The Science

All earthquake warning fundamentally relies on detecting an earthquake very quickly, and sending a warning out to people before they experience the shaking.
Earthquake in cross section
CC Attribution Share-Alike 3.0 License/Wikispaces
Earthquake Basics

Earthquake warning is possible because of two facts that are true of every earthquake. The first is that the earthquake is not felt everywhere at the same time. The shaking is felt first at the focus, which is a point deep in the Earth's crust below the epicenter, where an earthquake begins. The shaking spreads out in all directions from the focus, and the farther away you are from the focus or epicenter, the longer it will be before you feel the shaking.

P/S waves
CC-2.0-Attribution/USGS
P-waves and S-waves

The second fact is that all earthquakes radiate two types of waves: P-wavesPrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking. move quickly through the crust, about 20 times the speed of sound. They are very weak, however, and you can only feel them for earthquakes that are very strong or very close by. They don't cause any damage. S-wavesSecondary, or S, waves are the first waves to arrive from an earthquake that are large enough to cause damage. They are slower than P-waves but cause much greater shaking. move at about half the speed of P-wavesPrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking., but carry most of the shaking energy of an earthquake and cause the most damage.

Mexico EQW system map
CC-Attribution-NonCommercial-ShareAlike 4.0/SASMex
Earthquake Warning

Earthquake warning systems can be characterized by asking two big questions. First is whether they detect the shaking of the S-wavesSecondary, or S, waves are the first waves to arrive from an earthquake that are large enough to cause damage. They are slower than P-waves but cause much greater shaking. directly, or detect the P-wavesPrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking. and estimate how strong the S-wavesSecondary, or S, waves are the first waves to arrive from an earthquake that are large enough to cause damage. They are slower than P-waves but cause much greater shaking. will be. Although direct measurement of the S-waveSecondary, or S, waves are the first waves to arrive from an earthquake that are large enough to cause damage. They are slower than P-waves but cause much greater shaking. is generally more reliable, it is also much slower and produces a Blind ZoneA region, centered on the epicenter, where no warning is provided before serious ground motions arrive. A few seconds of delay producing a warning can produce a Blind Zone covering 400 square miles. around the epicenter. Mexico has an S-waveSecondary, or S, waves are the first waves to arrive from an earthquake that are large enough to cause damage. They are slower than P-waves but cause much greater shaking. based earthquake warning system.

Comparison of EQW types
CC-Attribution-NonCommercial-ShareAlike 4.0/SWS
Types of Earthquake Warning

The second question to ask about earthquake warning systems is whether they estimate shaking just where the seismometerdevice that measures ground motions, typically acceleration or velocity is (on-site warnings) or use a network of seismometersdevice that measures ground motions, typically acceleration or velocity to estimate shaking everywhere (networked warnings).

Asking these two questions lets us classify earthquake warning systems into four categories as shown in the table.

QuakeGuard with sensors and GPS
CC-Attribution-NonCommercial-ShareAlike 4.0/SWS
QuakeGuardTM Earthquake Warning

The QuakeGuardTM earthquake warning system is a hybrid system, bringing together all the advantages of On-site and networked P-wavePrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking. earthquake warning systems. Because each of the sensor stations in the QuakeGuard network is also a standalone system by itself, the QuakeGuard network does not suffer from the Blind ZoneA region, centered on the epicenter, where no warning is provided before serious ground motions arrive. A few seconds of delay producing a warning can produce a Blind Zone covering 400 square miles. problems that other networked earthquake warning systems have. Because the stations are designed to reject false alarms even in standalone mode, the QuakeGuard network does not suffer from the false alarms that plague other standalone systems. And because the QuakeGuard stations are networked together, they provide nearly twice as much warning to people farther from the epicenter as a standalone system could provide.

