lable at ScienceDirect

Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397

Contents lists avai

Journal of Rock Mechanics andGeotechnical Engineering

journal homepage: www.rockgeotech.org

Full length article

Determination and applications of rock quality designation (RQD)

Lianyang Zhang*

Department of Civil Engineering and Engineering Mechanics, University of Arizona, Tucson, AZ, USA

a r t i c l e i n f o

Article history:Received 16 September 2015Received in revised form1 November 2015Accepted 14 November 2015Available online 9 March 2016

Keywords:Rock quality designation (RQD)Rock mass classificationDeformation modulusUnconfined compressive strengthEmpirical methods

* Tel.: þ1 520 626 0532.E-mail address: lyzhang@email.arizona.edu.Peer review under responsibility of Institute o

Chinese Academy of Sciences.

http://dx.doi.org/10.1016/j.jrmge.2015.11.0081674-7755 � 2016 Institute of Rock and Soil MechaniBY-NC-ND license (http://creativecommons.org/licens

a b s t r a c t

Characterization of rock masses and evaluation of their mechanical properties are important and chal-lenging tasks in rock mechanics and rock engineering. Since in many cases rock quality designation (RQD)is the only rock mass classification index available, this paper outlines the key aspects on determination ofRQD and evaluates the empirical methods based on RQD for determining the deformation modulus andunconfined compressive strength of rock masses. First, various methods for determining RQD are pre-sented and the effects of different factors on determination of RQD are highlighted. Then, the empiricalmethods based on RQD for determining the deformationmodulus and unconfined compressive strength ofrock masses are briefly reviewed. Finally, the empirical methods based on RQD are used to determine thedeformationmodulus and unconfined compressive strength of rock masses at five different sites including13 cases, and the results are compared with those obtained by other empirical methods based on rockmass classification indices such as rock mass rating (RMR), Q-system (Q) and geological strength index(GSI). It is shown that the empirical methods based on RQD tend to give deformationmodulus values closeto the lower bound (conservative) and unconfined compressive strength values in the middle of thecorresponding values from different empirical methods based on RMR, Q and GSI. The empirical methodsbased on RQD provide a convenient way for estimating the mechanical properties of rock masses but,whenever possible, they should be used together with other empirical methods based on RMR, Q and GSI.� 2016 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting byElsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/

licenses/by-nc-nd/4.0/).

1. Introduction

Natural rockmass differs frommost other engineeringmaterialsin that it contains discontinuities such as joints, bedding planes,folds, sheared zones and faults which render its structure discon-tinuity. To determine the engineering properties of rock masses, itis important to consider the effect of discontinuities. As Palmström(2002) noted: “The engineering properties of a rock mass oftendepend far more on the system of geological defects within the rockmass than on the strength of the (intact) rock itself. Thus, from anengineering point of view, a knowledge of the type and frequencyof the joints and fissures is often more important than the types ofrock involved. The observations and characterization of the jointsshould therefore be done carefully.”

Becauseof thediscontinuousnatureof rockmasses, it is importantto choose the right domain that is representative of the rock massaffectedby the structure analyzedwhendetermining theengineeringproperties (Fig.1). The behavior of the rockmass is dependent on the

f Rock and Soil Mechanics,

cs, Chinese Academy of Sciences. Pes/by-nc-nd/4.0/).

relative scale between the problem domain and the rock blocksformed by the discontinuities. When the structure is significantlylarger than the rock blocks formed by the discontinuities, the rockmass may be simply treated as an equivalent continuum for theanalysis (Brady and Brown, 1985; Brown, 1993; Hoek et al., 1995;Zhang, 2005). Treating the jointed rock mass as an equivalent con-tinuum (i.e. the equivalent continuum approach), different empiricalcorrelations have been proposed for estimating the engineeringproperties of jointed rock masses based on the classification indicessuch as rock quality designation (RQD) (Deere et al., 1967; Coon andMerritt, 1970; Serafim and Pereira, 1983; Zhang and Einstein, 2004;Zhang, 2010), rock mass rating (RMR) (Bieniawski, 1978; Serafimand Pereira, 1983; Yudhbir and Prinzl, 1983; Nicholson andBieniawski, 1990; Mitri et al., 1994; Sheorey, 1997; Aydan andDalgic, 1998), Q-system (Q) (Barton et al., 1980; Barton, 2002), andgeological strength index (GSI) (Hoek and Brown, 1997; Hoek, 2004;Gokceoglu et al., 2003; Hoek and Diederichs, 2006).

Since in many cases RQD is the only information available fordescribing rock discontinuities, this paper focuses on the determi-nation of RQD and its utilization for evaluating the engineeringproperties (mainly deformation modulus and unconfined compres-sive strength) of rock masses. First, different methods for

roduction and hosting by Elsevier B.V. This is an open access article under the CC

Fig. 1. Simplified representation of the influence of scale on the type of rock massbehavior (after Hoek et al., 1995).

Fig. 2. Procedure for determination of RQD using coring (after Deere, 1989).

