PACIFIC Passive seismic techniques for environmentally friendly and cost efficient mineral exploration D1.3– Report comparing best practice in active and passive exploration methods Grant agreement number 776622 Due date of Deliverable 30/11/18 Start date of the project 01/06/2018 Actual submission date 13/12/18 Duration 36 months Lead Beneficiary DIAS Description Comparison of active and passive seismic methods for mineral exploration; definition of best practice to be applied during passive seismic exploration. Dissemination Level PU Public x CO Confidential, only for members of the consortium (including the Commission Services)
19
Embed
PACIFIC deliverable 1€¦ · gold, porphyry copper, nonconformity uranium, and Mississippi Valley-type (MVT) deposits also can ... Nafe-Drake curve (grey) for common rocks at a standard
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
PACIFIC
Passive seismic techniques for environmentally friendly and cost efficient mineral exploration
D1.3– Report comparing best practice in
active and passive exploration methods
Grant agreement number 776622 Due date of Deliverable 30/11/18
Start date of the project 01/06/2018 Actual submission date 13/12/18
Duration 36 months Lead Beneficiary DIAS
Description
Comparison of active and passive seismic methods for mineral exploration; definition of best practice to be applied during passive seismic exploration.
Dissemination Level
PU Public x
CO Confidential, only for members of the consortium (including the Commission Services)
D1.3– REPORT COMPARING PRACTICE IN ACTIVE AND PASSIVE EXPLORATION METHODS
List of figures ......................................................................................................................................... 3
List of tables........................................................................................................................................... 4
Figure 1. Illustration of active versus passive and inline versus offline waves adopted from http://www.parkseismic.com/. Blue arrows show the active surface-waves, red arrows show the passive waves, and black lines/dashed lines show body-waves. Indexes “i” and “o” define the direction of the waves corresponding to the direction of array (“i” means in line and “o” means offline). .................................................................................................................................................. 6
Figure 2. Sketch of cross-correlations between the station-pair in ambient noise method adopted from Weaver (2011). ............................................................................................................................. 7
Figure 3. Nafe-Drake curve (grey) for common rocks at a standard confining pressure of 200 MPa (Z) adopted from Salisbury and Snyder (2007). Also shown are values for ore minerals such as pyrite (Py), pentlandite (Pn), pyrrhotite (Po), chalcopyrite (Ccp), sphalerite (Sp), hematite (Hem), magnetite (Mgt), galena (Gn), and fields for host rock-ore mixtures. ................................................................... 8
Figure 4. Schematic representation of the data processing scheme for passive surface-wave tomography adopted from Bensen et al. (2007). Phase 1 shows the steps involved in preparing single-station data prior to cross-correlation. Phase 2 outlines the cross-correlation procedure and stacking, Phase 3 includes dispersion measurement and Phase 4 is the error analysis and data selection process. ............................................................................................................................................................. 10
Figure 5. Schematic representation of the data processing steps for extracting body-waves from noise data. ..................................................................................................................................................... 11
D1.3– REPORT COMPARING PRACTICE IN ACTIVE AND PASSIVE EXPLORATION METHODS
Table 1. Comparing advantages and disadvantages of active seismic methods (mainly the reflection method) with passive seismic methods (surface-wave noise tomography and body-wave reflection retrieval). ............................................................................................................................................. 12
D1.3– REPORT COMPARING PRACTICE IN ACTIVE AND PASSIVE EXPLORATION METHODS
Across the globe, the mineral industry is seeking new technologies to replace or complement old
geophysical methods in order to improve exploration efficiency at depth. Seismic methods provide
high-resolution images of geologic structures hosting mineral deposits and, in a few cases, can be
used for direct targeting of mineral deposits (Malehmir et al, 2012; Salisbury and Snyder 2007; Figure
1). While active seismic reflection techniques can be used to explore for ore deposits at depth, they
are expensive, and the methodology can be both dangerous and environmentally destructive (since
explosives are often used to achieve an adequate signal-to-noise (S/N) ratio). This brings new
opportunities and a motivation for geophysicists to develop and introduce new, more efficient high-
resolution passive seismic methods.
