next_inactive up previous


NanoseismicSuite manual


SeisServ, SonoView, TraceView and HypoLine


Benjamin SICK
benjamin.sick@geophys.uni-stuttgart.de

Partly based on HypoLine and SonoDet documentation from Manfred Joswig (manfred.joswig@geophys.uni-stuttgart.de)





Homepage http://www.nanoseismic.net


Contents


Quick start

Windows

Linux


Introduction

There are four main applications. The application server SeisServ and the three clients SonoView, TraceView and HypoLine which can be seen in figure 2.1. The client applications poll the server for data of a specific time and trace and synchronize the time of the currently selected data between each other.

Up to date information can be found on http://www.nanoseismic.net.

Figure 2.1: NanoseismicSuite
Image nanoseismicsuite


Installation and configuration

Windows

Figure 3.1: Optional installation of Java at setup.
Image setup_java

Linux


Data retrieval

Overview

Data is read in the Center for Seismic Studies (CSS) and MiniSEED format. In CSS data files must be referenced by an ASCII header file. It contains all necessary access information for each station and each component. The referenced time periods may range from some minutes for single events up to several hours or days for continuous recordings, there is no upper limit for the NanoseismicSuite. Furthermore parameters about the campaign and stations used for the data are stored in the knowledgebase.

We expect the signal output of geophones where voltage is proportional to ground velocity. Calibrating ADC counts to true ground velocity by the factor calib is based on a value which is specified in the plateau or at the corner frequency of the transfer function; in any case it must not be in the decaying part below the corner frequency. However, to conform to CSS header file definitions we demand the specification of a factor calibm [nm] which is valid for ground motion and got specified at period calper [sec]. The internally used calib is derived from calibm by multiplying the latter with $ 2\pi$ and dividing it through calper. When any of these two values is zero, no calibration is performed; instead digital counts are displayed. [Optional: calib *= lsbfak to correct where 1 ADC count >1.0 of unscaled data, e.g. Lennartz M24.]

Automatic generation

The header file can be created automatically from raw data in MiniSEED, SEG2 or Lennartz M24 digitizers and a default knowledgebase is created automatically on application startup if it does not exist. Changes to the knowledgebase can be made from the knowledgebase editor of SeisServ or directly in the corresponding XML files.

Furthermore IRIS (http://www.iris.edu) provides the tool rdseed (http://www.iris.edu/forms/rdseed_request.htm) which can convert SEED data records directly into the CSS format IRIS also provides multiple converters for ASCII, GGP, GSE, MARS, MT, SAC and SEISAN into the MiniSEED format (http://www.iris.edu/pub/programs/converters/) which in turn can be read by the NanoseismicSuite. This way it is possible to view almost all datasets in the NanoseismicSuite without the manual creation of a header file.

Header

The header file must contain information about start time, station and component, #samples, sampling rate, calibration factor and period, type of binary data representation, filename and path. The byte offset is used to store two or more waveforms in one file. This feature is mostly used for bridging shorter data gaps or compiling three-component and small array data. The NanoseismicSuite supports the decoding of header files conforming to the standards of the Center for Seismic Studies (CSS) - now: Center for Monitoring Research (CMR) in Arlington, VA. Two slightly different database structures, Version 2.8 and Version 3.0, are supported as well as our own, non-standard format derived from it. The latter is called Short Format and restricts the header content to just the necessary information. An overview of the different header file types is given in table 4.1.


Table 4.1: Header file formats
Type of header file Identifier in Batch File Header filename
CSS Version 2.8 "CSS vers. 2.8" *.c28
CSS Version 3.0 "CSS vers. 3.0" *.c30
Short Format "Short Header" *.sht


All header parameters are shown in listing 4.1. Table 4.2 shows an explanation of all short header parameters, 4.3 shows the supported binary file data types and listing 4.2 shows an example of all parameters. The entry for directory can be absolute as well as relative. All reference to time is exclusively done via epoch while the entry date is optional to ease human orientation. Some parameters in the original CSS header formats are of no interest here; they are marked by "-".


