 to 20 meters. 8 cm in depth. Data from the active seismic experiment Patterned ground ('raindrop' pattern) was [Ko.vach et al., 1971] indicate an unconsolidated fairly general, except near the top of Cone surficial layer 8.5 meters thick at the site of the Crater. The reasonfor the developmentof this Apollo lunar surface experiments package surfacetexture is not yet known, although it is (Alsep). probably related to the impact of small particles Surface characteristicsare markedly different on the lunar surface. within a zone extending outward about 350 Crater morphologiesrange from freshthrough meters from the rim of Cone Crater from what subduedto almost completelyobliterated. De- they are elsewhere.In this region, the soil is terminations of minimum crater sizes associated coarsegrained, and rock fragments are abun- SoIL AT APOLLO 14 SiT• 5647 dant; elsewherethe soil is fine grained, and disturbedareas. Smoothed and compressedareas rock fragments are sparselydistributed on the (e.•., transporter tracks and astronaut foot- surface. prints) are brighter at some sun angles,as is The crew reported a zone of firm, compact shown in Figure 4. In some instances,it was soil betweenstations B1 and B2 (Figure 1) but difficult for the astronauts to distinguishbe- said that this was • small, isolated patch in a tween small, dust-coveredrocks and clumps or generally powdery surface. Downslope move- clods of soil. The tops of many of the large ment of loose soil has obliterated some small rockswere free of dust, althoughfillets of soil craters on slopes. were common around the bottom. The astro- As has been observed during previous mis- nauts commentedthat the maior part of most sions,disturbed areas appear darker than un- large rocks a.ppearedto be buried beneath the

..-- • %• •.:•-•--.• ...... -:;.•.::. ':'."- ...... •.?.,..::-•• '"•.... •.., ...... • •..'•,, ...... - •;..-...... •.,.•..••.".';•" .-.:,... •.• •, • '... • ' . . • •-•--%. '...... ;;%.'•-,?:.. ?-' .= '? .... • •'½.. •.• •" ...•.. ? ..... •-. :.½..:....--•.:.,:., . •,•: .: ,• ß •. .,.• •.½..:,.... . • ...... •,, ... :..,• : , •. .•....?,...: .....•:• , ...... ":'"'•" '"'•-":2;-;•;- ..?...:•..• * • ....;;.,.•'- * . '-•"•-,;.;-' "'-' - -' •':". .½,;•.•,'•.•'.: .... :';':"/;•..-•'•;•...... "-:.... .•.-.',%..• .....'.., ½•"'•'•'•••:'?;•.. . .'"•,.½.,.: '.':•.,.....,.,.. - .. ' ;.%'...... ß.... -"--,.•' - .•.."-'•...... 5-:.•%"-.' ' .• • .,.:'.•,;•'..?...•f.--.•¾:•;•'% ...•::.-...•:,•..•,...•.•..;•.,•?..-. , . ,,:.:?,.:..:.:..,..---...• ... •.:?-'-",•;%.•;½ ...... ?---..?-:?-, .•;:•*; ...... •.; • •..• ..... Fig. 4. The transpo•er tracks on relatively level, firm soil in the vicinity o• station du•ng the secondextravehicular activity (AS14•-9058). 5648 MITCHELL ET AL. surroundinglunar surface. They also saw no hesion reflects coarser grain-size distribution, obvious evidence of natural soil-slope failures differentcomposition, orthe effects of prolonged on crater walls. exposureto atmosphereis not yet known. Grain-size distribution. Grain-size distribu- CHARACTERISTICSOF RETURNED SOIL SAMPLES tions for severalsamples are shownin Figure 5. Nearly 14 kg of fines (soil) were returned Rangesin distributionfor Apollo12 samplesare from the lunar surface. In addition, two single shownfor comparison.It can be seenthat the and one doublecore-tube samples were obtained. bulk sample,which is composedof surfaceand Some observationsrelative to the physicaland near-surfacematerial, and the surface layer compositionalcharacteristics of these materi- from the trenchhave particlesizes comparable als are useful for subsequentinterpretation of to thoseof the Apollo 12 samples.However, ma- the mechanicalproperties of the soil in situ. terial that fell out of the Cone Cr;•ter core Physical characteristics.The soil from the sample and the material from the lower layer Fra Mauro site is generallylighter in colorand of the trench are considerablycoarser, although showsmore color variability than that from the still somewhatfiner than the coarselayer found Apollo11 and 12 sites.The coloris morebrown in the Apollo 12 doublecore tube [Lindsay, than black, probably becausethere is more 1971]. Median grain sizesfor the samplesshown plagioclasein the Apollo 14 soil than in earlier in Figure 5 are summarizedin Table 1. samples.Thus far, a significantproportion of Core-tubesamples. A doublecore-tube sam- glassbeads has not been found,although the ple was obtainedat stationA (Figure 1). Near content of angular glassfragments is high. the rim of Cone Crater at station C', an attempt The Apollo 14 bulk sampleappears to be less to take a singlecore tube in a rock-strewnarea cohesive than the soil from previous Apollo failed when the soil sample (some or all) fell sites,even after removalof the fractioncoarser out of the core tube. At Triplet Crater (station than I mm by sieving.Whether this lesserco- G), two attemptsto recovera triple core sam- 120FSurfoce Ioyeroftrench (19 N)/,r Apol i012 / Comprehensivesomple7 /' ttdouble core tube somples (11somples} lOt'J_ • , ,,',,' •ondsurfoce somples (6 somples)

•=, I trench(mixture} (2IN) '• 801••• ax/Cone Croter

• • Bulk- sorepie . ' • . -• 4o- '- _ ,,._ --..

•orse Ioyerin the Apollo 12 - doublecore tube • somple / , II i I I I I I I•;• • • I I I I177+• • • I 010 I OJ 0.01 Groinsize, mm

[1971]. T•e •o]• p•t• of •e •po]]o 1• cu;•e• •;e to; ]•;ge •e •e• p•;• •e to; •mM] (•l-mm) •b•mple• (0.036 to 0.113 g;•m) of t•e ]•;ge •mp]e;. Soir, ,T APor,r,o 14 Sing 5649 TABLE 1.. Median Grain Sizes oœ Apollo 14 Soil Samples

Median Grain Size, 8ample Locationa mm

i. Comprehensive Lunar module, upper 1 ½m 0.050 Bulk sample, bag 2 Lunar module 0.055 Cone crater core Station C' 0.735 Surœace oœ trench Station G 0.087 Middle oœ trench Station G 0.068 Bottom o• trench Station G 0.410

asee Figure 1. ple failed, and two single core-tube samples cm thick. Grain sizes appear to increasefrom were obtained instead.In the first attempt for top to bottom. a single core tube, the astronaut reported that 2. The singlecore-tube sample from Triplet he had struck rock and had been unable to Crater containsat leastthree layers, ranging in penetrate deeperthan one core tube. Examina- thickness from 0.5 to 5 cm. tion of the stainless steel bit from the core tube 3. The calculatedbulk densitiesin the upper confi•rmedthis, becauseit was visibly dentedand and lower halvesof the doublecore tube (1.73 burred.A secondattempt was made approxi= and 1.75 g/cm•) are significantlyless than the mately 9 meters north at a site that appeared values of 1.98 and 1.96 g/cm• determinedby similar. The coarsenessof the soil at depth in Carrieret al. [1971]for the Apollo12 double this area, as revealedby the sample from the core-tubesamples. Similarly, the bulk density trench bottom, could account for the driving for the singlecore tube of 1.60 g/cm is lessthan difficulties. Data on the returned core-tube sam- the value of 1.74 g/cm• obtainedfor the Apollo ples are summarizedin Table 2. 12 singlecore tube. Accordingto Carrier et al. Data on the core tubes thus far available [1972], the Apollo 14 core-tube densities cor- [Mitchell et al., 1971; Lunar Sample Prelimi. respond to in situ densities on the lunar sur- nary Examination Team, 1971; Carrier et al., face of 1.45 to 1.60 g/cmL These smaller values 1972] indicate: of density are possiblydue to a lower effective 1. The double core-tube sample contains specificgravity for the Apollo 14 soil particles, from sevento nine layersranging from I to 13 which consist of more microbreccias and fewer