Map showing locations of red tagged buildings from South Napa Ea
CC-Attribution-NonCommercial-ShareAlike 4.0/SWS
Blind Zones

Earthquake warnings are a race between the electronic warning signals and the S-wavesSecondary, or S, waves are the first waves to arrive from an earthquake that are large enough to cause damage. They are slower than P-waves but cause much greater shaking. of an earthquake. The S-wavesSecondary, or S, waves are the first waves to arrive from an earthquake that are large enough to cause damage. They are slower than P-waves but cause much greater shaking. of an earthquake emanate from the epicenter at over 2 miles per second. So every second that the warning is delayed means that several more square miles feel the shaking.

If the signal gets to a user before the shaking begins, that user gets some benefit over what would be possible with a seismic switch, or simply by feeling the shaking. If the shaking gets to the user before the signal does, then no earthquake warning is provided. This race heavily favors the earthquake near the epicenter, where the seismic waves are felt within a few seconds of the start of the quake. This area near the epicenter is often called the “Blind Zone.” It can encompass hundreds of square miles in which nobody gets an earthquake warning. The Blind Zone is a feature of every other networked earthquake warning system, but the QuakeGuard earthquake warning system has no Blind Zone, because of a hybrid approach that uses a network of standalone earthquake warning stations.

US false alarm cartoon
CC-Attribution 2.0/Nicholson
False Alarms

Any time someone takes an action, based on an earthquake warning signal, that isn’t warranted by the shaking they experience, it is a false alarm. That action can be anything from simply dropping under a table to stopping a multi-million-dollar industrial process. The cause of the false alarm can be anything from a small or distant earthquake, to a truck going by the seismometerdevice that measures ground motions, typically acceleration or velocity, to malicious hacking of the earthquake warning signal. The cause is immaterial: false alarms cost money and erode confidence in earthquake warnings. The QuakeGuard earthquake warning system is designed to minimize false alarms through careful hardware and software design.

Modified Mercalli Intensity vs PGA/PGV
CC-Attribution-NonCommercial-ShareAlike 4.0/SWS
Magnitude vs. Intensity

When talking about the “size” of an earthquake, most people immediately think about its magnitude. Magnitude is a measure of the amount of energy released by an earthquake. But people don’t feel the energy of an earthquake, they feel shaking. The severity of that shaking is known as the “intensity.” A key feature of intensity is that, while an earthquake has a single magnitude, the intensity varies from location to location for any earthquake. The intensity goes down as you get farther from the earthquake, so that being 10 miles away from a magnitude 6 earthquake feels about the same as being 100 miles from a magnitude 7. In other words, what matters to people is not the magnitude of the earthquake, it’s the intensity!

Magnitude vs, Intensity diagrams
CC-Attribution-NonCommercial-ShareAlike 4.0/SWS
Magnitude Misses the Point

Most earthquake warning systems try to estimate the magnitude of an earthquake, and then use the magnitude to estimate the intensity that people in various locations will experience. QuakeGuard algorithms don’t look at the magnitude of an earthquake. Rather, they estimate the intensity directly from the P-wavePrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking.. In doing so the QuakeGuard system realizes two important benefits:

Graphic depicting how to Beat the S-wave to the surface
CC-Attribution-NonCommercial-ShareAlike 4.0/SWS
Speed

Estimating magnitude from the P-wavePrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking., and then intensity from magnitude, is a two-step process. What’s more, estimating magnitude can take several seconds. In contrast, estimating intensity directly from the P-wavePrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking. is a single step and can be done much more quickly, in less than half a second in most cases.

PGA vs prediction parameter
CC-Attribution-NonCommercial-ShareAlike 4.0/SWS
Accuracy

Going directly from P-wavePrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking. to intensity enables much more accurate estimates of intensity. The two-step process of going from P-wavesPrimary, or P, waves are the fastest moving waves produced by an earthquake and are first to arrive. They do not carry much energy and cause only slight shaking. to magnitude to intensity causes a compounding of errors that leads to a much less accurate estimate of intensity.