Measurement of cross-counter leader fractions in an 18NM64: - - PowerPoint PPT Presentation

Measurement of cross-counter leader fractions in an 18NM64: Detecting single and multiple atmospheric secondaries

Alejandro Sáiz,1 Warit Mitthumsiri,1 David Ruffolo,1 Paul Evenson,2 and Tanin Nutaro3

1Department of Physics, Faculty of Science, Mahidol University, Bangkok 10400, Thailand 2Bartol Research Institute, University of Delaware, Newark, DE 19716, USA 3Department of Physics, Faculty of Science, Ubon Ratchathani University, Ubon Ratchathani 34190,

Thailand

July 18, 2017

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 1 / 12

Outline

Introduction Neutron Monitors The Princess Sirindhorn Neutron Monitor The Leader Fraction L Cross-counter Leader Fractions Absolute times and cross-counter distributions Cross-counter leader fraction and counter separation Discussion Summary and Conclusions

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 2 / 12

Introduction Neutron Monitors

Neutron Monitors

◮ Cosmic rays produce

cascades of secondary particles on Earth’s atmosphere

◮ If energetic enough,

secondary particles reach ground level

◮ Neutron monitors measure

atmospheric neutrons from cosmic ray cascades

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 3 / 12

Introduction The Princess Sirindhorn Neutron Monitor

The Princess Sirindhorn Neutron Monitor (PSNM)

◮ PSNM operates since late 2007 ◮ Location: Doi Inthanon, Chiang Mai

province, Thailand

◮ Altitude: 2560 m above sea level ◮ High vertical cutoff rigidity (16.8 GV)

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 4 / 12

Introduction The Princess Sirindhorn Neutron Monitor

PSNM is a standard 18-NM64:

◮ 18 proportional counter tubes

(10BF3 gas) on a continuos row

◮ 30 tons of Pb as neutron

producer

◮ Polyethylene neutron reflector

and moderators One atmospheric neutron interacts with a Pb nucleus producing more neutrons, which are moderated by the polyethylene and finally detected in the proportional counters through: n + 10B − − → 4He + 7Li*.

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 5 / 12

Introduction The Leader Fraction L

Neutron time delays

1 10 100 1000

n (s

• 1)

140 120 100 80 60 40 20

t (ms)

(a)

10

2 3 4 5 6

100

2 3 4 5 6

1000

n (s

• 1)

2.0 1.5 1.0 0.5 0.0

t (ms)

dead time

(b)

8 8

The neutron monitor data acquisition system records time delays between successive neu- tron counts

◮ Long time delays: counts from

independent atmospheric neutrons

◮ Short time delays: mostly from

neutrons produced from the same Pb nucleus = ⇒ information about cosmic ray energy

◮ Cosmic rays with higher energy

produce higher energy atmospheric neutrons

◮ Neutrons with higher energy produce

larger numbers of neutrons in the Pb

◮ Multiple neutrons produced together

in the Pb show short time delays

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 6 / 12

Introduction The Leader Fraction L

Leader fraction L

Relevant parameter: the leader fraction L.

◮ Can be extracted from time delay histogram ◮ A “leader” is a count that did not follow a previous count ◮ L interpreted as the ratio between the number of atmospheric particles that

produced at least one count, and the total number of counts

◮ Therefore L ≤ 1

◮ L contains information about cosmic ray energy

L normally defined as “same-counter” leader fraction

◮ It does not include correlations in nearby counters ◮ It may include some correlation of different atmospheric secondaries from the

same CR primary

◮ In this work we study cross-counter time correlations (preliminary results)

Long-term variations in L since late 2007 presented in the poster session

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 7 / 12

Introduction The Leader Fraction L

Figure: Leader fraction L and count rate C at PSNM since late 2007 (Banglieng et al, this conference).

