Transcript String 21 Flashers and AMANDA

String 21 Flashers and
AMANDA
Michelangelo D’Agostino
UC Berkeley
London Collaboration Meeting
Thanks



Dave Hardtke: for doing the first work
on this
Kurt: for being the ice guru
John Pretz: for writing his wonderful,
bugless merging module for the rest of
us to use
M. D’Agostino
Outline



Are the String 21 flashers bright
enough to be seen by AMANDA?
What do these events look like?
Can such flasher events be used for
any calibration or ice property
measurements?
M. D’Agostino
The Verdict:
Yes!
 up to 60% of
flashes trigger
AMANDA
 Roughly 40 to
50% of the
AMANDA events
were stringtriggered

M. D’Agostino
closest
AMANDA
strings
What do the events look
like?
DOM 23 flashing
M. D’Agostino
What do the events look
like?
DOM 23 flashing
M. D’Agostino
What do the events look
like?
9 10
16
Stuck String
M. D’Agostino
What do the events look
DOM 23 z position
like?
M. D’Agostino
Up and Down String 21
DOM 12
DOM 23
DOM 32
M. D’Agostino
Analysis Details



in July, flasher runs were taken for all
DOM’s on string 21
10 Hz, full brightness, all 12 LED’s
run monolith offline, requiring a simple
multiplicity of 10 string 21 hits in a
5000 ns window
M. D’Agostino
Analysis Details


use John’s Icetray merging module to
look for AMANDA triggers with -20 s
< t < -5 s of a String 21 trigger
filter out muons (about 1/4 of events):
require that the flashing DOM be
present in the event and that ATWD
channel 3 has a digitized current pulse
M. D’Agostino
Up and Down String 21
Ryan’s dustlogger
data for String 21
DOM 23
what’s
going on
here? DOM
16
M. D’Agostino
Ice Properties: Fluence
Analysis


from Poissonian statistics, the average
number of received photons in the diffusive
regime is given by <N>=-log[P(0)], where
P(0) is the probability that the receiver is
not hit
1/λ, where λ is the propagation length, is
given by the slope of log[<N>d] vs. d,
where d is distance from the light source
M. D’Agostino
Ice Properties
3 closest strings
DOM 16 flashing
restrict z to +-5 m
of the flashing DOM
to probe a narrower
ice layer
corresponds to a λ
of 45.3 m +-11.8 m
photons all the
way to ~400 m
M. D’Agostino
Ice Properties
M. D’Agostino
Ice Properties
dip from
previous graph
M. D’Agostino
maximum from
previous graph
Ice Properties




we have a biased sample though, since
AMANDA had to be triggered to see the OM
hits; the data is L1 filtered as well
P(0) is an overestimate
<N> is an
underestimate that varies with distance
thus, these are upper limits on the
propagation lengths
flasher variability is more likely the culprit
for the suspicious dip (cf. Chris Wendt’s
talk); longer width runs are needed to tell
M. D’Agostino
Conclusions



the flashers are pretty damn bright
ice properties have been measured over a
greater horizontal distance and give
reasonable agreement
since flashers should be seen in ~9 IceCube
strings, such a method will be much more
useful in the future for measuring local ice
properties, especially with a unified trigger
system
M. D’Agostino
Conclusions

flasher coincidences might also be
useful in AMANDA simulation—a
simulation capable of reproducing the
depth variation in this simple flasher
system would surely have a better
handle on systematics (COGz?)
M. D’Agostino