An Introduction To Seismic Interpretation PDF
An Introduction To Seismic Interpretation PDF
An Introduction To Seismic Interpretation PDF
toSeismic·
. interpretation
R McQuillin M Bacon W Barclay
2) CURSO de INT6RPRl:TACION
SISMICA Ccap�tulot parciales)
In�. P. B�ccoccio:
An lntroduction
toSeismic
lnterpretation
. .
R McQuillin M Bacon W Barclay
An lntroductlon to Selemlc lnterprotatlon
2. DATA
ACQUISITION
In this chapter we consider the applicaiion of seismic treated by data processing and, as far as possible, must be
theory, as described in chapter 1, to the design of equip measured and compensated for during data acquisition.
ment for generating seismic pulses, and detecting and This is mainly achieved by layout design, on land by
recording the earth 's response to the passage of seismic proper design of geophone sprea<ls and arrays and at sea
waves through it. The aim here is not to treat the subject by use of well designed hydrophone arrays. Display of the
of data acquisition at the level which would be required recorded signa! will not, in most situations, give an easily
in a text-book written for geophysicists principally interpretable picture of geological strucn.ire. This record
concerned with field techniques, but to present enough needs further processing to achieve such cla1ification, and
background information on this subject to satisfy the needs these processing techniques are the subject of the tollow
of the seismic interpreter who may have little opportunity ing chapter.
to participate in field surveys, or at best gain experience of
. only a limited range of the methods current!y in- use.
Exploration seismc,logy is a remote-sensing technique
in which the aini is to record as detailed a picture 2.1 Layout design
as possible of subsurface geology. The product of In designing layout systems, the emphasis is placed on
a seismic investigatio11 is a geological model which can be eliminating unwanted signals or 'noise' of both the random
described as the s11m of a finite series of layers of varying and coherent variety. Use of multiplc sources, multiple
thickness, physical properties (density and seismic velocity) detectors per trace, and the summing of common reflec
and structural attitude. Interpretation of this model is in tion point traces (see figures 2/1 and 3/6) brings about a
terms of geological structure, lithological vanatton, distinct improvemént in signa! to noise ratio in the case
stratigraphy and, in oíl exploration, hydrocarbon of random noise. For spatially random noise, the improve
prospectivity. ment is proportional to J n where n is thc number of
Seismic data is acquired using a system consisting of detecting elements in the acquisition system, the signals
three main components: an input source, an array of from which are added together to providc the final record.
detectors and a recording instrument. The input source is For example, the summing of eight separate seismic signals
designed to generare a pulse of sound· which meets, as near (eight shots at same shot-point locat:ion), detected by·
as possible, certain predefined requirements of total energy, geophone spreads of t:wenty geophones per trace t.hen subject,
duration, frequency content, maximum amplitude and during processing, to twent}'.-four fol<l stacking, will provlde
phase. Reflected and refracted seismic pulses (the output an improvement ratio of ✓.s x 20 x 24 = 62 or 36dB. This
from the earth) are detected by an array of geophones or a may be compared with a single shot record, single-fold
hydrophone array, tl1cn recorded by a recording processed with, as before 20 geophones per spread, in which
instrument, and in both cases these ourput signals will be case the improvement ratio is J 20 = 4.5 or 13dB. The
modified by the response characteristics of that part of the former acquisition metl1od shows a relative improvement
system. Each seismic record is thus a time record of the of 23dB over the latter in signal to noise ratio enhancement.
