Ref: Length of Day changes:

If the earth undergoes a change in rotational rate this could manifest in a location as change in where the sun's shadow is cast (compared to some reference like the previous year's "mark"). This assumes that "noise" in the measurement(s) are removed and there are probably more than a few....

In other words, in a "clean" shadow difference signal (where the noise has been removed) it may be possible that a change in the position of the sun's shadow compared to the same point in the previous year (remember this is with all known "noise" removed...i.e. perihelion shifting about the sun,etc.) could be due to a LOD change. Of course, local earth movement seems even more likely. A ~1.5 sec/year over twelve months on average (recent ~LOD change rate)makes for requiring a very fine measurement capability...and probably requires an atomic clock to ensure "when" the measurment was taken.... ==>Is Shan using an atomic clock to ensure that he takes the measurment at the same time each day? If not, and his data is significant, then he is probably measuring ground movements (as he mentioned).

Ref: vibration measurements.
I was referring to instrumenting the shadow making structure and the shadow surface with accelerometers. This is done in order to detect and compensate for relative motion between these objects which inject noise into the signal by changing the shadow pointing vector geometry (unrelated to sun/earth angles). Let's say a train goes by nearby and shakes the shadow making object when the measurment is being done. The object will then cast a shadow according to this shaking and will contaminate the true measurement if looking for change in the sun ray angles...as you know. I did not mean to imply that Shan was measuring "vibrations" with a once a day measurement (although he also said "every 15 minutes)?? How many measurements does he take in a day??

Sure...there are maybe many more "noise" sources that could be identified and must be accounted for...but a statistical check makes sense as the first step....

Another thought is that Shan maybe measuring ground movements, as he said, that may be ULF or lower and outside the passband of any seismo's in the area (and therefore not being detected).

Movements of large masses of earth (via plate theory if you subsribe) could be expected to create ULF oscillations which form long distance waves with wave tops and troughs across the landscape. These could be active at a large distance away from him (and yet detectable by him). ==> Does any of Shan's data correlate well with any IRIS or other ULF stations that may be in the area? (See http://www.seismology.harvard.edu/~ekstrom/Research/Noise/index2.html & find a station acronym list on web-site somewheres...)

If so, then most all of Shan's data may be mirrored in a local ULF sensor site (where the "noise" from other ULF sources compared to those that are quake precursory may benefit from some type of local increase in the signal to noise (SNR) ratio for this type of ULF data). In this regard, your supposition that other instruments may be capturing this same data may be correct. One way to find out is to go to the linked web-site and try and locate any nearby stations (in India somewhere? Need Shan's lat and long).

Anyway, from what he has said so far, he cannot be dismissed out of hand. Whether his claims are significant is a leading question though. I'm sure it is appreciated that you are willing to undertake doing the cross-checking on his data.

Some outside,longshot speculation on my part is that the earth around his area is creating a high SNR for quake precursory ULF which he is measuring and is able to use to predict quake activity (assuming his predictions are significant)

His assumption that the area under the sun as being the most likely area to have a quake is quite interesting as it presumes that he is versed in the solar/terrestrial/quake papers. Is this true Shan? What caused you to think that that would be so ? Just curious....

I hope this clarifies my post a little.

• Harvard Seismology: Real Time Noise