L. Zhang / Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397390

determiningRQDare presented and the effects of different factors ondetermination of RQD are discussed. Then, different empiricalmethods based on RQD for estimating the deformationmodulus andunconfined compressive strength of rock masses are presented andbriefly discussed. Finally, these empirical methods are used todetermine the deformation modulus and unconfined compressivestrength of rock masses at five different sites including 13 cases, andthe results are compared with those using other empirical methodsbased on rockmass classification indices RMR, Q and GSI. This paperoutlines thekeyaspectsondeterminationofRQDandprovidesusefulinformation for effective evaluation of the deformationmodulus andunconfined compressive strength of rock masses based on RQD.

2. Determination of RQD

RQD was proposed by Deere (1964) as a measure of the qualityof borehole core. The RQD is defined as the ratio (in percentage) ofthe total length of sound core pieces that is 0.1 m (4 inch) or longerto the length of the core run. Besides the direct method for deter-mining RQD from coring, different indirect methods are alsoavailable for evaluating RQD.

For determination of RQD using core boring, the InternationalSociety for Rock Mechanics (ISRM) recommended a core size of atleast NX (size 54.7mm) drilledwith double-tube core barrel using adiamond bit. Artificial fractures can be identified by close fitting ofcores and unstained surfaces. All the artificial fractures should beignored while counting the core length for RQD. The correct pro-cedure for determining RQD from coring is shown in Fig. 2.

RQD can also be determined from discontinuity frequency ob-tained from scanline sampling. Correlations between RQD andlinear discontinuity frequency have been derived for differentdiscontinuity spacing distribution forms (Priest and Hudson (1976);Sen and Kazi, 1984; Sen, 1993). For example, for a negative expo-nential distribution of discontinuity spacings, Priest and Hudson(1976) derived the following relationship between RQD and lineardiscontinuity frequency l:

RQD ¼ 100e�ltðlt þ 1Þ (1)

where t is the length threshold. For t ¼ 0.1 m as for the conven-tionally defined RQD, Eq. (1) can be expressed as

RQD ¼ 100e�0:1lð0:1lþ 1Þ (2)

Fig. 3 shows the relations obtained by Priest and Hudson (1976)between measured values of RQD and l, and the values calculatedusing Eq. (2). For values of l in the range of 6e16 m�1, a goodapproximation to measured RQD values is found to be given by thefollowing linear relation:

RQD ¼ 110:4� 3:68l (3)

It is noted that Eq. (1) was derived based on the assumption thatthe length of the sampling line L is large so that the term e�lL isnegligible. For a short sampling line of length L, Sen and Kazi (1984)derived the following expression for RQDwith a length threshold t:

RQD ¼ 1001� e�lL � lLe�lL

he�lLðlt þ 1Þ � e�lLðlLþ 1Þ

i(4)

Fig. 4 shows the variation of RQD with the length of samplingline L for discontinuity frequency l ¼ 10 m�1 and length thresholdt ¼ 0.1 m. It can be seen that when L is smaller than about 0.5 m,RQD increases significantly as L increases. When L is larger than0.5 m, RQD changes slightly with L. Therefore, it is important to usesampling lines that are long enough so that e�L can be negligible.

Seismic velocity measurements have also been used to estimateRQD. By comparing the P-wave velocity of in situ rock mass withlaboratory P-wave velocity of intact drill core obtained from thesame rock mass, the RQD can be estimated by (Deere et al., 1967):

RQD ¼ �vpF

�vp0

�2 � 100% (5)

where vpF is the P-wave velocity of in situ rock mass, and vp0 is theP-wave velocity of the corresponding intact rock.

Fig. 3. Relationship between RQD and discontinuity frequency l (after Priest andHudson, 1976).

L. Zhang / Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397 391

Other empirical correlations similar to Eq. (5) were also pro-posed by researchers, including (El-Naqa, 1996; Bery and Saad,2012):

RQD ¼ 0:77�vpF

�vp0

�1:05 � 100% (6)

RQD ¼ 0:97�vpF

�vp0

�2 � 100% (7)

although Sjogren et al. (1979) and Palmström (1995) proposed thefollowing hyperbolic correlation between RQD and P-wavevelocities:

RQD ¼ vpq � vpFvpqvpFkq

� 100% (8)

where vpq is the P-wave velocity of a rock mass with RQD ¼ 0, andkq is a parameter taking into account the actual conditions of the insitu rockmass. Based on regression analysis of the data obtained for

Fig. 4. Variation of RQD with the length of sampling line L.

heavily fractured calcareous rock masses outcropping in southernItaly, Budetta et al. (2001) obtained as vpq ¼ 1:22 km/s andkq ¼ �0.69. Thus Eq. (8) yields:

RQD ¼ 1:22� vpF1:22vpFð � 0:69Þ � 100% (9)

It is noted that RQD varies with the direction of the borehole orsampling line. As an example, Fig. 5 shows the variation of esti-mated RQD by Choi and Park (2004) for a site in the west-southernpart of Korea on the lower hemisphere equal-angle stereo projec-tion net. The variation of RQD with direction can be clearly seen.Therefore, it is important to specify the corresponding directionwhen stating a RQD value.

