Some considerations on shallow seismic reflection …directory.umm.ac.id/Data Elmu/jurnal/J-a/Journal Of Applied... · Some considerations on shallow seismic reflection surveysq ...

Ž .Journal of Applied Geophysics 45 2000 127–139www.elsevier.nlrlocaterjappgeo

Some considerations on shallow seismic reflection surveysq

M. Feroci a, L. Orlando a,), R. Balia b, C. Bosman a, E. Cardarelli a, G. Deidda b

a Dip. Idraulica, Trasporti e Strade-UniÕersita ’La Sapienza’ di Roma, Area Geofisica, Via Eudossiana 18,`00184 Roma, Italy

b Dip. Georisorse e Territorio-UniÕersita di Cagliari, Cagliari, Italy`

Received 4 June 1998; accepted 26 May 2000

Abstract

High-resolution shallow seismic reflection surveys require more attention to the choice of source and configuration,receivers and recording geometry for optimizing data acquisition than conventional oil exploration surveys. Moreover, some

Ž . Žstandard processing techniques to increase signalrnoise SrN ratio need special accuracy for example, surgically precise.removal of early-time coherent noise and iterative, small time shift static corrections . This paper compares results obtained

using different sources at two test sites: explosive, cap, shotgun, hammer and weight drop. Data from experiments usinggeophones with different natural frequencies and using various acquisition geometries are also compared. In data processing,it is demonstrated how increasing the SrN ratio for high-resolution results requires special consideration in some common

Ž .processing steps F–K filter, first arrivals muting, elimination of air wave and static corrections . The comparison, based onshot gathers and stack sections, shows that attenuation of high frequencies by the earth is the most significant influence onthe spectral properties of the data, as expected the source itself also does have some influence on frequency content,depending to some extent on surface conditions. The high-velocity explosive sources produced the highest frequencyreflections and best SrN ratio, because they have higher energy related to higher burnrblast velocity and source

Žcontainment then the other sources and they are used in hole i.e. below ground surface where the air wave energy is more.attenuated but the shotgun also an explosive source was reasonably comparable to high explosive when used in hole.

Special care must be taken during processing to insure artifacts are distinguished from real reflection events. q 2000 ElsevierScience B.V. All rights reserved.

Keywords: Shallow seismic reflection; Sources; Processing

q This report is mostly from the PhD research of M. Feroci, whose contribution is the largest. The authors R. Balia andG.P. Deidda gave useful suggestions on data acquisition and worked on the field for the site of Cagliari. The other authorsgave their contributions in the entire project but particularly E. Cardarelli for data acquisition, C. Bosman for processing andL. Orlando for both. Manuscript preparation and editing are by M. Feroci and L. Orlando.

) Corresponding author. Fax: q39-6-445-850-80.Ž .E-mail address: [email protected] L. Orlando .

0926-9851r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.Ž .PII: S0926-9851 00 00024-0

( )M. Feroci et al.rJournal of Applied Geophysics 45 2000 127–139128

1. Introduction

In the last 20 years, the growing interest inengineering and environmental problems has in-creased the use of seismic reflection surveys in

Žthe study of shallow targets hydrogeological,engineering, environmental, archaeological, and

.geotechnical problems . The most importantconsideration connected with these methods isrecording reflections with broad bandwidthŽ .spectra shifted towards high frequency and to

Žattenuate as much coherent noise air wave and.ground roll as possible. To obtain that, it is

necessary to choose carefully the sources, geo-phones, geometry of acquisition, processing toapply to the data etc. Shallow seismic reflectionsurveys should not be considered routine, butone requiring special equipment and parametersfor each site and target. Consequently, manyauthors have concentrated their studies on theproblems connected with the method. In particu-

Ž . Ž .lar, Hunter et al. 1982 , Hunter et al. 1984 ,Ž .Pullan and Hunter 1990 , and Steeples and

Ž .Miller 1990 focus on data collection tech-niques designed to optimize shallow reflections.

