Dear Armando, Lorenzo, Glennys, all:
There is no answer yet to the causes of the mismatch vertical-inclined (beyond the statistical fluctuation you have mentioned). We are working here in Santiago at different levels, each evolving at a different pace:
- Reconstruction level, coming back to the basics:
a) Reviewing the muon 2D maps and what are the differences across models and primaries.
b) The substraction of the residual EM component
c) We are also planning to revisit the other elements of the reconstruction, including treatment of the non-triggered stations, tank selection etc.
d) Finally, we want to revisit the aperture calculations after the previous studies are concluded.
- At physics level:
a) Felix has studied the impact of a broken calibration curve in HAS. This is expected because a change in composition will change the Muon number. Note the fundamental difference wrt
to vertical showers, where the EM component dominates S38 in the calibration curve, and therefore composition plays a less important role.
For some of the topics we will be ready to report the progress during the next meetings. I cc the Santiago group just in case someone is not in this list and wants to add more information.
Cheers!
L.
PS: we are taking this opportunity to sync Offline-HAS with latest eFit developements. It will also be reported.
On 6 Dec 2022, at 13:39, Glennys R Farrar <
gf25@nyu.edu> wrote:
Thanks for bringing this back to attention!
The most striking thing is the events reconstructed both ways, and there are a lot of events in this dataset . Underestimating inclined energies goes in the direction of reducing the discrepancy with TA in the common band - could somebody use your p 12 fit
to the “adjustment factor” assuming inclined reconstruction is the problem and see how far that goes to reducing the Auger TA discrepancy?
Glennys
Sent from my iPhone
On Dec 6, 2022, at 3:26 AM, Armando di Matteo <armando.dimatteo@to.infn.it> wrote:
Dear Lorenzo,
I agree that *most* of the difference was a fluctuation, but
there also seemed to be a systematic event-by-event shift between
the two reconstructions in events with 57° < θ < 63°, which
cannot possibly be just statistics... Has that been
investigated?
(Maybe it would be more appropriate to use 60° < θ < 65° instead,
as IIRC I've heard that using the inclined reconstruction at 60°
is already a bit of a stretch whereas the vertical reconstruction
might easily be stretched by a few degrees more, but I wouldn't
expect that to make a major difference.)
Best,
Armando
On Tue, 2022-12-06 at 12:17 +0100, Lorenzo Caccianiga wrote:
Dear Armando, all
As far as I know there were no further developments, also
because we couldn't rule out that it was mostly due to a
statistical fluctuation.
Indeed, looking at the new data this would look to be the case,
as we have in 2021 2 new inclined events above 100 EeV which,
given the low stat, changes the numbers a lot:
until the end of 2021 we have 55 vertical events above 90 EeV (
49 until the end of 2020) and 5 inclined (2 until 2020) which
is still a low ratio wrt to the expected one but much better
than in your slides.
(the ratio now gets even better above 100 EeV, where we have 35
vertical and 4 inclined events)
I'm now working on preparing the update of the dataset until
the end of 2022 and we'll see how it goes with one additional
year.
I'm personally convinced that there's indeed a small effect
somwhere (on the energy scale?) which was enhanced by an
unfortunate fluctuation in the dataset that we just released,
but of course this is just my feeling!
I'm available for any further check, of course!
cheers
Lorenzo
Il giorno lun 5 dic 2022 alle ore 12:39 Armando di Matteo
<armando.dimatteo@to.infn.it> ha scritto:
Dear colleagues,
Last year I showed (slides attached) that above a few tens of
EeV
there seems to be either an underestimation of inclined
energies
or an overestimation of vertical energies, visible not only
by
comparing the two separate datasets with their standard
zenith
angle cuts, but also by reconstructing events with zenith
angles
between 57° and 63° using both reconstructions and comparing
the
two reconstructed energies event by event (see sl. 12).
Has there been any progress in investigating where this
discrepancy comes from and/or in proposing possible ways to
correct it?
Best,
--
Armando di Matteo <armando.dimatteo@to.infn.it>
Istituto Nazionale di Fisica Nucleare (INFN), sezione di Torino,
Turin, Italy