### Parton cascades and elliptic flow

Three new preprints address the question of what it takes to describe the measured elliptic flow in Au+Au collisions at RHIC in the framework of a partonic Boltzmann equation:

- Scalings of elliptic flow for a fluid at finite shear viscosity
- The “minimal” viscosity and elliptic flow at RHIC
- Towards a unified understanding of jet-quenching and elliptic flow within perturbative QCD parton transport

The first two studies use QCD-inspired elastic parton cross sections and adjust its value dynamically such that the effective kinematic shear viscosity (η/s) is kept constant, e.g. equal to 1/4π. The two preprints differ slightly in the prescription: *Ferini et al.* adjust the cross section as a function of the parton density; *Molnar* scales the cross section as function of time. Such scaling is, in fact, approximately realized in perturbative QCD. Note, however, that the values of the transport cross section needed to describe such a small shear viscosity are much larger than predicted by perturbative QCD.

Both studies find a semi-quantitative agreement with the RHIC data (*Ferini et al.*)

and hydrodynamics calculations (*Molnar*), respectively. The *Ferini et al.* study finds that scaling of v_{2} with the eccentricity is violated if a partonic freeze-out at the critical energy density is implemented, in agreement with recent data from PHOBOS and STAR.

The third preprint (*Fochler et al.*) reports new results from a parton cascade which includes both elastic (*gg ↔ gg*) and inelastic (*gg ↔ ggg*) collisions. As the same group reported earlier, the inelastic processes are found to be highly effective at increasing the transport cross section, resulting in a shear viscosity near the lower bound η/s = 1/4π. The new work shows results for the elliptic flow as well as jet quenching (*R*_{AA}), which are in semi-quantitative agreement with the RHIC data. It will be interesting to explore further how sensitive the results are to the specific implementation of the cross section for *gg ↔ gg* used by the authors.

All three studies identify the partonic elliptic flow with the observed hadronic elliptic flow. This may not be correct. If the valence quark number scaling v_{2}(p_{T}) is taken at face value, the partonic v_{2} saturates at 7.5% instead of 15%. As noted by *Ferini et al.* a significantly smaller parton cross section, which may be compatible with perturbative QCD, may then be sufficient to explain the RHIC data. In order to study this quantitatively, the quark recombination mechanism must be implemented within the parton cascades.