Occasionally, I get to find very interesting scientific apps for the Mac. In light of the recent events, SeisMac is definitely one of those. While not technically useful for the general public, it is a very important application for research.

All Mac laptops include the so-called Sudden Motion Sensor, a Micromechanical system (MEMS) accelerometer. This sensor is a very cheap integrated circuit which can be found in iPhones, iPod touch, the WiiMote, and many others gadgets. In the case of laptops, its main purpose is to park the hard drive head in case of a sudden drop, reducing the chance of damage to hard disk platters due to impact with the reading head. The sensor can however be queried and used for many other applications, from the facetious (like LiquidMac) to the more serious, like SeisMac. The idea is to use the Sudden Motion Sensor to detect earthquakes, effectively converting your laptop in a seismometer. It is interesting to see how sensitive the Sudden Motion Sensor can be. This is a tracking of my laptop with SeisMac

|SeisMac image|

The nervous, black oscillations you see on the left hand side are produced by a fan on my table, which I then removed. As you can see, these oscillations are uniform on the x, y and z axes. The single spike along the z axis you see in the center it's me doing a touchpad click to take a screenshot, and the series of strong impulses on the right it's again me, typing the filename. Don't be deceived by the stronger than usual gravity. I never calibrated the reading, so the absolute values can be wrong.

SeisMac is great, and you can configure its sensitivity to disregard human-generated events and catch the big deal when the quake strikes, but it does not beat another similar program, with a very interesting distributed project behind it: the Quake Catcher Network. The idea is simple, but very powerful: laptops and desktops volunteering to be part of the network constantly collect seismic data through the Motion Sensor, and alert a central server of any movement. If the laptop is moved, hit, or a local, non-seismic event occurs (such as a truck passing by), the signal will be considered by the central server as spurious due to its locality. However, if multiple unrelated laptops feel the same event, the server will validate it as an actual quake. Knowing the position of each laptop it is possible to obtain the intensity of the quake at each location, the speed of propagation, and the kind of oscillation. All these data provide precious details not only on the quake itself, but also on the terrain and the buildings, a very important dataset to improve prevention and emergency response.