The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here:

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

QCD and event generators

The theory group in Lund has been a pioneer in the field of event generators for high energy physics, with particular emphasis on Quantum Chromodynamics (QCD), since the inception of the field in the late 1970s. Today seven seniors, Christian Bierlich, Rikkert Frederix, Gösta Gustafson (emeritus), Leif Lönnblad, Malin Sjödahl, Torbjörn Sjöstrand (emeritus) and Korinna Zapp all work on various aspects of this field, and participate actively in the development of several event generators.

Precision calculations

Colour calculations

Multi-purpose simulation

Comparison to data

The aim of an event generator is to simulate the collision process between two incoming particles, giving rise to an varying numbers of outgoing particles - an event. Many physics mechanisms are at play in these collisions, but common is that quantum mechanical choices between different possible outcomes have to be applied at each stage of the collision. Such random choices are made with so-called Monte Carlo techniques, hence the full name "Monte Carlo Event Generators".

Event generators, for general or special purposes, today play a key role both in experimental and theoretical studies of particle collisions. An experimental researcher can use a simulation to compare theory with data obtained in large collider experiments, such as the Large Hadron Collider at CERN. A theorist can use it to understand a theoretical framework better, by varying the assumptions made in the simulation, or to provide predictions for new areas of experimental study.

The scientific work underlying event generators spans a broad range. In one end of the spectrum, researchers in this group develop new formalisms for QCD calculations, and new models for the physics which cannot be calculated from first principles. At the other end of the spectrum, the group researchers develop software, which is in turn validated and shipped to experimental collaborators.

Some event generators are developed for general purpose, meaning that their usage in principle extends to all types of collisions, often with several different approaches to the same problem. Others are more specialized to either one type of collision, or a particular method of calculation. Below we give a brief overview of the main activities of the group.

Specialized programs perform ultra-precise calculations

Precision calculations play an important role in the success of collider experiments such as those carried out at the Large Hadron Collider. These calculations are necessary to account for the effects of higher-order quantum corrections on the interactions between subatomic particles, which can have a significant impact on the outcomes of collisions. Conducting these calculations often requires sophisticated mathematical methods and advanced computational tools.

The theory group in Lund is actively involved in developing techniques to incorporate these quantum corrections into the predictions for collisions using MadGraph5_aMC@NLO, a parton-level event generator. By accurately simulating the interactions between subatomic particles and taking into account the complex quantum corrections, MadGraph5_aMC@NLO can provide precise predictions for the outcomes of collider experiments.

New formalisms put structure to QCD colour

One of the challenges of performing calculations in QCD, is that it gives quarks and gluons the quantum number "colour". In QCD there are three colours, and corresponding anti-colours. They have nothing to do with the colour of daily life, but are rather constructs which mathematically behave somewhat similar to ordinary colours. The challenge is, that when colour is taken into account, the number of ways which particles can be correlated with each other, grows extremely fast. The codes ColorMath (Mathematica) and ColorFull (C++), developed at the division, provide ways treating this complexity.

Almost any type of collision can be generated by PYTHIA

The generator studies in Lund were begun with the JETSET program, which later evolved into the current PYTHIA. It has been extensively used for more than 40 years, at essentially all major collider experiments over that time. The program attempts to describe all aspects of many different collision types, and therefore consists of a large set of physics modules combined in a consistent but flexible structure. Some parts are based on external input, but the bulk of the physics models have been developed within the PYTHIA collaboration, which nowadays has expanded also to include members outside of Lund.

Comparing theory and data

When event simulations are to be compared to experimental data, we must take care to make sure that the comparison is, as we call it, “apples to apples“. In simulated collision events one has in principle full information about all aspects of the collision, while in experiment, only a subset can be measured, and a lot has to be inferred. Often the simulation is used to make these kinds of inferences, and one can therefore end up essentially comparing simulation to simulation with a bit of data input. The group participates in a collaboration called RIVET, which attempts to bring experimental analyses in a state, so apples to apples is always ensured.