WP 2.2 Possible precursory seismic quiescence and b-value changes

Start date or starting event:

M0

Lead contractor:

WAPMERR

Participants:

WAPMERR

1

Objectives: Determine if the June 2000 earthquakes could have been predicted by precursory seismic quiescence or b-value changes, and what kind of seismicity patterns followed these main shocks.

2

Methodology/work description: Map the changes in seismicity rate and in the magnitude frequency distribution as a function of space and time in the South Iceland seismic zone, and evaluate the possible relations of such changes to the M=6.6 main shocks in June 2000.  It has been shown that seismicity rates and the b-values of the frequency-magnitude relationship can be changed for years by a major earthquake.  Such co-seismic changes are clearly due to a redistribution of stress by the major earthquakes, and can be correlated with the changes in the Coulomb fracture criterion.  Precursory changes in seismicity rates also have been documented in many outstandingly strong datasets, but their causal connection to the following main shock is only suggested by the coincidence of the two phenomena in time and space.  Precursory decreases in the b-value have been reported by several authors, but truly compelling cases are rare.  A co-seismic change was demonstrated with statistical significance in the M=7.2, 1992 Landers earthquake.  Close cooperation is planned with WP 2.1, but these two packages complement each other. We plan the following steps of analysis: (1) Quantitative analysis of the minimum magnitude of completeness as a function of space and time in the South Iceland seismic zone, together with an analysis of the homogeneity of reporting as a function of time.  This type of analysis will take some time and energy, but is necessary to ensure that artificial reporting rate changes do not influence the results. The result of this first step in our analysis is a catalogue that (a) starts at the earliest practical time, (b) extends to the mapped limits of an optimal minimum magnitude of completeness, and (c) is corrected for possible magnitude shifts that may have occurred as a function of time. (2) To measure seismicity rate changes as a function of time, earthquake swarms and aftershock sequences must be removed from the catalogue.  Usually, we use the parameters in the Reasenberg  algorithm, which were derived for California.  However, it could be that these are not adequate for the seismicity in Iceland, in which case it will cost time and effort to derive the appropriate constants.  The product of this second step is a declustered catalogue and a catalogue of clusters. (3) In a third step we will map seismicity rate changes as a function of time.  (a) We will examine how  the June 2000 main shocks have turned off and turned on seismicity in various volumes around them.  (b) We will map any possible seismic quiescence that may have preceded these main shocks and we will search all time and space covered by the catalogue for possible other periods of quiescence that might exceed in significance the precursory quiescence.  Quiescences near main shocks can only be claimed as possible precursors, if they exceed or at least equal in significance possible periods of quiescence not associated with main shocks.  The product of this phase will be case histories of seismicity rate changes related to main shocks in Iceland. (4) We will also construct maps of b-value changes with time and of b-value differences as a function of space.  Changes with time can seldomly be documented, thus we do not necessarily expect to find them.  However, b-value differences in space are ubiquitous and they seem to be related to asperities in fault zones.  We propose to test the hypothesis that in Iceland also, volumes that generate major earthquakes are special in that they show low values of the quantity defined as ‘local recurrence time’ (which is inversely proportional to the annual probability of a main shock of a given magnitude).  Up to now, correlation in minima in local recurrence times with main shock locations and asperities identified geodetically exist in central, southern and northern California as well as in Japan and Mexico.  The product will be an evaluation of the hypothesis that asperities may be mapped by minima in local recurrence times in Iceland.  (5) The final step will be modelling of the seismicity rate changes observed.  The basis of the models will be changes in the Coulomb fracture criterion.  We will seek to correlate such changes due to the main shocks with co- and post main shock rate changes.  This work will be done in a different work package.

3

Deliverables including cost of deliverable as percentage of total cost of the proposed project: 

4

Milestones: Delivery of the above items at the date indicated.