Seismic Aftershock Monitoring Network Optimization based on. Detection Threshold Estimation from Background Noise Measurement. Overview. Mini-Array ...
Institute for Geophysics University of Stuttgart Germany
Seismic Aftershock Monitoring Network Optimization based on Detection Threshold Estimation from Background Noise Measurement 1
Institute for Geophysics University of Stuttgart, Germany
T3-P110
1
2
Benjamin Sick , Nicolai Gestermann and Manfred Joswig
SAMS Network Configurations
Overview The Seismic Aftershock Monitoring System (SAMS) is an important method during the initial period of an On-site Inspection (OSI) to identify the possible source area of an underground nuclear explosion (UNE). A network of tripartite mini-arrays and single three-component seismic stations will be deployed during an OSI to detect and localize aftershocks in the vicinity of a possible explosion (Sick et al., 2012). The threshold of a seismic network depends on many aspects such as configuration, data quality and site conditions of each individual station. During an OSI a tradeoff between fast station deployment and precise site analysis has to be made. A first rough site characterization is possible with information about geology, local facilities and infrastructure e.g. roads. An individual characterization needs additional information to be acquired by field measurements. A method with a visualization tool will be presented which can be integrated into the SAMS and allows an inspector to estimate the detection threshold of the inspection area and to adapt the network configuration to the needs by densifying the network or relocating stations. A threshold reduction at locations of interest can be implemented and information about station data quality assists an inspector to focus the work on sites with optimum conditions.
1
Federal Institute for Geosciences and Natural Resources GEOZENTRUM HANNOVER
2
Federal Institut for Geosciences and Natural Resources, Hannover, Germany
Network Configuration 1
Network Configuration 2
We tested our algorithm for two different synthetic network configurations of mini-arrays for a Seismic Aftershoc Monitoring System (SAMS) during OSI within a fictitious rectangular inspection area (IA) of 1000 km2 size with side lengths of 35 km and 28.5 km. A rectangular restricted-area site (RAS) with side lengths of 2 km and 1.7 km was included. It was assumed that an array deployment is not possible within this RAS. The first network configuration (Network Configuration 1) covers the whole IA with 50 miniarrays (blue triangle) and a minimum distance between arrays of about 5 km. The second example network configuration (Network Configuration 2) includes 11 mini-arrays within the inspection area. 8 mini-arrays are concentrated in the vicinity of the RAS. The remaining 3 mini arrays are distributed over the IA. The noise levels at the mini-arrays are selected randomly with values between 0.5 and 51 nm/sec based on experiences from previous campaigns. A minimum of one detection at three mini-arrays was selected as the localization rule for both examples.
Location Threshold Configuration 2
Location Threshold Configuration 1
Background Noise Measurement
The equidistant distribution of miniarrays result in large variations of the location thresholds over the IA (red rectangle) depending on the noise conditions at the array sites. The northwest region of the IA including the RAS (black rectangle) would have a worse detection threshold than the remaining area. This network design is reasonable if no area of special interest exists and if no further information about noise conditions is available. A hexagonal network design could improve the location capability for the same number of mini-arrays.
Seismometer of mini-arrays can only be deployed at the surface because of treaty restrictions and time limitations. The noise conditions can vary considerable. The detection threshold will be estimated from the mean value and standard deviation of the absolute waveform data from a noise window (green line at the figure on the left). The maximum amplitude (red line) will be used to calculate an additional quality measure. Magnitude distance relation is calculated for the local magnitude ML (Gutenberg and Richter, 1956) with extension for small magnitudes (see figure below).
This network configuration is focused on the RAS as an area of special interest within the IA (red rectangle). The limited number of mini-arrays can be deployed during the first days of an OSI. The arrays in the vicinity of the RAS (black rectangle) lead to a low location threshold better than -0.9 ML. The good detection capability within the IA can be achieved with only 3 additional arrays at selected sites with a low noise level.
0,0
Ml
single station DE04 results 1,0
SNS actual DE04 results
yield 600 g
IA SNS typical night time performance
75 g
location thresholds: -1.97 - +0.07 ML
RAS location thresholds: -1.30 - -0.90 ML
RAS location thresholds: -0.30 - -0.15 ML
2,0
Example Noise data with calculated noise level (green line) and maximum amplitude (red line).
IA
location thresholds: -1.05 - +0.16 ML
(Joswig, 2008) 3,0 0,1
0,3 3000 1000
1,0 300
100
3,0 30
10
10,0 3
distance [km] 100,0 no of network stations
Location Rules The use of arrays for SAMS offers a variety of possibilities for the epicenter determination. Two types of phase onsets are assumed which require different signal to noise ratios (SNR) to fulfill the location process requirements : · Onset type 1 (OT1): arrival time can be determined at one array element (SNR ≥ 1.5) · Onset type 2 (OT2): Coherent energy (all array elements) to determine arrival time, back-
azimuth and apparent velocity of phase onset (SNR ≥ 3.0) Basic Location requirements · Two phases (P and S) of type OT1 at two mini arrays · Two phases (P and S) of type OT2 at one mini array · One phase (P or S) of type OT2 at two mini arrays · One phase (P or S) of type OT1 at three mini arrays Reliable depth estimation requires additional phase information.
Conclusions
Mini-Array Design Each mini-array consists of a threecomponent central station and three one-component satellite stations as shown on the figure on the right. Mini-arrays are optimized for fast deployment and can run many days from a single battery but still a detection of an event at only one mini-array provides the capability to locate it. The aperture of a mini-array is mainly predetermined by cable lengths and the time limitations for deployment.
Satellite Array Element Vertical-Component Seismometer
Central Array Element 3-Component Seismometer 100 m
· The noise conditions at the single mini-arrays have a significant impact on the detection and location capability of the network. · A few mini-arrays at sites with low noise conditions and good coupling to the ground are more important for the detection and location capability of the SAMS network than a large number of mini-arrays at sites with high noise levels. · The location capability of network configuration of the Seismic Aftershock Monitoring System (SAMS) of an OSI can be improved significantly if the knowledge about the noise conditions at the individual sites is taken into account. · A tool to estimate the detection and location threshold of individual mini-arrays of a SAMS network is crucial for planning and conduct of an OSI mission. · Results from this study have to be verified with real data from previous OSI exercises. References Gutenberg, B., and C. F. Richter (1956). Magnitude and energy of earthquakes. Annali di Geofisica, 9, 1-15. Joswig, M. (2008). Nanoseismic Monitoring fills the gap between microseismic networks and passive seismic. First Break. 26, 121-128. Sick, B., Walter, M. and Joswig, M. (2012). Visual Event Screening of Continuous Seismic Data by Supersonograms. Recent Advances in Nuclear Explosion Monitoring Vol. 2 - Pure and Applied Geophysics, PAGEOPH. Doi:10.1007/s00024-012-0618-x.
