As I said in my reply to amatecha the existence of EM phenomena caused by earthquakes is well established but that relationship isn't yet well quantified—we can't turn the effect on to just suit our testing. That's the advantage of amateurs monitoring 24/7. We'd eventually catch it this way I reckon.

I am not an expert in this field but what is described in the article is anomalous (long distance) FM transmissions.

I guess they used FM because existing FM diffusion had mandated the creation of lot of FM emitters (one every 70 miles/100 km) so it enables to cover a wide area.

No doubt FM stations could be used but the TX radius is only about 100 miles. I might be wrong but I doubt this would be enough to triangulate the point of EM interference and in many places there would not be enough transmitters. Anomalous long range FM transmissions would be possible if the atmosphere was ionized (like sporadic E-layer transmissions) but it's a moot point whether the EM effects of that type are sufficiently regular in an already unpredictable earthquake environment. It's possible the low frequency approach may be more consistently sensitive, that is it may still work when there was no ionization necessary for the anomalous FM transmissions.

Of coarse, that's only speculation on my part. Again, this needs further scientific investigation, and again amateurs could be helpful. The FM band is 88-108MHz but the amateur bands of 50-54MHz (6m) and 144-148MHz (2m) could be used as a feasibility study (the characteristics of those bands ought to be close enough to evaluate the idea).