Figure 1. Illustration of active versus passive and inline versus offline waves adopted from http://www.parkseismic.com/. Blue arrows show the active surface-waves, red arrows show the passive waves, and black lines/dashed lines show body-waves. Indexes “i” and “o” define the direction of the waves corresponding to the direction of array (“i” means in line and “o” means offline).
Over the last decade, seismic tomography based on interstation correlations of ambient noise has
developed into a standard tool for exploring and monitoring the Earth’s interior (Fichtner, 2015).
Ambient noise tomography is an example of wave-field interferometric imaging in which the goal is
to produce subsurface structural images by recording the ambient noise field of the Earth using
surface arrays of seismometers or geophones (Figure 1; Artman, 2006). In general this method is
based on the extraction of the surface-wave contribution to the seismic field from the cross-
correlation of seismic noise between the station pairs (Figure 1; Roux, 2009). Another application of
active seismic interferometry is to extract body-waves and retrieve the earth’s reflection response
from cross-correlations of seismic noise recordings (Draganov et al. 2007, 2009). Compared to
surface-wave extraction, body-wave extraction and reflection retrieval is a much greater challenge
because ambient noise is typically dominated by surface-wave energy (Nikata et al. 2015; Draganov
et al. 2009, 2007). The images produced with this technique are directly analogous to those produced
with conventional reflection seismic data (Artman, 2006).
2.1 Use of the active seismic method in mineral exploration
Salisbury and Snyder (2007) reported successful usage of 2-D and 3-D reflection seismic surveys for
the detection of (1) large massive sulphide (VMS) deposits, (2) massive sedimentary exhalative
(SEDEX) and (3) iron oxide copper gold (IOCG) deposits. However, other types of deposits like lode
gold, porphyry copper, nonconformity uranium, and Mississippi Valley-type (MVT) deposits also can
be detected.
In principle, most of the ores in all these deposits display higher acoustic impedances than their
common host rocks, in large part because of their high densities. They therefore plot to the right of
the Nafe-Drake curve (Figure 3), and it should be possible to detect them using high-resolution
reflection techniques if the deposits meet certain size and geometry requirements.
When the ore is in altered rock, as is commonly the case for hydrothermal ore deposits, then the target material may have lower density and lower seismic velocities due to the presence of hydrous or carbonate minerals. In some cases, the lower acoustic impedance may provide sufficient contrast to identify the ore zones.
Figure 3. Nafe-Drake curve (grey) for common rocks at a standard confining pressure of 200 MPa (Z) adopted from Salisbury and Snyder (2007). Also shown are values for ore minerals such as pyrite (Py), pentlandite (Pn), pyrrhotite (Po), chalcopyrite (Ccp), sphalerite (Sp), hematite (Hem), magnetite (Mgt), galena (Gn), and fields for host rock-ore mixtures.
Although the reflection method is mostly used to the direct detection of ore, there are reports on
successful usage of the method in detection of the structure that controls the ore (Ashton et al, 2018).
2.5 3.0 3.5
8
7
6
5
Density (g/cm3)
)s/
mk(
pV
02=Z
5.71
60.0=R
200MPa
Ccp
Sp
Gn
PoPn
Py
Ccg+pS/p
g+oP
g+yP
Mixed
Sulfides
4.0 4.5 5.0
etigolc
E
etitodir
eP
etinex
oryP
etilepat
eM
etilobihp
mA
orbba
G
etiluna
rgcifa
M
etisohtr
onA
orbbagat
eM
).caf
.hcsneer
g(
etilunar
g.tnI
tsihcsacim.zt
Q
etilunargcisleF
ssiengcisleF
.bihpma(
etinarG
etiroidonarG
etinitnepreS
etiztrauQ
setilrebmiK
stnemideS
Mgt
meH
D1.3– REPORT COMPARING PRACTICE IN ACTIVE AND PASSIVE EXPLORATION METHODS
Figure 4. Schematic representation of the data processing scheme for passive surface-wave tomography adopted from Bensen et al. (2007). Phase 1 shows the steps involved in preparing single-station data prior to cross-correlation. Phase 2 outlines the cross-correlation procedure and stacking, Phase 3 includes dispersion measurement and Phase 4 is the error analysis and data selection process.
