PAKISTAN

WATER AND POWER DEVELOPMENT AUTHORITY

DIAMER-BASHA DAM PROJECT

TRAINING ON THE
MICROSEISMIC MONITORING SYSTEM
AT M/S KINEMETRICS FACILITIES, SWITZERLAND

REPORT BY
SYED KAZIM MAHDI, DIRECTOR SEISMIC STUDIES

FEBRUARY 2008

OFFICE OF THE
GENERAL MANAGER & PROJECT DIRECTOR (TDP/GB)
 DIAMER-BASHA DAM PROJECT
TRAINING ON MICROSEISMIC MONITORING SYSTEM

1. SEISMOTECTONIC SETUP

The available instrumental/historical data indicate that the Diamer-Basha Dam Project is
positioned in a region which has been identified as seismically active in the past and
continues to release stresses in the form of moderate to large earthquakes. Seismic studies
during various stages have suggested that most of the seismicity occurs along the Jaglot
Syncline, about 36 km north of the site and along the Main Mantle Thrust (MMT), about
48 km south of the site. Near the proposed Main Dam in addition to many small extent
lineaments two very strong features have been identified that run in the transverse
direction to the previously known faults. One seems to run about 250 meters upstream the
site and traverses even the Chillas Complex in the north and Kamila Complex in the
south of the Dam site. The other one is at about ten km upstream from the first. Within
the scenario of the October 08, 2005 Mw = 7.7 Earthquake, it becomes important to be
very cautious regarding the seismic risk for such a Dam that is already located in serious
seismic environments. Moreover according to the recommendations of the International
Committee on Large Dams (ICOLD), “in such type of seismically active areas, the
seismic monitoring systems should be installed at least five years prior to the estimated
first impounding and should continue afterwards for an indefinite period of time”.

2. PURPOSE OF THE MICROSEISMIC MONITORING SYSTEM

For the acquisition of seismic data, a network of ten stations Microseismic Monitoring
System (MMS) distributed in an area of (40 * 40) km around the site (Map attached), was
commissioned by M/S Kinemetrics of Switzerland during August-September 2007. The
main purpose of the MMS is to establish the level of seismicity and to identify any
seismogenic structures at/around the proposed Dam site. As such the objective of the
MMS is to record the occurrence of earthquakes in time and space. The recorded
waveform data is analyzed to access source and spectral parameters. The distribution of
the earthquake epicenters is used to identify active geological structures and together with
the magnitude of each event, it can contribute to assessing the seismic hazard for a
particular area or site. It is critically important that the recorded data must be an accurate
and true reflection of the ground motion induced by the elastic waves propagating from
the source of an earthquake to the seismological station.

According to the estimates of the Consultants the initial duration for the deployment of
the MMS at the proposed Diamer-Basha Dam site is for eighteen months only, with the
following purposes:

1. Picking of ‘P’ and ‘S’ waves.

2. Location of earthquakes parameters.
3. Magnitude estimation.
4. Publishing of Bulletins (once per quarter).

2
3. MMS CONTRACT

M/S Kinemetrics (KMI) of Switzerland and their local agents M/S National Engineers of
Lahore have been awarded the following contract for the establishment of MMS at the
proposed Diamer-Basha Dam site:

a. Carry out the noise survey for the establishment of ten station MMS.
b. Construction of Vaults at the locations of ten stations.
c. Installation of equipment at the ten stations.
d. O & M of ten stations for first six months i.e. up to March 2008.
e. Gathering of data from the ten stations for first six months.
f. Processing of seismic data for first three months i.e. up to November 2007 at
the M/S Kinemetrics facilities in Switzerland.
g. Issue of Earthquake Bulletins.

4. TRAINING OF SEISMOLOGISTS

WAPDA Authorities approved training for the following two Seismologists on

Microseismic Data Analysis, at the M/S Kinemetrics office in Switzerland.

1. Syed Kazim Mahdi, Director Seismology, WAPDA Tarbela Dam Project.

2. Mr. Nayyar Majeed, Senior Seismologist, WAPDA Mangla Dam Project.

M/S Kinemetrics submitted a schedule for training of one Seismologist for 21 days.
However, training of two Seismologists from 07.01.2008 through 18.01.2008 was
approved by WAPDA Authorities. In Switzerland following training program was
arranged by M/S Kinemetrics.

