In the Aug. 29 issue of Science News Mag., in an article titled "Shaky Forecasts", is a description of a three-year old international effort to "standardize and make more scientific the way earthquake predictions are stated and tested," and to "test how well statistical quantitative models of earthquake likelihood hold up to what happens in real time."

The article is at: http://www.sciencenews.org/view/feature/id/46373/title/Shaky_Forecasts

I bring this up because I had not previously read of this group (the Collaborative Study for Earthquake Predictability - CSEP, link below), and the information might be of particular interest to Roger and others who are attempting to analyze the reliability of various prediction methods. At the CSEP link is a link to their "CSEP Tests". Roger may wish to look at those to see if any of the tests are essentially the same as his own.

Both John Vidale and Susan Hough are quoted in the article. There is also an intriguing bit about a successful prediction and the opinion that short-term, reliable prediction is achievable in the special case of locations where "the geology is relatively simple, as on transform faults along the East Pacific Rise."

Another seismologist's application of a prediction method is described in the article as "proving robust against almost three years of seismic measurements" - though that description is not defined. Does it mean she has exceeded chance, or the "null hypothesis"?

As a side note, in reference to a brief thread here regarding "tremor", the article notes the terms "slow quakes", "silent quakes", "aseismic slip", "nonvolcanic tremor", "slow transients" and "episodic tremor and slip" are all synonyms.

Mike W.

• CSEP