output signals which are generated at interfaces in a series Figu re 2/1 shows a typical marine multi-channel acquisi·
of stratigraphic layers because of the changes in acoustic tion system and illustrates how. data are acquired in a way
impedance which occur at such boundaries, modified firstly which allows stacking during processing. Although this
by transmission decay and noise interference in the earth íllustration shows only a marine system, acquisition of land
and then by detector and recorder response characteristics. data is based on identical principies. At sea, a survey ship
This can be summariscd as follows: tows a hydrophone streamer made up of a number of sec
tions, numbered one to forty-eight in the figure. Modern
R.ecorded signa/= So urce pulse * {Refiectivity *
streamers are fitted with 24, 48 or 96 such sections and
(Earth filter + Noise)/ • Detector response
each section consists of a group of hydrophones which are
• Recnrding instrument response,
pressure sens1t1ve sound detectors (see p.26) Signals
where * repre�ents convolution (see p.188)
received by the hydrophones in each section are summed
Assuming that we know the signal characteristics of the so that each section is considered to be an independent
seismic pulse a.nd the response characteristics of detector single detector. In figure 2/1 the reflections from a single
and recording instrutn�nt, then we can separate that part horizon are schematically portrayed as received in the first
vf the function contained in square brackets, and this is eight sections of the 48-5ection streamer. Let us assume
the earth 's impulse response. The earth 's reflectivity is what 'that the distance between sections is 50m and that the ship
we wish to mea.sure. The earth's filter is a variable function is travelling at Skm/h (approximately 4kts). If the first
of absorption and � �tentuation which can be compensated shot S 1 occurs at time t1, a reflection from depth point
for in data processing. Noise cannot be so adequately no.1 is received in section no.1 of the stre,imer and thence
9
Data Acqulsltlon
....,.,----------- 3 km approx---------�--
1 2 3 4 5 6 7 8
Shot point
. . . . . . . . .. . . ... . . . . . .
... . . . . . . .. .. .. . .. ...
. . . . . . .. . . . . . .
.. .. ....
.. . ... .
. . . .. . . .. .. . .. .. .. .. . .. .. .. .. .. .. .. . .. .. .. . . .. .. .. . .
. . . .
. ... . . . . .... .
1� Distance along ship's course
S1 1 2 3 4 5 6 7 8
. t1
S2 1 2 3 4 5 6 7 8 9
12
S3 1 2 3 4 5 6 7 8 9
t3
S4 1 2 3 4 5 6 7 8 9 10
14
S5 1 2 3 4 5 6 7 8 9 10
ts
Q) S5 1 2 3 4 5 6 7 8 9 10 11
E te
S7 1 2 3 4 .5 6 7 8 9 10 11
t7
tai----t-t-Se 1 2 3 4 5 6 ___
-----�-�__..._.__....__....._ 7 · 8__�
__
9__10.__..___.
11 12
Shot number ......
�é---- Hydrophone sections numbered 1 to 48 �:
tn 1./'1
18 19 20 21 22 23 24 25 26 27 28 29 30 31
19 20 21 22 23 24 25 26 27 28 29 30 31 32 48
1
e-t
recorded in channel no.l of the sc:ismic recording system that thc location of thc midway point betwccn source
vnboard ship. The common depth point (CDP) position and section no.2 of thc strcamcr is thc samc CDP location
is located mid-way bctwc:cn the locations of S 1 and section as for S1• This occurs at t :z , scc lower half of diagram, and
no.1; in thc lower half of the diagram at t 1 the scction it can be scen that thc distancc bctwcen S 1 and S1 is half
no.1 location highlil{htcd as a· largc dot. Thc ncxt shot S1 thc distancc bctwecn sections, that is 25m. Thc intcrval
is timcd so that thc $1\ip has· progrcssed to a position such bctwcen shots (t 2 -t 1 ) should be set thereforc at 11.25s'.
10
An lnlroduc1.on lo S.lnmlc lntet'pnltlrtlon
At t2, the signals rccorded 011 channel no.2 of the record Multíple, or secondary , reflections (as oppose<l to die
ing system are therefore those associated with CDP no.1. primary reflections on which data are being 5ought) d.n be
Shots S 3 to S 48 follow at the same interval and successive attenuated or even effectively eliminated by commm,
records are obtained on ship from CDP 1, until data from reflection point (or common depth point, CDP) stacking.
shot S48 is recordcd on channel no.48 of the recording The principie is to design the trace spacing such that che
system. It should be noted that as the ship progresses along second ary reflections have the appropriate residual normal
course, the seismic signa! reflected from CDP 1 will have moveout to be stacked out of phase and consequently
travelled an evcr incrcasing distance between shot-point and much reduced in amplitude, while the primary reflections
receiving strcamer scction. The change in geometry is are stacked in phase by application of the correct normal
corrected for during processing, and it is possible to add moveout velocity (see chapter 3). The formulae to be
together (stack). ali 48 records pertaining to CDP l. u tilised for the simplest cases (see figure 2/3) are as follows:
Obviously the same is true for the locations CDP 2, 3 etc.
