At lower intensities the shower generated is less intense as well and does not trigger beam dump. They actually scanned log data from past successful runs and found multiple such events which just weren't strong enough to cause a problem.
So lets assume that the total rate of beam-dust events is constant,
regardless of intensity. If there is a fixed radiation energy threshold involved the
event rate above this threshold rises with intensity.
That would explain the strong correlation of the 7 events (that triggered a dump)
For the other events given in the "Sub-threshold UFOs" diagram the same applies,
but with a detection threshold given by sensitivity and noise-floor of the BLMs and
shielding effects. If one would add this events the correlation would be weaker.
So if this (somewhat macroscopic) model were right, the correlation would disappear
if one would be able to detect all beam-dust interactions.
But if one looks at the nuclei-level this model is still wrong because more protons around
increase the possibility of interactions with the dust and in the end there will be always a correlation with intensity. Even if the dust-falling-rate would be constant.
Hope I got it right this time
I wonder if there are some non-linear effects too, depending on what amount
of dust gets evaporated, atomized, ionized or quarkized (?
Or can the cross section be viewed as constant (for a given shape and mass)?
I'm not sure because I did not understand what Myers were talking about,
but I believe Myers talked about possibilities to cure the dust-events.
I understand something like "clean it out with high beam intensity" and/or
something about degassing (in the LEP). Did he meant to use the heaters
in the pipes (or screens) to fix any plastic stuff around by melting it?
It is around 00:17:43 at
http://indico.cern.ch/materialDisplay.p ... fId=105780