8/13/2019 vol19n3-291-295

1/5

Ultrasonic Velocities and Related lastic Properties Hawaiian asaltic Rocks

MURLIH. M NGHN NI and GEORGE P. WOOLL RDL OR TORY ME SUREMENTS of the compressional-wave and shear-wave velocities of variousrock types give a useful basis for understandingthe geological significance of observed variationsin seismic-wave velocities associated with thecrust and upper mantle (B irch, 1961, 1964 .The rocks exposed at the surface in Hawaiiprovide excellent material for testing thevelocity-density relationships and the anisotrop effects in rocks believed to form thecrust and the upper mantle under the oceanfloor. These rocks exhibit wide ranges of porosity, density, and mineralogical composition, andtheir physical properties represent a number ofemplacement environments.

The Hawaii Institute of Geophysics programfor measuring .the -elasticproperties of the Ha>waiian rocks is confined as yet to normal pressure and temperature conditions, but it isplanned to expand the program to includestudies under high pressure and temperatureconditions to match those at depth down tothe level of the lower crust and the uppermantle.

The following paragraphs present the resultsof measurements of density and ultrasonicvelocities for some selected Hawaiian rockswhich are significant in the interpretation ofthe crustal and upper mantle structure fromseismic and gravity observations.

METHOD OFME SUREMENTThe pulse technique of measuring compressional-wave velocities described by Birch (1960)was the one employed in the present investiga

tion. Shear-wave velocities were measured usingthe technique described by Jamieson and Hoskins 1963 . Their method involves the use ofP-wave transducers, the conversion of the Pwave pulse into an S-wave pulse through a

1 Hawaii Institute of Geophysics Contribution No.88.

pyrex glass prism, transmittance of the S-wavethrough the sample, and reconversion of the Swave into a P-wave through a similar prism.Barium titanate transducers were employed inmaking all measurements. In addition to cores,samples cut into cuboids of suitable dimensions(2 X 2 X 2 inches) were used for measuringvelocities in three mutually perpendicular directions .

RESULTS ND DISCUSSIONThe elastic properties of some of the Hawaiian basaltic rocks that were studied are given

in Table 1. The rocks represented are tholeiiticand alkaline olivine basalts, trachyte, amphibolite, _ankaramite,dunite, and -eclogite. Basaltsrepresent the largest group of rocks studied. Themaximum compressional-wave velocity (7.0krri/sec ) under surface conditions is found inan amphiboliric intrusive dike rock having adensity of 3.0 gm /cc, This rock is found in theKoolau caldera. Another specimen (eclogite ) ,also from the Koolau caldera, has a rather lowdensity (2.81 grn/cc and velocity (6.2 km/sec),In hand specimen the eclogite appears to beporous ( .> 5 , and under the microscope itshows some alteration zones formed as a resultof reaction with the magma during its eruption.Without this contamination the eclogite wouldhave had a density and velocity equivalent tothat found for the amphibolitic rock of theKoolau caldera. The significance of high density(3 .0 grn/cc or higher) and high velocity (7.0krn/sec or higher ) values is evident in thatthese materials truly represent, in terms of seismic velocities, the constituents of mantle-likematerial in this caldera. An ultrabasic rock witha surface density of 3.0 gm/cc and velocity of7.0 km/sec, when placed in the upper mantleenvironment under the oceans (approximatelya depth of 10-12 km , would have a velocityof 7.5-7.8 krn/sec ,

291

8/13/2019 vol19n3-291-295

2/5

292 PAO FIC SCIENCE, Vol. XIX, July 1965TABLE 1

ELASTIC PROPERTIES OF SOME HAWAIIAN BASALTS

J.MODULUS F.BULK OF YOUNG S

E MODULUS RIGIDITY ) MODULUS POIS- V. DENSITY) dyn es/em dynes/em dynes /em K SON SSAMPLE km /see km /see g ee X 10-11 X 10-11 X 10-11 RATIO)J.-

Olivin e basalt 4.95 2.56 2.0 3.15 1.31 3.45 2.40 0.3174.634.82Olivine basalt 4.65 2.50 2.30 3.05 1.47 3.80 2.07 0.292Olivine basalt 5.654.38 3.10 2.36 4.03 2.27 5.23 1 78 0.2645.47Olivine basalt 5.08 3.02 2.40 4.6 2.18 5.56 2.11 0.296

ankaramite) Olivine basalt 5.52 2.76 2.60 5.27 1.98 5.27 2.63 0.330

6.06-- --_ -2.94- - --2.81 6 29 - -2.43 - - - 6.45 _ 0.3285.82 - - --- 2.59 --5.86Amphibolite 6.90

6.75 3.53 2.95 8.5 3.67 9.63 2.32 0.3126.76Hawaiite 4.20 2.51 2.59 2.4 1.63 4.0 1.48 0.224Trachyte 5.18 2.83 2.60 4.22 2.08 5.4 2.15 0.298

