Periodic Reporting for period 2 - DISCOVERHEP (Turning noise into data: a discovery strategy for new weakly-interacting physics)
Reporting period: 2022-10-01 to 2024-03-31
For decades, the Standard Model of particle physics has defined our understanding of the fundamental constituents of matter and the ways in which they interact. With the discovery of the Higgs boson in 2012, all of the components of the Standard Model are now in place, and yet at the same time we know that it is not a complete description of the Universe. The most obvious exception is the lack of a description for how particles interact with the gravitational force, but even neglecting this limitation, there are several open questions that the Standard Model does not provide an answer to.
One key ingredient that is beyond any Standard Model explanation is the existence of dark matter. We have seen clear evidence for the presence of dark matter in numerous different astronomical observations, and we have even measured the dark matter abundance in the Universe to be five times that of the normal matter that we are familiar with in our everyday lives. However, despite all of these observations in the Universe, we have yet to find any experiment-verified means by which the properties of dark matter and its interactions with the Standard Model can be described.
One of the leading classes of theories posits that dark matter is a particle which only weakly interacts with the Standard Model; if this is the case, dark matter could be produced very rarely in collisions at the Large Hadron Collider (LHC) at CERN in Geneva. Discovering such dark matter would require the analysis of enormous datasets, and the LHC experimental community is actively searching for such possibilities.
The DISCOVERHEP project starts from this baseline, but turns the normal dataset analysis strategy on its head. When the two proton beams of the LHC are collided, this results in many individual proton-proton collisions. The normal analysis strategy focuses on a single one of these collisions, typically looking at those which are very energetic: such energetic collisions were impossible to produce at previous particle colliders, and thus the discovery of dark matter may just be waiting for us to reach a high-enough energy. Instead, the DISCOVERHEP project discards the high energy collision, and focuses on analysing all of the low energy collisions: perhaps dark matter was not seen at previous colliders because it so rarely interacts with the Standard Model, and thus the discovery of dark matter may be waiting for a larger dataset of low-energy collisions.
One key ingredient that is beyond any Standard Model explanation is the existence of dark matter. We have seen clear evidence for the presence of dark matter in numerous different astronomical observations, and we have even measured the dark matter abundance in the Universe to be five times that of the normal matter that we are familiar with in our everyday lives. However, despite all of these observations in the Universe, we have yet to find any experiment-verified means by which the properties of dark matter and its interactions with the Standard Model can be described.
One of the leading classes of theories posits that dark matter is a particle which only weakly interacts with the Standard Model; if this is the case, dark matter could be produced very rarely in collisions at the Large Hadron Collider (LHC) at CERN in Geneva. Discovering such dark matter would require the analysis of enormous datasets, and the LHC experimental community is actively searching for such possibilities.
The DISCOVERHEP project starts from this baseline, but turns the normal dataset analysis strategy on its head. When the two proton beams of the LHC are collided, this results in many individual proton-proton collisions. The normal analysis strategy focuses on a single one of these collisions, typically looking at those which are very energetic: such energetic collisions were impossible to produce at previous particle colliders, and thus the discovery of dark matter may just be waiting for us to reach a high-enough energy. Instead, the DISCOVERHEP project discards the high energy collision, and focuses on analysing all of the low energy collisions: perhaps dark matter was not seen at previous colliders because it so rarely interacts with the Standard Model, and thus the discovery of dark matter may be waiting for a larger dataset of low-energy collisions.
The DISCOVERHEP project revolves around the central idea that new physics (dark matter or otherwise) may be present at low energy, it is just a very rare process, thus we need to have access to the largest possible low-energy-collision dataset. This is done by exploiting the fact that the LHC produces many proton-proton collisions each time the two proton beams are collided, what happens in each of these collisions is independent of all of the others, and the overwhelming majority of the collisions result in low-energy physics processes. It is therefore possible to remove the high-energy collision that caused the data to be recorded; the remaining low-energy collisions form a large and previously neglected dataset that may be the key to discovering rare new physics.
