The History of BUU Codes at Giessen

Our interest in transport calculations started with a sabbatical visit of Ulrich Mosel at Michigan State University during spring/summer of 1986. During this time also the Giessen PhD student Wolfgang Bauer spent some time there, under the guidance of George Bertsch, but working for a degree at Giessen. While George had written one of the very first BUU codes Wolfgang Bauer rewrote this code and during Mosel's visit at MSU the first calculations of particle production, in this case photon production in heavy-ion reactions, were done and later published. With Wolfgang Bauer the very first Giessen BUU code returned to Giessen.

From then on a number of very good young scientists have worked on it. Wolfgang Cassing joined the effort at this point. First, Koji Niita (Japan) rewrote again large parts of the code and implemented pion production, photon production and a good description of the nuclear ground state, Angel de Paoli and Gustavo Batko (both Argentina) used it to calculate etas and kaons produced in heavy-ion collisions and Gyuri Wolf (Hungary) did a series of important early papers on dilepton production in heavy ion reactions which provided some of the motivation for the construction of HADES, the dilepton spectrometer at GSI. Very early on, a parallel development to construct a manifestly covariant BUU code was started. Giessen PhD students Volker Koch and Bernhard Blaettel wrote the first such code and used it to analyse BEVALAC data on flow and particle production (1990). Klaus Weber in his thesis (1992) did work on a covariant description of the momentum dependence of the mean field at high energies, where the Walecka model becomes too repulsive. Tomoyuki Maruyama (Japan) followed up on this and did the first numerical implementation of an explicit momentum dependence that corrects this deficiency of the Walecka model at high energies. Giessen PhD student Andreas Lang performed the first thermal analysis of heavy-ion reactions with this code; he analyzed the degree of equilibration during heavy-ion collisions and showed that full equilibration is reached on in the latest stages of a collision when the nuclear density has already decreased below its equilibrium value. Another Giessen PhD student, Stefan Teis, then did a very complete analysis of pion and eta production in heavy-ion collisions measured by the TAPS group at GSI (1997). He also analyzed the antiprotons and very-high energy pions produced in such reactions. At about this time, diploma student Alexander Hombach did the very first application of a BUU calculation to photon-induced reactions on a nucleus, analyzing the eta-production data obtained by the TAPS group at MAMI. Alexander Hombach, in his PhD thesis, did a corresponding detailed analysis of flow observables in such collisions. Teis, together with another PhD student, Martin Effenberger, rewrote parts of the code and implemented significantly more resonance excitations into the code. At roughly the same time (1998), PhD student Jochen Geiss, applied the earlier version of the code to an analysis of ultrarelativistic heavy-ion collisions, in particular to the problem of production of strangeness and charmonium in such collisions; the interest in these questions is still alive today.

At this point, as a spin-off the HSD project was established, focusing on High Energy Heavy-Ion Collisions. This project is maintained by W.Cassing and E.Bratkovskaya totally detached from the following development (cf. its ​website for details.).

Starting with the work of Hombach and Effenberger the code was applied to elementary collisions on a nucleus, as e.g. of photon- and pion-induced reactions. Effenberger, in his thesis, showed the first grand predictions for photon-nucleus collisions leading to dileptons. He also did the first high-energy calculations with photons where shadowing becomes important. In the following years 2000-05, Thomas Falter took this theme up again and worked on the description of high-energy non-resonant electron-induced reactions in the DIS regime, improving the treatment of shadowing in the code. Together with Kai Gallmeister*** a picture for the time-dependence of fragmentation was established. At the same time some activities in heavy-ion collisions were still taken up. Alexei Larionov*** (Russia) has worked on the description of heavy ion collisions in the SIS energy region and above; as a major topic, he studied the influence of medium-dependent cross sections. Together with diploma student Markus Wagner he investigated strangeness production in heavy ion collisions and, in particular, the 'strange horn' in the K/pi ratios. PhD student Jürgen Lehr included off-shell nucleons in his code version and was the first who worked on both photon- and electron-induced processes in the resonance region. The last results based on the the original Effenberger-Teis version were published by Pascal Mühlich in the year 2007. He worked on pion- and photon-induced processes both in the resonance and high energy region, thereby focusing on possible signals for changes of meson properties within the medium. During all of this time Stefan Leupold (then a postdoc and later Assistant Professor at Giessen, now at Uppsala) provided essential theoretical input both to our implementation of transport theory in GiBUU and to the study of in-medium properties of hadrons.

