Start: June 1996 (month 4)
End: February 1998 (month 24)
Responsible partner: IMOR.DG
Cooperative partner: UUPP.DGEO
The basic option of the SIL seismic system techniques is to use microearthquakes to bring to the surface information from the source areas of earthquakes. Based on detailed microearthquake analysis it is possible to monitor active faults and movements on these, as well as stresses and stress changes in their surroundings. The smaller the earthquakes are which can be used the closer we are to continuous monitoring of such features, and the more detailed information we obtain of the spatial conditions. Therefore it is so significant to be able to obtain automatically as detailed and secure information as possible [41,,].
A new software tool is being developed, ACIS, to be introduced into the automatic procedures of the SIL system. ACIS is an acronym for Reducing manual checking by Automatic Correlation of Incoming Signals. As has been shown in the work on multievent analysis for detailed mapping of faults most seismic events correlate well with each other within some areas. A geographically indexed database is being created where different classes of earthquakes are stored. As new earthquakes are recorded by the network, the system automatically looks for similar waveforms in the database, and if found, takes the onset and the first motion direction picks from there. If no existing entry in the database correlates with the new event, the event is checked interactively by the network operators. This approach will improve the accuracy of the automatic analysis and reduce the need of work for interactive checking of the data without loss of useful signals. Preliminary testing has demonstrated that the approach described here is possible. It is expected that the first version of the algorithm will be ready for testing within the SIL system in October 1998. It is the intention that after testings it will be introduced in the routine procedures of the SIL system and thus become a basis for enhanced alert detector algorithms. One of the consequences of a more accurate real-time hypocenter location is that real-time fault plane solutions will be much more reliable, making real-time monitoring of stress changes from microearthquakes possible in practice [].
Work has been carried out for studying and refining the alert thresholds for the SIL related alert system in Iceland. An alert detector monitoring directivity, large amplitudes and background noise in both unfiltered and filtered bands of the seismic waveform data is operated on all the SIL stations. It has been tuned for different types of sensors as the SIL system operates according to need with 1 second, 5 seconds and broadband sensors [6,7,42].
The continuous seismic signal at the SIL site stations is bandpass filtered in three channels and the 1 minute mean amplitude is calculated and sent to the SIL center, where the interpretation of characteristics of the tremor is carried out and linked to the alert system. These frequencies show to be useful in discriminating noise of different origin. The lower frequencies are typical for harmonic volcanic tremor, while the highest frequency seems to be expressing noise created by very intensive activity of very small earthquakes, although these are not discriminated as such. Such an activity is more typical in the approaching of an eruption and may possibly be of significance in the introductionary phase of earthquakes. Much work remains to be done to analyze the noise and how it is related to other activities of the crustal forces. This noise monitoring is already now used to monitor volcanic activity [].
The experience described above also provides a good basis for ongoing work in designing a new detector in the SIL system which will be aimed at detecting and automatically evaluating complicated earthquakes (i.e. slow quakes having corner frequencies of the order of 1 Hz or earthquakes which appear as composed of low and high frequency motion) which are often observed in Iceland, especially from areas close to the volcanoes. A graphic tool has been developed to help visualizing the effects of changing the triggering parameters and algorithms of the detector. This tool is very significant and efficient for the ongoing development of the new detector algorithms. This work is carried out in close cooperation with UUPP.DGEO.