Sivers vs. Collins effect in azimuthal single spin asymmetries in pion production in SIDIS

Recently it has been argued that the transverse momentum dependent twist-2 Sivers distribution function does not vanish in QCD. Therefore both, the Collins and Sivers effects, should be considered in order to explain the azimuthal single spin asymmetries AUL in pion production in semi-inclusive deeply inelastic lepton scattering of a longitudinally polarized target. On the basis of presently available phenomenological information on the Sivers function we estimate that for those asymmetries AUL in the kinematic region of the HERMES experiments the Sivers effect can be neglected with respect to the Collins effect. It is argued that the same feature holds also for the COMPASS and CLAS experiments. This justifies theoretical approaches to understand the HERMES data on the basis of the Collins effect only.