The RQD can also be estimated using the correlation betweenRQD and volumetric discontinuity frequency lv (Palmström, 1974;ISRM, 1978):

RQD ¼ 115� 3:3lv (10)

where the volumetric discontinuity frequency lv is the sum of thenumber of discontinuities per unit length for all discontinuity sets,which can be determined from the discontinuity set spacingswithin a volume of rock mass as (Palmström, 1982):

lv ¼ 1s1

þ 1s2

þ 1s3

þ/ (11)

where s1, s2 and s3 are the mean discontinuity set spacings. Randomdiscontinuities in the rock mass can be considered by assuming arandom spacing sr for each of them. According to Palmström(2002), sr ¼ 5 m can be assumed. So the volumetric discontinuityfrequency lv can be generally expressed as

Fig. 5. Variation of estimated RQD (%) with scanline direction (after Choi and Park,2004).

L. Zhang / Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397392

lv ¼ 1s1

þ 1s2

þ 1s3

þ/þ Nr

5(12)

where Nr is the number of random discontinuities.The use of volumetric discontinuity frequency lv for estimating

RQD provides a quite useful way in reducing the directionaldependence of RQD. It is also possible to do core boring, scanlinesampling and/or wave velocity measurements at different di-rections and then evaluate the overall RQD of the rock mass.

3. Empirical methods based on RQD for estimating rock massproperties

3.1. Estimation of deformation modulus

Based on field studies at Dworshak Dam, Deere et al. (1967)suggested that RQD be used for determining the deformationmodulus of rock masses. By adding more data from other sites,Coon andMerritt (1970) developed a relation between RQD and themodulus ratio Em/Er as shown in Fig. 6, where Em and Er are thedeformation moduli of the rock mass and the intact rock,respectively.

Gardner (1987) proposed the following relation for estimatingEm from Er by using a reduction factor aE which accounts for thefrequency of discontinuities by RQD:

Em ¼ aEEraE ¼ 0:0231RQD� 1:32 � 0:15

�(13)

This method was adopted by the American Association of StateHighway and Transportation Officials in the Standard Specificationfor Highway Bridges (AASHTO, 1996). For RQD >57%, Eq. (13) is thesame as the relation of Coon and Merritt (1970), while forRQD < 57%, Eq. (13) gives Em/Er ¼ 0.15.

It is noted that the RQD-(Em/Er) relations by Coon and Merritt(1970) and Gardner (1987) have the following limitations (Zhangand Einstein, 2004):

Fig. 6. Variation of Em/Er with RQD proposed by Coon and Merritt (1970).

(1) The range of RQD < 60% is not covered or only an arbitraryvalue of Em/Er is selected for the whole range.

(2) For RQD ¼ 100%, Em is assumed to be equal to Er. This isobviously unsafe in design practice because RQD ¼ 100%does not mean that the rock is intact. There may be discon-tinuities in rock masses with RQD ¼ 100% and thus Em maybe smaller than Er even when RQD ¼ 100%.

Zhang and Einstein (2004) expanded the database shown inFig. 6 by collecting more data from the published literature (seeFig. 7). The expanded database covers the entire range(0 � RQD � 100%) and shows a nonlinear variation of Em/Er withRQD. The rocks for the expanded database include mudstone, silt-stone, sandstone, shale, dolerite, granite, limestone, greywacke,gneiss, and granite gneiss. Again, one can see the large scatter of thedata, especially when RQD >65%. Zhang and Einstein (2004) dis-cussed the possible reasons for the large scatter, including testmethods, directional effect, discontinuity conditions and insensi-tivity of RQD to discontinuity frequency (or spacing). Using theexpanded database, Zhang and Einstein (2004) derived thefollowing RQD-(Em/Er) relation for the average trend:

aE ¼ EmEr

¼ 100:0186RQD�1:91 (14)

The average RQD-(Em/Er) relation gives aE¼ 0.95 at RQD¼ 100%,which makes sense because there may be discontinuities in rockmasses at RQD ¼ 100% and thus Em may be smaller than Er evenwhen RQD ¼ 100%. By plotting the RQD-(Em/Er) relations by Coonand Merritt (1970) and Gardner (1987) also in Fig. 7, one canclearly see that it is not reasonable to assume a constant Em/Er valueat the low RQD region.

3.2. Estimation of unconfined compressive strength

Kulhawy and Goodman (1987) suggested that, as a firstapproximation, the unconfined compressive strength scm of rockmasses be taken as 0.33sc when RQD is less than about 70% and

Fig. 7. Expanded data and different relations between Em/Er and RQD (Ebisu et al.,1992).