DATES DESCRIPTION/TRAINING EXPERT/SEISMOLOGIST

05.01.2008 Arrival at Zurich Switzerland Dr. Muhammad Idreessi
06.01.2008 Arrival at Preverenges Switzerland Dr. Muhammad Idreessi
Location of M/S Kinemetrics office
07.01.2008 to Training on seismic data analysis Dr. Muhammad Raeesi
11.01.2008 By SEISAN Software University of Bergen
Norway.
12.01.2008 & Weekly off Dr. Muhammad Idreessi
13.01.2008
14.01.2008 & Training on seismic data analysis Dr. Nikko
15.01.2008 By ANTELOPE Software Austria
16.01.2008 to Training on seismic data analysis Dr. Mladen Zivcic
18.01.2008 By ANTELOPE Software Republic of Slovenia
19.01.2008 Departure for Zurich , Switzerland Dr. Muhammad Idreessi
20.01.2008 Arrival at Dubai
21.01.2008 Arrival at Lahore
22.01.2008 Briefing to G. M. (TS) WAPDA Dr. Izhar ul Haq
On Training at Switzerland.

3
5. SEISMIC ANALYSIS SOFTWARE “SEISAN”

The SEISAN Seismic Analysis Software is a complete set of programs and a simple
database for analyzing earthquakes from analog and digital data. With SEISAN it is
possible to locate local and global earthquakes by entering phase readings manually or
pick them with a cursor, locate events, edit events, determine spectral parameters,
azimuth of arrival from 3-component stations and compute epicenters. The system consist
a set of programs tied to the same database. Using the search programs it is possible to
use various criteria to search the database for particular events and work with this subset
without extracting the events. Most of the programs can operate both in a conventional
way (using a single file with many events), or in a database manner.

The data is organized in a database like structure using the file system. The smallest basic
unit is a file containing original phase readings (arrival times, amplitude, period, azimuth,
and apparent velocity) for one event. The programs are mostly written in FORTRAN and
few in C. Some source codes are given, so the user may be able to fix bugs. The
programs have been compiled and linked with system compilers and linkers on SUN,
GNU compiler on Linux and MaxOSX, and Digital Visual FORTRAN 6.0 under
Windows. SEISAN runs under Sun Solaris, Linux, MacOSX, Windows95/98, and
Windows NT/2000. The SEISAN system is built of programs made by many different
individuals. It is freely available for all non-commercial use through Internet. However,
the main drawback of the program is that it is still under development stages, and its
seismological parameters locations dose not match with the International organizations
like ISC, USGS & EMSC.

6. SEISMIC ANALYSIS SOFTWARE “ANTELOPE”

Antelope is effective state-of-the-art real-time system software used for transporting,

processing and analyzing the seismic data. It took nearly 20 man-years to develop this
licensed product that is currently used by major data centers, such as IRIS and ORFEUS
in Germany, as well as by many seismic network operators around the world like USA,
UK, China, Russia, Australia, Austria, India, UAE and Saudi Arabia. Currently, there are
more than 60 licensed instances of Antelope running worldwide, including licenses at
30+ seismic networks/arrays.

Antelope has been used to create global scale "virtual" seismic networks that are
composed of many unique individual regional networks and single stations from global
networks. In order to integrate real-time environmental data flows, Antelope makes
extensive use of both public and private Internet facilities and the standard TCP/IP
communication protocols that assure a minimal impact upon the operations of the station
owners. Antelope provides important capabilities for environmental data integration and
data sharing that are relevant to all of the earth sciences.

As an example of the Antelope implantations, projects range from single station

configuration usage such as that in Romania to larger levels of > 1000 stations (5000+
channels) such as the UCSD IGPP Virtual Seismic Network (VSN) project
(http://eqinfo.ucsd.edu/vsn/index.html) or the IRIS-run US array of more than 400 Aspen

4
portable stations that send data in real-time using VSAT communications
(http://eqinfo.ucsd.edu/usarray/index.html).

Key Features:

 Open-architecture modular design concepts throughout

 Distributed real-time data acquisition and processing capability

 Distributed real-time system monitoring and control capability

 Comprehensive automated seismic event information in near real-time

 Network size independent - system scales only with hardware used

 Records data in real-time to non-volatile disk ring buffer

 Size of ring buffer limited only by maximum file size

 Real-time system uses client/server TCP/IP paradigm

 Supports all telemetry using either standard duplex serial interfaces or standard
TCP/IP Level 4 protocol over multiple physical interfaces

 Unique set of on-line and off-line processing tools

 Information system interfaces and functionality

 Offers tools with RDBMS for rapid access to earthquake information

 Provides a rich development toolkit

Highly configurable and adaptable to differing monitoring system requirements

Antelope Seismic Analyais Software is a distributed open-architecture UNIX based

acquisition, analysis and management software system designed to provide a
comprehensive set of environmental monitoring data and processed information in real-
time. Finely engineered as an all-in-one system, Antelope integrates state-of-the-art
technology and scientific advances, making it ideal for monitoring seismic events from
local/regional, national and global networks and arrays. Antelope takes full advantage of
the extensive support services provided by the UNIX environment and standard TCP/IP
network utilities over multiple physical interfaces.

