Thanks for the feedback; I did find one major mistake in my Monte
Carlo, though I doubt it will be significant.
MS>From: "AXPBO::SPURIO - TEL. 630 5226" <Maurizio.Spurio@bo.infn.it>
MS> First, it seems to me not true that in the Lipari code the NC are
MS> identically zero. In fact, ~5% of my UGS+ID events, and almost the same
MS> fraction in the IU (Surdo) sample were produced by NC.
I am sorry if my explanation was unclear; the NC are not all zero, but
the elastic scattering part (process 1) of the NC cross sections are
set to zero in the following lines in SIGNU_INI:
C...NC - elastic scattering
SSNU (JA,LNU,2,1,1) = 0.
SSNU (JA,LNU,2,2,1) = 0.
Resonant and DIS cross sections for NC are non-zero in Lipari.
MS> Second, ~1 year ago Antonio find some bugs in the Lipari generator, and
MS> we use now the corrected code. One copy of the corrected CAR file with
MS> the Lipari code is in:
MS> AXPGS::DISK$MACRO:[SPURIO]INTERNAL.CAR
Thanks, I have ftp'ed this file and find it identical to the one I am
using. I also have a few bug fixes from Antonio that I keep in a
cradle, and the latest Bartol flux as determined during Stanev's visit
to Lecce in June.
MS> Third, all the data cards should be the same: this is the output for
MS> our data card, assuming a neutrino threshold of 0.1 GeV.
My initialization was identical to yours, except for one important
respect -- I was using the wrong structure funtions -- I was using the
Duke-Owens set selected in Paolo's STR_FUN_INI_0. I have changed it
tonight to use Morfin-Tung set 1, as is standard in the rest of the
collaboration. This lowered my DIS cross sections by a few percent,
but the overall interaction rate dropped just one percent (because
most of the neutrino flux is in the region where DIS is not so
important). I will regenerate the events with the new structure
functions (this will take a day or two) but I do not expect much
effect beyond the 1% change in interaction rate.
MS> In 79 year, we expect in MACRO:
MS> 79y * 5.31kton*(111.49+41.83) (kton year)**-1 = 64316 events
With the new structure function, my calculation gives 63,886 events,
a discrepancy of less than 1% from yours.
AS>From: Antonio.Surdo@le.infn.it
AS>...
AS>ask to Bob if this is the case). Practically, in GMACRO with the Lipari nu
AS>interaction generator, the outgoing lepton 4-momentum (and that of the
AS>delta in the resonance case) is explicitely generated, then Lund is invoked
AS>to produce the rest of secondaries (if present). If this is also the
AS>mechanism in case of NEUGEN generator, the reson for discrepancy must be
AS>another one.
In NEUGEN, the hadronic multiplicity is picked according to the
so-called 'KNO distribution'. Then pi's and K's are added until the
charge and multiplicity are as desired. Then the entire hadronic
cluster is decayed according to phase space. So the event will be
fatter (in the center of hadronic mass, anyway) than would be an event
hadronized along the line of flight of an energetic parton recoil.
In general, I agree with the tone of these questions, that my analysis
is not very mature and a lot of checks need to be made before we draw
any conclusions. I will be working on that until the next neutrino
group meeting at the end of November, and I welcome any further
suggestions.
Bob