CDP stacking is valuable not only as a means of increasing
Source Receiver
signa! to noise ratio but also, during processing (see 1-•············ X ·•·•········-1
chapter 3), of allowing differentation between primary
reflections from geological structure, and multiple reflec·
tions in sea and rock layers. Multiple reflections can then be
suppressed to improve the quality of the final seismic
section display. In land survcys, shot·point locations are
surveyed at fixed interval� and groups of geophones are
peggcd into thc ground with a group interval which is
equivalen( to. the section interval of th.e marine strearner.
Obviously, surveys on land cannot be .conducted_ with the (a)
speed of a marine survey, and tim.ing of shots is irrelevant
to a static layout, nevertheless, the geometrical principies
are identical.
Cohcrent noise, of the types illustrated in figu re 2/2,
c:m be reduced in two ways depending on whether it is a
direct near horizontal wave originating near ground leve!,
in thc sea or ncar seabed, or is a reflected near vertical
travelling wave. In general, lowest velocity direct waves
arrive la test, a factor which can be utilised in design criteria
for detector arrays. With reflected noise waves, lowest
velocity waves arrive earliest and chis can be u tilised in thc
design of optimum trace spacing for attenuation stacking.
1-··········•· X • • • •X2
...., . . . x, . . ·► l....;• ...... _.I
• • •• •• •• •• •• •• --
TIMEª
¡¡
�
;;
1
Figure 2/3: Ray path multiple reflection geometry (not adjusted
far refractior. according to Snell 's Lawl.
Figure 2/2: Coherent seisniic noise types. Far primary and multiple a) First o rder or simple multiple.
reflections the average velocities increase and the normal moveouts b) Second arder or peg leg multiple, that is, T = T m 1 + Trn 2 .
decrease with time (alter S.O. Brasei in unpublished report, Design Such multip!es are common in marine survey dati due to repeate d
of seismic field techniques. Atlantic Richfield, 1971)'. roflections betwP.en seabcd and sea surface.
11
Dala Acqulslllon
✓
.140
To 2 + � _ To x2
= 2( ) (JTo2
,,, .120
4 4V1 2 2 + V 22
2 x2
� .100
J 2
V, 2 - To + V22
2... .� 080
:,
&b60
!!l
ii. The residual moveout second order (asymmetri 040
cal) or peg-leg multiple equation (see figure
2/3 b for dcfinition of symbols,)
o l.!-...J.1111!!!�=:::!===::I:�L_l__.L._¡___L__¡__¡____¡
O 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6
6TM1 +·T,12 - í:,Tprlmari•
x2 2
Shot to detector distance in teet
For optimum attenuation, the trace distances are chosen wanting, and non-uniform trace spacing is impracticable,
such that the Í'>t differences are approximately equal to the so the design adoptcd is invariably a compromise offering
multiple period divided by (fold of stack-1). Figu re 2/4 Iimited attenuation. Howevcr, it is important to be aware of
is adapted from the original paper on stacking by Mayne• the factors which control attenuation of multiples, in
demonstrating the principie. In practice, unless expanded particular if an area is to be resu1vcyed with the aim of
spreads are shot (see figure 2/5) knowledge of the type of acquiring improved data in a situation where multiples are
multiple to be suppressed, and of their velocities, is usually · known to pose interpretation problems. Figure 2/6 is an
illustration of a single-fold as against multi-fold comparison;
• ·w. Harry Mayne, 'Common reflection point horizontal data the distinct multiple suppression and improvement in the
stacking techniqucs'. Geophysics, vol. 27; (1962) pp.927-938. signa! to noise ratio of the primary reflections is appareht.
