Geotechnical Hazards L2

In fault networks, several faults can accommodate deformation. In earthquakes, stresses induced can increase in nearby faults bringing them close to failure or they can reduce, making the faults move away from failure risk.

This can be modelled with the kinematic source model

The seismic moment is considered the energy release of an earthquake in Nm

or 

the mu term is known as the shear rigidity and is often known as 3.3x10^10N/m2

It is a way of quantifying earthquake size

Cut off happens when the frequency the seismograph measures to relate it to the moment magnitude falls below the corner frequency - it is beyond this frequency that seismographs can't observe the changes in source strength

Fault - 3D surface representing zone of weakness in rocks.

Faults can be split into segment areas, with areas split and made up of segments according to barriers in rupture propogation or changes in geometry. Combinations of these segments can define ruptures, with the magnitude proportional to the panel area.

An area associated with different level of seismic activity, magnitudes of earthquakes etc., but with a uniform probability of earthquake occurrence. 

Area source can be split into cells with possible ruptures mapped out according to the known characteristics of the site.

Represented at a single location

Could be used to model small area faults, or where the closest distance of a rupture to a site is needed and this will not change. Can be computationally more efficient to use points.

Areas associated with lower levels of seismic activity which can't be discounted. 

Source scaling can relate area to magnitude.

i.e. 

or vice versa. Even rupture length can relate to magnitude

Rupture width down the fault dip is limited due to the depth of the seismogenic layer.

This leads to a break in scaling the area to the magnitude (the magnitude levels out)

It is often difficult to know if the maximum displacement is observed - many observations are needed for this. This is why approximations are used often relating u and x.

Epicentral - 2D dist (x,y)

Hypocentral - 3D dist (x,y,z) between site and earthquake

magnitude

strike, dip, depth

fault mechanism

In 1D only magnitude and epicentre/hypocentre needed

If a kinematic distribution is used, coupled with waveform modelling, we need:

magnitude

strike, dip, depth

fault mechanism

spatio-temporal slip distribution over rupture surface

Following a series of earthquakes, average rate can be found:

with  

And total uncertainty

You can find the history of slips from stream channels. 

i.e. for this one channel 2 has been subject to two ruptures, 1 and 3 have been subject to one and the other streams none at all

Knowing the size of the slip, (A), and u, the moment magnitude can be found.

A trench can be excavated across an active fault, assessing offsets at the fault.

Where carbon rich material has been deposited on either side of the soil, the timing of an earthquake can be assessed and bound to assess rupture history

 can learn rupture history by looking at differential settlements across sediment history

Recurrence intervals

Magnitudes from field investigations

These are likely to be very uncertain

Max stress is hz component

Max component in vertical

When the faults are smaller as the geometry cannot be easily found for these faults

For large sources when geometry can be easily defined