\begin{lstlisting}[caption=Header parameters, label=header_overview]
''CSS vers....
...bm, calper, dattyp,
directory, filename, byte-offset, [lsbfak]
\end{lstlisting}


Table 4.2: Parameters of header file
Name Description
date Start day, first 4 digits (year), following 3 digits (day of the year)
epoch Start time as unix timestamp (seconds after 1970/01/01)
stat Station name
chan Station's channel
nsamp Number of samples
samprate Sample rate
calibm Amplitude scale factor
calper Period scale factor
dattyp Data type, see table 4.3
directory Data directory
filename Data filename
byte-offset Byte offset in data file
[lsbfak] Lsbfak (optional)


Table 4.3: Possible values of parameter dattyp
Name Description Byte order Comment
i2 INTEL short integer Little endian CSS equivalent
i4 INTEL long integer Little endian CSS equivalent
s4 SUN long integer Big endian CSS equivalent
f4 INTEL single precision real Little endian CSS equivalent
t4 SUN single precision real Big endian CSS equivalent
d0 Ascii float numbers in columns   Offset: lines
m4 Mini-SEED format Big Endian Steim1-compressed 4kB blocks
m1 Mini-SEED format Big Endian Steim1-compressed 1kB blocks
m5,ms Mini-SEED format Big Endian Steim1-compressed 512B blcks
m42 Mini-SEED format Big Endian Steim2-compressed 4kB blocks
m12 Mini-SEED format Big Endian Steim2-compressed 1kB blocks
m52,ms2 Mini-SEED format Big Endian Steim2-compressed 512B blcks
nx Nanometrics X5 format Little Endian Steim1-compressed 4kB blocks
pd Geotech PDAS 100 Little endian 14+2 bit gain ranged


\begin{lstlisting}[caption=header.sht, label=listheader]
2007227 1187178104.000 ...
...z 720000 400.0 0.00185 1.0 i4 .\data\SNS1\data 08151141.CZS 240
\end{lstlisting}

Knowledgebase

Further meta data of a measurement campaign is stored in the knowledgebase. The parameters of the knowledgebase are stored in the following two XML files in the subdirectory knowbase below the directory of the header file:

  1. campaign.xml for campaign related information as start, end and campaign name
  2. stations.xml for stations related information as names and locations of stations and which stations form a SNS

Campaign knowledge

In table 4.4 are all campaign parameters explained. In listing 4.3 is the source of an example campaign.xml file with all parameters.


Table 4.4: Elements of campaign.xml file
Element name Description
campaign name The name of the campaign
campaignstart The date and time of the campaign start (format: yyyy/mm/dd hh:mm:ss)
campaignend The date and time of the campaign end (format: yyyy/mm/dd hh:mm:ss)
globalsamplerate The sample rate which should be used in all applications
latitude The latitude of a reference point used later for the positions of the stations
longitude The longitude of a reference point used later for the positions of the stations
height The elevation of a reference point used later for the positions of the stations
halfsize Scale of epimap in HypoLine
x-offset The longitude offset for the campaign location
y-offset The latitude offset for the campaign location
layerModel name The name of the layer model
layerModel ratio P-velocity to V-velocity ratio of layer model (vP/vS)
layer vP P-velocity of layer
layer d Thickness of layer
layer wavetype Wavetype of layer (Pg, Pn, Unknown)

language=XML
\begin{lstlisting}[caption=campaign.xml, label=listcampaign]
<?xml version=''1.0...
...
<wavetype>Pn</wavetype>
</layers>
</layerModel>
</campaign>
\end{lstlisting}

Stations knowledge

In table 4.5 are all stations parameters explained. In listing 4.4 is the source of an example stations.xml file with all parameters.


