greatitself.
Continue
research33 min studyLesson 627 in Physics

Define jets that survive unresolved radiation

Connect real–virtual cancellation to a measurement function and perform a transparent anti-kt clustering decision with declared distance units.

Before you begin
On this page

The question for today

What you’ll understand

  • State infrared and collinear safety as limits of an observable.
  • Apply anti-kt pair and beam distances in the correct order.
  • Distinguish calculability from hadronization and detector corrections.

A quark or gluon can radiate arbitrarily soft or nearly parallel partons, so a detector-level jet cannot simply mean one parton. A useful jet definition groups nearby energy flow in a way that changes continuously under unresolved radiation. Infrared and collinear safety allow singular real-emission contributions to cancel their virtual counterparts for a suitably defined observable. They make perturbative predictions possible, while still leaving finite corrections, scale dependence, hadronization and experimental response to be assessed.

The measurement must share the unresolved limit

Represent an observable by a measurement function applied to every partonic final state. Adding a particle of energy approaching zero should leave its value unchanged in the limit. Replacing a particle by exactly parallel daughters whose four-momenta add to the parent should also recover the same value. These limits ensure the observable assigns the same weight to configurations that a finite-resolution measurement cannot distinguish, allowing the cancellation of their singular contributions when real and virtual terms are combined.

Counting every parton separately fails the collinear test: one parent becomes two even at zero opening angle. A total energy sum passes both tests, and many carefully designed jet observables do as well. Safety is a property of the complete observable, including clustering, cuts and tags. An infrared-safe jet algorithm does not automatically make every feature computed inside those jets safe; raw particle multiplicity, for example, retains sensitivity to collinear fragmentation. Identified species can instead require fragmentation-based descriptions.

On+1(p1,…,pn,λk)→λ→0On(p1,…,pn),On+1(…,zp,(1−z)p)→collinearOn(…,p)\mathcal O_{n+1}(p_1,\ldots,p_n,\lambda k)\xrightarrow{\lambda\to0}\mathcal O_n(p_1,\ldots,p_n),\qquad \mathcal O_{n+1}(\ldots,zp,(1-z)p)\xrightarrow{\mathrm{collinear}}\mathcal O_n(\ldots,p)
The soft momentum scales with positive λ; the collinear identity uses a massless parent and0<z<1. These measurement limits express IRC safety in suitable phase-space regions; singular boundaries and cuts require consistent limiting treatment.↔ Scroll wide equations horizontally.

Anti-kt compares pair and beam distances

Sequential recombination algorithms calculate pair distances and distances to the beam, select the smallest, and repeat. If a pair wins, combine its four-momenta into a new pseudojet using the chosen recombination scheme. If a beam distance wins, declare that object a completed jet and remove it. Anti-kt weights its distances by inverse transverse momentum squared, which makes a hard object attract nearby soft radiation before soft objects merge among themselves at comparable angular separations.

The angular separation uses rapidity and azimuth, with the azimuth difference wrapped across its periodic boundary. Rapidity is the natural longitudinal-boost variable; pseudorapidity is a useful massless approximation, not an identical definition for a massive pseudojet. The radius parameter is dimensionless and controls angular grouping. With two isolated objects, a separation smaller than the radius leads to merging before the harder object’s beam distance wins. Nearby competing hard objects can alter boundaries, so a general jet need not be a perfect circle.

dij=min⁡(pTi−2,pTj−2)ΔRij2R2,diB=pTi−2,ΔRij2=(yi−yj)2+(Δϕij)2,pijμ=piμ+pjμd_{ij}=\min(p_{Ti}^{-2},p_{Tj}^{-2})\frac{\Delta R_{ij}^2}{R^2},\qquad d_{iB}=p_{Ti}^{-2},\qquad\Delta R_{ij}^2=(y_i-y_j)^2+(\Delta\phi_{ij})^2,\qquad p_{ij}^{\mu}=p_i^{\mu}+p_j^{\mu}
These are anti-kt distances with E-scheme four-vector recombination. If pT is in GeV, every distance has units GeV⁻²; R,y and Δφ are dimensionless. The inclusive algorithm always selects the smallest distance among all pair and beam candidates.↔ Scroll wide equations horizontally.