Ž .Knapp and Steeples 1986 discuss instrumenta-tion issues, i.e. as the dynamic range must be

Ž .high 16–18 bit to record the low energy ofreflection signal and as the importance to applythe analog filters before ArD signal conversion

Ž .for lower dynamic range. Widess 1973, 1982 ,Ž . Ž .Kalweit and Wood 1982 and Knapp 1990

refers to vertical resolution, as it depends onbandwidth, on the frequency content and on thephase of the signal. The high-resolution goal inshallow seismic surveys puts special require-

Žments on the choice of source to use Singh,1984; McCann et al., 1985; Miller et al., 1986;

.Pullan and MacAulay, 1987; Miller et al., 1992 .In fact, it is not possible to record high-frequency

Ž .data )80 Hz if the source does not generateand propagate high frequencies. Furthermore, ithas been pointed out that the source affects notonly the frequency content of the record, butalso the quantity of energy generated and, above

Ž .all, the signalrnoise SrN ratio. Miller et al.

Ž .1986, 1992, 1994 have made field compar-isons of various sources placed in sites withdifferent geology and have come to the conclu-sion that the quality of recorded data dependsgreatly on the depth of the water table and onnear surface geology. With their experiments,they have demonstrated that filling the shot holewith water allows a higher SrN ratio in therecords due to containment and improved cou-

Ž .pling. Pullan and MacAulay 1987 observedthat the source is influenced from the soil.

Ž .Meekes et al. 1990 refer as the superficialsources produce stronger air wave and groundroll compared to the hole-source. Experiments

Žconducted with high explosives Ziolkowski and.Lerwill, 1979 demonstrated that the resolution

decreases as the energy of the source increases.The question of choosing a source is still criticalsince it is not always possible to use an invasive

Ž .source shot holes : because of location in popu-lated areas with utility and contamination issuesand because shot holes are difficult and costlyto install. Therefore, continued experimentationsin different geologicrhydrologic settings canprovide us with a broader experience base tohelp determine the optimum source and config-uration.

The importance of the geophones and geo-phone plants is also to be taken into account as

Ž .pointed out by Palmer 1987 and Maxwell etŽ .al. 1994 .It is clearly possible to increase the SrN

ratio through data processing. But, as SteeplesŽ .and Miller 1990 have pointed out, in high-res-

olution seismic prospecting, some standard pro-cessing operations require special attention. Forinstance, static corrections must be very care-fully determined, since comparably small timeshifts can lead to greater than 1r2 l phaseshifts in high-frequency data and misalignmentduring stack can severely affect the resolutionof the final section. Refracted events can beinterpreted as reflections on the stack sectionunless they are correctly identified and carefullyeliminated in the initial processing phase. Theapplication of filters also requires great atten-

( )M. Feroci et al.rJournal of Applied Geophysics 45 2000 127–139 129

tion: if not eliminated or carefully trackedthrough the processing coherent noise can bealiased or aligned leading to apparent coherency

Žin the stacked section Steeples and Miller,. Ž .1990 . Steeples and Miller 1990 also point out

the importance of an accurate velocity modelsince it can vary rapidly in the horizontal andvertical direction in the shallow subsurface.

In consideration of the above, this paper ana-lyzes the possibility of increasing the SrN ratioinitially during data collection, and later duringprocessing. The results obtained from experi-ments with unique, mostly easy-to-use engineer-

Ž .ing low energy sources, in geological situa-tions where the water table is less then 3 m, arereported here. The results have been analyzedby qualitatively comparing the records obtainedwith the various sources.

2. Data collection

The experiments were conducted at two siteswith different lithological situations.

Ž .The first site, situated near Cagliari Italy , isa dry lake-basin where the surface soil is

clayey–silt with sandy intercalation, and sub-horizontal layering. The detailed stratigraphy ofthe site encompasses lacustrine clay and siltŽ . Žfrom 0 down to 6–8 m , loose sands between

.6–8 and %20 m , intercalation of sandstonesŽ .and marl between 20 and %50 and finally

Ž .marl 50–70 m . These formations rest onMiocene sandstone bedrock.