Table 4.5: Elements of stations.xml file
Element name Description
sns id The SNS name
north,west,east,center id The station name (must be the same as stat from header file)
x The longitude position relative to the referencepoint + offset
y The latitude position relative to the referencepoint + offset
z The elevation relative to the referencepoint
static Not used yet
clock Not used yet
traceZ The vertical component of a station
traceEW The horizontal east-west component of a station
traceNS The horizontal north-south component of a station
active switch to turn station/SNS on (true) or off(false) in SonoView

language=XML
\begin{lstlisting}[caption=stations.xml, label=liststations]
<?xml version=''1.0...
...</active>
</center>
<active>true</active>
</sns>
</stations>
\end{lstlisting}

Data import

If data is already available in the CSS format, the header file can be directly loaded from within SeisServ with the Load Data button. If data is available in MiniSEED, SEG-2 or Lennartz M24 data format, a header can be created within the same dialog. For further information see the SeisServ section 5.0.1.


Usage of the software

After starting the software, SeisServ starts up as shown in figure 5.1. After start-up data can be loaded from the database or a header file can be loaded with the button Load Data.

If the data is successfully loaded, SeisServ shows for every trace the name, start and end time as in figure 5.2.

After loading data, SonoView can be started from Menu $ \rightarrow$ Tools $ \rightarrow$ SonoView. SonoView is shown in figure 5.3. From within SonoView, TraceView can be started from Menu $ \rightarrow$ View $ \rightarrow$ Show TraceView. TraceView is shown in figure 5.4. Both SonoView and TraceView are by default started automatically. HypoLine can be started from Seisserv Menu $ \rightarrow$ Tools $ \rightarrow$ HypoLine and is shown in figure 5.5.

Configurations of all applications are saved in the user directory of the current user in a subfolder nanoseismicsuite.

Figure 5.1: Start screen of SeisServ.
Image seisserv_start

Figure 5.2: SeisServ with loaded data.
Image seisserv_loaded

Figure 5.3: SonoView
Image sonoview

Figure 5.4: TraceView
Image traceview

Figure 5.5: HypoLine.
Image hypoline_socket


SeisServ

SeisServ reads seismic data and meta-data, it provides this data to the other modules and allows editing of the meta-data, e.g. the geometry of seismic stations.

SeisServ supports two modes of header loading, database and file based. Furthermore, headers for raw data in MiniSEED, SEG2 or Lennartz M24 digitizers can be created. All of the above are available from the "Load Data" dialog.

Figure 5.6 shows SeisServ with some explanations. In figure 5.7 and figure 5.8 the campaign and stations knowledge base editor is shown. SeisServ provides data for SonoView and TraceView and manages the knowledge base. It is possible to change or create a knowledge base in SeisServ with the knowledge base editor accessible from the button Knowledgebase.

Figure 5.6: SeisServ with explanations.
Image seisserv_explained

Figure 5.7: Campaign knowledge base editor.
Image knowledgebase_editor_campaign

Figure 5.8: Stations knowledge base editor.
Image knowledgebase_editor_stations

Loading from database

Configure and test connection as shown in Figure 5.9 with appropriate values.

Explanation of options is as follows:

Figure 5.9: SeisServ preferences dialog with example database configuration.
Image seisserv_db1

Load data as shown in Figure 5.9. The data from the database can be filtered by dates, campaigns and stations:

Figure 5.10: SeisServ database loading dialog.
Image seisserv_db2

SonoView

After loading the data, SonoView is the first application to use in a typical event screening scenario. It visualizes super-sonograms in a manner to maximize the visible data on one screen. An arbitrary amount of SNSs and time spans can be loaded. An analyst can scroll fast through the continuous data in SonoView and mark suspicious events for further processing steps. Figure 5.11 shows SonoView with some explanations.

Figure 5.11: SonoView with explanations.
Image sonoview_explained

Preferences

The preferences dialog can be reached from the Main Menu $ \rightarrow$ File $ \rightarrow$ Preferences. It is shown in 5.12.

Figure 5.12: SonoView preferences.
Image sonoview_pref1 px Image sonoview_pref2

The preferences are:

View

The View options are in the Main Menu $ \rightarrow$ View.

All options can also be called with keyboard shortcuts which are also shown in the help menu (reachable by F1), see figure 5.13.

