View source for Energy resolution (ATLAS)

[[Category:ATLAS]]
[[Category:Event reconstruction]]
[[Category:Electron/photon reconstruction]]

''This article only describes the situation for electrons and photons and could be expanded.''

For electrons and photons in ATLAS, the ''relative'' '''energy resolution''' <math>\sigma_E/E</math> of the [[electromagnetic calorimeter]] can be parametrised as

<math>
\frac{\sigma_E}{E} = \frac{a}{\sqrt{E}} \oplus \frac{b}{E} \oplus c,
</math>

where the ''sampling-term'' coefficient <math>a</math>, the ''noise-term'' coefficient <math>b</math> and the ''constant-term'' <math>c</math> are [[pseudorapidity]]-dependent parameters. The sampling term contributes mostly at low and intermediate energy and depends on [[calorimeter shower]] and sampling fluctuations. The noise term is dominated by the [[pileup]] noise in the standard LHC data taking (i.e. at an average number of interactions per [[bunch crossing]] of the order of a few dozen). At higher energies, the relative energy resolution tends asymptotically to the constant term, <math>c</math>, which in ATLAS has a design value of 0.7%. For electrons and [[Photon conversion|converted photons]], interactions with the material in front of the calorimeter also contribute to the <math>a</math> and <math>b</math> terms.   

A relative uncertainty of 10% on the sampling coefficient <math>a</math> is assumed based on [[Test beam|test-beam]] data. The constant term <math>c</math> is extracted from <math>Z \to e^+e^-</math> measurements.  In standard LHC data taking conditions, the energy resolution at low energy is dominated by the noise coefficient <math>b</math> and measuring the sampling coefficient <math>a</math> to reduce its uncertainty is difficult.