A multi-stage apporach for jet evolution

The PHYS-WG has completed developing the multi-stage approach for parton showers through different regions of phase space in relativistic heavy-ion collisions. The MATTER event generator is applied to describe jet showers at high virtuality scale where rare-scattering multiple-emission dominates parton splitting. Two event generators, LBT and MARTINI are utilized to describe parton elastic and inelastic scatterings inside the medium at low virtuality scale where multiple-scattering induced single-emission dominates parton evolution. These two event generators are shown comparable for parton observables at the time scale that is comparable to the realistic QGP lifetime. And an AdS/CFT based event generator is adopted to describe parton energy loss at the thermal scale in the strong coupling limit. This unified approach was numerically realized in a Patch Code we developed and validated in a static medium for jet evolution at partonic levels, where we investigated the relative weights of the different energy loss schemes for the medium modifications experienced by the jet by varying the parton energy, medium length and separation virtuality scale. Our results serve as an important benchmark for developing a multi-stage scheme of jet evolution in heavy-ion collisions, and have been documented and published in Phys. Rev. C96 (2017) 024909.

The PHYS-WG has closely collaborated with the SD-WG to embed our Patch Code of the multi-stage jet evolution into the JETSCAPE framework.


Hadronization models for jets

The PHYS-WG is developing 3 different hadronization models to convert partons into hadrons.

The Colored Hadronization Model requires a full record of parton color information during jet showers, based on which a long string for each shower is constructed and fed back to Pythia for breaking into final state hadrons. 

The Colorless Hadronization Model does not require tracking the color information during parton evolution. Instead, colors are assigned to the final state partons based on their phase space configuration. Partons are then fed back to Pythia for breaking into hadrons. 

The Recombination (Coalescence) Model determines the probablity of partons combining into hadrons based on the overlap (Wigner functions) between parton wavefuctions and hadron wavefunctions. Partons that do not combine are returned to Pythia for string fragmentation.


Physics results of jets in p-p and A-A collisions

The PHYS-WG has succeeded in tuning the JETSCAPE framework such that it provides good descriptions of most jet observables in proton-proton collisions at both RHIC and the LHC,  such as single inclusive hadron spectra, single inclusive jet spectra,  jet shape, etc. This validates the physics function of the JETSCAPE software, as well as provides a crucial proton-proton baseline for studying the nuclear modification of jets in nucleus-nucleus collisions. Our results in proton-proton collisions have been documented and under preparation for a second collaboration publication. Effects of hadronization on jet observables, which was usually ignored by the heavy-ion community, will also be systematically discussed in this publication. Preliminary results of jet observables in nucleus-nucleus collisions are also shown to be consistent with experimental data. Details can be found in the following presentations.

Quark Matter 2018 Presentation by Kolja Kauder 

CIPANP 2018 Presentation by Abhijit Majumder 


Bayesian extraction of the jet transport parameter (in collaboration with the STAT-WG)

The PHYS-WG has worked together with the STAT-WG to establish a statistical analysis framework that helps in extracting the precise values of the jet transport parameter from model to data comparison. The most general and sophisticated parametrization of the jet transport parameter has been designed with five parameters, taking into account of its dependences on both the energy (or virtuality) scale of the jet parton and the temperature scale of the thermal medium. The newly developed multi-stage approach is implemented to calculate the jet quenching observables on 80 Latin hypercubes evenly distributed across the 5-diemensional parameter space. This takes over a million CPU hours on the Open Science Grid (OSG), and the collected results are utilized to train the Gaussian Process (G.P.) emulator. The G.P. then serves as a fast surrogate of the realistic physics model calculation and is used to compare to experimental data across the multi-dimensional parameter space, and extract the posterior distribution of this parameter space using the Bayesian analysis. This leads to a paradigm shift in the methodology of parameter extraction, from which we obtain a 90% credible region of the jet transport parameter. The separation scale in the multi-stage jet evolution has also been extracted for the first time. Its dependence on collision system has been explored. This will lead to a significant step forward towards the precise quantitative understanding of jet energy loss in QGP. Another publication is also under preparation based on this new statistical framework and its physics findings. 


Upcoming developments

  • Incorporate hadronic rescatterings (SMASH) into JETSCAPE
  • Incorporate heavy quarks into JETSCAPE
  • Realize concurrent simulation of hydrodynamic evolution and jet propagation, and investigate jet-induced medium excitation in heavy-ion collisions


 ↑ back to top