6 F0L0 1 F0L0
'"o
"'
. -15 g,
20
1...------------------------------,---------'
Figure 2/6: Sicgle fold and rnultifold seismic section comparison.
On the single fold section pronounced multiples interfere with
reflections from a target horizon at between 1 .05 and 1. 15s tv110-
way time.
(CovrtlPtV: Mobil Oíl CBnede L td).
12
An lntrod11. 'Ion to Seiamlc lnt,wpretatlon
__Subsurface eommon
_
�· depth point Ctwerage
'
S.,,.ad
G,
5ocatton_-
G,
05- -
o•
. 'º -1l
•o-
� �
.s
g
f f
o o
J 'º �
30
A B
Figure 2/5: Expanded spread shooting for velocity determination are also in wide use. In practice, economics and operacional
and identification of multiples. a) The same zone of common depth consideratíons combined with the unpreuictability,
point (CDP coverage) is tested seven times by shooting at locations
S1 to S7 the geophone spreads G 1 to G 7 located with varying
throughout a survey area, of the strength and characteris
offsets. b) Application of normal moveout and datum reflection tics of coherent noise, dictate that the sekcted design will
flattening makes identification of multiples easier because of their be a compromise rather than ideal for each individual shot
residual moveout. location. It should be noted here that the first direct
(Courtesy: Mobil Oil Can;,da Ltd).
arrivals, or first-breaks, provide valuable near surface refrac
tion information, and in both land and marine work these
are not suppressed in the field. Where they interfere with
Suppression of dircct no1se waves is attained by using desired shallow reflections they are muted during the data
an appropriate number of detectors, spaced areally or processing stages as described in the following chapter.
linearly at pre-determined intervals. The problem of surface
noise is greater m land surveys than marine surveys and Source arrays can be used for coherent noise suppres
usually it is necessa[)' to conduce noise spread or 'walk sion either as an alternative to geophone arrays, or more
away' surveys (see figure 2/7) in each new area using single commonly, as a complement.. The same principies of design
phones per traer over the total length of the planned as for detector arrays are involved and this is most irnport
spread. Once the period T of noise has been established it a.nt for proper utilisation of low energy surlace sources such
i� possible to plot an array response graph (figure 2/8) as dinoseis, vibraseis and thumper.
which indicates the attenuation achieved by a particular So far we have discussed design criteria in terms of the
array within a particular wavelength band. If AL and A5 spacing of detectors, thc number of groups (or sections in
are respectively the longest and shonest coherent noise marine work) to be used and the leve! of fold or stacking
wavelengths it is desir�d to suppress, then simple design multiplicity. A final consideration is that of the length of
criteria indicate that the number of geophones N should be spread. Design criteria here are not so specific, but the
greater than 2AL/A5, the length of the array L equal to length of spread used is usually related to the depth of geo·
,'\
l .n,L and the detector spacing 6L equal to 1.SALIN. In the logical objective; long spreads are used to investigate deep
above only a uniformly spaced linear array has been structure, short �preads to obtain highly resolved data on
considered; non-unifonnly spaced linear and areal arrays shallower objectives.
13
Data Acqulsltlon
20
30
(/)
"ai 40
.o
fil
-o
.f
e 50
'°
Q Desired attenuation, -13.69 db
in wavelength interval, 30-150 ft
::,
e
Q)
Geophone group interval
::: 60
<( necessary , 25 ft
80
90
100--t---r-----.------.-----.---.-----,----,-, --,---,----.----.-�----l----4-
25 30 35 40 45 50 55 60 65 70 80 90 100 150 300
Wavelenqth in feet
Figure 2/8: The response of a geophone spread designed to attenuate
unwanted noise in the wavelength band 30ft to 150ft.
(Courtesy: Sercel).
strike of controlling faults bccame known is. illustrated in ely narrow and except for special shallow, high frequency
figu re 2/9. input, high resolution surveys, it is limited to around
5-l00Hz. Commonly, at exploration, depths around
10,000ft or deeper, frequencies higher than 4 CH: are rarc.