TABLE2. Data on Apollo !4 Core-Tube Samples

Core EstieatedDensity Sample #eight, Sample Length, Density in Tube, Recovery, In Situ, Location grams cm g/ca3 % g/c=•

StationA b 39.S 7.S f 1,73 63 .60

Station 80.7 16 1.60 >49 .45 Station .•Ge 169.776.0 $1.if12 f 1.75

½a•/e•et al. [1972]. oUpperpart of doublecore tube. dLowerpart oœdouble core tube. Single core tube near Triplet crater. inglecore tube near Triplet crater; loose and undisturbed sample. etermined by Heik•n [1971] by X radiography. 5650 MITCHELL ET AL. crystalline fragments than the Apollo 12 soil 1. From less than 10% to 75% of the dif- particles. ferent samplesare composedof glass. 4. Some of the sample from Cone Crater 2. There are abundant glassand lithic frag- may have mixed with the samplefrom Triplet ments in the coarserpart of the • l-ram frac- Crater, becausethe same tube was usedat both tion, but the finer part is dominatedby mineral locations. particles. 3. Plagioclaseand pyroxene are the most Soil composition. The results of studies by abundant, the plagioclaseto pyroxene ratios the Lunar Sample Preliminary Examination varying from 16:1 to 1:2. This ratio varies Team [1971] have establishedthat the Apollo both from sample to sample and from one size 14 soil samplesvary in composition,as well as fraction to another.Some olivine is present. in grain size.The fractionscontaining particles 4. Highly angular glassfragments finer than finer than i mm have the followingcharacteris- 62.5/•m in diameter are present, and there are tics: only a few glassspheres.

. ...

. .

Maximum depth in lunarsoil with one hand Test number Depth,cm 5a 50

la 44 2:0 42

Fig. 6. Apollo simple penetrometer.The pilot pushedthe penetrometerinto the lunar soil three times. The arrowsindicate, the maximum depths he achievedwhile pushingwith one hand. The full black-and-whitestrips are each 2 cm long (S-70-34922). So• •T A•oœ•o14 S•TE 5651 penetrationsinto the lunar surface, thus pro- Lithic fragments,similar to the lithic clasts viding information on soil property variations found in the fragmentalrock samples,comprise with depth.The lunar modulepilot first pushed 70 to 100% of the 1- to 2-mm fraction of the the penetrometer into the lunar soil as far as soil. Most of the rocks and soil samplesclosely possiblewith one hand, called out the penetra- resembleeach other chemically. tion depth, and then pushedit deeperwith both ADHESIVE AND COHESIVE CHARACTERISTICS hands. This procedurewas repeated twice more at points approximately4 meters apart. Objects brought into contact with the lunar A photographof the penetrometeris shown surface or with flying dust tended to become in Figure 6. It can also be seenin the left mid- coated, although the layer generally remained dle of Figure 7, where it was later used as an- thin. The equipmenttransporter sprayed dust chor for the geophonecable of the active seis- around; however, thick layers did not form on mic experiment. The penetrometer consistsof any of the component parts, and no 'rooster a 68-cmdongaluminum shaft, which is 0.95 cm tail' dust plume was noted behind the trans- in diameterand has a 30ø (apex angle) cone porter wheels. tip. The penetrationdepths are given in Figure Although dust adhered readily to the astro- 6 and Table 3. When the pilot pushedwith both •tauts' suits, it could be brushedoff easily, ex- hands, no difficulty was experiencedin pene- cept for that part that had been rubbed into trating deeperinto the lunar surface,which in- the fabric. Dust sprinkledonto the thermal de- dicatesthat the penetrometerhad not struck a gradation sample array was easily brushedoff, rock during the one-handpart of each test. The althoughthat part of the dust that filled in the forces associated with the one-handed and two- recessednumber depressionsof the sample handed modes shown in Table 3 were estimated tended to cohere.In this case,the dust formed from simulationsperformed by an experimenter into the pattern of the numbers; then, when in an astronaut's space suit flying in a 1/6-g the sample was tapped, the dust remainedin- trajectory in a KC 135 airplane. tact and bouncedout of the depression,retain- The resistanceof the lunar soil to penetra- ing the shape of the depression. tion by the penetrometer consistsof two com- That the surface soil at the Fra Mauro site ponents,point resistanceand skin friction along possessesa small cohesionis demonstrated in the shaft: several ways; for example: F= F•,q--F. (1) 1. Clumping of soil in the vicinity of the lunar module footpadsand elsewhere. where F is the lunar soil resistance,F• is the 2. The numbersformed by dust on the ther- point resistance,and F s is the skin friction. mal degradationsample. Both components are functions of soil 3. Retention of deformed shapesand near- strength, density, and compressibilityand of vertical slopes in astronaut footprints and frictional characteristics at the penetrometer- transporter tracks. soil interface. Of the two types of resistance, point resistanceis by far the more sensitiveto It appears, however, that the cohesionis less variationsin soil properties. than would be anticipated from the results of A general expressionfor skin friction is previousmissions, as is noted subsequently.The sourceof lunar soil cohesionhas not yet been L = (2) determined; however, there are several possi- bilities: e.g., primary valencebonding between whereL is the unit skin friction, D is the shaft freshly cleavedsurfaces, electrostatic attraction, diameter, and z is the penetrationdepth. and bondingthrough absorbed layers. The unit skin friction is commonlycalculated by usingthe equation PENETRATION RESISTANCE OF THE LUNAR SURFACE • = '7zK tan a (3) At the Alsepsite in Figure 1, the Apollo sim- where 7 is the unit weight of the soil, K is the ple penetrometerwas usedto obtain 3 two-stage coefficientof lateral earth pressure,and tan a 5652 MITCHELL ET AL.