The above statement is easy to claim from a theoretical perspective, but experimentally it comes with many challenges. While the different collisions are independent, in order to observe those collisions we need to use an experimental apparatus (a series of particle detectors with different properties), and that apparatus is not a perfect device. It is therefore a complicated process to trace back each observation of a signal in each of the detectors to the collision it came from, and in some cases it is not always possible to do so in a reliable manner; in these cases, the uncertain collisions must be discarded. This process is referred to as reconstruction, and the traditional reconstruction software used by LHC experiments assumes that only one of the collisions is of interest.
The ATLAS Experiment is one of the two large general-purpose experiments at the LHC, and is the experiment to which the DISCOVERHEP group is affiliated. The main work performed in the first half of the DISCOVERHEP project has been to adapt the normal ATLAS reconstruction software from assuming there is one collision of interest, to instead processing all of the recorded collisions, and resolving any possible conflicts that arise from this process between different collision interpretations. This effort is mostly complete, and preliminary studies have demonstrated both the feasibility and utility of such an approach in studying low-energy collisions at the LHC with dramatically increased statistical sensitivity compared to traditional approaches.
The above statement is easy to claim from a theoretical perspective, but experimentally it comes with many challenges. While the different collisions are independent, in order to observe those collisions we need to use an experimental apparatus (a series of particle detectors with different properties), and that apparatus is not a perfect device. It is therefore a complicated process to trace back each observation of a signal in each of the detectors to the collision it came from, and in some cases it is not always possible to do so in a reliable manner; in these cases, the uncertain collisions must be discarded. This process is referred to as reconstruction, and the traditional reconstruction software used by LHC experiments assumes that only one of the collisions is of interest.
The ATLAS Experiment is one of the two large general-purpose experiments at the LHC, and is the experiment to which the DISCOVERHEP group is affiliated. The main work performed in the first half of the DISCOVERHEP project has been to adapt the normal ATLAS reconstruction software from assuming there is one collision of interest, to instead processing all of the recorded collisions, and resolving any possible conflicts that arise from this process between different collision interpretations. This effort is mostly complete, and preliminary studies have demonstrated both the feasibility and utility of such an approach in studying low-energy collisions at the LHC with dramatically increased statistical sensitivity compared to traditional approaches.
The development of the central DISCOVERHEP methodology, by which every recorded collision is reconstructed and independently analysed, represents a fundamental transition in the interpretation and relevance of the LHC dataset. Before the development of this approach, the LHC proton-proton collision physics programme was primarily of relevance to the study of high-energy phenomena, with very infrequent “special runs” aimed at studying low-energy physics. The DISCOVERHEP project instead provides a way by which both low-energy and high-energy phenomena can be studied simultaneously at the LHC, just by changing how the recorded data is reconstructed.
The DISCOVERHEP methodology is already relevant now, but the importance of such a development will grow with time, as the LHC and then the High Luminosity LHC (HL-LHC) produce more and more simultaneous proton-proton collisions. In case future colliders are built, then this new methodology may influence the design of the associated particle detectors: the sensitivity to low-energy physics using such an approach is limited by the ability to identify the collision each signal in the detector comes from, and an appropriately designed detector could perform well in studying both low-energy and high-energy physics objectives.
With the creation of the unique low-energy collision dataset nearing completion, work has begun on the analysis of this dataset for different possible types of new physics, with an emphasis on possible candidates that could describe the way in which the Standard Model interacts with dark matter. In the second half of the project, the DISCOVERHEP team is eager to continue the search for dark matter and to expand the use of this new methodology in searches for other possible types of new physics.
The DISCOVERHEP methodology is already relevant now, but the importance of such a development will grow with time, as the LHC and then the High Luminosity LHC (HL-LHC) produce more and more simultaneous proton-proton collisions. In case future colliders are built, then this new methodology may influence the design of the associated particle detectors: the sensitivity to low-energy physics using such an approach is limited by the ability to identify the collision each signal in the detector comes from, and an appropriately designed detector could perform well in studying both low-energy and high-energy physics objectives.
With the creation of the unique low-energy collision dataset nearing completion, work has begun on the analysis of this dataset for different possible types of new physics, with an emphasis on possible candidates that could describe the way in which the Standard Model interacts with dark matter. In the second half of the project, the DISCOVERHEP team is eager to continue the search for dark matter and to expand the use of this new methodology in searches for other possible types of new physics.