2.2 The seismic pulse This means that thin bed resolution becomes increasingly
At the beginning of this chapter we referred to the earth difficult with increased depths of exploration. For a 40Hz
filter as one of the convolution factors of the seismic pulse wavelet, and an interval velocity of lS,000ft/s, beds thinner
which contributed to the final signa! output. The earth 's than 94ft, the quarter-wavelength one-way time width
filter attenuates thc seismic pulse in three ways: absorp would be irresolvabie. The band-limiting nature of the ·
tion of energy through conversion into heat; spherical earth 's filter and its frcquency attcnuation with depth is
divergtnce of the wave front; and reflectivity losses at graphically illustrated in figure 2/10. At 12,000ft, the 40Hz
acoustic interfaces. The attenuation involves amplitude and amplitude factor 0f a wav�form is one third of the 201-iz
frequency and these are interrelated. In normal seismic factor, while that for a l00Hz wavefonn is about one
prospecting, the bandwidth of useful frequencies is extrem- fortieth. lt shou!d be appreciated d1at che ideal seismic
15
An lntroouctlon to Selsmlc lntorpnitr.':ln
17
Data Acqulsltlon
1000
- -�
¡_......-;--...,-,,:;;
�
o ¡...- .& \..ll
��-'"'
o� " al1'''-e
'é.\Gr\
- o º"('
�� f,0"/o
01n
{3 100 º º'" 1--{
H �-¡¡<s �
.o t:::¡;:;", 9 .g:-:í.'E>º,1¡,:-11'0\e'-
·C
�
3 seis 0óeis
.
o :::::---:--:;S'¡;_. ., 9\ �
-
-�
c,a s e-¡_
�%-;: /t o �
E ;::::::-
i- �
to'º'
......-...... ,.,..- 11 30"-.)
' � º
� o
f\
�
ox.:: - - - - - -x- A-.¡...
!:
c1>5' - º - p-;.;, ?
,s
Cll
10
L,......--:::P . , ''" º".) ,...o
'
\f'J _..
ss?
r- ,� 'y_j
/
el o r:,º1<-.l
''y.J � 3 1<-.l
o 'v- ''ª"
o/ S?ª
1 ª
,,1 1
...., 'f..)
a
-
1 11
106 108
1
10s
1
103 104 107
___l.,
.•
<(7
o
.,•, observed bubble oscillation period versus depth; 300in at 2000 psi
of air. R-W is the Rayleigh-Willis curve. Near surface, the bubble
period shows a divergent decrease due to the mass unloading effect
5 which occurs as the bubble approaches the air/water interface.
0
/
Observed curve
"'¡ , used under lícence by severa! contractors and has other
//
10
V
(¡)
15
oxygen are separately fed into the rubber sleeve and ignited
.S J by a sparking plug. The high energy pulse is generated as
-
20 the sleevc expands; after expansion,-the sleeve contiacts
;.e:
// with venting of gases to the surfacé, decompression being
//
5
�2 assisted by cooling and pumping. Multiple arrays such as
'3
0 shown in figure 2/18 are generally required to provide
sufficient total energy output for dcep penetration,
1,.,
/
35
40 2.4.3 M a x ip u lse
/ Maxipulsc is one of the simplcst seismic sourccs currently
45 in use. A small cylindrical charge of nitrocarbonitrate
50 ,/ explosivc is projected by water pressurc down a hose to a
submcrged gun, striking its detonator on a firing \Vhcel as
-� -400 .120 .140 .160 it is ejcctcd, see figure 2/20. A percussion cap is activatcd
Bubble periocl in seconds and this fires a delay fuse which detonates thc main charge
B
aftcr sufficient pcriod for the gun to havc becn towed a few
mctres from the explosion point.