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 8 / 12

Cross-counter Leader Fractions Absolute times and cross-counter distributions

Cross-counter time delays

t (ms)

2 4 6 8 10

)

• 1

N (ms

7

10

8

10

9

10

=1 ∆ =2 ∆ =3 ∆ =10 ∆ =17 ∆

Long-time histogram Short-time histogram Long-time fit

Figure: Unnormalized distributions of cross-counter time delays t at counter separation ∆ of 1, 2, 3, 10, and 17. Each histogram represents ∼ 2600 h of data.

◮ Definition of absolute times: Fit

individual oscillator sequence times to GPS time. Accuracy ∼ ±3 µs

◮ Smaller dead-time effect ◮ Distributions can be constructed

for any ordered pair of counters

◮ For simplicity, we accumulate all

counter combinations at the same separation ∆

◮ Fit to the exponential “tail”

gives cross-counter leader fractions L∆

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 9 / 12

Cross-counter Leader Fractions Cross-counter leader fraction and counter separation

Cross-counter leader fraction L∆ as a function of ∆

∆

2 4 6 8 10 12 14 16 18 ∆

L

0.8 0.85 0.9 0.95 1

L∆ increases with ∆ but never reaches L∆ = 1 (no correlation). A value ∼ 0.997 seems independent of ∆ for counter separation > 5.

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 10 / 12

Discussion

Interpretation of a constant L∆ = 1

◮ Time correlations in distant counters probably from correlated atmospheric

secondaries produced by the same CR primary ⇒ “atmospheric leader fraction” in a counter pair ∼ 0.997

◮ Total secondary leader fraction for the whole monitor: 0.99718 ∼ 0.953 ◮ Monte Carlo simulations of independent atmospheric secondaries does not

show significant cross-counter correlations (P.-S. Mangeard, private communication):

1 10

2

10

3

10

Neutron, 10MeV, ZEN=0, AZI=0

Tube 2 4 6 8 10 12 14 16 18 Tube 2 4 6 8 10 12 14 16 18

Neutron, 10MeV, ZEN=0, AZI=0

1 10

2

10

3

10

4

10

Neutron, 100MeV, ZEN=0, AZI=0

Tube 2 4 6 8 10 12 14 16 18 Tube 2 4 6 8 10 12 14 16 18

Neutron, 100MeV, ZEN=0, AZI=0

1 10

2

10

3

10

Neutron, 1GeV, ZEN=0, AZI=0

Tube 2 4 6 8 10 12 14 16 18 Tube 2 4 6 8 10 12 14 16 18

Neutron, 1GeV, ZEN=0, AZI=0

◮ Bare counters in PSNM station (no Pb, no reflector) show same-counter L

between 0.996 and 0.998 ⇒ Bares measure atmospheric leader fraction?

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 11 / 12

Summary and Conclusions

Summary and Conclusions

◮ In previous work, we succeeded relating same-counter time delays with

variations in CR spectrum

◮ Here we implement an absolute timing system for cross-counter time delay

analysis

◮ Correlations between distant counters consistent with multiple atmospheric

secondaries from the same CR primary

◮ We estimate the secondary leader fraction ∼ 0.953, which gives a secondary

multiplicity of ∼ 1.049

◮ Variations due to solar effects still under study

Thank you for your attention

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 12 / 12

Summary and Conclusions

Chance coincidences

R(t) is the probability of time delay ≥ t for a neutron count in one counter tube

◮ n(t) is the probability density function: n(t) ≡ −dR/dt ◮ α is the probability per unit time of having a new count

If all the counts were independent: dR dt = −αR,

• r

d dt ln R = −α, so R = e−αt and n = α e−αt = ⇒ a straight line in a log-linear plot of n(t)

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 13 / 12

Summary and Conclusions

Following counts

Short time delays are dominated by counts of neutrons produced from the same Pb nucleus = ⇒ “following” counts

◮ Total distribution is not the sum of follower distribution and chance

coincidences: distributions are not independent!