Antelope consists of two major sub-systems:

 ARTS, the Antelope Real Time System

 AS IS, the Antelope Seismic Information System.

5
The current generation of Antelope provides full functionality for seismic network and
array operations and control, including real time data acquisition from field digitizers,
interactive control of field equipment, system state of health monitoring, real time
automated data processing (detection, picking, seismic event association, seismic event
location, archiving). It also offers interactive and batch processing, information system
functions, automated distribution of raw data and processed results, batch mode seismic
array processing and a powerful development toolkit for extending and customizing the
system.

The Antelope Seismic Information System uses the relational database (RDBMS)
formalism and the CSS v. 3.0 schemas for information organization. Antelope runs on the
Sun Microsystems' Solaris operating system on both SPARC and Intel architectures. In
addition to providing specific functionality for seismic monitoring systems, Antelope
provides a robust and versatile substratum of generic functions that can be used to
support other non-seismic monitoring applications.

ANTELOPE is a on-line licensed software of M/S Kinemetrics and its price for
installation and configuration is around U. S. $ 150000/-

7. MAIN POINTS OF THE TRAINING

1 During the visit to M/S Kinemetrics Switzerland, the WAPDA

Seismologists were trained to process the data through SEISAN and
ANTELOPE software.

The SEISAN is free Internet available software and has been recently
developed by University Professors & Students in Norway. It is still under
testing stages and not been adopted by the seismic observatories
throughout the world for data processing/analysis. Its location parameters
are sometimes sent to the University for Verifications etc. During the
training it was known that the data processing through SEISAN is manual.
The ten stations (3 components) data is required to be first divided into
daily and hourly basis. Later all the 30 channels are merged together for
event by event manual locations and magnitudes determination. It is a very
lengthy process. The main drawback of the whole process is that the final
results at this stage do not match with the results of other internationally
recognized software. Hence the results are non reliable and cannot be
useful for further Seismic Hazard Analysis.

On the other- hand the ANTELOPE is internationally established software

and most of the seismic observatories through out the world are adopting it
for data processing & cataloguing. Its results are accurate and do match
with the results of other international observatories like ISC, USGS &
EMSC. Main features have already been described above.

6
2 Seismograph stations of MMS at the Diamer-Basha Dam are not well
distributed for locations of epicenters in the southern-eastern side and one
to two stations (preferably six components) are required to be added.

3 The MMS was healthy, at the time of the last visit to replace batteries.
However, because of no inter-connection and central recording the current
status is unknown, until the stations are visited again.

4 The seismic data analysis and interpretation is done by two experts hired
by M/S Kinemetrics. Dr. Nikko in Austria and Dr. Mladen Zivcic in
Slovenia. They work together through Email system. They carry out the
analysis and interpretation through ANTELOPE Software, which is
available at their facilities in Austria and Slovenia. Results produced so far
by them are of International Standards.

5 The level of seismic activity is up to some extent adequate to allow

significant conclusions after first report of three months is issued. In three
months approximately 250 events have been recorded within the network.

6 The smallest magnitude for which events are located within the network
seems to be satisfactory.

7 The consultants of WAPDA have not procured the ANTELOPE Software,

which is required for processing and analysis of the Diamer-Basha Dam
Project seismic data.

8 Sometimes the data flow is not continuous at one or two stations because
of un-charged batteries. The Solar Panels have not been installed at the
field stations. This may result in the loss of valuable data specially when
the batteries are discharged and cannot be replaced immediately.

9. The results may be useful only if the hypocenter locations are precise and
the magnitudes are realistic and reliable. For identifying active faults the
epicenters should be located with errors not larger than about 2 to 3 km
within the network, and depths should have errors not exceeding about 5
km. The seismic velocity structure in the crust beneath and around the
Diamer-Basha site is not known. Thus, it is necessary that a local velocity
model is derived because otherwise the objectives of the Project cannot be
achieved. This can be carried out by using explosions in quarries at the
proposed Dam site, as well as by simultaneous inversion of hypocenters
and velocity model. The experts hired by M/S Kinemetrics are
experienced enough do this with the help of ANTELOPE Software.

