Seismic record section of vertical-component seismograms from the earthquak
which occurred near Havre de Grace, MD on October 17, 1996. Records from
seismographic stations in New Jersey and southern New York are plotted.
The station are at azimuth range of 42 to 51 degree from the epicenter.
Station code and azimuth are indicated at the end of each trace. Dotted lines indicates
theoretical arrival times of Pn and Sn waves which are the first arrival P and S waves
that traveled mainly through the top of the upper mantle just underneath the Moho discontinuiety.
Pg and Sg indicate prediected arrival times of P and S waves multiply reflected in the crust.
Acutally, Sg marks the arrival of Lg wave train which are the most prominant seismic waves observed
at distances few tens of killometers out to about two thousands killometers from the source.
To: Linda Tinsman, Earth Science teacher at
North East High School in Cecil County, Maryland
I believe that the location for the earthquake in question was determined by myself soon after the earthquake
using seimic records from about 10 stations in Delaware, New Jersey and New York. You can find locations of these
stations from the map and tables in this web site. However, I don't think I used stations in Pennsylvania,
perhaps it was not available at the time.
The above seismic record section gives us an idea about how the seismic signals generated by the earthquake
were propagated and recorded at various stations by displaying records in distance and time.
It also give us some hint on the strength of the earthquake, since it is not easy to discern the first
arrival P waves on these records. Of course, nearby stations should still show more clear (that is more strong and sharp) onset arrivals.
We used to use measure arrival times and amplitudes of several primary seismic phases at many station observations
surrounding the earthquake source and use computer programs to determine hypocenter of the earthquake sources.
If the Earth is homogeneous everywhere, then we can determine the source location very accurately.
However, the Earth is heterogeneous and, in particular, the Earth's crust where most of the signals
shown above have traveled from its source to each station crosses complex geologic structures, we always have some uncertainties on its location.
For instance, the location that you can find from the table (click location to see it) indicates that the precision of the location is about +-8.2 km,
which is pretty large, since it indicates that the true epicenter maybe in an area nearly 64 squre killometers wide. If we include the depth of
the source, the solution is 5.4 km +-9.16 km, then the true hypocenter maybe inside 540 cubic killometers.
These are very difficult events to locate, becauase
1) earthquake is small, (magnitude 2.3), so the seismic signals it generated were weak,
2) there are not many seismographic stations around the earthquake epicenter, obviously if there were number of nearby
stations from this small earthquake, they would have provided more clear observations, and hence improve our location precision.
In general, due to low seismicity in Eastern United States, State or Federal governments does not spend enough funds to operate seismographic stations.
If you use "distance" from each station to locate the event using simple geometrical method, then you can use travel time difference, S-P in seconds, and multiply it by some constant (say 8 km/s) will give you approximate location. Please try it again and let me know.
Here is a useful web site illustrating how local earthquakes are located Studying Earthquake, USGS.