Abstract. The theoretical predictions for the subtraction constants lead to a very accurate dispersive representation of the ππ scattering amplitude below 0.8 GeV. The extension of this representation up to the maximum energy of validity of the Roy equations (1.15 GeV) requires a more precise input at high energies. In this paper we determine the trajectories and residues of the leading Regge contributions to the ππ amplitude (Pomeron, f and ρ), using factorization, phenomenological parametrizations of the πN and N N total cross sections at high energy and a set of sum rules which connect the high and low energy properties of ππ scattering. We find that nonleading Regge terms are necessary in order to achieve a smooth transition from the partial waves to the Regge representation at or below 2 GeV. We obtain thus a Regge representation consistent both with the experimental information at high energies and the Roy equations for the partial waves with ℓ ≤ 4. The uncertainties in our result for the Regge parameters are sizable but in the solutions of the Roy equations, these only manifest themselves above KK threshold.
MotivationLow energy pion physics has become a precision laboratory: the chiral symmetry properties of the Standard Model can now be compared with low energy precision experiments, using Chiral Perturbation Theory (χPT), dispersion theory and numerical simulations of QCD on a lattice. In principle, as illustrated by the prediction for the magnetic moment of the muon, physics beyond the Standard Model can show up at low energies, provided the quantity of interest can not only be measured accurately, but can also be calculated to sufficient precision. In this context, the interaction among the pions and in particular, the elastic ππ scattering amplitude play a crucial role.The Roy equations [1] provide a suitable framework for the low energy analysis of the ππ scattering amplitude, as they fully incorporate the basic properties that follow from analyticity, unitarity and crossing symmetry. They express the real parts of the partial waves as integrals over the imaginary parts that extend over all energies. The high energy behaviour of the imaginary parts thus enters the analysis, even if the Roy equations are evaluated only at low energy. More specifically, the contributions from the high energy region are contained in the so-called driving terms. Near the ππ threshold, these terms are very small. Accordingly, the results for the threshold parameters are not sensitive to the high energy contributions. The driving terms, however, grow with the energy. In [2,4], where the threshold parameters and the coupling constants of the effective chiral SU(2)×SU(2) Lagrangian relevant for ππ scattering are determined to high accuracy, the Roy equations are solved only below 0.8 GeV. On general grounds [1] these equations are valid in the region 1 s ≤ 68 M 2 π , i.e. for √ s ≤ 1.15 GeV, but in order to use them we need an accurate evaluation of the driving terms. This calls for a better understanding of the imaginary ...