|Observatories and Research Facilities for EUropean Seismology|
|Volume 2, no 2||August 2000||Orfeus Newsletter|
Broadband Stations in the Hashemite Kingdom of Jordan
In 1983 the Government of Jordan established a seismological observatory to monitor earthquakes in Jordan and adjacent areas to assess hazards posed by seismic events in the region. The project initially was funded jointly by the government of Jordan and the US Agency for International Development. Eight short-period stations were deployed initially in the vicinity of Amman under a joint working arrangement between the Natural Resources Authority (NRA) of Jordan and the US Geological Survey (USGS). Subsequently, the network was expanded in several stages to its current configuration of 26 short-period stations covering the entire country. This expansion was undertaken as part of the Program for Assessment and Mitigation of Earthquake Risk in the Arab Region (PAMERAR), with funding from the Government of Jordan, the Arab Fund and the Islamic Development Bank. At present, the National Seismic Network is operated by the Jordan Seismological Observatory (JSO) from its headquarters at the NRA in Amman. The map of Figure 1 shows the distribution of stations in the network.
In 1998 the NRA, the USGS and the University of California Lawrence Livermore National Laboratory (LLNL) entered into an agreement to deploy two broadband seismic stations in Jordan. The purpose of this addition to the National Seismic Network was to obtain long-period constraints on earthquake source mechanisms in the Dead Sea rift (e.g., moment magnitudes, moment tensor descriptions) and other regions of the Middle East, to characterize regional seismic wave propagation and to improve estimates of crust and upper mantle structure in the region. The two stations are deployed at Hittiyah (HITJ) in the southern part of the country and Ruweishid (RUWJ) in the eastern part of the country (see Figure 1).
The instruments deployed at the sites initially were Guralp CMG-ESP 30-second 3-component broadband seismometers recorded by Reftek 72A-07 24-bit data acquisition systems synchronized to UTC with RT-111A GPS clocks. Figure 2 shows a general view of the Hittiyah site (the vault is below ground just out of sight to the right). Figure 3 shows the deployment of the instrument and recorder on the pad within the Hittiyah vault. The locations of the stations are provided in Table 1. Signals from the seismometers initially were sampled at 100 samples per second. After approximately six months of recording it became apparent that there was little usable signal energy above 20 Hz, and the sampling rate was reduced to 50 samples per second. In November 1999 the CMG-ESP instrument at the Hittiyah station was replaced with a 100-second CMG-3T. The data are recorded locally to 1 GB SCSI disks. The disks are swapped during service visits every 2-4 weeks and returned to the JSO headquarters in Amman where the data are downloaded to a SUN UltraSparc workstation for analysis.
Table1 Locations of the Broadband Stations
The Hittiyah station is favorably placed to study events in the Gulf of Aqaba. It lies to the east of the Dead Sea rift; data from this station combined with data from station EIL on the west side of the rift should permit interesting studies of the effects of the rift on propagation, and should provide good constraints on mechanism through long-period full-wave synthetic modeling. Hittiyah also is in a position to observe explosions from the large phosphate mines of central and southern Jordan. It is just 37 kilometers from the large mine at Shediyah. The Ruweishid station is in an area not covered by any other broadband stations. It is in a good location to constrain the structure of the northern Arabian plate, and to observe the less frequent seismicity of eastern Jordan.
Figure 4 shows sample recordings of large earthquakes made with these stations. The events shown are the June 27, 1998 Adana, Turkey (mb 6.3) earthquake, a magnitude 5.3 earthquake north of Egypt (May 28, 1998), both well recorded at Hittiyah, and a magnitude 6.6 earthquake in southern Iran (March 4, 1999) recorded at both Hittiyah and Ruweishid. The data in Figure 4 have been filtered into the 10 to 50 second period band to emphasize the broadband nature of the recordings.
The long-period data may be used with waveform modeling either to invert for velocity structure or to invert for event parameters (depth, mechanism, moment). An example of inversion for event parameters is shown in Figure 5. Here, synthetic seismograms have been computed for comparison with the Hittiyah recording of the Egyptian earthquake. The velocity model has been held fixed to a model suitable for the region, but the depth, mechanism and moment of the source have been varied to obtain a good match between the synthetic seismograms and the observed data. The lowest residuals occur for a depth around 25 kilometers. This depth and the magnitude (Mw =5.3) estimated by modeling the waveforms from a single station compare well with the Harvard CMT values (which are a network estimate). This example shows the value of even a few broadband regional stations for interpreting larger seismic events. For large, well-characterized events it is possible to estimate the (1-D) velocity structure for the path by holding the source parameters fixed, and varying the layer velocities defining the model until a match is achieved between the synthetic seismograms and the data.
At the short-period end of the spectrum, the data are useful for detecting and characterizing small local and regional events. Figure 6 shows a collection of observations, made at Hittiyah, of explosions probably occurring at the Al Hasa phosphate mine. These observations were assembled using a waveform correlation and clustering technique for identifying populations of similar events. Explosions, once attributed to the originating mine, can be used to constrain travel-time models for purposes of locating natural seismicity.
In general, the data from these new stations should provide new insights into regional structure, and, perhaps, contribute to global tomographic studies. The data are suitable for receiver function analysis to constrain crust and upper mantle structure. Continuous data for recording periods during 1998 and 1999 have been made available to GEOFON for distribution.
AcknowledgementsFor the Lawrence Livermore National Laboratory authors: This work was performed under the auspices of the U.S. Department of Energy by the University of California Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48. Manuscript no: UCRL-JC-139887