Fig. 7. The penetrometer, in position as anchor for the geophone cable, is shown in the left middle ground. Approximately 14 cm. of the penetrometer shaft are visible above the lunar surface. The deep footprints and slope failures in the vicinity of the small crater in the foreground suggestsofter soil in this area (AS14-67-9376). is the coefficient of friction between the shaft friction. However, the value of F, has also been and the soil. investigatedby laboratory simulationsand has Analysis of data by Vesi• [1963] for co- been found to be negligiblecompared with the hesionless soils shows that actual unit-skin-fric- uncertainty associatedwith the forces exerted tion values can be as much as five times greater by the astronaut.Thus, in the followinganalysis, than values calculatedby using equation 3 and values have been determined two ways: (1) that • ratio of unit skin friction to unit point F, -- 0, and (2) •,/(F•,/A) -- 0.3%, where A bearingof 0.3% is the least that can reasonably is the penetrometer cross-sectionalarea. be assumed.If this were the casefor the pene- Before Apollo 14, the strength of lunar soil trometer measurements,a significantpart of the had beendetermined quantitatively by direct in- applied force would have been carried by shaft place force-deformationmeasurements only by SOIL AT APOLLO 14 S• 5653

TAB'LE 3. Apollo SimplePenetrometer Depths

Unit of Penetration Deptha, )/odeof Force ForceF, ResistanceF/A a, Test cm App1i cation N N/cm2 i i• L ß i

la 44 One-Handed 71 to 134 100 to 188 lb 62 T•o-handed 134 to 223 188 to 314

2a 42 One-handed 71 to 134 100 to 188 2b 68 T•o-handed <134 to 223 <188 to 314

3a 50 One-handed 71 to 134 100 to 188 3b 68 Two-handed <134 to 223 <188 to 314

aCross-sectionalarea of the penetrometer,0.71 cm2;all forceis assumed to be carried in point bearing (F• = 0). means of the soil mechanicssurface sampler experimenton Surveyors3 and 7. On the basis of bearing-capacitytheory for a shallow fiat plate, Scott and Roberson[1968] calculatedthe '.oF No skin friction followingvalues. _• ------Unitskin friction =0.3 percent - •. ofthe unit-point bearing Friction angle• 37ø-35ø Cohesionc 0.35-0.70 kN/m •' •X• resistance Bulk densityp 1.5 g/cm8 2.0 -%•XN% /•ForF/A.-188 N/cm2 While the bearing-capacitytheory for an in- • X -N •.-' / Z:42 •(t.t 20) strument like the Apollo simple penetrometer is not as well developedas that for a shallow • •.o fiat plate, theory can be usedto gain somein-

sight into the shear-strengthparameters of the ._ • 0.$ soil in situ at the Apollo 14 landing site. The point-bearingcapacity of the penetrometer is given by 0.4 Surveyors r/.a = + (4) • and• whereNc and Nq are functionsof friction angle O.2 •, p is the soil density,and g is the acceleration F•F/A: I• N/• of gravity. z: 50cm(t•t A uniqueset of valuesof • and c cannotbe obtained from this equation, becausethere are 35 4O 45 three unknowns(c, •, p). However,if a value Friction angle,{,deg. of p is assumed,then for eachvalue of • there is only one value of c, and a curve of • as a Fig. 8. Combinationsof cohesionand friction function of c can be calculated. This has been angle to accountfor penetrometerresistance. The values of • and c at the Apollo 14 Alsep site done in Figure 8 using a value of p -- 1.8 probably lie within the indicated bands.The pa- g/cms (variationsin p have little effecton the rameters were calculated from the penetrometer calculations)and values for Nc and Nq (incam- data by means of the bearing-capacity theory. 5654 i•ITCHELL ET AL. pressibledeformation assumed) from Meyerho• The site chosenfor trenchingwas the western [1961, 1963]. The two cases(Fs -- 0 and rim of a small crater approximately6 metersin )•s/(Fp/A) -- 0.3%) notedabove are shown. diameter and 0.75 meter deep. A trenching The bandsof valuesfor • as a functionof c tool consistingof a 21.6-cm-long(8.5-inch) by result from the range of depths for the three 15.2-cm-wide(6-inch) blade orientedat 90ø to one-handedtests and from the uncertainty asso- a handle was used like a backhoeby the astro- ciated with the force appliedby the astronauts naut. A view of the completedtrench from the (Table3). The setof valuesfor • and c for the northeastis presentedin Figure9, whichshows in situ soil at the Apollo 14 Alsepsite must lie the steepesttrench wall. AstronautShepard re- somewherewithin these bands. Also plotted in portedthat diggingwas easy and estimatedhis Figure8 'is • as a functionof c for data.from first cut to be to a depth of approximately15 the Surveyor3 and 7 sites.The band for the cm with sidewallsat an angle of 70ø to 80ø. penetrometerparameters, if no skin frictionis With the next cut, the walls were steepenedto assumed,falls above and to the right of the 80ø to 85ø, but at that point, they started Surveyor parameters,which implies that the caving.It is not clear .from the varioustran- values of the soil-strengthparameters at the scriptsand debriefingsjust how steepthe walls Apollo 14 landingsite may be greater than remained over-all, but it appearsthat, for the those at the two Surveyor sites. However, if rest of the excavation,slopes steeper than ap- skin friction is a significantfactor, the param- proximately60 ø couldnot be maintained. eter values at the different landing sites may The excavationpassed through three distinct be approximatelythe same.The former situa- layers.The upper 3 to 5 cm were dark brown tion is not unreasonable,however, because the and fine grained.Next, a very thin layer (0.5 Surveyordata are from surfacetests, whereas cm thick or less) of black glassyparticles was the penetrometerdata are from subsurface encountered.Beneath this layer was a material tests, and in general the lunar soil strength muchlighter coloredand coarsergrained. This probablyincreases with depth. stratificationis not readily visible in Figure 9; The detailed variation of unit penetration however,the differencein grain sizeis marked, resistancewith depth in the lunar soil is prob- as is seenby the distributioncurves in Figure 5 ably very complex,as a result of layers and for samplesfrom the upper and lowerlayers. pocketsin the soil with varyingbulk density, On the basisof previousestimates of lunar grain-sizedistribution, cohesion, and other fac- soil friction angleand cohesionat other sites, tors. An infinite number of possibledistribu- it wasanticipated that trenchingto a depthof tions of unit penetrationresistance as a func- 60 cm with near-vertical sidewalls should be tion of depth would fit the penetrometerdata possiblewithout difficulty. If one assumescon- obtainedduring the Apollo 14 mission,because servativelya homogeneoussoil with a density only end-pointvalues are known,although the of 1.9 g/cm•, a cohesionof 0.35kN/cm •, anda pilot did indicatethat penetrationresistance frictionangle of 35ø, then a stabilityanalysis increasedwith depth. Thus the data are in- indicatesthat a vertical cut shouldbe possible sufficientlysensitive to reveal lateral inhomo- to a depthof 85 cm beforesidewall failure de- geneitiesat depthamong the three penetration velops.Analysis of the Apollo14 photographs sites. showsa maximumtrench depth in the rangeof SOIL MECHANICS TRENCH 25 to 36 cm for sidewallslope angles increasing from 60 ø to 80 ø . At station G (Figure 1), Astronaut Shepard Combinationsof cohesionc, density7, slope attempted to excavatea 60-cm-deeptrench with one vertical sidewall to: (1) exposethe heightH, slopeangle •, and friction angle• in situ stratigraphy; (2) provide a meansfor that correspondto incipientfailure for homo- estimationof soil-strengthparameters by using geneousslopes are shownin Figure 10. These stability analyses;(3) provide data on varia- curvesare basedon the assumptionthat curved tions in soil texture and consistency;(4) enable (circular) failure surfacespass through the toe samplingat depth,and (5) determinethe ease of the slopeand that soil is homogeneous.A with which surface material could be excavated. band of values that probably encompassesthe Sore •T APollo 14 S• 5655

...... ,•.:. .•y.' :..- . .f:.•,,..•..: ,.;-;'....ß • %-.:?½ .....ß ?;:..:, ;,,..:'...... ;, .- •.$.;•: .;.:•-' ,..,'-,,.., '*;':::./•'.i•:; •½:!:,:-... ;,-;'-•:.,:,, _ ',..'".:.... .,.;•;i:"' %-,,...... :t.' "'.,.•...,, -:"•'*'{ :.'p.;f-"';.,.....,.. ...o...... :,-' "• :,;.• '..,,•':, ,.•-*.•,...... '"'••••.... •X..-..';:•'*:,,,-,•o,-....;:,;*.'-,•,,t,:..,*.,.;,.•:•' ....;:, -,, ...',:: ...