The three values of V p are for transm ission in th ree mutually perpendicular d irections of prop agat ion through the samespe imen

Figure 1 shows the velocity-density relationof the Hawaiian basalts and other volcanicrocks. These are primarily olivine basalts withvarying amounts of olivine and othe r ferromagnesian silicates. The velocities of the olivinebasalts vary from 4.5-6.5 krn/sec with corresponding changes in densities from 2.2-2.8gm / cc. As will be seen, the values are largelydependent on vesicular structure, the amountof interstitial glass present, and the am ount ofolivine present.Figure 2 shows the variation in ultrasonicvelocities in four olivine basalts of comparabledensity but having different percentages of olivine. The inclusions of dunites and olivine -richrocks in some of the volcanic flows on the islandof Hawaii, especially in the 1801 flow, are not

uncommon, and have been described by Rosser al. 1954). These rocks, composed principally of olivine, give velocities of 6.5-7.2krn/sec density 2.8-3.1 grn/cc under atmospheric conditions of temperature and pressure.Of all the Ha waiian rocks whose transmissio nvelocities have been measured including eclogite ), the dunites and olivine-rich inclusionsgive the highest values. The measured seismicvelocities of the upper mantle under the Ha waiian Archipelago Furumoto and Woollard,p. 315 in this issue; Eaton, 1962) are of theorder of 8.2-9.0 krrr /sec . Eatons deduced velocity distribution under a typical volcano is basedupon local earthquake data on Hawaii. The values reported by Furumoto et al. are based up onexplosion seismic refraction measurements. In-

8/13/2019 vol19n3-291-295

3/5

Ultrasonic Velocities-MANGHNANI and WOO RD 9

8r - ... r -

FIG. 1. Velocity versus density relationship in Ha waiian rocks. 1 and 2, amphibolitic rocks; 3, eclogite ;4, hawaiite; 5, ankaramite; 6, vesicular basalt; 7, py.roxen ite; 8, tholeiite.

asmuch as most of the magm as connected withHawaiian volcanic eruptions are believed tooriginate at the depths of earthquake foci ator about 60 km, the seismic velocities belowthe Moho are important from the poi nt of viewof the chemical composition of the upper mantle. Although it is not the pur pose of this r epor tto discuss the pros and cons of various modelsof the upper mantle , it can be said that thepresence of olivine-rich nodules in some ofthe Hawaiian flows, the increasing transmissionvelocity of the olivine basalts ankaramitewith increasing olivine content see also Birch,1961 ), and the seismic velocities of the layerbelow the Moho discontinuity 8.2-9.0 krn/sec)all indi cate that the upper mantle below theHawaiian Arch ipelago is most probably composed of material resembling dunite. Theplanned high pressure and temperature measurements of velocities of the dunires and olivinerich basalts at the Hawaii Institut e of Geo-

150hO 5 - - - 90 tOOVO LUMETRICl

: 6]

FIG. 2. Velocity versus olivine content in basalts.

I

physics should clarify some of the unsolvedqu estions.Amphibolitic rocks Fig. 1, 1 and 2) andan eclogite Fig. 1, 3 ) having higher densitiesthan basalt show higher velocity values, asstated earlier. Although the curves of Figure 1have been d r a w n _ L a p p . ~ n . t- --porosity ,,alues-as- deduced from wet and drydensity determinations, these are obviously notall true porosities and permeability is a parameter that is also incorporated . The effect of thisfactor will require further study. In gross form,though, the velocity-dens ity relationships shownfor volcanic rocks are essentiallg-correcr. Asm igh t be expected, the velocity in a rock is significantly lowered when the porosity increases.Th e dispersion in the three curves can be attributed to the differences in mineral composition, crystallinity of the material constitutingthe rock, the size and geometry of vesiclespresent, and the undetermined effect of perm eability due to micro fractures.Horizontal variations in seismic velocities ofthe upper layers of the basaltic lavas on theHawaiian Islands, as observed by the HawaiiInstitu te of Geophysics group Furumoto andWoollard, p. 315 in this issue) and others, arereal and thus signify variations not only inmineralogical composition and density but invesicularity and permeability as well. Problemspertaining to the degree of vesicularity to beexpected at depth can best be understood bystudying the variation in seismic velocity inmaterial of variable vesicularity and density.The ideal approach to the solution of this pro b-

320.8. 6.4BULK DENSI TY ljIm ce

B 3- 10.,.C > 10

POROSITYA

8/13/2019 vol19n3-291-295

4/5

294 PACIFIC SCIENCE, Vol. XIX, July 1965

3.0A-7- 14

A- 12 _. ...9;12 .....1 7

.... 10-13A lO1 6

.0 2.0

OENSITY

eo00 5 . 6Glass .

