Dispersive analysis of the pion vector form factor without zeros

TL;DR

This paper updates the dispersive analysis of the pion vector form factor using $e^+e^- o\pi^+\pi^-$ data, showing that assuming no zeros simplifies uncertainties and highlights discrepancies in muon magnetic moment calculations.

Contribution

It introduces a dispersive approach assuming the form factor has no zeros, reducing systematic uncertainties and providing new insights into data set discrepancies.

Findings

Data are compatible with zero-free form factor.

Discrepancies in muon g-2 contributions are amplified by dispersive constraints.

The pion charge radius helps distinguish between data sets.

Abstract

We perform an updated analysis of $e^+e^-\to\pi^+\pi^-$ cross-section data using a dispersive representation of the pion vector form factor. We show that the available data are compatible with the assumption that the form factor is free of complex zeros and that under this assumption the largest systematic uncertainty in a previous analysis can be eliminated. We investigate both a constrained Omn\`es representation as well as a hybrid phase-modulus representation and we quantify the discrepancies in the hadronic vacuum polarization contribution to the anomalous magnetic moment of the muon based on different $e^+e^-$ data sets. We find that the dispersive constraints exacerbate these discrepancies. Together with the assumption of the absence of zeros, the pion charge radius becomes a useful observable to discriminate between the different data sets. This provides an opportunity for future improved lattice-QCD determinations to probe the discrepancies independently of full computations of hadronic vacuum polarization. We also reevaluate the two-pion contribution to Euclidean windows and we observe that systematic discrepancies between the data sets persist even at very long distances.