The following sources were used at this site:Ž .GEL-1 35 g explosive, seismic cap only,

Ž . wMinibang shotgun 8-gauge like the BetsyŽ .xseisgun described by Miller et al. 1986 , ham-

Ž .mer 7 kg used on a steel plate, weight dropŽ . Ž .Dynasource Miller et al. 1986 . To enhancethe high frequency for all these sources, the datawere recorded with 100-Hz geophones along a140-m profile using the same type of off-endgeometry, geophone interval 2 m, minimumoffset 6 m, maximum offset 52 m. Some datawere also recorded with geophone interval 0.5

Žm using the Minibang minimum offset 1.5 m,.maximum offset 13 m . All profiles have 1200%

coverage.The most important noise problem encoun-

tered using surface sources was the dominantŽ .presence of the air wave Figs. 3 and 5 , which

Ž .Fig. 1. a Shot gather from the Cagliari site. The data were collected using explosive source and 100-Hz geophones, offendŽ . Ž .geometry minimum offset 6 m, maximum offset 52 m and 2-m geophone interval. Gain trace balance is applied forŽ . Ž . Ž . Ž .displaying. b Frequency spectra of shot of a related to 1–8 traces indicated with 1 , 9–16 traces indicated with 2 andŽ .17–24 traces indicated with 3 . The spectra show absolute value in linear scale.


Ž . Ž .Fig. 2. a Shot gather from the same site of Fig. 1, collected using cap source. b Frequency spectra as in Fig. 1.

has high energy and spectra overlapping thereflection spectra. Therefore, some experimentswere conducted to find ways to minimize airwaves during the recording phase. In particular,the Minibang shotgun used had its plate modi-fied by the authors, so its barrel and a small partof the plate itself is buried in a 60-cm shot hole.To attenuate the air wave, experiments werealso conducted using an array of six 100-Hzin-line geophones. The pattern was chosen fol-

lowing the formula suggested by Verna and RoyŽ .1970 , using an air wave velocity of 340 mrs.Records were obtained using the Minibangsource and the same geometry as the other testrecords.

The second site, located near the FiumicinoŽ .Airport Rome, Italy has as target the deltaic

series, the first 100 m of which are character-ized by sub-horizontal layering. The experi-ments at this site were all conducted along the

Ž . Ž .Fig. 3. a Shot gather from the same site of Fig. 1, collected using Minibang shotgun source. b Frequency spectra as inFig. 1.


Ž . Ž . Ž .Fig. 4. a Shot gather from the same site of Fig. 1, collected using weight drop Dynasource source. b Frequency spectraas in Fig. 1.

same 140-m profile with 100-Hz geophones anddifferent types of hammers, namely two ironhammers of 7 and 0.8 kg, respectively, and a

Ž .wooden hammer of 4 kg 25=14 cm . The firsthammer was used on steel plate and the last twohammers were used on wooden plate. Off-endgeometry with minimum offset 3 m and maxi-mum offset 26 m were used with geophoneinterval 1 m. To define the influence of geo-phone interval on the record, data were col-

lected with the Minibang and the 7 kg hammerwith geophone interval 2 m, minimum offset 6m, and maximum offset 52 m. In all profiles,fold is 12. No high explosives were allowed atthis site.

A Geometrics ES2401 24-channel, 15-bitŽ .ArD, Instantaneous Floating Point IFP con-

version seismograph with a dynamic range of114 dB was used at both sites. All shots wererecorded with a sample interval of 0.2 ms and a

Ž . Ž .Fig. 5. a Shot gather from the same site of Fig. 1, collected using hammer source. b Frequency spectra as in Fig. 1.

()

M.F

erocietal.r

JournalofA

ppliedG

eophysics45

2000127

–139

132

Ž . Ž .Fig. 6. F–K cumulative spectra of 25 shots acquired with Minibang buried 60 cm below the surface a and Minibang used normally b .


record length of 409 ms with no analog filtersapplied. To make static corrections and deter-mine the interval velocity of the shallow layers,a refraction profile was also recorded at both

Žsites. For processing the program Seistrix 3 by.Interpex was used.

3. Records analysis

Records for each site were analyzed by com-paring the frequency spectra obtained with thedifferent sources. The results from each site areanalyzed separately.

Ž . Ž . Ž .Fig. 7. Shot gathers acquired along the same line using single-geophone a and six-geophone array 100 Hz b traces 1–6Ž . Ž .connected to single geophones and traces 7–12 connected to strings 50-cm geophone spacing . c Frequency spectra of

Ž . Ž . Ž .geophones 7–12 of a . d Frequency spectra of geophones 7–12 of b . The source was Minibang and the channel spacingwas 2 m.