22
--�---------·--··------�--------�-------------------
Data Aco!.!lllltlon
ing sour-::e arrays which can b•� dfoctive in attenu?.ting 2.5.1 Lan d geophones
coni:1e,1t 1101sc. Figure 2/21 illustrates the principie of a geophone designed
to measurc particle motion by conversion to electrical
2 4.7 Sparker ,ystems energy. Arrival of a compressional wave sets the earth 's
In �p·,rker source systems, acoustic energy is gene;� ied by surface in motion; the geophone case with rigidly attached
dectr •.al discharges in sea \'..-atcr. Generators are t ,ed tn co:l is oupled to the ground and moves in sympathy
chargc cpacitor banks which can then be trigg 0red to whereas the magnét which is suspended on springs remains
discl1arl'e high voltage (3-lOKVA) through spark tir �ffoctive!y s·,ationary because of its inerti:i. Mo·.-c;n::;.t of
arr.iys toweJ in thc water. Low energy sparkers, 100 the coil wiihin the magnetic field induces an electrical
joul�-5 kilojoule, are widely used in shigle channcl seismic voltage acros.� the coil which is proporcional to the velocity
profiling as pan of sha!low geological stndies, engineering d the coil -vitii respect to the magnct. An altemative, and ·
sit-. surv cys etc, (see McQuillin and Ardus, J 977). High more .::ommo11 desÍ"'II is to mount the magnet rígidly and.
energy sparkers, cnc�gy up to ZOO k,lojoule. are used as a suspenc. the cnil arvund an inercial mass; the prindple is
sourcc for convcntional seismic work, one advantage being identícal and ;;pct; geophones are calkd moving coi! geo
t'1e relatively low operaticnal cost of using this type of phones. In both c:Jses the ou!put is independent of
system. In rr.cent years, a more significant use of the frequency at fn:�;uencies above thc natural frequency of the
s;,arker as a source for mt.!t:,hannel seismic 2cquisition is suspended clcment, bclow natural frequency the response is
in the field of shallnw gas detection, in particular for dril! frequency depen<lcnt. Modero gcop!�ones 1•se a dual coil
site �Jrvcys. J-kre sparkers ar� operated in the 3-5 kilo system, the duaí, series-connected coils red•· � externa!
joule range and data acq.iisition aims for high resolution · interference.
data in the 0-ls two-way rcflectio:i time range. Sections In land surveys, it is important that goo..: •;•1g is
are prriccssed co give true amplitude rec0very (see chapter 3) obtained between the ground surface :md the geophones.
disp!..y as wel! as conventional displays. Thi." ca�es are usualiy either heavy with a flat base, or light
with a coupling spike which is pushd into the soil. Geo
2.5 Geophones and hydrophonc� phones are usually depluyed in rroups at each detector
In scismic surveying, two types of accustic derccting location, the phones bcing connected ir, series The group is
transducer· are uscd, gcophoncs on land and hydrophones callcd a string. and each �tring is attached t<> a ,:akeout in
in marine cm:di,ions and in mud-fille,� roorcholes, We have the main cable which fceds to a specific channel in the
seen in chapter 1 th�t in reflection surveying we are recording system. For spced and efficiency during a survcy
a�tempting to record trains of P-waves, or compressional of a seismic line, sevéral cables are laid out so that the
waves as they pass a specified point. Thus pressure sc:nsitive numroer of strinu,s dcpk,yed is greater dlan the number of
ph,,nes are ideally suited for marine work, recordi,1g as channeb oeing rccorded. After e;;:ch seismic shot, or a fixed
they do thc pressure changes above or below ambíent water number of shots, a string or a number of strings are pickc<.!
pr.-:ssure, On land, pressure-sensitive phones cannot be used up from the end of line already surveyed and 'leap-frogged'
as it is generally impractica! to bury the phones in such a to the other end. The recording systcm incorporares a roll 0
way :·hat they would have adcquate fü.. id coupling with along' swi,ch which is used to drop-out and pick-up i:hc
surróunding material. The u.se of pressure sensitive p!:one� appropriatc cable take-outs.
at or near ground surfacc · is further complicated ·by tite ·
fad that the air to ground ini:erferenc:e acts as a ph2se 2.5.2 Hydro phon-::s
change refkction boundary at which differcntíal pressure í'he principie of the marine hydrophone is very simple.