L. Zhang / Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397 393

then linearly increase to 0.8sc when RQD increases from 70% to100% (see Fig. 8), where sc is the unconfined compressive strengthof the intact rock. The Standard Specifications for Highway Bridgesadopted by AASHTO (1996) suggested that scm be estimated usingthe following expression:

scm ¼ asscas ¼ 0:0231RQD� 1:32 � 0:15

�(15)

The variation of scm/sc with RQD based on Eq. (15) is also shownin Fig. 8. It can be seen that the general trend of these two relationsbetween scm/sc and RQD is about the same: scm/sc is constantwhen RQD is smaller than a certain value and then linearly in-creases when RQD increases. Obviously it is inappropriate to as-sume that scm/sc is constant when RQD varies from 0 to a certainvalue (70% for the relation of Kulhawy and Goodman (1987) and64% for the relation of AASHTO (1996)). For example, for a very poorquality rock mass (RQD < 25%) and a fair quality rock mass(RQD ¼ 50e75%), different scm/sc values should be expected.

It is also noted that the scm/sc versus RQD relation in Eq. (15) isthe same as the Em/Er versus RQD relation in Eq. (13), which maynot be appropriate. Researchers have studied the relation betweenscm/sc and Em/Er and found that they can be related approximatelyby the following equation (Ramamurthy, 1993; Singh et al., 1998;Singh and Rao, 2005):

scmsc

¼ as ¼�EmEr

�q

¼ aqE (16)

where the power q varies from 0.5 to 1 and is most likely in therange of 0.61e0.74 with an average of 0.7. It can be seen that theAASHTO method (Eq. (15)) uses the upper bound value of q ¼ 1.

It needs to be noted that the relation between scm/sc and Em/Er(Eq. (16)) is derived based only on triaxial test data on jointed rockmass specimens with different joint frequencies, orientations andconditions (Ramamurthy, 1993; Singh et al., 1998; Singh and Rao,2005) and has not been tested against field cases. The power q inEq. (16) may vary significantly for different rock types anddiscontinuity conditions. Nevertheless, using the average value of

Fig. 8. Comparison of scm/sc versus RQD relations by Kulhawy and Goodman (1987),AASHTO (1996), and Zhang (2010), respectively.

q ¼ 0.7 and the Em/Er versus RQD relation in Eq. (14), Zhang (2010)derived the following scm/sc versus RQD relation:

scmsc

¼ as ¼ 100:013RQD�1:34 (17)

Fig. 8 shows a comparison of the scm/sc versus RQD relation (Eq.(17)) with the suggestions, respectively by Kulhawy and Goodman(1987) and AASHTO (1996). Eq. (17) covers the entire range(0 � RQD � 100%) continuously. For RQD >70%, Eq. (17) is in goodagreement with the suggestions of Kulhawy and Goodman (1987)and AASHTO (1996). For RQD <70%, however, Eq. (17) is differentfrom the suggestions of Kulhawy and Goodman (1987) andAASHTO (1996), with Eq. (17) considering the continuous variationof scm/sc with RQDwhile the suggestions of Kulhawy and Goodman(1987) and AASHTO (1996) both assuming constant scm/sc values.

4. Comparative analysis and discussion

4.1. Comparative analysis

To evaluate the empirical methods by Zhang and Einstein(2004) and Zhang (2010) for estimating the deformationmodulus and unconfined compressive strength of rock massesusing RQD, they are applied to five sites with detailed geotech-nical information available: the Sulakyurt dam site in centralTurkey (Ozsan et al., 2007), the Tannur dam site in south Jordan(El-Naqa and Kuisi, 2002), the Urus dam site also in centralTurkey (Ozsan and Akin, 2002), a high tower site at Tenerife Is-land (Justo et al., 2006), and an open pit mine site in the vicinityof Berlin, Germany (Alber and Heiland, 2001). The results arecompared with those from other empirical methods based on thecommonly used rock mass classification indices RMR, Q and GSIin order to indirectly check the accuracy of the empiricalmethods based on RQD. Table 1 lists the properties of rocks at thefive sites which cover a reasonable but clearly limited range ofrock types. Tables 2 and 3 list the empirical methods based onRMR, Q and GSI for estimating the deformation modulus andunconfined compressive strength of rock masses, respectively. Itis noted that these empirical methods were developed based ondatabases of different sources and, as shown below, may givevery different estimation values.

Fig. 9 summarizes the estimated deformation modulus valuesfrom the empirical method based on RQD by Zhang and Einstein

Table 1Summary of rock properties at five sites (after Alber and Heiland, 2001; El-Naqa andKuisi, 2002; Ozsan and Akin, 2002; Justo et al., 2006; Ozsan et al., 2007).

No. Rock Er (GPa) sc (MPa) RQD (%) RMR Q GSI References

1 Granite 31.5a 74 8.5 24 0.08 19 Ozsan et al.(2007)2 Diorite 19.5a 60 1.5 21 0.05 16

3 Limestone (L1) 24.8a 31 54 57 4.23 52 El-Naqa andKuisi (2002)4 Limestone (L2) 10.4a 13 50 59 5.29 54

5 Limestone (R1) 29.6a 37 48 59 5.29 546 Limestone (R2) 21.6a 27 45 54 3.04 597 Marly limestone 22.4a 28 44 55 3.39 508 Andesite 41.9 93 41 34 0.56 41 Ozsan and

Akin (2002)9 Basalt 40 142 15 38 0.63 42.510 Tuff 11.6 24 10 21 0.11 3111 Basalt (d1) 60.9 69 77 59 6.6 52 Justo et al.

