Earthquake Monitoring

• Seismic Gap Theory:

  • seismic gap is segment of active fault thats not slipped for a long time compared to other segments along same structure
  • over long periods of time, displacement on any segment must be equal to that experienced on other parts of fault
  • large/long-standing gap is considered to be fault segment most likely to suffer from future earthquakes
  • eg. Parkfield, California on San Andreas Fault; 1857-1966, 6 magnitude 6 earthquakes occured every 22yrs so predicted event between 1983-1993
• Seismic Velocity:

  • P-wave velocites from minor tremors (<2) decrease some time in advance of major events then rise immediately before it (duration of P-wave anomaly related to magnitude)
  • from evidence a graph can be drawn to predict magnitude of event, eg. 1973 Blue Mountain Lake, New York - may not be accurate as only collectes limited evidence
• Seismic Velocity:

  • water saturated rocks under stress increase in volume before fracture (opening up of micro-cracks increase porosity)
  • water migrates into micro-cracks/water pressure reduces temporarily reducing P-wave velocities until water moves in rock restoring porewater pressure
  • length of time to restore pressure indicator of magnitude + major earthquake occurs shortly after porewater pressure restored
• Pattern of Foreshocks:

  • foreshocks are tremors of at least M2.5 on Richter-Scale and take place prior to mainshock M5.4 or greater on Richter-Scale
  • 50% of major earthquakes preceded by foreshocks + 5-10% of small earthquakes are foreshocks (lead to false warnings)
  • East Pacific Rise transform fault show foreshock activity before main event
  • eg. increase in foreshock activity enabled successful evacuation of 1 million people day before Feb 4th 1975 M7.3 Haicheng Earthquake by China State Seismological Bureau
• Groundwater Levels:

  • gradual lowering of water levels over a period of months/years before a seismic event
  • accelerated lowering of water levels in final few months/weeks proceeding earthquakes + rebound where water levels increase rapidly few days/hours before main shock
  • water levels need to be adjusted for tidal cycles/seasonal variations in water abstraction
  • eg. China, 100+ wells >1000m deep to moniter groundwater
• Groundwater Levels:

  • eg. Tangshan Mine; 1923-73, stable groundwater level + 1973 onwards, pumping rate decreased + 2 days-3hrs before M7.6 event (killed 240,000-650,000) pumping rates suddenly increased from 25 to 75m$^3$/sec$^-1$
• Groundwater Temperatures:

  • water temperature monitered to +/- 0.0001$^oC$ accuracy (deep wells suitable as protected from rainfall/seasonal effects)
  • eg. day before 1923 Great Kanto Earthquake, Japan (M7.9), geysers became active at Atami Hot Springs
  • in Japan, monitering at USA Volcanic Region showed 0.3$^oC$ increase over 6months then decrease immediately before seismic event
• Groundwater Chemistry:

  • chemical composition of groundwater affected by seismic events; levels of chloride/sulphate in solution monited in Kobe June 1993-Jan 1995; June 1993-July 1994, chloride/sulphate levels constant (13.7-14.1ppm) + July 1994-Jan 1995, steady rise to 15ppm + Kobe Earthquake (M7.2) occurred 17th Jan 1995
  • not useful for predictionas levels didn't peak until end of Feb 1995, decreasing through March to former levels
• Ground Deformation:

  • prior to earthquake, ground is uplifted/tilted due to swelling of rocks caused by strain building up on fault
  • involves measuring small changes in ground level using Tiltmeters, Strainmeters, Creepmeters, GPS and Laser EDMs (Electronic Distance Measurers)
  • eg. Rapel Reservoirs, Chile 3rd March 1985 (M7.9); 8 months prior, levels measures show differences due to tilting (maximum tilt 13cm over 20km baseline)
• Radon Gas:

  • stress released as micro-fractures open in rocks at depth (groundwater flows in) - radon trapped within rock escapes through micro-fractures into groundwater solution or as CO2 released from stresses it acts as carrier for radon
  • increased levels of radon gas precursor to earthquake - can be monitored in deep springs/wells (short half-life so unlikely to seep to surface from rocks at depth)
• Radon Gas:

  • eg. Kobe, Japan (M7.2); during 1993, radon gas levels stable at 20Bq/L + end of Nov 1994, levels increased to 60Bq/L + Jan 1995, rapid increase to 250 Bq/L, levels dropped to 30 Bq/L 7 days before event
• Electrical Resistivity:

  • is the resistance to flow of an electrical current; rocks poor conductors of electricity (water more efficient)
  • micro-cracks develop, groundwater flows into these cracks causing electrial resistivity to decrease (conductivity increases)
  • 5-10% drop in resistivity observed prior to earthquake + values rapidly return to normal after seismic event