A finite partonic answer is not a complete detector prediction

IRC safety removes a particular obstruction to fixed-order calculation; it does not imply a small perturbative correction. Narrow radii, jet vetoes and endpoint distributions can create large logarithms that need resummation. Hadronization and underlying-event effects shift reconstructed jet momentum, with different dependence on radius and kinematics. Pileup and detector calibration add further effects. Predictions and data must use matched object definitions, event selections and the same treatment of stable particles or fiducial acceptance.

Parton-shower generators approximate repeated radiation and provide hadronization models, but their output should not be mistaken for the definition of perturbative safety. An unsafe observable does not become safe because a finite cutoff makes a generator return a number. Conversely a safe observable can still need substantial nonperturbative corrections at low momentum. The vector instrument below checks four-momentum-component addition only as a planar analogy; it has no clustering list, rapidity coordinate, beam distance or jet-radius control.

σ[O]=∫dΦn (B+V) On+∫dΦn+1 Rreal On+1+⋯\sigma[\mathcal O]=\int d\Phi_n\,(B+V)\,\mathcal O_n+\int d\Phi_{n+1}\,R_{\mathrm{real}}\,\mathcal O_{n+1}+\cdots
B,V and Rreal denote Born, virtual and real contributions with the required subtraction or regulator treatment understood. The separate integrals can be divergent; matching unresolved measurement limits permits their cancellation in the consistently combined IRC-safe observable.↔ Scroll wide equations horizontally.

Work it through

Make one anti-kt clustering decision

An event has only two objects with transverse momenta100GeV and10GeV, separation ΔR=0.20 and radius R=0.40. Compute all candidate distances and determine the first clustering action.

  1. The beam distances are d1B=1/100²=1.0×10⁻⁴GeV⁻² and d2B=1/10²=1.0×10⁻²GeV⁻². The inverse momentum square selects the harder object in the pair minimum.

  2. The angular factor is(0.20/0.40)²=1/4. Hence d12=(1.0×10⁻⁴)/4=2.5×10⁻⁵GeV⁻², smaller than both beam distances.

  3. The algorithm first recombines the two objects by four-vector addition. With no other objects, the resulting pseudojet will subsequently be declared a jet. No assumption of scalar energy addition is needed for this decision.

The result

The pair distance2.5×10⁻⁵GeV⁻² wins, so the two objects merge before either is declared a completed jet.

The living notebook

Explore the model

Interactive lab

Interactive lab · vectors

Check directional components before combining objects

For fresh runs Reset and set BOTH angle0°, magnitude2; then Reset and set angle90°, magnitude2. Predict projections(2,0) and(0,2), so equal magnitudes do not mean equal vectors. This planar projection lab is only an addition reminder; it does not run anti-kt or supply jet momenta and rapidities.

Check directional components before combining objectsPlayback traces a fixed vector and its projections. Angle and magnitude set the geometry; the moving marker does not change them. Current readings: x component: 1.393; y component: 0.975; Magnitude: 1.7. All parameters are adjustable with the labeled controls below.−7.46−3.7303.737.46−3.5−1.7501.753.5xyvvₓ = 1.39
Vectorx / y projectionsTrace marker
x component
1.393
y component
0.975
Magnitude
1.7
t = 0 s

Playback traces a fixed vector and its projections. Angle and magnitude set the geometry; the moving marker does not change them. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.

Look forJet recombination adds four-vectors rather than treating every momentum magnitude as a scalar sum.

Pause & reason

Check your understanding

Which observable fails collinear safety in a purely partonic calculation?

Your turn

Try it yourself

Keep the same100GeV and10GeV objects and R=0.40, but change their separation to ΔR=0.60. Which distance wins first, and how many jets result in this two-object event?

Take a moment to make a prediction or work it out on paper. Open a hint when you need one.

Give me a hint

The beam distances stay fixed while the angular factor becomes(0.60/0.40)².

Compare with the solution

The pair distance is1.0×10⁻⁴×2.25=2.25×10⁻⁴GeV⁻². The smaller hard beam distance1.0×10⁻⁴ now wins, so the100GeV object is declared a jet and removed. The10GeV object is then declared its own jet. The result is two jets. This conclusion uses the event’s having only two objects; a competing third hard object could change a more general clustering sequence.

Keep the idea

Take it with you

  • IRC safety matches unresolved real-emission and lower-multiplicity measurement limits.
  • Anti-kt chooses the smallest pair or beam distance at every step.
  • Perturbative safety does not remove hadronization or detector effects.

Read further

Original sources and deeper explanations to follow at your own pace.

Optional progress, saved in this browser when available. Revisit any lesson at any time.

Lesson not marked complete.