approache$ zero. For these reasons land gcoph.ones are of Within the phor,e a piezoelectric transducer ,iroduces
the type sensitive to particle moti0n. voltages in response to · pressure changes caused by the
passage through surroundh1g water of seismic pressure
Output waves. Figure 2/22 shows thé frt:quency response of a
typical hydrophone. For static cable recording, good
response can be obtdncd with a simple hydrophone
element, but current methods employ continuous profiling
using towed streamers and substancial noise is generated
--· Coil through vibration of the cable (strumming) and sudden
acceleration/deccleration effocts produced by hea11�
activg on the towing vessel, the vcsse!s mo\·emcnts being
tr:m:mitted to the stream�r. Various methods are employt:d
---- Magnet to reduce: these cffects:
l. Shíp motion is decoupled from the streamer by
using an elastic non··:.ctive lcad-in section; thís
absorbi, t.'.e ship's heave motion allówing the
cable to be towed at a constant speed through
the witter.
//10.
//10.
1li
ton.(1977) y Lines y Ulrych (1 977).
En algunos casos, parecería que el principal efecto so
bre la amplitud y los espectros de fase de la traza sísmica es a
tribuible a la respuesta del instrumento. Se puede aplicar a la
amplitud una correcci6n preliminar y, en particular, aplicarla a
las características de respuesta de fase de los filtros de instru
mentos de campo como lo publicaron los fabricantes.
La deconvoluci6n de los datos antes de las correcciones
de NMO debería ser seguida por la deconvoluci6n después de la suma
de tra-zas. Invariablemente, el proceso de suma modifica el espec
tro de amplitud de las trazas de entrada. En consecuencia, es nece
sario aplicar esta segunda deconvoluci6n "de ajuste" después de la
suma para restaurar el espectro de amplitud y efectuar cualquier
ajuste secundario necesario al espectro de fase.
Luego de la suma de trazas, se pueden aplicar procedi
mientos convencionales de filtrado y puesta en escala de amplitud
para producir una sección sísmica convencipnal. La salida de banda
·ancha· �o puesta en escala (unscaled) proveniente del proceso de su
ma, deconvolucionada, es la entrada al proceso de inversi6n. Si la
suma no pudo suprimir un porcentaje satisfactorio de ruido, puede
ser dtil aplicar un filtro en abanico (o F-K) (Embree y otros, 1963)
siempre que los datos respondan a ese tipo de tratamiento.
El ancho de banda de la señal debería oscilar entre tres
y cuatro octavas. Generalmente, el ancho de banda se extenderá a
75 Hz o más. El ancho de banda de los datos sísmicos marinos con
vencionales, registrados a intervalos de 4 meses, es limitado por
el filtro "antialias" a una límite superior cerca de 62 Hz. El ex
tremo de baja frecuencia del espectro se encontrará entre los 5 y
15 Hz segdn la respuesta de los ge6fonos y/o cualquier filtro cor
ta-bajos aplicado en el registro de campo. La deconvoluci6n habrá
extendido el extremo de baja frecuencia del espectro en forma leve,
dentro de la zona de pendiente (roll-off) de.respuesta de frecuen
cia de los ge6fonos y/o los filtros de campo; pero las frecuencias
muy bajas, que representan la informaci6n de velocidad rudimentaria
deben obtenerse de otra fuente.
ANALISIS DE VELOCIDAD DERIVADO DE REFLEXIONES PARA BAJAS FRECUEN
CIAS
Si fuera posible registrar datos sísmicos confiables en
el orden de los 6 Hz, como ocasionalmente ocurre, la inversi6n no
produciría adn una aproximaci6n a un perfil s6nico mejor que la del
perfil de 6 a 250 Hz de la Fig. 4. Las componentes de velocidad de
O a 5 Hz, que están ausentes y que forman una funci6n suave y don
tinua de velocidad instantánea en funci6n de tiempo, pueden recupe-
//11.
209
• -SECCION SISMIC.f.\
}< '� ) .