(2006)12 Basalt (d2) 5.3a 15 42.5 38 3.4 3913 Limestone 25.7 41 50 57 2.4b 52 Alber and

Heiland(2001)

a Estimated using the modulus ratio (MR ¼ Er/sc) values from Hoek andDiederichs (2006).

b Estimated from GSI using the correlation GSI ¼ 9lnQ þ 44.

Table 2Empirical relations based on RMR, Q and GSI for estimating deformationmodulus Emof rock masses.

Authors Relation EquationNo.

Bieniawski (1978) Em ¼ 2RMR� 100 ðRMR > 50Þ (18)

Serafim andPereira (1983)

Em ¼ 10RMR�10

40 ðRMR � 50Þ (19)

Nicholson andBieniawski (1990)

EmEr

¼ 0:0028RMR2þ0:9eRMR=22:82

100 (20)

Mitri et al. (1994) EmEr

¼ 1�cosðpRMR=100Þ2 (21)

Read et al. (1999) Em ¼ 0:1ðRMR=10Þ3 (22)

Barton (2002) Em ¼ 25 log10Q (23)

Barton (2002) Em ¼ 10ðQsc=100Þ1=3 (24)

Hoek et al. (2002) Em ¼ ð1� D=2Þffiffiffiffiffiffiffisc100

q10

GSI�1040 ðsc � 100 MPaÞ (25)

Gokceogluet al. (2003)

Em ¼ 0:0736e0:0755RMR (26)

Gokceogluet al. (2003)

Em ¼ 0:1451e0:0654GSI (27)

Sonmezet al. (2004)

EmEr

¼ s0:4a ; s ¼ eGSI�100

9 ; a ¼ 12 þ e�GSI=15�e�20=3

6 (28)

Hoek (2004) Em ¼ 0:33e0:064GSI (29)

Hoek andDiederichs (2006)

Em ¼ 100

1�D=21þeð75þ25D�GSIÞ=11

(30)

Hoek andDiederichs (2006)

EmEr

¼ 0:02þ 1�D=21þeð60þ15D�GSIÞ=11 (31)

Sonmez et al. (2006) EmEr

¼ 10ðRMR�100Þð100�RMRÞ=4000 expð�RMR=100Þ (32)

Note: The empirical relations before 1989 use RMR76 which is smaller than RMR89by 5. D is the disturbance factor.

L. Zhang / Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397394

(2004) and other empirical methods based on RMR, Q and GSI forall 13 cases at the five sites. It can be seen clearly that the estimatedvalues from the empirical method based on RQD are mostly at orslightly smaller than the low values (lower bound) of the estimatedvalues from the different empirical methods based on RMR, Q andGSI. So the empirical method based on RQD by Zhang and Einstein(2004) tends to give conservative estimation of the deformation

Table 3Empirical relations based on RMR, Q and GSI for estimating unconfined compressivestrength scm of rock masses.

Authors Relation EquationNo.

Yudhbir and Prinzl (1983) scmsc

¼ e7:65ðRMR�100Þ

100 (33)

Ramamurthy et al. (1985);Ramamurthy (1996)

scmsc

¼ eRMR�10018:75 (34)

Trueman (1988); Asef et al.(2000)

scm ¼ 0:5e0:06RMR ðMPaÞ (35)

Kalamaras and Bieniawski(1993)

scmsc

¼ eRMR�100

24 (36)

Hoek et al. (2002) scmsc

¼ eGSI�1009�3D

12þ1

6

�e�

GSI15�e�

203

�(37)

Bhasin and Grimstad (1996);Singh and Goel (1999)

scm ¼ 7gfcQ1=3 ðMPaÞ;where fc ¼ sc/100 for Q > 10and sc > 100 MPa, otherwisefc ¼ 1; and g is the unit weightof the rock mass in g/cm3

(38)

Sheorey (1997) scmsc

¼ eRMR�100

20 (39)

Aydan and Dalgic (1998) scmsc

¼ RMRRMRþ6ð100�RMRÞ (40)

Barton (2002) scm ¼ 5gðQsc=100Þ1=3 ðMPaÞ (41)

Hoek (2004) scmsc

¼ 0:036eGSI30 (42)

Singh et al. (1997) scm ¼ 7gQ1=3 ðMPaÞ (43)

modulus of rock masses compared to the different empiricalmethods based on RMR, Q and GSI.

Fig. 10 summarizes the estimated unconfined compressivestrength values from the empirical method based on RQD by Zhang(2010) and other empirical methods based on RMR, Q and GSI for all13 cases at the five sites. The estimated values from the empiricalmethod based on RQD are essentially in themiddle of the estimatedvalues from the different empirical methods based on RMR, Q andGSI.