. f.,;-:"•'-ø'.. - ' "--.:' ;;...... ?#*'•..:.....:-•'•::.::..':./:";....•'%N%", :;::..' '..-.,½"' '.."- ... ' ',•' '• ...... ••,.'-..'..'½' ". - *"•,':-' ':*{-....•':4½'. o'-'..'/ -. • ..• . '-,4.- ", ....

-•;.i. '•:...... ,, •'½'-'!,:..-i,..,;.. -' ' ..... ':'..... ,,.:::.;:";' ...... "':' .•t' *'-"'.. ;•.:,.•; -.."'":'::::-,..-::,."•:'..,, -I.½• .-"•,:.,,,.r,•:•.•..:,:...:. ....*• ,,. .. -...... •,;.,...... ,,..,?, .:.,.. . ,...,•:,...... -.-.?. •.-• .•.'.....,.....,...... "}..• .,,:a...:;.:.-.•,.:,•;::.½...... :• ,...:--...... :...... ,:::::..(½.,--.::,t:...... •.... :. ,, .....•...... •.:.....•...... --•.; :...,.**,...... ,.,.-,..-.•½•;,... ;,.,...•....-,...... ,:. ..._..,.•.a.:,.•."•.• •. ,.-•. .•**.•,,;;• ,,,,.,.,...,.--•--•;:,.,,:,•,....,.•, ./,-;•...... :...% :'-: ..4.*.,.,'•*½,.•.,•½:::...•..•: '*;...... ,,...... :• c:-"-.'::: .,...,,.:,,..:.."'-"- "-'-' ...,....:..•:::: "::.X...... '*- - ..ß ...... ' .. .•.,•,...... **- :.;.•'' --:•.., ...... *.:--,•-• •...... :'.- :"*'-...'• ..,::•;,•...... ,...... - .,,:::.*-:•:-:.*•a-•--.::• ...... :,...... -,: .--•---,*,. :-,.:,..... ,-...... -'--:,• .-.:.,. .-...... - --•.•.. . ,...... , t,-....?...... ,. •...... -*.--•...... i'•;,...... - ,...... ß,-:-,,., ..;..;-,•*•,.:.f-...... ,.,...... -•. •..**-.-,-::...... *, ...... •J ,:• ...... ,.. "-•'*:'••"':;•'.'*• '¾::a':"'"•a'•,,½'" "'".. •'.*.-:.:'•:•?.''.-, .,.,i*:*.....":•;: ...... ,....:.-' '.-;,,.-.,.,;: "•'..:..--;:.'.".•:..'•"•'•'"""'•'<•••••••,:...... -•:,:•.,:,..• ...'.-.Z.:'.,::.,..:• ••,-':'-, ...- ,.-'";,:.":: ..,.•';.'•:'-'•'"'*"""'-". -..'-•/;,..;.a:•}::•.. ½,. .,..,;:.•,;? ,•..,* ,.... -,.j:',•'**',":"•.½.*;::: :;/•;...... •,;;-. ....:."**'""--'a,.•:.. * ...... -,•.,a,'. .,..,..: ....:1•!':•""*::. ':':..:,,;{"•½;--. -,':'•b.... ';*;."-'%'"**½J"'"'•m•:!:;•J*'. *..•½.'.-... • ":".::.t. -*-.*:.""-•};{%'.,...... '

.-'*"..;.'-.-., '•':'*'•?•?,-•,•-•:;...... ,,,,1•.;:.•-.'*,--.•, a'-:;:::,:" ..,:.?•'.*-:. . , -•"- .....•-o-?;:•:• • ---•,**:..--..,:::,:;.,..-.:,-::;---'-•a-: '.?:"< ,:.t: ..•'?+::.t}•:::,-. ....ß.., . -.-*...,:.-o:•:;.';'-•:::;.:.*•.:,' :,..,:,,;t:,; ...... ,.' ...;,::.-•:..-::;,i• :•, ...... :.•-:,.:-•'•! .::,:;,'.--.-:½. '½:•,.;z .:;.,;*; -.:.::.,, . ',•.•; .. ,,'...... ;.,.•,'-" ...... :;•,•.-: • ':,::..-...•. -:-.:•,;: .• ...... •...... ,,:...... ,...... :/(..:,:::• .... •;i.,,,-,:,...... ,.....;;;,,.,,.:-:;:._...- -,;':j,.... ,-,...... - ,...,,.:-..,?...... ,•:•;**;:.?:,.:%:,...,:•,,.- ..... :.,.:...½.:.:-...... :., "...... *.'':."tlr.'.'...' .: **•:-%.,,:. .•:;,•..,...... ,-..: ' ...... :.... ::.:.....-:-:-: - - .r.;:..•:;. ....•.::...-;.•.•::.--::•.,. :-:•.....' ' - ""' x...... ' .'*-:"' - '.. '-:•:*-::'?:-•½. %-• ,.:,,,::.::.:.?&;;'?.... .-½:.:.- "*"- "...... ;:., *"'-.:*:": •.... -..-:

-..::::'.**z:--*.-*::;:'::-.-•,:•..-::,•::•.:,•'•.:-:..... ';;;;;..,.&•? •::*:•-,..-;::---,',:'-.',..:;•,:'•'::¾'" .:.4:;-•. •;•;;' i ..... "--' ..-.•"•""...... - '"'. •.!'•! "-:•" ':'•"•'•:-•-•'•''?•'-".-*-;":;•:'•'...... ',.i.•.,,,,,_.'.,:..-.-: :•-:-:":;•'"'¾ ....:. ....*'::•' ::.:•-'•:'"" ...... '--*'?':':*•':"" ?.,½:,•...... :- ....•'-?';'"•' ,...... "'•'' !•' •':• •'-½'...... '•½•;-•:-'"•"('""':.- ...... ½":"'½:/;-.-. --!:•:':&::'::',i'...... i½:.*.;•;•."-;-' . ":'.....,.,ß...... :•',.•:;•; ....,,:...... '.... *"'-:*',.;.:..,-..•' -.-• Fig. 9. View of the soil mechanicstrench from the northeast. The steepest sidewalls a•re estimated to be 65ø to 80ø. A pile of excavated material is at the end of the trench at the left in the photograph (AS14-64-9161). conditionsin the Apollo 14 trench is shownin requiredwould be evenless, as can be seenfrom Figure 10. Figure 10. It shouldbe noted,however, that the For the slopeangles and correspondingtrench computed values of cohesionare lower-bound depthsnoted above, an assumedunit weightof estimates;that is, they are the minimum values 3 kN/m 8 in the lunar gravity field, and an as- requiredto maintain stability for the conditions sumed friction angle of 35ø , the required cohe- shown.Nonetheless, it doesappear that the soil sion values for stability fall in the range of beneaththe surfaceat the locationof the Apollo approximately0.03 to 0.10 kN/m •. Thesevalues 14 trench is less cohesive than would have been are considerablyless than the values estimated anticipatedon the basisof previousdata. This using previousluna. r-surface data. For higher lower value for cohesionis compatiblewith the values of friction angle, the values of cohesion coarser grain sizes encountered. 5656 1VIITCHELL ET AL.

