Agreement between the Xmax distributions measured by the Pierre - PowerPoint PPT Presentation

Agreement between the Xmax distributions measured by the Pierre Auger and Telescope Array Observatories J. Bellido, J. Belz, S. Blaess, V. de Souza, W. Hanlon, D. Ikeda, P. Sokolsky, Y. Tsunesada, M. Unger, A. Yushkov, for the Pierre Auger and TA Collaborations Presenter: Vitor de Souza University of São Paulo 1

We present the solution for a decade-long controversy. Take away message TA and Auger composition measurements (Xmax) agree within the systematics 18.2 < log 10 (E/eV) < 19.0 2

Don’t jump into conclusions TA Collaboration, ICRC2017 Auger Collaboration, ICRC2017 Different detectors and analysis. 3

Auger bias Auger ≠ – Xmax at the atmosphere TA bias – Approx. unbiased TA Xmax ATM – Xmax in the detector ≠ – Enhance statistics Xmax DET 4

Auger bias ≠ TA bias 5

How to compare Xmax from Auger and TA ? N = 1953 N = 625 Xmax Auger Xmax TA Xmax always in g/cm 2 6

Auger Data Mean = 738 Mean = 744 RMS = 60 RMS = 66 Xmax Auger Xmax TA (Auger) TA Detector Simulation TA Analysis 7

AugerMix Mean = 738 RMS = 60 Xmax TA ( AugerMix ) Xmax Auger SAME RESULTS WITH EPOS-LHC TA Detector Simulation TA Analysis 8

Now we can compare the distributions TAData AugerMix Xmax TA (TAData) Xmax TA Are these distributions compatible ? Xmax TA (AugerMix) 9

Get P1 data Compatibility tests: P1 = 5.34 x 10 -21 Anderson-Darling and TAData Kolmogorov-Smirnov AugerMix P2 < 10 -5 Normalize P1 probabilities Xmax TA Get 10 5 P1 MC probabilities P1 MC Distribution Count (N) how many times 10 5 MC realizations of AugerMix P1 MC < P1 data Compatibility Probability P2 = N/10 5 10

11

Systematic Uncertainties Xmax Systematic Uncertainty TA = 20.3 g/cm 2 Auger ~ -10 and +8 g/cm 2 Shift Xmax to match the mean of the distributions. Recalculate the compatibility tests. 12

Xmax Shift versus Systematic Uncertainties 13

14

Compatibility between TA data and AugerMix 15

Repeat the same analysis but now calculates the compatibility probability P2 between TA data and pure proton Xmax distributions. 16

Comparison of TA data with AugerMix and Protons 17

Conclusion 18.2 < log 10 (E/eV) < 19.0 TA Xmax distributions are compatible to AugerMix Xmax distributions within the systematic uncertainties. 18

Conclusion 18.2 < log 10 (E/eV) < 19.0 Proton TA Xmax AugerMix TA Xmax distributions are as compatible to pure proton composition as they are to AugerMix. 19

Xmax Compatibility Table 18.2 < log 10 (E/eV) < 19.0 Compatible Incompatible Proton Iron Mixed Auger only only Auger !!! TA 20

Request to the UHECR community Proper comparison demands use of detector simulation and analysis chain of both experiments. Please refer to this common (Auger/TA) analysis in your studies. Do not propagate an unsustainable controversy. 21

Extras 22

Compatibility tests ● Kolmogorov-Smirnov and Anderson-Darling ● Calculate the probability (P1 data ) that the two distributions were generated from the same parent distribution DET Xmax TA (AugerMix) versus DET Xmax TA (TAData) ● Draw 10 5 MC distributions of DET Xmax TA (AugerMix) – Use fit Gauss ⊗ Exponential – Calculate 10 5 P1 MC – Get P1 MC distribution -> Compatibility distribution ● Using the distribution P1 MC calculate the probability P2 of finding a P1 worse than P1 data P2 = Compatibility Probability 23

Comparison of TA data with AugerMix and Protons 24

Mean Xmax Comparison WG - UHECR 2016 25

Mean Xmax Comparison WG - ICRC 2015 26