4.2. Discussion

Determination of the deformation modulus and unconfinedcompressive strength of jointed rock masses is an important andchallenging task in rock mechanics and rock engineering. Theempirical methods based on RQD provide a convenient way forestimating the deformation modulus and unconfined compressivestrength of rock masses because, in many cases, RQD is the onlyavailable information about discontinuities in routine site in-vestigations. However, care should be taken when applying theempirical methods based on RQD for determining the deformationmodulus and unconfined compressive strength of rock massesbecause RQD is only one of the many factors that affect thedeformability and strength of jointed rock masses. Other factorssuch as discontinuity orientation and discontinuity surface condi-tions can also have a great effect on the deformability and strengthof jointed rock masses.

To apply the empirical methods based on RQD for determiningthe deformation modulus and unconfined compressive strength ofrock masses, the following aspects should be noted:

(1) When RQD is the only information available about rockdiscontinuities, the empirical methods based on RQD canbe used to estimate the rock mass deformation modulusand unconfined compressive strength but care should betaken when applying the estimated values. The empiricalmethods based on RQD should be used only for a firstestimation.

(2) When RQD and other information are available for deter-mining the rock mass classification indices RMR, Q and GSI,the empirical methods based on RQD should be usedtogether with the empirical methods based on RMR, Q andGSI to evaluate the rock mass deformation modulus andunconfined compressive strength. The estimated values fromthe empirical methods based on RQD can be compared withthe ranges of the estimated values from the empiricalmethods based on RMR, Q and GSI to get an idea on the effectof RQD on the deformability and strength of rock masses.

5. Conclusions

This paper reviewed the methods for determining RQD andevaluated the empirical methods based on RQD for estimating thedeformationmodulus and unconfined compressive strength of rockmasses. The conclusions are as follows:

(1) There are different methods for determining RQD. It isimportant to consider the effect of different factors such assampling length and direction on RQD when using a methodto determine RQD.

(2) The empirical method based on RQD by Zhang and Einstein(2004) for estimating deformation modulus of rock massestends to give low (conservative) values compared to thedifferent empirical methods based on RMR, Q and GSI.

Fig. 9. Estimated rock mass deformation modulus values from the existing empirical methods based on RMR, GSI or Q and the method based on RQD.

Fig. 10. Estimated rock mass strength values from the existing empirical methods based on RMR, GSI or Q and the method based on RQD.

L. Zhang / Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397 395

(3) The empirical method based on RQD by Zhang (2010) forestimating unconfined compressive strength of rockmasses tends to give values in the middle of the valuesfrom the different empirical methods based on RMR, Q andGSI.

(4) The empirical methods based on RQD provide a convenientway for estimating the deformationmodulus and unconfinedcompressive strength of rock masses but, whenever possible,they should be used together with other empirical methodsbased on RMR, Q and GSI because RQD is only one of the

L. Zhang / Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397396

many factors that affect the deformability and strength ofjointed rock masses.

Conflict of interest

The author wishes to confirm that there are no known conflictsof interest associated with this publication and there has been nosignificant financial support for this work that could have influ-enced its outcome.

References

Alber M, Heiland J. Investigation of a limestone pillar failure, part 1: geology, lab-oratory testing and numerical modeling. Rock Mechanics and Rock Engineering2001;34(3):167e86.

American Association of State Highway and Transportation Officials (AASHTO).Standard specifications for highway bridges. 16th ed. Washington DC: AASHTO;1996.

Asef MR, Reddish DJ, Lloyd PW. Rock-support interaction analysis based onnumerical modeling. Geotechnical and Geological Engineering 2000;18(1):23e37.

Aydan Ö, Dalgic S. Prediction of deformation behavior of 3-lanes Bolu tunnelsthrough squeezing rocks of North Anatolian fault zone (NAFZ). In: Proceedingsof Regional Symposium on Sedimentary Rock Engineering, Taipei, China; 1998.p. 228e33.

Barton N, Loset F, Lien R, Lunde J. Application of Q-system in design decisionsconcerning dimensions and appropriate support for underground installations.In: Bergman M, editor. Subsurface Space: Environmental Protection, Low CostStorage, Energy Savings. Proceedings of the International Symposium (Rock-store’80), vol. 2. Pergamon Press; 1980. p. 553e61.

Barton N. Some new Q-value correlations to assist in site characterization andtunnel design. International Journal of Rock Mechanics and Mining Sciences2002;39(2):185e216.

Bery AA, Saad R. Correlation of seismic P-wave velocities with engineering pa-rameters (N value and rock quality) for tropical environmental study. Interna-tional Journal of Geosciences 2012;3(4):749e57.

Bhasin R, Grimstad E. The use of stress-strength relationships in the assessment oftunnel stability. Tunnelling and Underground Space Technology 1996;11(1):93e8.

Bieniawski ZT. Determining rock mass deformability: experience from case his-tories. International Journal of Rock Mechanics and Mining Sciences & Geo-mechanics Abstracts 1978;15(5):237e47.