2.2.2 Passive seismic reflection using body-wave extraction
In reflection retrieval methods, body-wave data needs to be extracted in a pre-processing step.
Unfortunately body-waves derived from ambient noise have a systematic problem. The potential
non‐ideal distribution of the sources of the most prominent phase, i.e. surface-waves, causes specific
artefacts that travel at higher apparent velocities, arriving earlier than the predicted arrival time of
the surface-wave (Nakata et al. 2015). These artefacts cover a time window when the direct/
refracted body-wave is also expected. Fortunately, the main frequency content of the surface-waves
is much lower than that of the body-waves and can therefore be suppressed by simple band passing.
On the other hand, compared to surface-wave extraction, body-wave extraction is a much greater
challenge because ambient noise is typically dominated by surface-wave energy (Nakata et al. 2015).
Nonetheless, retrieving reflection noise data allows extraction of velocity information and
construction of depth images with higher resolution than for surface-wave tomography (Draganov et
al. 2009) and the images produced with this technique are comparable to those produced using active
seismic reflection methods. Nikata et al. (2015), Ryberg (2011), Draganov et al. (2009, 2007) and Roux
et al. (2005) suggested alternative procedures to extract body-waves from noise data but in general
the processing is similar to those for extraction of surface-waves except for some additional steps
(Figure 5; refer to section A.2 in Annexes of this report for more detail).
D1.3– REPORT COMPARING PRACTICE IN ACTIVE AND PASSIVE EXPLORATION METHODS
Figure 5. Schematic representation of the data processing steps for extracting body-waves from noise data.
2.3 General comparison of active and passive methods
The main differences between active and passive seismic methods are 1) the use of active sources in the former and their absence in the latter, 2) the time taken to acquire data, and 3) environmental impact and cost. Planning and designing the survey, data acquisition, quality of the results and difficulty in result interpretation are additional differences. Table 1 summarizes the comparison of advantages and disadvantages of active versus passive seismic method.
In passive methods, the processing steps strongly depend on the data and the target(s) of the project and a unique processing sequence does not exist. In active data, whilst the details of acquisition are also driven by the target, the basic acquisition geometry and processing steps are very mature - including the availability of industry standard software packages.
2.3.1 Best passive seismic practice for the PACIFIC project
Since a unique procedure does not exist for the passive seismic method and the processing steps strongly depend on the nature of the data, we will apply two different strategies in the PACIFIC project to find the best practice. First, we will build a realistic synthetic model and try to solve the problem and find the structures within that model by extracting the body-waves and retrieving the reflection image by applying the different procedures suggested in this report. Then, we will extract part of the real data acquired at the Marathon site and apply the steps suggested by different studies summarized in this report to see which one yields the best outcome.
D1.3– REPORT COMPARING PRACTICE IN ACTIVE AND PASSIVE EXPLORATION METHODS
Table 1. Comparing advantages and disadvantages of active seismic methods (mainly the reflection method) with passive seismic methods (surface-wave noise tomography and body-wave reflection retrieval).
