I can't figure out if they're putting their motion detectors in the cell-phones to detect vibration, or networking every turned on cell phone through GPS to detect longitudinal/latitudinal tremor waves. I'm hedging on both.

They could develop an app that would filter out noise and report large,rhythmic accelerations to the network.Wouldn't help the owners of the detecting smart-phones,though.

So... a couple years ago I had a very similar idea for ground based ionospheric mapping.

Use the antenna already present in everyone's car, (calibrated to some experimentally derived standardized value that varies from car to car due to variable antenna gain), plus a couple of simple circuit boards with FPGA's and GPS capabilities and that should do it, more or less.

Program the antennas to sweep through frequencies and record a spectrograph, report their position at the time of data acquisition, and analysis can do the rest.

Then you realize... To do this on any kind of large, meaningful scale, you'd probably have to contract with existing telecom towers here and there and have the cars broadcasting the data, because the owners aren't going to upload it often. You'd need a lot of cars outfitted, too.

And then, you realize that people don't even like the idea of broadcasting data to any organization that keeps records of their whereabouts.

If anyone can pull it off, go for it.

Monitoring the accelerometer all the time consumes the battery, So perhaps you will want an accelerometer that can detect interesting signals by itself, but it would also have to tell if it was sitting on a table and not swinging in a handbag or pocket.

If phones in an area pick something up, then the network could do a quick push notify to wake up other phones in the area to get detailed measurements.

and am_Unition it sounds like you are reinventing the ionosonde!