What Is a Seismometer? Measuring Motion From Inside the Motion
By the BrainSnail editorial team. How these articles are written and checked, and how to tell us when one is wrong.
An instrument sitting on the ground moves with the ground, which makes measuring ground movement awkward, since there is no stationary reference to measure against. Every seismometer solves that the same way: it contains a mass suspended so loosely that it lags behind, and the relative movement between the mass and the frame is the signal.
The inertial principle
Suspend a heavy mass on a very weak spring inside a rigid frame, and when the frame is shaken quickly the mass tends to stay where it is, because inertia resists sudden acceleration and the weak spring transmits very little force. The frame moves and the mass does not, so the gap between them changes, and recording that gap records the ground motion. The key quantity is the instrument's natural period, meaning how long the mass takes to oscillate freely, and it must be long compared with the ground motions being measured, which is why sensitive long-period seismometers use elaborate suspensions to achieve periods of many seconds with a compact instrument. Early designs traced the motion mechanically onto smoked paper with a stylus, which is why seismograms were drawn as continuous lines on rotating drums. Modern instruments use force feedback: rather than letting the mass move, an electromagnetic system pushes it to stay in place, and the current required to do that is the output, which gives a far wider range and a flatter response.
What the record contains
A seismogram from a distant earthquake shows several distinct arrivals, and the separation between them is what locates the event:
- •P waves, compressional waves that travel fastest and arrive first, moving material back and forth along the direction of travel and able to pass through liquid
- •S waves, shear waves arriving later, moving material perpendicular to travel, and unable to pass through liquid, which is how the liquid outer core was discovered from the shadow zone where S waves do not arrive
- •Surface waves, slower still and much larger in amplitude, which travel along the surface and cause most of the destruction
- •The time gap between P and S arrivals, which grows with distance and therefore gives the distance from the station to the earthquake
- •Distances from three or more stations, which intersect to fix the location, since one distance alone defines only a circle
- •Amplitude corrected for distance, which gives magnitude, with the modern moment magnitude scale based on the physical energy released rather than on the older amplitude-based scale
What the network sees beyond earthquakes
A global seismic network is a general-purpose vibration sensor and detects a great deal besides earthquakes. Nuclear tests produce a distinctive signature, and seismic monitoring is a principal verification mechanism for the test ban treaty, with a dedicated international network operating for that purpose and with the discrimination between an explosion and a natural earthquake resting on the relative size of different wave types. Volcanic tremor precedes eruptions and is a primary forecasting tool. Landslides, glacier calving, meteor entries, mine collapses and large industrial explosions all appear. Ocean waves produce a continuous background hum called microseism that was long treated as noise and is now used to image the subsurface, since the noise itself carries information about the structure it travelled through. The pandemic produced a widely reported result when the global reduction in human activity lowered background seismic noise measurably, making quieter natural signals detectable in cities for the first time.
Early warning
The separation between wave types enables something genuinely useful. Because P waves travel faster than the destructive S and surface waves, a station near the epicentre detecting the P arrival can transmit a warning electronically that outruns the damaging waves, giving locations further away a warning of seconds to tens of seconds. That is not much and is enough to stop trains, halt surgery, open lift doors at the nearest floor, shut valves and let people take cover, and systems operating in Japan, Mexico, Taiwan and the American west coast do exactly that. The limits are inherent: the area nearest the epicentre, which suffers most, gets no warning at all, magnitude must be estimated from the first seconds of a signal and is sometimes wrong, and false alarms erode compliance. Dense low-cost sensor networks, including accelerometers in phones, are being used to supplement professional instruments, trading individual quality for coverage, which suits a problem where speed of detection matters more than precision.
The takeaway
A seismometer suspends a mass so weakly that it lags behind when the ground moves, and the relative motion between mass and frame is the signal, with modern instruments measuring the force needed to hold the mass still instead. The gap between fast P waves and slower S waves gives distance, and three stations fix a location. S waves failing to cross the core revealed it is liquid. Because P waves outrun the destructive ones, detection near the epicentre can warn distant places by seconds.