Low Cut: , 10, 15, 25, 35, 50, 70, 100, 140, 200, 250, 280, 400
Notch: 50, 60, 150, 180
High Cut: 32, 64,125, 250, 500 or 1000 Hz
The first recommendation for cases when you are having trouble getting sufficient signal to noise would be to increase your signal via stacking the data with multiple source events or get a more powerful seismic source. This will usually produce better results than the application of filters.
Another approach would be acquire data when the noise sources are less present. That may mean collecting data at night when the area is closed or the traffic is less. Early morning can be better for areas where the wind tends to increase during the day.
The selection of filters is very site dependent and can depend on a variety of factors as well as the type of survey being performed.
1) Typically the Notch filters are to remove noise due to electrical power lines (50 or 60 Hz and their harmonic frequencies depending on the country you are in).
2) Low cut filters are generally used for noise due to wind and moving vehicles, but care must be taken not to remove too much bandwidth from generated seismic signal. Often the noise sources have the same frequencies as the seismic data you are interested in and can’t be effectively removed using frequency filtering.
3) High cut filters can be used to remove noise from high frequency vibratory signals such as compressors or airplanes.
In general it is best to record the data without any frequency filters and filter in post processing or only on the displayed data in our software. It will be a matter of experimentation to determine the best filters at your site.
Modern 24-bit seismographs (Geode, Stratavisor, ES-3000, etc) have a much wider range of signal amplitudes that they can record accurately. This means that they can still accurately record smaller seismic signals even in the presence of larger noise signals. Therefore there is a reduced need for analog filters that are applied prior to digitization of the signals.
Digital filters are more flexible and can be more specifically applied to the noise that is recorded rather than the “Broader Brush” of analog filters. Digital filters also have the benefit of being able to go back to the original data if the wrong filter is applied, which is not the case with Analog filters. The general approach in the seismic industry is now to record everything – including the noise – and the filter out what you don’t want later.