1
1
¡I 1 ( ! I
l FIG.xr.-10
s
0-6
1-0
1-S
215
CAP. XII: F AL L A M IE N T O
I - EVIDENCIAS DE FALLZ\MIENTO
lidild de los dütos, que a veces es tan gri:.nde que las reflexio
nes por debajo de la falla están completamente ausentes, confo�
mando una zona obscera o de silencio ("sombra" de la fé,lla). En
ocasiones el pl3no de falla en sf genera, como se verá más ade
lante, reflexiones que a menos que la sección sea migrada, pue
de no verse clara J.a asociaci6n con la falla de dichas reflexio
nes.-
II - 1 - En registros de campo:
-\-
P.E., PE.2
.) 1
1
1 ;1
1
-�ª-�-
1 ¡
1 / 1
.. 1 11 ;¡. /{
1 /1 / 1
// 1
' ':'
¡
1 ,/ 1 / 1 I
1
1 I/ 1 /
/ /
/
¡
/
/
/
f
I¡ ¡ , 1/
v I¡ f
'/ I
� V /
PEí PEí PEi P¡E,Í / I
1 / /
/
/ //
¡ /¡
�"
PEs
FIG . .x:rr- 3
II - 2 - En secciones sísmicas:
'
1
-4--
'I
i-. e�
I.S 1
-�·
FIG. 10
IV �AC0Nru1IENTO
1� 417
u
11
l:u;n,�nL-.::i.tL:i p,11:a <.:>:,Ln., c,'l:;o:.; ,�n ,;'.l C.::i.¡_,. )'.V-punto III-1.l, Je-van
'
,1 t,nni.c�nto ](,c¡ic.J111Pnt1• po::;Lctior z; lu dcpositac5-6n; juid-,,rn,.... nl0 12_
9,v]o :iquf cJ Jo�; rno\'i1:lic11t-.o.s ;:iur1i_cos dc.l. t�rciario . .l\dern(.is la
, �
lh�l f:or,1c i (in l':Xi :; t-r•n i-r.� pcnni t i.ó dilucidar 1 a biíurca.ci6n en �rs el
� s<,·gundo y t..0rccr ntvcl cont:andci desde abajo, ya que b.i'lb:t.a dudas
)s � de cómo con U nua.!1,1 hacict la izquierdél el segundo nivel partien
J□
do desde la derecha, donde es di:._;cordante.-
·1
.:
de la posibilidad de vincuJaci..ón con afloramientos, complemcn
tétncln el cjc'mpl.o ya citado de la. Fig. l. La primer refu:cncia
(FtrJ. XVTT-2';) :�:, un buc11 ej0.rnplo de cosi net.3. v:i.nculacj_ón en
t1.r:: e] ¿;[lot·drrús_,nt.c en superficie y un rl'ivcJ sí.smico 2:1 profu�'..
dicli,,l. E] scqunclo tcjcrnplo (I•'.i.q. VllJ--10) está lejo:; de los aflo
J (7 r:1¡,1i_c,n lo:; m:i'.'; prfr,-;_i_r,,o:0, pcrc en subr,uelo se observa nílid;;un12n-:..,�
el é,�;Lilo pci11i¡H..',1t10 el e, fallétmicnlo invc,rso, bien conocidn hacia la
dcrc'.clJ¿:1 de 1::J ,,i'c:ctón s:í.1.;mic2, c,n afloramientos a�� lai; Sir.!!Tas·
Pampeanas etc C..Snbba-.:3an Luis. i:':l úllun 2jemplo (Fiq.XVIJ:-26),
presentado como cOJi1:r,11;Le 1c,s :iustamc!nLe la falta ::i.b,�oluti:l de a
f.lor,nnicnto1.;, surn;:ida .:.J. unü quietud horizont¡..ll él.e la to;:Jo,p:¿\�_'[;:-.
y aún de leda l,j colunna lcrcj_aria:- ·ante las fu(:r,:r•:; t:__j_,;],,c.::tcio-·
ne:; profundtts. -