0.20 probewas calculated.The influencearea is given by

= ( + D•) Ni where At is the total influence area of all par- .15 ticles larger than i•, i• is the smallestparticle size that will stop penetration,D, is the diameter of particle i, D•, is the diameterof the probe,and N, is the numberof particlesin the size range i (850 size ranges were used to de- scribe the Apollo 12 size-versus-frequencydis- • .IO tribution curve). The ratio of this influence area to the total samplearea representsthe probabilityof strik- ing a critical particle .in a unit volume.The Probable conditions for analysisis conductedto any desireddepth, and .o5 Apollo14 the cumulativeprobability of not encountering trench a critical obstructingparticle is calculated. The probability analysiswas used to determinethe likelihoodthat the three penetrometer penetrationscould be made to the full depth of the instrument (68 cm) without encounteringa 04O •o 60 7o 8o rock equal to or larger than the diameter of Slopeangle,/1, deg the instrument(0.95 cm). The prQbabilityfor the three events was calculatedto be 51%. Fig. lO. Stability numbers for homogeneous slopes. The four core-tube events were analyzedin a similar manner. The combined probability of the four tubes being driven to their respective IN SITU PARTICLE-SIZE DISTRIBUTION depths without encounteringa fragment equal A preliminaryanalysis has been made to com- to or greater than the core-tubediameter (1.95 pare behavior during surface penetration and cm) was 67%. In fact, one of the core tubes trenching with that predicted on the basis of (tube 2022) hit an obstruction.The probability the Apollo 11 and 12 particle-sizedistribution that one or more core tubes would hit an ob- curvesgiven by Shoemakeret al. [1969, 1970]. structionwas only 33'%. For the analysis, the assumptionshave been The Apollo 11 and 12 sizedistribution curves made that particle-sizedistribution does not were also used to predict the number of par- vary with depth and that the volumetric ratio ticlosof a given sizethat shouldbe foundin the of variouslysized particles equals the area ratio soil mechanics trench. The results indicate that observedat the surface,or only a few particleslarger than 1 cm would have been expectedin the excavatedsoil. How- ,4,/,4 = v,/v ever,many rocksin the 1- to 2-cmrange were where A, is the area of particlesof size i in the reported, and numerousparticles of this size surfacesample, A• is the total surfacearea o.• can be observed in the photographsof the the sample,V, is the volumeof particlesof size trench. Furthermore,the grain-sizedistributions i in the sample,and V• is the total volume of determinedin the Lunar Receiving Laboratory the sample. Sufficientdata are not available at for material from the bottom of the trench presentto determinereliably the probablevaria- (Figure 5) showed more than 20% of the tions in size-versus-frequencydistribution with sampleto be coarserthan I cm. This is strong depth. evidence that the determinationsof particle The total area of influence of all particles size versus frequency made at the surface of large enough to interfere with a penetrating the Apollo 11 and 12 sites are •notgood indi- SO•L AT APOLLO 14 S•T• 5657 cators of the size distributions likely to occur esses,as would be indicatedby nearly random with depth at other sites. local variations and near homogeneity on a LATERAL VARIATION OF SOIL PROPERTIES regional basis. Variations in crater morphologyat the Apollo Study of the variability of soil propertieswith 14 landing site indicate the following age se- lateral position is important in order to ascer- quence of craters along the geologictraverses tain both the magnitudeand randomnessof the (Figure 1), beginningwith the oldest,according variations that may be encountered It is of to interpretations by Eggleton and Offield interest to know whether variations can be [1970] and by Trask [1966] as reported by associateddirectly with differencesin geomor- Swannet a/. [1971]. phologicalfeatures or whetherthe influenceson the soil of original geomorphologymay have 1. Highly subduedcraters characterizedby been obliterated by random lunar-surfaceproc- very gentle depressiol•sat (a) the landing site

Fig. 11. The transporter in use during the second extravehicular activity (AS14-68-9404). 5658 MITCHELL ET AL. of the lunar module; (b) west of the lunar soils.With this quantity known and the results module at the Alsep deploymentsite; and (c) of tests on lunar soil simulants, it has been north of station A. possibleto estimate the in-place void ratio 2. Moderately subduedcraters at (a) North densitypL, and friction angle •L for the soil on Triplet Crater; (b) the 50-meter-diameter the moon. This procedureis describedin the crater east of station F; and (c) the 10-meter Appendix. crater at station A. The results of the analysisare presentedin 3. Cone Crater and the sharp 45-meter Tabl• 4. From the indicatedranges in lunar crater at station E. soil properties, it can be seen that within re- 4. The sharp 300-meter crater at station C gionsof comparablegeologic age there is a small and a small 10-meter crater next to which a but distinct difference between average prop- football-sizerock was collectedduring the first erties of soil located in intercrater areas on extravehicularactivity. firm, level ground and soil located in soft pockets and fresh crater rims and slopes.No Information relating to soil properties asso- appreciabledifferences in lunar soil properties ciated with these features can be obtained by between regions of different geologicage are inference from the observed interaction of discernible,however, on the basis of the com- various objectsof known geometryand physi- puted values. cal characteristics with the lunar surface. One It appears,therefore, that, within the upper suchtype• of interactionthat has'been useful few centime•tersof the lunar surface at the for study of soilproperties is that of the wheels Fra Mauro landing site, the initial consistency of the modularizedequipment transporter with of the lunar soil resultingfrom a major geologic the surface. event of short duration, such as the primary The transporter(Figure 11) is a two-wheeled, impactof a largemeteorold or th• impactof ricksha-type vehicle equipped with pneumatic ejecta,has been modified by long-durationproc- tires that was used to carry equipmentduring essesthat have created a steady-statecondition. both periods of extravehicular activity. The Such processesmay includethe continuousflux maximum weight of the vehicle plus payload of micrometeoroidsand the steady downslope in lunar gravity was 128 N (28.9 lb), including movementof material, superposedon the con- a maximum weight of rock and soil samplesof tinuous formation and extinction of small and 38.7 N (8.7 lb). During the secondperiod of medium-sizeimpact craters.The continuousin- extravehicular activity, the transporter tra- teraction of such events with the lunar surface verseda distanceof approximately2.8 km, and is further evidencedby the followingobserva- the total weight fluctuated between89 and 120 tions: (1) formation of raindrop patterns on N, dependingon the amountof samplecollected. relatively level surficial fine-grainedmaterial, The transporter tracks in the vicinity of but low frequency of occurrenceof such pat- station B can be seen in Figure 4. The astro- terns on steeper slopes; (2) rounding 6ff of nauts estimatedthe track depths to vary be- boulders and development of fillet material tween 0.5 and 2 cm and noted that the largest against outward-slopingrock surfaces;and (3) sinkagesoccurred around soft crater rims. A the presenceof small fine-grainedhummocks a closeupphotograph of a transporter track in few centimetersin size. Irrespective of age, the immediate vicinity of station A is shownin however,the lunar soil at the surfaceappears Figure 12. The estimatedtrack depth at this in generalto be firmer at level intercrater areas location is 1.0 cm. than at small fresh crater rims and slopes. The facts that the transporterwas equipped with pneumatic tires of known geometry and LUNAR SOIL MODULUS load-deflection characteristics and that the lunar The results of the Apollo 14 active seismic soil at the Apollo 14 site is predominantlygran- experiment,reported by Kovach et al. [1971], ular make possiblethe estimationof the average indicatedthat the seismicvelocity (P wave) in changein the penetrationresistance with depth, the upper 8.5 meters of the Fra Mauro site is designatedby GL, on the basis of the analysis 104 m/sec. Accordingto elastic theory for a developedby Freitag [1965] for cohesionless homogeneous,isotropic medium, the P-wave So•, •T A•o•,•,o 14 S•T• 5659