Seismic Methods Comparison
Active seismic reflection
Passive surface-
wave noise
tomography
Passive body-wave
reflection
Typical Targets
Horizontal to shallow-
dipping units with density
contrasts; laterally
restricted targets such as
cavities or tunnels at
depths
Structures with
lateral and vertical
velocity contrasts
Horizontal to dipping
units with density and
velocity contrasts
Disadvantage
- Need for an active source
- High expense - Can potentially be
dangerous - Environmentally
destructive - Lots of pre-processing - Low signal to noise - Sometimes difficult to
interpret - Special array design
- Depends on noise source properties
- There are no general rules for processing
- Lower resolution
- Depends on noise source properties
- Data dominated by surface-wave energy
- There are no general rules for extracting the body waves
- Sometimes difficult to interpret
- Special array design
Advantage
- Can directly target the ore in favourable circumstances
- Clear protocols and industry-standard software tools are available
- Higher resolution - Can give high resolution
structural information
- Lower cost - Absence of an
active source - Can target the host
rock with velocity variation
- Lower environmental impact
- Lower cost - Absence of an active
source - Industry standard tools
can be used - Similar procedure to
active reflection methods
- Possible high resolution, if high frequencies exist
D1.3– REPORT COMPARING PRACTICE IN ACTIVE AND PASSIVE EXPLORATION METHODS
Reflection image retrieval based on the extraction of the body-waves from noise is more challenging than surface-waves suggested flows included.
A.2.1. Nakata et al. (2015)
Processing sequence suggested by Nakata et al. (2015) for body-waves extraction is as follow:
(1) Downsample the data to reduce the computational cost and focus on body-waves up to 15 Hz.
(2) Then apply seismic interferometry by computing power-normalized cross correlation (cross coherence) between receivers A and B in the frequency domain.
(3) Daily correlation functions.
(4) To isolate body-waves and mute surface-waves, apply a time window with Gaussian-shape tapers to each daily correlation and stacked correlation to suppress signals.
(5) Then compute a second cross correlation between each daily correlation and the corresponding trace in the bin of the appropriate distance.
(6) Using a noise suppression filter to further improve the SNR of body-waves. This step involves an adaptive covariance filter (ACF), which is designed for ambient-noise analysis.
A.2.2. Reberg (2011)
Reberg (2011) suggested the following procedure for extracting the body-waves from noise data:
(1) Split the data into one hour slices.
(2) Excluding those time windows with shots and earthquakes.
(3) One‐bit normalize the data.
(4) Cross‐correlations of station pairs in the frequency domain and subsequently stacked.
The correlation function, defined for positive and negative correlation times, represents seismic waves traveling from station 1 to station 2 and vice versa. When cross correlating one station against all other stations, positive and negative correlation times correspond to pseudo shot and pseudo receiver gathers, respectively.
(6) Applying a Normal‐Move-Out (NMO) correction to the data.
The NMO correction removes (“flattens”) the hyperbolic shape of the travel-time of a seismic reflection caused by a horizontal reflector.
A.2.3. Draganov et al. (2009)
(1) Obtain poststack time-migrated images of the subsurface by following a standard processing scheme consisting of statics correction.
D1.3– REPORT COMPARING PRACTICE IN ACTIVE AND PASSIVE EXPLORATION METHODS
In order to retrieve the response that would be recorded by the passive array of geophones, Draganov et al. (2009) used the following procedure:
(1) Energy-normalized each noise panel separately.
(2) Extracted the first trace from each noise panel and correlated this trace with all the other traces in the same panel.
(3) The summation result was then band-pass filtered between 2 and 10 Hz to obtain a so-called common-source gather with the retrieved source position corresponding to the location of the first geophone.
(4) The above procedure was repeated in such a way that they retrieved source positions at all the geophone positions of the passive array.
To improve the clarity of reflection arrivals in data sets, they performed the following additional processing steps:
(5) Resorted the traces in the retrieved common-source gathers into common-offset panels.
(6) The traces in each common-offset panel were summed and normalized for the number of summed traces, producing a single output trace per common-offset panel.
(7) The output traces from the different common-offset panels were sorted into a so-called common-offset stack panel.
(8) The resulting common-offset stack panel was further filtered in the frequency-wave number domain (f-k filtering) to eliminate the surface-waves and then band-pass filtered between 13 and 33 Hz.
A.2.5. Roux et al. (2005)
The processing performed on 1-day noise recording on one seismometer consists of:
(1) Eliminating high-amplitude events by truncating the recording amplitudes at three times the standard deviation of the ambient noise signal.
(2) Equalization of the noise spectrum in the frequency intervals.
(3) Applying plane wave beamforming.
(4) Select station pairs whose relative locations lie along lines having certain azimuths.