What is a Spectrogram?

A spectrogram is a means for viewing the frequency content of a seismogram as it changes with time. Once each minute, we calculate the frequency spectrum of the seismogram between 0 and 10 Hz. The spectral amplitude values are converted to color with deep blues representing low values, ranging through greens and yellows to deep red for the high values. Each minute is thus displayed as a horizontal colored line representing by its changing color the differences in shaking intensity at different frequencies from 0 to 10 Hz. By plotting these horizontal lines adjacent to one another as they are calculated we can see a time sequence of the frequency spectrum.

How to Read the Display

The spectrograms displayed are from a few of the seismograph stations routinely recorded by the Northern California Seismograph Network. The spectrograms show a record of the frequency content of ground motion at a particular seismograph station in Northern California during a 24-hour period. The spectrogram is "read" from top to bottom (this is the direction that time increases). Each horizontal line represents in color the amount of ground motion at frequencies ranging from 0 to 10 Hz. Each horizontal line represents the frequency spectrum for 1 minute of data.

The corresponding data trace is plotted along the right-hand axis.

The vertical lines are not part of the spectrogram but are present to indicate equal intervals of frequency. Time is indicated at the left side of the plot in local Pacific time and at the right side in Universal (or Greenwich) time.


When an earthquake occurs the spectrogram will show ground motions that typically last from several tens of seconds to many minutes depending on the size of the earthquake and the sensitivity of the seismograph.

On these spectrograms you may see local earthquakes in Northern California and earthquakes elsewhere in the world. Almost any earthquake in the world having a magnitude greater than 5.5 will be seen on these spectrograms.

Illustrative Examples »

How the Data Channels are Named

Each data channel has a three part name such as MSL VHZ NC. The first part identifies the station. The middle part describes the data. The last part identifies the seismic network. The station name and network uniquely identify the location where the data are being recorded. The data descriptor tells a) what is being measured (velocity, displacement, acceleration), b) what sort of instrument is doing the recording (digital, hi-gain analog, etc.), and c) the orientation of the sensor (vertical, horizontal-north-south or horizontal-east-west). For example, VHZ is a high-gain (sensitive) analog velocity sensor, sensing vertical movement.