Seismic Spectrograms - A new way to look at wiggles
February 13, 2014
by Steve Malone
A spectrogram is a visual way of representing the signal strength, or “loudness”, of a signal over time at various frequencies for a site. Not only can one see whether there is more or less energy at a particular frequency, but one can also see how energy levels vary over time. In order to visualize the three variables here (time, frequency and signal strength) one can not use a simple wiggly-line plot like a seismogram (amplitude vs. time) but rather needs to plot a third parameter. Traditionally for spectrograms, this third item is color. Thus spectrograms use color for signal strength, warm for strong and cold for weak. Frequency is represented on the vertical axis and like seismograms, time is on the horizontal axis. The graph here illustrates an example of a smallish earthquake followed by a larger earthquake on a spectrogram from a single station. The black line above the colorful spectrogram is a seismogram. See how much more information is present in the spectrogram. When spectrograms from more than one station are used together for the same time period their value is further enhanced.
Because interpreting spectrograms is quite complicated we have written a comprehensive document with extensive examples to help the public understand them. This document is based on material provided by Jackie Caplan-Auerbach and Seth Moran, both very experienced users of spectrograms, particularly in volcanic situations. Don't worry that, even after studying these examples carefully you find yourself sometimes being confused. There are many cases in which even the most experienced analyst may have trouble determining what is really going on. Although spectrograms can be confusing to read our feeling is that the more data the public can see the more educated they become and this decreases the likelihood of wild speculations, predictions and irresponsible comments.
Within reason we are happy to follow up and answer questions or address concerns that people have concerning spectrogram signals. However, we strongly suggest that when you have a question, first do as scientists do and try to figure it out yourself. You should look for other examples of your observation at different times or on different stations. Look at the examples and see if you can find anything similar. Think about the different things that can shake the ground and come up with your best interpretation and if still confused send us a comment or question. Be patient because our small staff has many responsibilites of which responding to comments and e-mailed question is not the highest priority.
Also, be warned, spectrograms can be facinating and lots of time can be "wasted" trying to figure out every different looking signal. I used to feel I was wasting time reviewing them but once I became fairly experienced I realized spectorgrams are just another way of watching the earth wiggle and its got lots of ways of doing that. Please check out our "What is a Spectrogram" page before even looking at them.