Can toads detect forthcoming earthquakes (and landslides?)

The mainstream media is reporting a story this morning that breeding toads in Italy apparently relocated from their nests five days before last year's L'aquilla earthquake in Italy. This is picking up on a paper published in the Journal of Zoology today by a team from the Open University.

The paper examines toad numbers at San Ruffino lake, 74 km from the earthquake focus:


The authors observed that on 28th March, 90 male toads (Bufo Bufo) had gathered for the breeding season . However, on 1st April, five days before the event, 96% of the male toads disappeared. This disappearance is reported to coincide with a burst of very low frequency radio wave disturbance of the ionophere, detected by a Russian receiver network. Numbers did not return to near normal levels until 15th April, reportedly two days after the last M>4.5 aftershock.

A Bufo Bufo toad. Similarities in appearance to climate change deniers Nigel Lawson and Christopher Monckton are apparently entirely coincidental. Image from here

There are three things that intrigue me about this report:

1. Why would toads develop an evolutionary ability to detect earthquake precursors (the Independent article speculates on this below, so I will return to it)? On the whole sitting in a pond of water is probably not a bad place to be in an earthquake, even one with mudslides, especially in a location like this lake, which is not obviously landslide-prone.
2. The Independent article makes the following statement, quoting Rachel Grant, the Journal author: "Russian scientists suggested that the toads may have been able to detect the release of radioactive radon gas from the ground, or the presence of charged particles in the ionosphere of the night sky, Dr Grant said. If so, it may be an evolved ability to protect the slow-moving animals from the frequent mud slides caused by earthquakes, she added." Hmmmm! For this to be effective the toads would need to know where to go should a mudslide be likely. I am not sure where this would be or how a toad would know.
3. The toads were located 74 km from the focus of what was a small earthquake. Can toads really detect precursors from this distance, and again why would there be an evolutionary advantage in being able to so-do, given the low probability of a landslide at such a large distance?

None of this undermines the observed behaviour or indeed the observational science of the paper. However, once again the focus is upon the potential for prediction of earthquakes, which is frankly a sideshow, when the real need is to focus on preparedness.