3.1. First test site: Cagliari

Figs. 1a–5a show, as an example, a shotgather for each source after gain applicationŽ .trace balance . Some differences, based onqualitative observations can be pointed out: the

Ž .record for the explosive Fig. 1a has a goodreflection at 45 ms, compared to the others,probably because of its greater SrN ratio. If weconsider a mean velocity of 2000 mrs, it couldidentify the limit between intercalation of sand-stone and marl and solid marl, located at depthof 50–70 m. There is, however, a high ampli-

Žtude ground roll. The cap record, instead see.spectra in Fig. 2a , is richer in high-frequency

content, allowing better detection of the twoevents immediately following the first arrivalŽ .between 35 and 50 ms ; however, the SrN

Ž .ratio based on continuity of reflections is lowerin some parts.

Ž . ŽIn the shotgun Fig. 3 and dynasource Fig..4 records, there is a strong disturbance from

Ž .the air wave. The hammer shot gather Fig. 5has low coherent signals compared with theother sources. It can be noticed that low energysources, such as cap, minibang and hammer, atshort offsets, refracted and air waves dominatedand superimposed to reflected wave.

A frequency analysis has been performedŽcombining the spectra of traces 1–8 closer to

. Ž .the shot point , 9–16 mid range and 17–24Ž .long offset . Figs. 1b–5b show the spectra forthe various sources. For all sources, most of theenergy of the traces closest to the shot point isconcentrated at frequencies of about 50 Hz andcan be attributed to the ground roll. The peak inthe long-offset traces is around 120 Hz and canbe attributed to both the air and reflection waves;the air wave is less evident in the case of theexplosive, but very strong with both the shotgunand the hammer. At frequencies above 100–150Hz, all sources show low energy. It can also benoted that for all sources, except the cap, eventhe traces that are farthest from the shot pointhave a remarkably low frequency content com-pared to the mid range traces. This analysisshows that most of the energy is attributable to

Ž .the coherent noise ground roll and air wave .For the sources used in the experiments and

the type of soil at the sites, frequency contentabove 200 Hz is negligible. Whereas it is possi-ble to eliminate most of the ground roll throughfiltering, because most of its spectrum is below80 Hz, it is not possible to filter the air wavebecause of its spectral overlap with reflectionsremaining strong, especially for the Minibang

Ž . Ž . Ž .Fig. 8. Shot gathers collected in Fiumicino Airport with different types of hammers: a iron hammer weight 7 kg , bŽ . Ž . Ž .small iron hammer weight 0.8 kg and c wooden hammer weight 4 kg . Geophone interval 1 m.


and Dynasource records. Where records are col-lected using a geophone interval of 2 m, the airwave is also spatially aliased in F–K domainŽ .Fig. 6b , making it difficult to eliminate it from

the shot gathers using velocity filters easilyproduces artifacts in the alias filtered record.

Fig. 6 compares the F–K spectra of the shotsŽ .by the Minibang buried a and used in normal

Fig. 9. Amplitude spectra of the reflected signal of shot gathers in Fig. 8, after it was chosen with a window on the record:Ž . Ž . Ž . Ž .a average amplitude spectra of big iron, small iron and wooden hammers, b iron hammer weight 7 kg , c small iron

Ž . Ž . Ž .hammer weight 0.8 kg and d wooden hammer weight 4 kg . The spectra show relative values in dB.


Ž .setup b . Note that the aliased air wave, whichŽ .is very strong in spectrum b , is considerably

Ž .reduced in spectrum a . Therefore, the use ofŽ .the buried Minibang modified allows a consid-

erable improvement in the quality of the recordsfor the attenuation of the air wave as pointed

Ž .out by Miller et al. 1994 .We tried to attenuate the air wave during

acquisition with the use of six-geophone arraysŽ .arranged according to Verna and Roy 1970 ,

taking into consideration that the air wave has avery wide frequency spectrum, with dominantfrequencies between 100 and 350 Hz and avelocity of 340 mrs. On this basis, the fieldrecords were collected along three coincidinglines of 12 shots each, using in-line strings ofsix 100-Hz geophones spaced 40, 50 and 60 cm.The shot gathers, shown in Fig. 7b, had traces1–6 connected to single geophones and traces7–12 connected to strings of geophones eachspaced 50 cm. For comparison, Fig. 7 shows

Ž .one record obtained with single-geophones aŽand one record obtained with strings b — only

.traces 7–12 connected to strings . The use ofstrings considerably attenuates specific fre-quency components of air wave, but no signifi-cant difference was noted in using three differ-

ent distances. As a further effect, there is aconsiderable attenuation of the ground roll,linked to the fact that the used pattern attenuates

Ž .low frequencies Fig. 7c and d in events with avelocity of 200–250 mrs — such as 50–70 Hz,the dominant frequencies of the ground rollfiltered by our geophones.