.- .

. .

Fig. 12. Close-up photographof transportertrack in the vicinity of station A. The transporterhas evidently rolled acrossslightly firmer soil, becausethe track narrows,in- dicating less sinkage.A small white rock has been pushedinto the soil by the tire in the lower left-hand comer of the photograph.The indention at the upper left-hand comer of the photographwas probably produced by the edgeof the camerawhile it wasbeing pisi- tioned (AS14-77-10358). velocityv• is relatedto Young'smodulus E by known that Young' modulusat small deforma- tion is related to confiningpressure (rs according to v,'- p(1E(1-•-•,)(1- --•) 2•,) 11/2 (7) E = Kp,(as/p,)" (8) wherep is density,and v is Poisson'sratio. Combinationsof modulus,density, and Pois- where K and n are parametersdependent on son'sratio correspondingto a velocity of 104 soiltype, andpa is terrestrialatmospheric pres- m/sec are shownin Figure 13. From a number sure in the same units as (rs. Kulhawy et al. of studies on cohesionlessterrestrial soils, it is [1969] have shownthat for well-gradedsands 5660 MITCHELL ET AL. TABLE 4. Variation in Lunar Soil Properties with Lateral Position as Dete•ined From Transporter Tracks

Geology Traverse Terrain G.L, • .,b $L' Stations a Type N/cma eL g/Zc•a deg

Highly SubduedCraters Lunar module site Level 0.83 to 0.47 0.68 to 0.70 1.85 to 1.81 43.4 to 41.0 Alsep site, sta. A, B Firm 0.73 0.69 1.84 42.8 Lunar module site Soft spots 0.34 to 0.27 0.74 to 0.75 1.79 to 1.77 39.6 to 38.5 Crater rims O. 30 O. 74 1.78 39.0

Moderately SubduedCraters A, G, B2 Level 2.02 to 0.47 0.62 to 0.71 1.91 to 1.81 47.1 to 41.0 Firm 0.91 0.68 1.85 43.2 A, B1, B2 Soft spots 0.47 to 0.20 0.71 to 0.77 1.81 to 1.75 41.0 to 37.2 Crater rims 0.30 0.75 1.77 38.5

Sharp Craters B3, C' Level 0.83 to 0.34 0.68 to 0.74 1.85 to 1.79 43.4 to 39.6 Firm 0.65 0.70 1.83 42.1 C' Soft spots 0.47 to 0.20 0.71 to 0.77 1.81 to 1.75 41.0 to 37.4 Crater rims 0.34 0.74 1.78 39.4

Values of lunar soil properties given as either ranges or averages. Accordingto descendinggeologic age. Thesevalues of densityare somewhathigher than those estimated on the basisof core- tube data (Table 2). They can be attributed in part to the value of 3.1 assumed for specific gravity in the analysis. The results of recent (1972, unpublished) tests on two Apollo 14 soil samples suggest that a value of 2.95 is more appropriate. typical values of K and n are 300 and 0.50, moduli correspondingto confinementat a depth respectively.From test data given by Namiq of 4.25 meters on the lunar surface (half the [1970] for a lunar soil simulant, n. can be de- depth of the estimatedthickness of the uncon- ducedto be 0.51, and K increasesfrom 34 at a solidatedlayer). If the lateral pressureis taken void ratio of 0.8 to 130 at a void ratio of 0.6. as one-half the vertical pressure,probably a From these values, one can computeaverage reasonablelower bound, the resulting modulus values are in the range of about 0.1 X 10' to 0.75 X 10' kN/m •'. It can be seen that these 2.4 -- values fall within the range, but at the lower end, of those shown on Figure 13. There is

2.0 reasonto believe that, had the terrestrial measurements been carried out under high-vacuum conditions,the modulusvalues would have been somewhathigher owing to the absenceof ad- sorbedfilms betweenparticles. The importanceof theseresults is that special or unusualsoil propertiesneed not be assumed in order to account for the low seismic veloci- o 0.8 ties. On the moon, the low-gravity conditions result in low confinement,which in turn gives 0.4 a low seismicvelocity.

