Surface seismic waves and amplitude with distance

[-sandro-]

OK so we all know that seismic waves decrease in amplitude at about 1/r, the decay is slower than P and S-waves but it's there.
So how do you explain that when I see examples of seismograms of the same earthquake from stations at different distance you can sometimes notice that surface waves amplitudes increase as you get away from the source?

Example: http://www.sciencebuddies.org/Files/3030/3/Geo_img046.gif

 

[davenn]

Hi Sandro

the amplitude of the surface waves at the Golden, Colorado station and the Colombian station don't look overly different in amplitude in that example.
You may be fooled by the fact that there is almost no P and S wave amplitude at the Colombian station

There's also no indication of the gain of any of the stations shown
In all my experience, ( I run a seismograph station) and can compare my data with may other stations
You will also probably find that the gain of the Palmer, Alaska station is reasonably low else the signal would have overloaded the recorder

IF we all were set at the same gain, there would be 2 main reasons that would cause possible amplitude differences

1) differences in rock densities between the quake epicentre and the stations concerned
2) the type of ground that the station is situated on ... ie ... soft soils, solid bedrock etc
soft soils will tend to amplify the lower frequency surface waves

2 of my 5 seismometers can be seen online here ...
My seismos

There's some surface waves present there at the moment ( at the time of writing this post)
from some Chilean quakes ... The P and S waves are non-existent

cheers
Dave

 

[-sandro-]

Ok, so what about this video example? It clearly shows amplitudes are higher at longer distances for surface waves.

 

[davenn]

That's still an unrealistic example as its just an animation

The real info on the grams, as in your first example are the increasing times of the difference in the P and S waves
It also shows well the spreading out of the surface waves in time this called dispersion. ( they spread out in duration with increasing time). This is because the velocity of the Surface Waves ( the Love and Rayleigh waves) are frequency dependent.
The velocity of the Body waves, the P and S Waves, is not frequency dependent and therefore there is no dispersion of those waves so their duration shown on the seismogram is constant, even at distance

In the real world, on real seismograms, and I have access to many records for any given large quake, I don't see the effect you are describing.

Finally, just to be clear, Your amplitude ... you are referring to the height of the wave record Plus and minus movement from the centre line ? <--- that is amplitude
Not the length of time the waves go on for ?? <-- That is dispersion

Nothing else can be obtained from the grams. There is no seismometer info given, no info on ground type at their locations etc.
The amplitude of the Surface Waves of a large shallow, crustal quake at the source, are huge. They are what cause the damage. The P and S Waves arrivals are difficult to even recognise if the quake is very close


cheers
Dave

 

[-sandro-]

Hi, yes by amplitude I mean the actual physical amplitude of a wave as you described. :)
Tomorrow I'm going to send an email to my professor because I clearly remember he said that (while showing the video) in class.

 

[davenn]

The other thought I had whilst I was away...

Seismometers designed for recording teleseismic events ( long distance > 1000 km) tend to have frequency responses that peak in the freq range of the surface waves 10 to 100 seconds
The top seismogram on my www site I gave above is designed (filtered) for that. It peaks at ~ 30 seconds

The bottom seismogram is of a short period sensor, it peaks at around 0.1 second ( 10 Hz). It won't record long distance surface waves overly well, if at all, on large quakes greater than ~ 2000 km or greater.

both of your references are mainly showing you the differences in propagation speeds of the P, S and Surface waves and how the arrival times at different stations changes with elapsed time

I have been working with seismograms for more than 20 years, and I'm still learning, I don't know it all, never will haha ! Gram interpretation is a complex subject. But as the years go by, patterns emerge, and one starts recognising the arrival of different waves, the direct traveling ones as well as those that bounce off the core and sometimes those that have traveled the opposite way around the Earth before getting to the seismometer

cheers
Dave

 

[jim hardy]

I'll be curious to see if yours sees that Alaska 'quake a couple hours ago...

 

[davenn]

Yes I recorded it Jim, but because it was quite deep, ~ 108km, it wasn't huge on the seismographs
Deep quakes don't produce much in the way of surface waves
It was totally outclassed by 6.9, 6.3 and 6.2 quakes much closer to me in the 1.5 hours prior to the Aleutian Is M8.0 event

here's a low gain long period channel on my system that shows the different events
http://www.sydneystormcity.com/zlo1.gif

Am at work at the moment ... give me a little time and I will save an image and label the events :)

Dave

 

[davenn]

OK here we go

you can see on that indicated line that the surface waves from the M 8.0 were minimal
Had this been a shallow quake < 30 km the surface waves would have been almost full amplitude ( top to bottom of the screen) even on this low gain channelEDIT ... some updates ... its been now downgraded to a M 7.9
It produce a local Tsunami of ~ 0.2 - 0.7 ft
This quake was quite a long way back (and deep down slip ) from the trench so hence why it didn't produce any significant tsunami
cheers
Dave