Tests of Power Corrections to Event Shape Distributions from e+e- Annihilation
P. A. Movilla Fernández, O. Biebel, S. Bethke
TL;DR
The paper tests non-perturbative power corrections to differential event-shape distributions in $e^+e^-$ annihilation by combining dispersive power-corrections (characterized by a universal $\alpha_0(\mu_I)$) with two-loop perturbative QCD and NLLA resummation across thrust, the $C$-parameter, and jet broadening observables. It uses data from JADE and LEP/SLC spanning roughly from 14 to 183 GeV and performs simultaneous fits for $\alpha_s(M_{Z^0})$ and $\alpha_0(\mu_I)$ under four matching schemes, including updated jet-broadening corrections. The results show $\alpha_0(2~\mathrm{GeV}) = 0.502^{+0.046}_{-0.032}$ (exp. syst) $^{+0.074}_{-0.053}$ (theo. syst) with $\alpha_s(M_{Z^0}) \approx 0.107^{+0.006}_{-0.005}$, supporting universality of $\alpha_0$ within uncertainties and improved consistency when revised power corrections for jet broadening are used. However, $B_W$ shows tension with other observables, and excluding it yields an $\alpha_s(M_{Z^0})$ in better agreement with the world average, reinforcing the overall validity of the power-corrections approach and its energy-dependent predictions.
Abstract
A study of differential event shape distributions using e+e- data at centre-of-mass energies of 35 to 183 GeV is presented. We investigated non-perturbative power corrections for the thrust, C-parameter, total and wide jet broadening observables. We observe a good description of the distributions by the combined resummed QCD calculations plus power corrections from the dispersive approach. The single non-perturbative parameter α_0 is measured to be α_0 (2 GeV) = 0.502 +- 0.013 (stat.) ^{+0.046)_{-0.032} (exp. syst.) ^{+0.074}_{-0.053} (theo. syst.) and is found to be universal for the observables studied within the given systematic uncertainties. Using revised calculations of the power corrections for the jet broadening variables, improved consistency of the individual fit results is obtained. Agreement is also found with results extracted from the mean values of event shape distributions.