3.2. Second test site: Fiumicino

Fig. 8 illustrates the records obtained at thesite near Rome, Fiumicino Airport, with varioustypes of hammers. Note that also here groundroll and air wave dominate the records. Identifi-able reflections can only be seen within the first70 ms of Fig. 8a. Fig. 9 shows the frequencycontent of the traces 17–24 within windows as

Žshown in the figure no coherent noises as air.wave and ground roll were included . The maxi-

mum frequencies recorded are between 300 and350 Hz. It can be noted that all three sourceshave dominant frequency of reflection around150 Hz and records collected with the 0.8-kgsteel hammer have a lower energy in the high-frequency band.

The comparison between the records ob-tained with the shotgun and the iron hammer at

Ž . Ž . Ž .Fig. 10. Shot gathers collected in Fiumicino site : a iron hammer weight 7 kg ; b Minibang. Geophone interval 2 m.


Ž .Fig. 11. Amplitude spectra of the reflected signal of shot gathers collected in Fiumicino site Fig. 10 , after it was chosenŽ .with a window on the record: iron hammer weight 7 kg and Minibang. Geophone interval 2 m. The average amplitude

Ž .spectra relative values in dB are also shown.

Ža geophone interval of 2 m was also made Fig..10 . The reflected signal was isolated with a

window on the record and the spectra are shownin Fig. 11. The reflections show dominant fre-quencies between 100 and 200 Hz and theMinibang source shows more energy in the200–450-Hz range, probably due to air wave,compared to the hammer.

4. Conclusions

Comparing the frequency content of thesources used at the first site, it can be noted that

there are no substantial differences betweenthem. The explosive and, in particular, the capshows a greater high-frequency content. In gen-eral, the amplitude of the reflected signal is lowcompared to the coherent noise with all thesources that have been used here. The modifica-tion of the Minibang plate and the burying ofthe barrel has reduced the air wave consider-ably. Moreover, shooting into a shot hole, get-ting the sources below the first 60 cm of thehighly absorbent weathering layer, helps im-prove the quality of the records . Nevertheless,field operations for routine prospecting in theseconditions are more expensive, and sometimes,


especially in urban and contaminated areas, it isnot possible to make holes at all.

Regarding the comparison between differenttypes of hammers at the second site, there arenot substantial differences in frequency content,but significant SrN difference. Reflection fre-quency were dominant around 150 Hz — higherthan at the first site — confirming the fact thatthe response is highly dependant on the ground.The records made with the iron hammer havehowever a better SrN ratio. When the geo-

Žphone spacing was increased to 2 m from 1 m.for the hammer comparison tests , the dominant

frequency with both the hammer and shotgunsource drops due to the greater distances trav-eled through the ground.

Therefore, it seems that the sources do notinfluence the response much in terms of fre-quency content — for which the ground filterplays a very important role — but that theynevertheless can dramatically effect the SrNratio, especially in consideration of the air wave.The use of an array of geophones allows consid-erable attenuation of the coherent noise andinvolves, perhaps, less work on the field com-pared to the buried Minibang; nevertheless, partof the useful signal may also be eliminatedalong with the noise, the higher frequencies inparticular. In fact, there may be time shiftsamong the geophones of the string caused byboth the apparent velocity of the reflected waveand by the static variations in the weatheringbetween one geophone and the other in thesame string, especially if the weathering layer isvery inhomogeneous.

In conclusion, the type of ground and thethickness of weathering exert the greatest influ-ence on the quality of the results. On the basisof the experiments above, it can be said that thebest results are obtained with sources in shotholes, which allow a greater transmission ofenergy and considerable noise reduction. Sur-face sources are more practical and economicaland, at times, are the only acceptable solutionsŽ .e.g. in inhabited areas . In these cases, special

Ždata collection techniques such as arrays of

.geophones can be used to improve the SrNratio. Moreover, a targeted processing can im-prove the quality of the results, even if often atthe expense of the depth of investigation andfold multiplicity.