CONCLUSIONS 0 I I I I I I 1.0 1.2 1.4 1.6 1.8 2.0 2.2 The results of the Apollo 14 missionhave Density- (;i/cm $) provided new insightsinto the properties and behavior of the soil at the surface of the moon. Fig. 13. Young's modulus as a function of density and Poisson's ratio for a seismic velocity of Some of the more significantfindings resulting 104 m/sec. from the analysesherein are as follows: So•. AT A•o•o 14 S•T• 5661 1. A greater variation exists in the charac- pected for a terrestrial silty fine sand layer in teristics of the soil at shallow depths (a few the lunar gravitationalfield. centimeters)in both lateral and vertical direc- APPENDIX tions than had previouslybeen supposed,and DETERMINATION OF SOIL PROPERTIES soil with a particle-size distribution in the FROM MET TRACK DATA medium- to coarse-sandrange may be encountered at shallowdepths. It is assumedthat the functional relationships 2. There is evidencethat venting from the describingthe tire-soil interaction under lunar soil voids of descent-engineexhaust gas after environmental conditions and low velocities are engineshutdown caused fine particle movement. essentiallythe sameas the relationshipsdescrib- 3. Particles erodedby the exhaustplume of ing the correspondingtire-soil interaction on the lunar moduledescent engine during landing earth. On this basis, the normalized sinkage were distributed over a large area around the z/d, in which z is the penetrationdepth of the final touchdownpoint. tire and d is the tire diameter,is related to the 4. As had beenthe casein previousmissions, dimensionlessquantity dust was easily kicked up and tended to adhere to any surfacecontacted. Ns•= G(bd)S/•w • (9) 5. Soil density determined from a double core-tubesample taken at station A was about in which G is the average changeof conepene- 1.60 g/cm'. These values are lower than those tration resistancewith depth, b is the tire sec- obtainedfrom the Apollo 12 core tubes, but are tion width, W is the tire load, $ is the tire de- well within the range thought to be character- flection,and h is the tire sectionheight, by the istic of the moon over-all. samefunction relationship as the corresponding 6. The resultsof simplepenetrometer meas- quantities are related to sand under terrestrial urementssuggest that the soil in the vicinity of environmental conditions. the Apollo 14 Alsep may be somewhatstronger The dimensionlessquantity Ns•, termed the than soil at the landing sites of Surveyor 3 sand-mobility number by Freitag [1965], ex- and 7, and comparisonof the results of anal- pressesthe soil strength and density character- yses using penetrometer,transporter-track, and isticsthrough the quantity G. The averagepen- trench data suggeststhat soil strength increases etration-resistancegradients GL of the lunar with depth. soil at various locationsalong the geological 7. Computationsof soil cohesionat depth at traverse were determined using transporter- the site of the soil mechanics trench result in track depth data (estimated from Hasselblad lower-bound estimatesconsiderably lower than camera stereo-pairs), known tire characteris- were expectedfrom the resultsof previousmis- tics, and the z/d versusNs• diagram shownin sions (0.03 to 0.10 kN/m' as opposedto 0.35 Figure 14. The solid cur•e representsthe best to 0.70 kN/m'). fit for terrestrial wheel-soil interaction test data 8. The determinationsfrom the Apollo 11 (180 tests) reported by Green [1967]. These and 12 sites of surfaceparticle size versusfre- tests were performed with pneumatic tires of quency are not good indicatorsof the size dis- differentgeometries and size on a uniform dune tributions likely to occur with depth at other sand from the Arizona desert, designatedYuma sites. sand.Figure 14 also showswheel-soil interaction 9. Analysis of transporter-track depths has data obtainedfrom tests performedwith trans- indicatedangles of internal friction in the range porter tires inflated to the same pressureand of 40ø to 45ø (if a cohesionlesssoil is assumed) subjectedto the same load range as those on for soil near the surface, and that the soil is the lunar surface. A lunar soil simulant consist- lessdense, more compressible,and weakerat the ing of a crushed basalt, Waterways Experi- rims of small craters than in level intercrater mental Station (WES) mix, was used for these regions. tests. 10. The magnitudeof the in situ modulusof Valuesof GL (Table 4) determinedfrom Fig- elasticity, based on the measuredseismic-wave ure 14 were convertedto correspondingvalues velocity, is compatible with that to be ex- in the earth's gravitationalfield G• by assuming 5662 MITCHELL ET AL. that penetration resistancevaries in direct pro- of gravity, B is the least dimensionof founda- portion to gravity level; i.e., Gr -- 6 G•,, in tion base (dimensionof foundation), z is the accordancewith the analysis of Costes et al. vertical distance from the surface to the base [1971]. of the foundation,No, N7, Nq are bearing ca- The behavior of the lunar soil simulant and pacity factors that dependon the angle of in- that of the actual lunar soil were assumed to ternal friction • of the soil, and so, •, s• are be the same if both are at the same void ratio. factorsthat dependon the shapeof the founda- Accordingly, G• values were used to estimate tion. the in-place lunar soil void ratios e•, (Table 4) If it is assumedthat c, N•, p, N7, and N• are at correspondinglocations from the diagram of independentof depth, then from equation 10 G versus void ratio for the lunar soil simulant (WES. mix) shownin Figure 15. Corresponding dq,/dz = S•pgN, values of soil density have been determinedas- But, from the definitionof G, suming that the specificgravity of the lunar soil particles averages3.1. aq,/az- Values of friction angle for different values so that Nq can be found from of G and density were estimated as follows [Cosreset al., 1971]: The resistanceto pene- Nq- G/S•pg (11) tration of the lunar soil can be expressedby the Values of friction angle were determined, Terzafthi [1943] simplified bearing-capacity using equation11 and N, versus• diagrams equation for foundations: developedby Meyerho/[1961] for shallowconi- cal-shapedfoundations and are listed.in Table q, -- sccNc-[- -•s2 • pgBN• -[- sqpgzNq (10) 4. where qz is the bearing capacity of the founda- It couldbe arguedthat the bearing-capacity tion (i.e., the resistanceto penetration of the factorsN•, N, and N, will increasewith depth, lunar soil at depth z), c is the soil cohesion, sincethe density of the soil will tend to increase p is the bulk densityof soft,g is the acceleration with depth, and both the cohesionc and •

0.15

0.10

0.05

O0 I0 20 30 40 50 G(bd •/= • SandMobility No. Nsu = W h

Fig. 14. Tire sinkage versus sand mobility number for two soils. The curve represents the best fit to first-passtests on Yuma sand at Towed Point [Green, 1967]. The points repre- sent first-passtests with transporter tire on lunar soil simulant (WES mix) at Towed Point. 8OIL AT APOLLO 14 5663 RelativeDensity, Dr' (percent) suchthat p and e are not sensitiveto the varia- 300ø I0 • 30 ,10 5o Go 70 14)0 I I I I I I tions in Gr values that result by multiplying 200- - $0.0 the correspondingG• valuesby G, rangingbe-

I00 - tween 3 and 6. The relationshipbetween Nq and 80- ß •b is suchthat • is not sensitiveto the resulting •. 60 -

40- variationsin Nq given by equation11. The • '•/•' o - I0.0 co valuesassociated with G• rangingbetween 3 '•// o 20- - 5.0:• and 4 are lower than the .• values resulting from taking G• = 6 by I to 1• deg. Cor- io- responding variations in density are of the 8- - 2.0 • 6- .•, order of 2%. 4

o ._ Acknowle.dgments. Professor W. N. Houston, Dr. H. J. Hovland, and •. T. Durgunoglu assisted - 0o5•in simulation studies and data analyses at the University of California, Berkeley. Dr. Stewart W. 0.8 Johnson, Richard A. Werner, and Ralf Schmitt 0.6 participated in phases of the work carried out at 0.4 the Manned Spacecraft Center, Houston. G. T. o.3 I I I I I I I o. Cohron assisted in the transporter-track analyses, 1.0I 0.4 4).1 1.0 0.9 0.8 0.7 0.6 O.S Void Ratio - (e) which were done at the Marshall Space Flight Center, Huntsville, Ala. The assistanceof these Fig. 15. Variation of penetration resistance individuals is gratefully acknowledged. gradient of lunar soil simulant (WES mix) with The studies reported herein were supported, in soil consistencyand packing characteristics.The part, by NASA contract NAS 9-11266, principal moisture content • is 0.9% for the solid curve investigator support for. soil mechanics experi- and 1.8% for the dashed curve. ment (S-200), and NASA contract NAS 9-11454, co-investigatorsupport for soil mechanicsexperi- ment (S-200). increasewith incressingdensity, and because the geometryof the zoneof plastic equilibrium REFERENCES within the soil mass changeswith depth. Be- Carrier, W. D., III, S. W. Johnson, R. A. Werner, causeof this, the derivative of q, in equation and R. Schmidt, Disturbance it, samples re- 10 with respect to z will not be as shown in covered with the Apollo core tubes, in Proceed- equation 11, but in generalwill include addi- ings of the Second Lunar Sci.e•ce Conference, tional terms as follows: vol. 3, pp. 1959-1972, MIT Press, Cambridge, Mass., 1971. Cartier, W. D., III, S. W. Johnson,L. H. Carrasco, dq,/dz = G = A -•- (B -•- S•pN•)g (12) and R..Schmidt, Core sample depth relationships: Apollo 14 and 15, in Proceedingsof the in which A and B are quantities independent Third L•nar ScienceConference, vol. 3, in press, of q and are obtainedby differentiatingc, p, q•, MIT Press, Cambridge, Mass., 1972. No, N7, and N• with respectto z. Thus the Costes, N. C., W. D. Carrier III, J. K. Mitchell, ratio G, of the penetration-resistancegradient and R. F. Scott, Apollo 11 soil mechanics in- of the soil at 1-g gravity level to that for the vestigation, Apollo 11 Preliminary Science Re- port, NASA Spec. Publ. 214, pp. 85-122, 1969. same soil at 1/6-g gravity level is less than 6. Costes, N. C., G. T. Cohron, and D.C. Moss, This observationis corroboratedby the anal- Cone penetration resistance test: An approach ysesof Hous'to•and Namiq [1971], who found to evaluating in-place strength aud packing that a ratio of 3 to 4 was appropriate for the characteristicsof lunar soils, in Proceedingso! the Second Lunar Science Conference, vol. 3, lunar soil simulant used by them. The actual pp. 1973-1987, MIT Press, Cambridge, Mass., value in any casewill be highly dependenton 1971. the magnitude of the cohesioncontribution to Eggleton, R. E., and I. W. Offield, Geologic maps the strength. of the Fra Mauro region of the moon, Map 1- Within the rangeof G, e, and • valuesesti- 708, U.S. Geol. Survey, Washington, D.C., 1970. Freitag, D. R., A dimensional analysis of the mated to be representativeof the in-placeprop- performance of pneumatic tires on soft soils; erties of the lunar soil, the functional relation Tech. Rep. 3-688, 140 pp., U.S. Army Eng. of G versus p or e of lunar soil simulantsis Waterways Exp. Sta., Vicksburg, Miss., 1965. 5664 MITCHELL ET AL.

