Pileup subtraction using jet areas

Abstract: One of the major challenges for the LHC will be to extract precise information from hadronic final states in the presence of the large number of additional soft pp collisions, pileup, that occur simultaneously with any hard interaction in high luminosity runs. We propose a novel technique, based on jet areas, that provides jet-by-jet corrections for pileup and underlying-event effects. It is data driven, does not depend on Monte Carlo modelling and can be used with any jet algorithm for which a jet area can be sensibly defined. We illustrate its effectiveness for some key processes and find that it can be applied also in the context of the Tevatron, low-luminosity LHC and LHC heavy-ion collisions.

• The paper introduces a jet area-based method that subtracts pileup by adjusting jet transverse momentum using a noise level measured event-by-event.
• It employs a parameter-free approach to extract the noise level via the median pₜ/A among jets and validates the method across various experimental setups.
• The method significantly improves measurement precision in contexts including dijet events, Z' mass reconstruction, and top quark analyses.

Pileup Subtraction Using Jet Areas

The research paper "Pileup subtraction using jet areas," authored by Matteo Cacciari and Gavin P. Salam, presents a distinct methodology aimed at addressing the key issue of pileup in high-energy physics experiments, a phenomenon particularly prevalent at the Large Hadron Collider (LHC). Pileup refers to the occurrence of numerous additional soft proton-proton (pp) interactions overlapping with the primary hard scatter event within the same bunch crossing. The presence of pileup poses significant challenges to the accurate measurement of hadronic final states and requires novel solutions for effective data analysis.

The authors propose a data-driven correction technique based on jet areas that can effectively counteract pileup without reliance on Monte Carlo simulations. The method is designed to work with infrared-safe jet algorithms and is versatile enough to be applied to any algorithm for which a meaningful jet area can be calculated. Two novel aspects form the foundation of this method: the identification of each jet's susceptibility to diffuse noise via its calculated area, and a parameter-free method to measure the noise level, represented by $\rho$, in a given event.

Methodology and Principles

The proposed approach adjusts the transverse momentum ($p_t$) of jets post jet clustering based on the pileup correction formula given by:

$\Delta p_t = A \rho \pm \sigma \sqrt{A} - L$

Here, $A$ represents the jet area, $\rho$ the noise level contributed by minimum-bias events, $\sigma$ the standard deviation of this noise distribution, and $L$ a term accounting for the potentially lost or gained jet components due to pileup. The procedure assumes the fluctuation-term $\sigma$ is smaller than $\sqrt{A}\rho$ and often neglects $L$, facilitating pileup subtraction through:

$p_{tj}^{(\text{sub})} = p_{tj} - A_j \rho$

The determination of $\rho$ is a critical step and it is extracted using the median $p_t / A$ value among all detected jets, isolating the baseline noise level from the hard interaction's contributions.

Experimental Validation and Applications

The authors validate their method across different experimental setups, exhibiting its robustness and applicability beyond the high-luminosity environment of the LHC. Several use cases were presented:

1. Dijet Events at LHC: The method was applied to simulated dijet events, demonstrating substantial improvement in the precision of transverse momentum measurements after pileup subtraction.
2. Leptophobic $Z'$ Mass Reconstruction: Implementing the method enabled the correction of mass peak shifts and spread due to pileup, restoring measurement fidelity.
3. Top Quark Reconstruction at Tevatron: Despite moderate pileup, the subtraction technique notably improved mass peak centrality and reduced resolution degradation.
4. Heavy-Ion Collisions: The methodology adapted to the high background environment of Pb collisions, affirming its efficacy in significantly noisy conditions.

Implications and Future Perspectives

The proposed methodology provides a robust, flexible, and efficient solution for pileup correction, independent of specific detector setups or simulation adjustments. Its parameter-free nature, coupled with the precise, event-specific noise measurement, offers significant promise for jet energy calibration, particularly at facilities like the LHC where pileup is a persistent concern. The approach's adaptability suggests potential for cross-application in other high-energy physics experiments facing similar noise challenges.

Looking ahead, the work paves the way for further optimization in experimental setups and Monte Carlo tuning specifically concerning soft pp interaction modeling. In addition, the direct event-by-event noise quantification enables enhanced analysis of underlying event structures, providing insights that could refine both theoretical models and practical techniques in jet physics. As experimental conditions evolve, the methodology could be the basis for future advancements in both theoretical and experimental high-energy physics.