7
 10. During the training there was a great deal of discussions concerning the
minimum magnitude that should be resolved within the network. On one
hand, one does not want to reduce the quality of the data by including
small events recorded by only a few stations. On the other hand, one
would like as much information as possible. As per standards adopted by
experts a minimum of six stations must have recorded event, in order to
include it in the analysis. After long discussions it was agreed to generate
a supplementary catalog of earthquakes recorded by only 4 or 5 stations
(that would have lower quality locations), and could be useful in defining
the frequency-magnitude relationship used in seismic hazard assessment.

11. The fact that some events with magnitude 1 have been located shows that
the noise level at the stations is low. It is difficult to ascertain without
information about the Local Attenuation Equations, a Local Magnitude
Scale that is equivalent to the International Magnitude Scales. Presently
Attenuation Equations derived for California are being used for
Magnitude determination. At this initial stage of the analysis with a
preliminary Scale established, it is not certain how close the results come
to be comparable to Scales elsewhere. However, this is a point that needs
attention. Calibrating the Scale by comparison to teleseismic magnitudes
may be useful, in addition to investigating the Local Attenuation
Equations.

12. The process of acquiring instrumentation, its development, operation and

extracting useful data is a significant investment. To stop the operation of
the network just after eighteen months dose not appear in the best interest
of WAPDA and Pakistan.

RECOMMENDATIONS

1. Solar Panels at all the Ten Seismic Stations may be installed to secure the power
supply. This is necessary to avoid the loss of data. Solar panels are standard
for supplying power to seismograph stations around the world.

2. The process of obtaining results by data analysis through SEISAN software is too
time-consuming and also the results are still non-reliable. WAPDA should
arrange for the procurement of configured ANTELOPE Software at the earliest.
Till than the processing and analysis of the Diamer-Basha Dam seismic data by
the team of Dr. Nikko and Dr. Zivcic be continued.

3. The ANTELOPE Seismic Processing and Analysis software (like most of the
modern seismic data processing and analysis software), operate on the UNIX &
LINUX computer systems. The Seismologists in WAPDA know only the
WINDOWS system and are not familiar with UNIX & LINUX systems of
computer operations. For the smooth operations of modern seismic analysis
software, the Seismologists of WAPDA should be provided training in UNIX &
LINUX operating systems.
8
4. It is essential that a local model for seismic velocities be developed. Not
including this, the results would be confusing. Kinemetrics team of Dr. Nikko
and Dr. Zivcic should be engaged in developing the velocity model and careful
calibration of the magnitude scale, which is at present not within their scope of
their work. The Kinemetrics team has the expertise and experience necessary to
cost effectively analyze the records gathered by the network and to produce a
high quality data catalog.

5. As per present level of recording at the seismic stations, the records will be
clipped for events of M > 4.0. In this case as per new International state of art
seismic monitoring systems, at least two six channel seismic stations (three
channels for seismometers & three channels for accelerometers), may be added
along the south-eastern side of the Diamer-Basha Microseismic Monitoring
System. These would be helpful in determining the orientation of active faults
and in the derivation of the acceleration attenuation equations.

6. In seismically active areas like Diamer-Basha where macro earthquakes are

relatively common, instant data retrieval and processing is required after a felt
earthquake, so that location and orientation of the fault systems are studied.
However, such provisions are missing in the present setup. It is required that all
the ten stations are inter-connected locally and then to a Central Station through
a network at Chillas and later all the data (as and when required) is transferred
to Tarbela or Lahore via Satellite for further final processing. Another main
benefit is that the data become available in real time. Besides the data flow is
faster and more consistent.
7. As per standard ICOLD recommendations and taking into consideration the
vicinity of high level seismicity, it is strongly recommend that the Diamer-
Basha Microseismic Monitoring System be operated for in-definite period of
time, best jointly with the Tarbela network.

8. The operation & maintenance including data acquisition & administration should
remain entrusted to the contractor M/S National Engineers. Under the situation
the contractor will remain responsible for proper O & M, timely data retrieval
even under difficult weather conditions; accuracy of the recorded data and
validity of the interpreted results. This will do away with the probability of data
loss due to skipped site visits by maintenance personnel. For administrative
control of all the seismic assignments at site one trained Senior Seismologist
may be posted at Chillas.

9
 DIAMER – BASHA DAM PROJECT
MICROSEISMIC MONITORING SYSTEM

10