1: %% ****** Start of file slactemplate.tex ****** %
2: %%
3: %%
4: %% This file is part of the APS files in the REVTeX 4 distribution.
5: %% Version 4.0 of REVTeX, August 2001
6: %%
7: %%
8: %% Copyright (c) 2001 The American Physical Society.
9: %%
10: %% See the REVTeX 4 README file for restrictions and more information.
11: %%
12: %
13: % This is a template for producing manuscripts for use with REVTEX 4.0
14: % Copy this file to another name and then work on that file.
15: % That way, you always have this original template file to use.
16: %
17: \documentclass[slac_one]{revtex4}
18: \usepackage{graphicx}
19: \usepackage{fancyhdr}
20: \usepackage{epsfig}
21:
22:
23: \pagestyle{fancy}
24: \fancyhead{} % clear all fields
25: \fancyfoot{} % clear all fields
26: %\fancyfoot[LE,LO]{Insert PSN Here}
27: \renewcommand{\headrulewidth}{0pt}
28: \renewcommand{\footrulewidth}{0pt}
29: \renewcommand{\sfdefault}{phv}
30: \renewcommand{\baselinestretch}{1.2}
31: \setlength{\textheight}{235mm}
32: \setlength{\textwidth}{178mm}%{170mm}
33: \setlength{\topmargin}{-20mm}
34:
35: \def\et {\ensuremath{E_{T}}\xspace}
36:
37: \input{babarsym}
38:
39: \begin{document}
40:
41: %Title of paper
42: \title{{\small{Hadron Collider Physics Symposium (HCP2008),
43: Galena, Illinois, USA}}\\ %% Please keep this conference title here
44: \vspace{12pt}
45: Commissioning of Particle ID at ATLAS and CMS with Early LHC Data} %% Paper title goes here
46:
47:
48: % Repeat the \author .. \affiliation etc. as needed
49: %
50: % \affiliation command applies to all authors since the last
51: % \affiliation command. The \affiliation command should follow the
52: % other information
53:
54: \author{T. Berger-Hryn'ova}
55: \affiliation{CERN, 1211 Geneve 23, Switzerland}
56: %
57: \author{on behalf of the ATLAS and CMS collaborations}
58: %\affiliation{FNAL, Batavia, IL 60510, USA}
59:
60: \begin{abstract}
61: This paper describes latest results on lepton (electron, muon and tau)
62: and photon particle identification at the ATLAS and CMS experiments,
63: with emphasis on how the particle identification can be validated and its
64: performance determined using early LHC data.
65: \end{abstract}
66:
67: %\maketitle must follow title, authors, abstract
68: \maketitle
69:
70: \thispagestyle{fancy}
71:
72: % body of paper here - Use proper section commands
73: % References should be done using the \cite, \ref, and \label commands
74: % Put \label in argument of \section for cross-referencing
75: %\section{\label{}}
76:
77: \section{INTRODUCTION} % Section title should be in all capitals.
78:
79: ATLAS and CMS are the two general-purpose experiments, which have recently completed installation at
80: the CERN Large Hadron Collider (LHC). Many physics processes which are
81: currently out of reach at existing colliders will become accessible at the LHC.
82: These processes range from the production of scalar Higgs bosons or of new vector gauge bosons to that
83: of supersymmetric particles or of TeV-scale resonances
84: resulting from presence of large extra-dimensions.
85: Charged leptons often provide distinctive signatures for such new processes, but they also
86: appear in the final state of many standard-model (SM) processes involving electroweak (EW)
87: bosons or top quarks. These SM processes constitute the dominant background to new signals and therefore need to be well
88: understood. They will also be used as calibration processes to understand in detail the performance of the detector.
89:
90: Thus, lepton (and photon) identification is very important for the physics program of
91: ATLAS and CMS. %, where the expected backgrounds are a factor of ten worse than they are at TeVatron.
92: With a bunch-crossing rate of 40$\,$MHz at design luminosity, of which only
93: 200-300$\,$Hz are planned to be recorded to mass storage, particle identification is
94: extremely important also at the trigger level and it will be crucial
95: to commission it as early as possible.
96:
97: The main data samples used to understand the detector performance and
98: the lepton identification are well-known EW processes, such as $Z\to ll$ and
99: $W\to l\nu$. They provide important benchmark channels for calibration,
100: alignment and monitoring of the detector performance.
101: However, at the expected start-up luminosities of $10^{31}-10^{32}$\cms, the total
102: bandwidth for EW processes will be below 1$\,$Hz, compared to a few tens of Hz
103: expected at $10^{33}$\cms, which will severely limit the statistics of such
104: samples. On the other hand, at these lower luminosities, the triggers can be operated with lower
105: \et-thresholds than those planned for the LHC design luminosity. This would potentially give access to
106: other more abundant data samples, such as direct $J/\psi$, $\Upsilon$, etc.
107:
108: This paper describes the commissioning strategies in the following order:
109: identifications of photons, electrons, muons and $\tau-$leptons.
110: For a detailed description of the ATLAS and CMS detectors and of their current commissioning status please
111: see Refs.~\cite{thomas,atldet,luca}.
112:
113: \section{Photons and Electrons}
114:
115: Photon and electron identification in ATLAS and CMS is extremely challenging: the electron
116: to jet ratio is $\approx 10^{-5}$ at 40\gev. Photon and electron reconstruction in both
117: experiments starts by the detection of clusters in the electromagnetic calorimeters.