The courageous PhD student Oliver Buss*** during ​his PhD (2004-08) did a complete rewrite of the code and transformed it to a modular, modern FORTRAN version. Since this point we label the code by "GiBUU". While O. Buss was rewriting the basic ingredients as e.g. the propagation routines, the first person to join the new code structure was Thomas Falter with his high-energetic electron-induced reactions. Later, also A. Larionov and K. Gallmeister joined the initiative. A. Larionov took care of the implementation of strangeness production and improved the baryon-baryon cross sections. In a joined effort, O.Buss and K.Gallmeister implemented the local ensemble method. This algorithm was not included in the preceding Effenberger-Teis version and improved the speed of the full ensemble simulations. K.Gallmeister was also a major player in the speed-up of the core routines and the basic Makefile design. Furthermore, he managed to replace FRITIOF by a modified PYTHIA version for the treatment of higher-energy DIS processes. In 2006, Tina Leitner*** joined the project and implemented her model for neutrino-nucleon interactions and shared the work-load in the implementation of the off-shell potential. Lately, also Theo Gaitanos*** (nuclear fragmentation), David Kalok (nucleon spectral functions), Birger Steinmüller (ground state properties) and Janus Weil*** (dileptons and vector meson production, gfortran compatibility) contributed to the new code. Gaitanos and Larionov implemented another, relativistic version of the mean field and connected the code with a statistical method to describe fragment formation and decay.

On the physics side O. Buss showed that the observed in-medium changes of the 2 pi strength could be well explained by a state-of-the-art treatment of final state interactions as they were possible now with the GiBUU code. He also published very detailed results on pion-nucleus interactions and Tina Leitner obtained the first reliable results on neutrino-nucleus reactions, both for inclusive and exclusive reactions. For both of these topics the theoretical input by Luis Alvarez-Ruso (then a postdoc at Giessen, now at Valencia) was absolutely essential. Kai Gallmeister, on the high-energy side, went on to extract prehadronic interactions from a detailed comparison with the EMC and HERMES experiments. Later, Olga Lalakulich (Russia) joined this neutrino activity, concentrating on a good treatment of background terms in the pion-production channel. At the same time, Murat Kaskulov (Russia) worked with the code on an analysis of nuclear transparency measured at JLAB, with an eye on color transparency and Ivan Lappo-Danilevsky used GiBUU to investigate scaling in the total cross sections for interactions of leptons with nuclei. More recently (summer of 2012) Kai Gallmeister has implemented a better description of QE scattering of leptons into GiBUU as well as the initial 2p-2h excitations. Alexei Larionov is doing essential work with antiproton-induced reactions on nuclei and Theo Gaitanos has implemented the fragment-formation routines from a statistical model into GiBUU; he is also performing (together with A.L.) work on the double-step reaction to produce double hypernuclei with antiproton beams at the experiment PANDA at FAIR.

More recently, Alexei Larionov worked again on dilepton production in close connection to new data from the HADES experiment. This connection was also taken up by Kai Gallmeister who contributed studies of Coulomb-effects in heavy-ion collisions. A. Larionov also implemented the relativistic chiral doublet-model into GiBUU and K. Gallmeister is using this now to analyze photo-eta production on nuclei.

In a sideline of the main branch, Kai Gallmeister extended the model to include so called Hagedorn resonances, i.e. resonances with arbitrary large values for quantum numbers like baryon number, strangeness, charge (or isospin), and lately even charm. These resonances present an alternative presciption of hadronization in heavy ion collisions with a special regard on the concept of detailed balance. Calculations on strangeness production in very low energy heavy ion collisions (HADES) and determination of transport coefficients of a thermalized medium in box simulations have been performed.

Over the last few years Kai Gallmeister has been essential for continuously improving the code, maintaining the different versions and providing patches whenever shortcomings were detected.

Since the code has been made public, multiple external users are running the code, mainly for electron and neutrino induced reactions and also for hadron- and heavy-ion-induced reactions as, e.g., in the HADES experiment.