Brady BHG, Brown ET. Rock mechanics for underground mining. London, UK:George Allen and Unwin; 1985.

Brown ET. The nature and fundamentals of rock engineering. In: Hudson JA, editor.Compressive rock engineeringeprinciple, practice and projects, vol. 1. Oxford,UK: Pergamon Press; 1993. p. 1e23.

Budetta P, de Riso R, de Luca C. Correlations between jointing and seismic velocitiesin highly fractured rock masses. Bulletin of Engineering Geology and theEnvironment 2001;60(3):185e92.

Choi SY, Park HD. Variation of rock quality designation (RQD) with scanlineorientation and length: a case study in Korea. International Journal of RockMechanics and Mining Sciences 2004;41(2):207e21.

Coon RF, Merritt AH. Predicting in situ modulus of deformation using rock qualityindexes. In: Determination of the in situ modulus of deformation of rock. ASTMInternational; 1970. p. 154e73.

Deere DU, Hendron AJ, Patton FD, Cording EJ. Design of surface and near surfaceconstruction in rock. In: Proceedings of the 8th U.S. Symposium on Rock Me-chanicseFailure and Breakage of Rock. New York: American Institute of Mining,Metallurgical and Petroleum Engineers, Inc.; 1967. p. 237e302.

Deere DU. Rock quality designation (RQD) after twenty years. U.S. Army Corps ofEngineers Contract Report GL-89e1. Vicksburg, USA: Waterways ExperimentStation; 1989.

Deere DU. Technical description of rock cores for engineering purposes. Rock Me-chanics and Engineering Geology 1964;1(1):16e22.

Ebisu S, Aydan O, Komura S, Kawamoto T. Comparative study on various rockmass characterization methods for surface structures. In: Rock Character-ization: ISRM Symposium, Eurock ’92. London, UK: Thomas Telford; 1992.p. 203e8.

El-Naqa A, Kuisi MA. Engineering geological characterisation of the rockmasses at Tannur Dam site, South Jordan. Environmental Geology 2002;42(7):817e26.

El-Naqa A. Assessment of geotechnical characterization of a rock mass using aseismic geophysical technique. Geotechnical and Geological Engineering1996;14(4):291e305.

Gardner WS. Design of drilled piers in the Atlantic Piedmont. In: Smith RW,editor. Foundations and excavations in decomposed rock of the Piedmontprovince. New York, UK: American Society of Civil Engineers (ASCE); 1987.p. 62e86.

Gokceoglu C, Sonmez H, Kayabasi A. Predicting the deformation moduli of rockmasses. International Journal of Rock Mechanics and Mining Sciences2003;40(5):701e10.

Hoek E, Brown ET. Practical estimates of rock mass strength. International Journal ofRock Mechanics and Mining Sciences 1997;34(8):1165e86.

Hoek E, Carranza-Torres C, Corkum B. Hoek-Brown failure criterion e 2002 edition.In: Proceedings of the 5th North American Rock Mechanics Symposium & the17th Tunneling Association of Canada Conference (NARMS-TAC 2002). Toronto,Canada: Mining Innovation and Tech.; 2002. p. 267e73.

Hoek E, Diederichs MS. Empirical estimation of rock mass modulus. InternationalJournal of Rock Mechanics and Mining Sciences 2006;43(2):203e15.

Hoek E, Kaiser PK, Bawden WF. Support of underground excavations in hard rock.Rotterdam: A.A. Balkema; 1995.

Hoek E. Personal communication. Rocscience; 2004.ISRM. Suggested methods for the quantitative description of discontinuities in rock

masses. International Journal of Rock Mechanics and Mining Sciences & Geo-mechanics Abstracts 1978;15(6):319e68.

Justo JL, Justo E, Durand P, Azanon JM. The foundation of a 40-storey tower injointed basalt. International Journal of Rock Mechanics and Mining Sciences2006;43(2):267e81.

Kalamaras GS, Bieniawski ZT. A rock mass strength concept for coal seams. In:Proceedings of the 12th Conference on Ground Control in Mining, Morgantown;1993. p. 274e83.

Kulhawy FH, Goodman RE. Foundations in rock. In: Bell FG, editor. Ground Engi-neer’s reference book, chapter 15. London, UK: Butterworths; 1987.

Mitri HS, Edrissi R, Henning J. Finite element modeling of cablebolted stopes in hardrock ground mines. Presented at the SME annual meeting. Albuquerque, USA.1994. p. 94e116.

Nicholson GA, Bieniawski ZT. A nonlinear deformation modulus based on rock massclassification. International Journal of Mining and Geological Engineering1990;8(3):181e202.

Ozsan A, Akin M. Engineering geological assessment of the proposed Urus Dam,Turkey. Engineering Geology 2002;66(3e4):271e81.

Ozsan A, Ocal A, Akin M, Basarir H. Engineering geological appraisal of the Sula-kyurt dam site, Turkey. Bulletin of Engineering Geology and the Environment2007;66(4):483e92.