◮ Distribution of chance coincidences gets affected by followers, and vice versa

The “conventional” way to estimate following counts:

◮ Number of counts during a short time window: multiplicity ◮ Contaminated by chance coincidences

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 14 / 12

Summary and Conclusions

Leader fraction

If β(t) is the probability per unit time of a following count with delay t since the previous count from the same production event, then dR dt = −R (α + β(t)) ,

• r

d dt ln R = − (α + β(t)) , so R = e−αte−

t

0 β(t′) dt′

We define the nuclear part of R(t) as: Rn(t) ≡ e−

t

0 β(t′) dt′

Then n = α e−αtRn − e−αt dRn dt For long time delays, dRn/dt ≃ 0 and Rn ≃ constant: Rn ≃ Rn(∞) ≡ L = ⇒ leader fraction, and n ≃ α L e−αt

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 15 / 12

Summary and Conclusions

Cross-counter leader fraction

If βij(t) is the probability per unit time of a following count with delay t since the previous count from the same production event, then dRij dt = −Rij (αj + βij(t)) , or d dt ln Rij = − (αj + βij(t)) , so Rij = e−αjte−

t

0 βij(t′) dt′

We define the nuclear part of Rij(t) as: Rn,ij(t) ≡ e−

t

0 βij(t′) dt′

Then nij = αj e−αjtRn,ij − e−αjt dRn,ij dt For long time delays, dRn,ij/dt ≃ 0 and Rn,ij ≃ constant: Rn,ij ≃ Rn,ij(∞) ≡ Lij = ⇒ leader fraction, and nij ≃ αj Lij e−αjt

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 16 / 12

Summary and Conclusions

0.7870 0.7860 0.7850 0.7840 L 01/01/2008 01/01/2009 01/01/2010 01/01/2011 01/01/2012 01/01/2013 01/01/2014 Date (UT) 2.30 2.25 2.20 2.15 18NM (10

6h

• 1)

566 564 562 560 558 P (mmHg) 10 8 6 4 2 Ew (mmHg) 0.0280 0.0270 0.0260 0.0250 3B/18NM (a) (c) (d) (e) (b)

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 17 / 12

Summary and Conclusions 2.28 2.24 2.20 2.16 2.12 18NM (10

6h

• 1)

01/01/2012 01/02/2012 01/03/2012 01/04/2012 01/05/2012 Date (UT) 566 564 562 560 P (mmHg) 8 6 4 2 Ew (mmHg) 5.26 5.24 5.22 M1/M2 4400 4200 4000 3800 Thule (h

• 1)

0.7864 0.7860 0.7856 0.7852 0.7848 L (a) (b) (c) (d) (e) (f)

(a) Count rate at polar neutron monitor (Thule) (b) Count rate at the Princess Sirindhorn neutron monitor (c) Leader fraction at the PSNM (d) Atmospheric water vapor pressure at Doi Inthanon (e) Atmospheric pressure at Doi Inthanon (f) Count multiplicity ratio at the PSNM

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 18 / 12

Summary and Conclusions

Measurement of L during the 2000–2007 Latitude Surveys

(GV)

c

P

2 4 6 8 10 12 14 16

Leader fraction

0.82 0.825 0.83 0.835 0.84 0.845 0.85 0.855

MC Min & Max DATA - 2001 DATA - 2002 DATA - 2003 DATA - 2004 DATA - 2005 DATA - 2006

Figure: Leader fraction L as a function of geomagnetic cutoff rigidity Pc for the 6 surveys that recorded time-delay histograms. The change of L with Pc confirms that it conveys spectral information. At low Pc it is also clear how L changed between years of high solar activity (e.g., 2001) and low solar activity (e.g., 2006) due to solar modulation.

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 19 / 12

Summary and Conclusions

Corrections to the equation

◮ The electronics have a dead time tdead ≃ 0.1 ms (time delays shorter than

tdead are not recorded)

◮ There is an electronics overflow at toverflow ≃ 142 ms (time delays longer than

toverflow are recorded modulo toverflow) Both effects vary with α. We include the corrections in order to estimate L more accurately: n ≃ α L eαtdead

• 1 +

1 eαtoverflow − 1

• e−αt.

Alejandro Sáiz Cross-counter leader fractions ICRC2017, July 18, 2017 20 / 12