Green, A. J., Development and evaluation of lated lunar soil, Ph.D. thesis, 218 pp., Univ. of mobility. numbers for coarse-grainedsoils, Per- Calif., Berkeley, 1970. formance of Soils under Tire Loads, Tech. Rep. Scott, R. F., and F. I. Roberson, Soil mechanics 3-666, 5, U.S. Army Eng. Waterways Exp. Sta., surface sampler, Surveyor Project Final Re- Vicksburg, Miss., 1967. port, Part 2, Tech. Rep. 32-1265, pp. 195-207, Heiken, G., Apollo 14 core tubes, Memorandum Jet Propul. Lab., Pasadena.,Calif., 1968. to TN 12/Curator, Planet. and Earth Sci., Scott, R. F., W. D. Carrier III, N. C. Costes,and NASA Manned Spacecraft Center, Houston, J. K. Mitchell, Mechanical properties of the Tex., April 29, 1971. lunar regolith, Apollo 12 Preliminary Science Houston, W. N., and L. I. Namiq, Penetration Report, NASA Spec. Publ. 235, pp. 161-182, resistance of lunar soils, J. Te•'ramech., 8, 59- 1970'. 69, 1971. Shoemaker, E. M., N. G. Bailey, R. M. Batson, Jaffe, L. D., Blowing on lunar soil by Apollo 12: D. H. Dahlem, T. H. Foss, M. J. Gro.lier, E. N. Surveyor 3 evidence, paper presented at Second Goddard, M. H. Halt, H. E. Holt, K. B. Larson, Lunar Science Conference, NASA, Houston, J. J. Rennilson, G. G. Schaber, D. L. Schleicher, Tex., Jan. 11-14, 1971. H. H. Schmitt, R. L. Sutton, G. A. Swann, Kovach, R. L., J. S. Watkins, and T. Landers, A. C. Waters, and M. N. West, Geologic setting Active seismic experiment,•Apollo 14 Prelimi- of the lunar samples returned by the Apollo 11 nary Science Report, NASA Spec. Publ. 272, mission, Apollo 11 Preliminary Science Report, chap.7, pp. 163-174,1971. NASA Spec. Publ. 216, pp. 41-84, 1969. Kulhawy, F. H., J. M. Duncan, and H. B. Seed, Shoemaker, E. M., R. M. Batson, A. L. Bean, Finite element analyses of stressesand move- C. Conrad, Jr., D. H. Dahlem, E. N. Goddard, ments in embankments during construction, M. H. Hair, K. B. Larson, G. G. Schaber, D. L. Rep. TE 69-4, 169 pp., Geotech..Eng., Univ. of Schleicher, R. L. Sutton, G. A. Swann, A. C. Calif., Berkeley, 1969. Waters, I-I. E. Holt, J. J. Rennilson, R. F. Scott, Lindsay, J. F., Sedimentology of Apollo 11 and W. D. Carrier III, N. C. Costes, and J. K. 12 lunar soils, J. Sediment. Petrol., 41, 780- Mitchell, Preliminary geologic investigation of 797, 1971. the Apollo 12 landing site, Apollo i2 Prelimi- Lunar Saml•le Preliminary Examination Team, nary ScienceReport, NASA Spec. Publ. 235, pp. Preliminary examination of lunar samples, 113-182, 1970. Apollo 14 Preliminary Science Report, NASA Swarm, G. A., N, G. Bailey, R. M. Batson, R. E. Spec. Publ. 272, chap. 5, pp. 109-132, 1971. Eggleton, M. H. Halt, I-I. E. Holt, K. B. Larson, M. C. M cEwen, E. D. Mitchell, G. G. Schaber, Meyerhof, G. G., The ultimate bearing capacity J. B. Schafer, A. B. Shepard, R. L. Sutton, of wedge-shaped foundations, Proceedings of N.J. Trask, G. E. Ulrich, H. G. Wilshire, and thee Fifth Inter•atio•al Conference o• Soil E. W. Wolfe, Preliminary geologic investiga- Mechanics a•d Foundation Engineering,vol. 2, tions of the Apollo 14 landing site, Apollo 14 pp. 105-•09, Dunod, Paris, 1961. Preliminary Science Report, NASA Spec. Publ. Meyerhof, G. G., Some recent research on the 272, pp. 39-86, 1971. bearing capacity of foundations, Can. Geotech. Terzaghi, K., Theoretical Soil Mechanics, 510 pp., J., 1, 16-26, 1963. John Wiley, New York, 1943. Mitchell, J. K., L. G. Bromwell, W. D. Carrier III, Trask, N.J., Size and spatial distribution of N. C. Costes, and R. F. Scott, Soil mechanics craters estimated from Ranger photographs: experiment, Apollo 14 Preliminary ScienceRe- Rangers 8 and 9, NASA Tech. Rep. 32-800, pp. port, NASA Spec. Publ. 272, chap. 4, pp. 87- 252-263, 1966. 108, 1971a. Vesi•, A. B., Bearing capacity of deep founda- Mitchell, J. K., W. N. Houston, T. S. Vinson, tions in sand, Rep. 39, pp. 112-153, Highway T. Durgunoglu, L. I. Namiq, J. B. Thompson, Res. Board, Nat. Res. Council, Washington, and D. D. Treadwell, Lunar surface engineer- D.C., 1963. ing properties and stabilization of lunar soils, report, NASA contract NAS 8-21432, 223 pp., Space Sci. Lab., Univ. of Calif., Berkeley, 1971b. (Received December 10, 1971; Namiq, L. I., Stress-deformationstudy of a simu- revised June 15, 1972.)