Acknowledgements

The authors wish to thank Prof. M. Bernabinifor his useful suggestions during the entire pro-ject, and Prof. D. Steeples for the finalmanuscript review. A special acknowledgementis for Prof. E. Brizzolari, who suggested andstarted this research but unfortunately died be-fore the conclusion. The authors wish also tothank referees Anonymous, S. Pullan and R.Miller for their critical revisions and usefulsuggestions.

References

Hunter, J.A., Burns, R.A., Good, R.L., MacAulay, H.A.,Gagne, R.M., 1982. Optimum field techniques forbedrock reflection mapping with the engineering seis-mograph. Curr., B, Geol. Surv. Can. 82-1b, 125–129.

Hunter, J.A., Pullan, S.E., Burns, R.A., Gagne, R.M.,Good, R.L., 1984. Shallow seismic reflection mappingof the overburden–bedrock interface with the engineer-ing seismograph — some simple technique. Geo-

Ž .physics 49 8 , 1381–1385.Kalweit, R.S., Wood, L.C., 1982. The limits of resolution

Ž .of zero phase wavelets. Geophysics 47 7 , 1035–1046.Knapp, R.W., 1990. Vertical resolution of thick beds, thin

Ž .beds, and thin-bed cyclothems. Geophysics 55 9 ,1183–1190.

Knapp, R.W., Steeples, D.W., 1986. High resolution com-mon-depth-point seismic reflection profiling: instru-

Ž .mentation. Geophysics 51 2 , 276–294.Maxwell, P.W., Faber, K., Edelman, H.A.K., 1994. Mod-

ern geophones: do they meet the demands of shallowseismic measurements? SEG 64th Annual Meeting. LosAngeles Technical Program: Expanded Abstract. pp.559–597.

McCann, D.M., Andrew, E.M., McCann, C., 1985. Seis-mic sources for shallow reflection surveying. Geophys.Prospect. 33, 943–955.

Meeks, J.A.C., Schieffers, B.C., Ridder, J., 1990. Opti-mization of high-resolution seismic reflection parame-


ters hydrogeological investigation in the Netherlands.Ž .First Break 8 7 , 263–270.

Miller, R.D., Pullan, S.E., Steeples, D.W., Hunter, J.A.,1992. Field comparison of shallow seismic sources.

Ž .Near Chino, California. Geophysics 57 5 , 693–709.Miller, R.D., Pullan, S.E., Steeples, D.W., Hunter, J.A.,

1994. Field comparison of shallow P-wave seismicŽ .sources near Houston, Texas. Geophysics 59 11 ,

1713–1728.Miller, R.D., Pullan, S.E., Waldner, J.S., Haeni, F.P.,

1986. Field comparison of shallow seismic sources.Ž .Geophysics 51 11 , 2067–2092.

Palmer, D., 1987. High resolution seismic reflection sur-veys for coal. Geoexploration 24, 397–408.

Pullan, S.E., Hunter, J.A., 1990. Delineation of buriedbedrock valleys using the optimum offset shallow seis-

Ž .mic reflection technique. In: Ward, S.H. Ed. ,Geotechnical and Environmental Geophysics. Soc. Exp.Geophys. 3, pp. 75–87.

Pullan, S.E., MacAulay, H.A., 1987. An in-hole shotgunsource for engineering seismic surveys. Geophysics 52Ž .7 , 985–996.

Singh, S., 1984. High-frequency shallow reflection map-ping in tin mining. Geophys. Prospect. 32, 1033–1044.

Steeples, D.W., Miller, R.D., 1990. Seismic reflectionmethods applied to engineering, environmental andgroundwater problems. SEG Ser.: Invest. Geophys. 5Ž .I , 1–30.

Verna, R.K., Roy, A., 1970. A graphical method forcomputing geophone group response. Geophysics 35Ž .4 , 704–707.

Widess, M.B., 1973. How thin is a thin bed? GeophysicsŽ .38 6 , 1176–1180.

Widess, M.B., 1982. Quantifying resolving power of seis-Ž .mic system. Geophysics 47 8 , 1160–1173.

Ziolkowski, A., Lerwill, W.E., 1979. A simple approach tohigh resolution seismic profiling for coal. Geophys.Prospect. 27, 360–393.

Some considerations on shallow seismic reflection …directory.umm.ac.id/Data Elmu/jurnal/J-a/Journal Of Applied... · Some considerations on shallow seismic reflection surveysq ...

Documents