Figure 5.13: Sonoview help menu.
Image sonoview_help

TraceView

Detected events from SonoView can be further analyzed in TraceView which visualizes the seismograms of these events together with a map of the measurement area with locations of the SNSs. It provides a two-dimensional neighbourhood overview of SNSs which can not be provided by the one-dimensional listing of SonoView. TraceView shows the seismograms of the currently selected SNS and the five adjacent SNSs based on geographic context. Basic filters and scalings can be applied to the seismograms and geo-referenced images of the measurement area can be shown in the map (e.g. satellite images). Figure 5.14 shows TraceView with some explanations.

Figure 5.14: TraceView with explanations.
Image traceview_explained

All options can be called with keyboard shortcuts which are shown in the help menu (reachable by F1), see figure 5.15.

Figure 5.15: TraceView help menu.
Image traceview_help

HypoLine

HypoLine allows a detailed analysis of events. It is used for the localization and magnitude estimation. Localization is done by time difference of arrival (TDOA) hyperbolae and S-P distance circles based on one-dimensional velocity models. HypoLine supports at the moment the processing of data from up to six SNSs which it gets from TraceView. This subset of SNSs is no restriction for weak events because they are anyway only visible at the surrounding stations. For further processing of single events tools as e.g. Geotools, Seisan, Pitsa or SeismicHandler can be used. HypoLine allows a first coarse localization and identification with interactive and graphical techniques for very weak events. Screenshots are given in figure 5.16, 5.17 and 5.18.

All options can be called with keyboard shortcuts which are shown in the help menu (reachable by F1).

For further information on HypoLine:

Figure 5.16: Screen layout of HypoLine showing a candidate event at the threshold of processing capabilities. The seismograms were acquired by the four SNS stations sketched in the zoom map.
Image hypoline1

Figure 5.17: Processing results for the candidate event of figure 5.16. The sonograms guide the phase picking for the four weak onsets, the jackknifing gives four triple junctions (red dots in the zoom map). For the adjustment of the epicenter, additional information from the tS-tP circle (dotted green circle segment) and the two array beams for P and S onset (yellow fans) is considered.
Image hypoline2

Figure 5.18: Application of jackknifing to beamforming for the candidate event of figure 5.16 The four Pc and Sc phases determine four tri-partite beams for P and S. Their spread is extremely sensitive again misadjustment. The lower two traces display beam overlays according to the selected time differences, the map of hypoline indicated accuracy by width of the yellow beam fans. Note the highly reliable results despite the low SNR at single traces.
Image hypoline3


Nanoseismic Monitoring Fundamentals

SNS - Seismic Navigation System

In figure 6.1 the typical layout of one SNS is shown. One three-component station in the center and three one-component vertical stations deployed as a tripartite array. This guarantees the optimal recording.

Figure 6.1: Layout of SNS
Image sns

Sonograms

Sonograms are spectrograms of seismic signals with the following properties:

In figure 6.2, the processing steps of a sonogram are shown. The horizontal axis is the time domain and the vertical axis is the frequency domain. The frequency domain is logarithmically scaled. As the spectrogram is a 3-dimensional plot, the power of each pixel is shown as a coloring based on the shown color map.

Figure 6.2: Composition of sonogram and color map
Image sonosteps

Super-sonograms

SonoView shows super-sonograms, which are explained in figure 6.3. The super-sonograms combine all four vertical traces of one SNS into one diagram. It uses each pixel of the normal sonograms to form a super-pixel. With these super-pixels a new sonogram is built. This allows to see the recording of one SNS in one row and check on array wide signal coherency.

Figure 6.3: Composition of super-sonogram
Image supersonocompilation

About this document ...

This document was generated using the LaTeX2HTML translator Version 2012 (1.2)

Copyright © 1993, 1994, 1995, 1996, Nikos Drakos, Computer Based Learning Unit, University of Leeds.
Copyright © 1997, 1998, 1999, Ross Moore, Mathematics Department, Macquarie University, Sydney.

The command line arguments were:
latex2html -dir html -split 1 -title 'NanoseismicSuite Documentation' -address 'Benjamin Sick (benjamin.sick@geophys.uni-stuttgart.de)' document_png.tex

The translation was initiated by on 2015-02-23


next_inactive up previous
Benjamin Sick (benjamin.sick@geophys.uni-stuttgart.de)