Well...one might just argue that Shan's method is a "Length of Day change detector" of sorts.

The link below is to a web-page which may have a link to a "polar meter" which measures changes in the rotational axis of the earth. One could argue that in addition to tracking Chandler's Wobble...one could also detect sudden changes in the earth's rotational rate (perhaps if the axis is deviated too).

BTW, some argue that Chandler's Wobble causes earthquakes and vice versa....

In Shan's case, decoupling the Chandler Wobble from his data should be possible (though maybe not easy...rigid earth vs non-rigid earth assumptions about the earth's axis to his location,etc). What's left could be considered "LOD changes" hypothesized to be caused by air flows across the earth's surface causing friction which either speeds or slows the earth's rotation rate (as I understand this theory...since the jury is still out...).

{Aside: BTW, I believe astronomers can easily collect LOD data by checking their location against a fixed star and checking for +/- deviations from their "known" lat/long. This is probably done routinely somewhere in astromony land....}

So, in one vein of thought, Shan's method appears to have some technical merit as an indicator of large forces acting in his region of the earth. However, Shan would also have to instrument the "shadow maker" structure and sun surface in order to subtract out relative and (resonant) vibrational motion between the two which could give bogus readings if not accounted for.

The other problem is that his predicted areas appear to be large areas of real estate which would tend to place his predictions down into the background rate (possibly). Of course, with the correct EQ catalogs and time, one could sort through the data to see if a statistical difference is noted between Shan's predictions and the background seismicity rates.

Any takers? A list of all previous predictions provided by Shan could expedite this process. The "where" would have to have an agreed defined boundary for a statistical test. This would create extra work in doing catalog searches using uneven geographical boundaries, etc...(i.e. "the Med" boundaries )

Further digging will tell....

• Polar Meter - targeting data for Chandler's Wobble motions