30.Ob20 - ,;-- :tn-- ; -;c-- -; :, - .: - ---:FIG. 3. Velocity versus density in cores from AlaeCrater, Hawaii.. lem would bet o drill through lava material andobtain the cores to examine vesicularity, density,chemical composition, and other physical properties. Figure 3 shows the physical measurements on some drill cores obtained from AlaeCrater, island of Hawaii, by the U. S. Geological Survey Volcano Observatory scientists. AI-- t hough- t here is-a---direcT r ela(ionship-between the velocity and density of these cores there issignificant variation in the velocities of cores(Fig, 3, A 9 15 AlO 16 AlO l3 A 9 12 A-11 7 A 7 12 etc.) with little or no change indensity (2 .6-2.7 gm /cc ) . The chemical composition of these cores is about the same (DallasPeck, personal communication ), but there is alarge variation in the glass content in the rockmateri al. Because the modulus of rigidity ofglass is much lower than the average value ofequivalent crystalline rock material, the velocity of propagation of elastic waves will be lowered as the percentage of glass increases. Therelationship between velocity and glass contentfor these cores is shown in Figure 4. The lowvelocities of the Pacific Ocean upper crustallayer (3 .8-4.2 krn/sec ) reported by Raitt( 1956 ), and by Shor and Pollard 1964), andverified by the unpublished studies by WesternGeophysical Company, suggest that the glasscontent may be influencing the seismic velocityto a large extent. Certainly, the work by Moore of the U. S. Geological Survey (personal communication) indicates essentiallyzero porosity for basalts extruded on the oceanfloor. Quick chilling could produce interstitialglass, however.

FIG. 4 . Velocity versus glass content in cores fromAlae Crater, Hawaii.ANISOTROPY

There is a significant degree of anisotropyin the transmission of compressional waves in. Hawaiian rocks. This is shown in Table 1 andFigure.1 , _where_velocities, obtained .in _threeperpendicular directions through the same sample are recorded. In some rock types there isas much as 8-10 difference between themaximum and minimum velocities observed.Although anisorropism in vesicular lavas is related to dimensional orientation of the vesicles,in whole rock material it is related principallyto mineral orientation. Both olivine and plagioclase feldspars are of special interest in the interpretation of seismic refraction data. Th is portion of the study now underway at the Instituteof Geophysics is still in its initial stage.

SUMMARY1. Density, rock structure, porosity, permea

bility, and glass content control the seismicvelocities of the vesicular basaltic lavas.2. Density, glass content, and mineralogical

compositi on of the non-vesicular flows and intrusives control their seismic velocities and related elastic properties.3. The alkalic basalts (e.g ., trachyric type)have a low velocity of transmission as comparedwith the tholeiitic olivine basalts.4. The average value of Poissons ratio for

Hawaiian basalts is 0.29.

8/13/2019 vol19n3-291-295

5/5

Ultrasonic Velocities-MANGHNANI and WOOLLARD 2951961. The velocity of compressional

waves in rocks to 10 kilobars, 2. Geoph.Res. 66:2199-2224.

1964. Density and composition of mantle and core. Geoph. Res. 69:4377-4388.EATON J . P: 1962. Crustal s tructure and vol

canism in Hawaii. In: Crust of the Pacific[Basin, Am. Geoph. Union GeophysicalMonogr, 6,pp . 13-29.

JAMIESON J. C, and H. HOSKINS 1963. Themeasurernenr of shear-wave velocities in solids Using axially polarized transducers. Geophysics 28:87-90.

RAITT, R. W. 1956. Seismic-refraction studiesof the Pacific Ocean Basin, 1. Crustal thickness of the Central Equatorial Pacific. Bull,Geo Soc. Am. 67:1623-1640.

Ross, C S., M. D. FOSTER and A. MYERS1954. Origin of dunites and of olivine-rich-- - inclusions iii--biisaliic rocks. --Am.-Mln. - 693-737.EFERENCES

ACKNOWLEDGMENTSThe studies reported were supported in partby the Office of Naval Research under Contract

ONR 3748 05 , and were greatly assisted bythe cooperation of the U. S. Geological SurveyVolcano Observatory group, and by Dr. LBarnes and Dr. G. A. Macdonald, of the HawaiiInsti tute of Geophysics, who contr ibuted samples of rock for the study. We thank Drs. D. LPeck and H. A. Powers of the Hawaiian Volcano Observatory for their kind help and forproviding some of the samples studied.

5. Dunites and olivine-rich inclusions in theHawaiian basalts have the highest transmissionvelocities observed in this study. This materialis most probably derived from the upper mantle.6. The Hawaiian rocks exhibit significant degrees of anisotropy due to the differences in themineral composition and orientation, and thegeometry and orientation of the vesicles.

BIRCH F. 1960. The velocity of compressionalwaves in rocks to 10 kilobars, 1. J. Geoph.Res. 65: 1083-1102.

SHOR , G. G., and D. D. POLLARD 1964. Mohole site selection studies north of Maui, Geoph. Res. 69:1627-1637.