Palmström A. Characterization of jointing density and the quality of rock masses.Internal report. Norway: A.B. Berdal; 1974 (in Norwegian).

Palmström A. Measurement and characterization of rock mass jointing. In:Sharma VM, Saxena KR, editors. In-situ characterization of rocks. Lisse: A.A.Balkema; 2002. p. 49e98.

Palmström A. RMida rock mass characterization system for rock engineeringpurposes. PhD Thesis. Oslo, Norway: University of Oslo; 1995.

Palmström A. The volumetric joint countea useful and simple measure of the de-gree of rock mass jointing. In: Proceedings of the 4th Congress of InternationalAssociation of Engineering Geology, vol. 2; 1982. p. 221e8. New Delhi.

Priest SD, Hudson J. Discontinuity spacing in rock. International Journal of RockMechanics and Mining Sciences & Geomechanics Abstracts 1976;13(5):135e48.

Ramamurthy T, Rao GV, Rao KS. A strength criterion for rocks. In: Proceedings ofIndian Geotechnical Conference, vol. 1; 1985. p. 59e64. Roorkee.

Ramamurthy T. Stability of rock mass. Indian Geotechnical Journal 1996;16:1e73.Ramamurthy T. Strength and modulus response of anisotropic rocks. In: Hudson JA,

editor. Compressive Rock EngineeringePrinciple, Practice and Projects, vol. 1.Oxford, UK: Pergamon Press; 1993. p. 313e29.

Read SAL, Richards LR, PerrinND. Applicability of theHoekeBrown failure criterion toNew Zealand greywacke rocks. In: Vouille G, Berest P, editors. Proceedings of the9th International Congress on Rock Mechanics, vol. 2; 1999. p. 655e60. Paris.

Sen Z, Kazi A. Discontinuity spacing and RQD estimates from finite length scanlines.International Journal of Rock Mechanics and Mining Sciences & GeomechanicsAbstracts 1984;21(4):203e12.

Sen Z. RQD-fracture frequency chart based on a Weibull distribution. InternationalJournal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts1993;30(5):555e7.

Serafim JL, Pereira JP. Consideration of the geomechanical classification ofBieniawski. In: Proceedings of International Symposium on EngineeringGeology and Underground Construction, vol. 1. Rotterdam: A.A. Balkema;1983. II.3e44.

Sheorey PR. Empirical rock failure criteria. Rotterdam: A.A. Balkema; 1997.Singh B, Goel RK, Mehrotra VK, Garg SK, Allu MR. Effect of intermediate principal

stress on strength of anisotropic rock mass. Tunnelling and Underground SpaceTechnology 1998;13(1):71e9.

Singh B, Goel RK. Rock mass classificationsea practical approach in civil engi-neering. Amsterdam: Elsevier Ltd.; 1999.

Singh B, Villadkar MN, Samadhiya NK, Mehrotra VK. Rock mass strength parametersmobilized in tunnels. Tunnelling and Underground Space Technology1997;12(1):47e54.

Singh M, Rao KS. Empirical methods to estimate the strength of jointed rockmasses. Engineering Geology 2005;77(1e2):127e37.

Sjogren B, Øvsthus A, Sandberg J. Seismic classification of rock mass qualities.Geophysical Prospecting 1979;27(2):409e42.

Sonmez H, Gokceoglu C, Nefeslioglu HA, Kayabasi A. Estimation of rock modulus:for intact rocks with an artificial neural network and for rock masses with anew empirical equation. International Journal of Rock Mechanics and MiningSciences 2006;43(2):224e35.

L. Zhang / Journal of Rock Mechanics and Geotechnical Engineering 8 (2016) 389e397 397

Sonmez H, Gokceoglu C, Ulusay R. Indirect determination of the modulus ofdeformation of rock masses based on the GSI system. International Journal ofRock Mechanics and Mining Sciences 2004;41(5):849e57.

Trueman R. An evaluation of strata support techniques in dual life gateroads. PhDThesis. Cardiff, UK: University of Wales; 1988.

YudhbirWL, Prinzl F. An empirical failure criterion for rockmasses. In: Proceedings ofthe 5th International Congress onRockMechanics, vol.1; 1983. B1e8.Melbourne.

Zhang L, Einstein HH. Using RQD to estimate the deformation modulus of rockmasses. International Journal of Rock Mechanics and Mining Sciences2004;41(2):337e41.

Zhang L. Engineering properties of rocks. Amsterdam: Elsevier Ltd.; 2005.Zhang L. Estimating the strength of jointed rock masses. Rock Mechanics and Rock

Engineering 2010;43(4):391e402.

Dr. Lianyang Zhang is an associate professor and thedirector of the Sustainable Materials and RecyclingTechnologies (SMART) Laboratory at the University ofArizona, USA. His research interests cover rock mechanicsand rock engineering, deep foundations, geoenvironmentalengineering, sustainable construction materials, andrecycling and management of wastes. He is the author oftwo books and has published more than 100 technicalpapers.