118: Electron candidates are also required to have an inner-detector track loosely matched to the cluster.
119: In ATLAS, there is also a
120: dedicated low-\pt electron reconstruction algorithm, which extrapolates inner-detector
121: tracks to the EM calorimeter.
122: The reconstruction of photons and electrons is quite challenging in both
123: experiments because of the $\approx 0.4-2.4 X_0$ of material in the inner
124: detectors. For example, in ATLAS, electrons with \et$=$25\gev lose on average $30-60\%$
125: of their energy before reaching the EM calorimeter, and
126: between $10$ and $60\%$ of photons convert into an electron-positron pair.
127: For a more detailed description of the electron and photon reconstruction
128: in ATLAS and CMS, see Ref.~\cite{charlot}.
129:
130: Despite some significant differences in detector technologies and intrinsic performance
131: (e.g. the sampling LAr EM calorimeter of ATLAS and the homogeneous PbWO$_4$ crystal EM calorimeter of CMS),
132: the performances expected for electrons and photons are similar for both experiments
133: in terms of efficiency and accuracy. For example
134: electrons of 50\gev energy are expected to be measured with a $1.5-2.5\%$ energy resolution in ATLAS
135: and with a 2$\%$ energy resolution in CMS.
136: Photons with a 100\gev energy are expected to be measured with a $1.0-1.5\%$
137: energy resolution in ATLAS. For $70\%$ of the photons which do not convert too early in the tracker,
138: the energy resolution is expected to be better in CMS, e.g. $\approx 0.8\%$ due to the superior intrinsic
139: accuracy of the crystal calorimeter.
140:
141: Very high rejection of the large background from hadronic jets is another important aspect of
142: electron and photon identification at the LHC. In ATLAS,
143: for a photon reconstruction efficiency of $85\%$, one can obtain a jet rejection
144: of $\approx 10000$ averaged over all jets types (quark and gluon jets).
145: For tight electron selection and for \et$>20$\gev,
146: the identification efficiency of isolated electrons from $Z$ decays is $65\%$ (it drops to $\approx 25\%$
147: for non-isolated electrons from $b/c-$decays) with a jet rejection of
148: $\approx 10^5$. After such tight electron selection of inclusive electrons with \et$>$20\gev
149: the remaining sample
150: is expected to contain $~10\%$ of isolated and $~65\%$ of non-isolated
151: electrons, with the remaining $~25\%$ expected to consist of residual background from hadronic jets dominated
152: by charged hadrons. The use of multivariate
153: methods improves the rejection by $~50\%$ for a fixed efficiency or improves
154: the efficiency by $5-10\%$ for a fixed rejection.
155:
156: Obviously, those expected performances will have to be validated with real data.
157: Both experiments are planning to use
158: the early data to understand the tracker (alignment, material distribution), to perform detailed intrinsic calibrations
159: of the EM calorimeters, as well as
160: to measure the trigger efficiency. All these aspects are crucial to obtain a detailed understanding of the
161: electron and photon identification performances.
162:
163: \subsection{Photons}
164:
165: Photon identification is particularly challenging, since there do not exist
166: any prominent di-photon resonances at high mass which could be
167: used for calibration and efficiency measurement purposes. Thus, the understanding of photon identification
168: in the early data will have to rely heavily on
169: the understanding of electrons, using simulations to account for the
170: small differences between them. As soon as a few \invfb of data become available
171: $Z\to ee\gamma$ and $Z\to \mu\mu\gamma$ decays will be used to improve the understanding of high-\pt
172: photons further.
173: In CMS the achievable statistical error on the photon efficiency is estimated
174: to be $0.1\%$ with 1\invfb of data.
175: At much lower \pt, one could use $\pi^0$ and $\eta$ decays to $\gamma\gamma$
176: to understand photon identification in early data.
177: Converted photons
178: from $\pi^0/\eta$ decays will be extremely useful to obtain a detailed mapping of
179: the tracker material. This is strongly correlated with a precise
180: EM calorimeter inter-calibration, as discussed below.
181:
182: \subsection{Electrons}\label{el}
183: A detailed understanding of electron reconstruction and identification at the LHC will rely strongly
184: on the presence of
185: ``standard candles'', such as the Z and W resonances, for the calibration of the energy scale,
186: for the EM inter-calibration, and for a precise understanding of the detector and trigger efficiencies, etc.
187:
188: Both experiments are planning to use the precise knowledge of the $Z$ mass
189: to perform an accurate EM calorimeter inter-calibration. Since the ATLAS EM calorimeter is locally very uniform by
190: construction, only an inter-calibration of large regions is required.
191: The initial spread from region to region is conservatively assumed to be
192: approximately $1.5-2\%$. Provided the material distribution
193: in front of the EM calorimeter is well-understood, the specified inter-calibration precision of 0.7$\%$
194: between regions of 0.2$\times$0.4 in $\eta\times\phi$
195: can be achieved in the case of ATLAS with 100\invpb of data.
196: In the CMS EM calorimeter, the energy response may vary with a spread of $~3-4\%$
197: from crystal to crystal, and therefore a precise
198: inter-crystal calibration has to be performed. Initially, single-jet triggers will be used to
199: inter-calibrate $\phi$ rings at fixed $\eta$ with an accuracy of a few percent, as is shown in Fig.~\ref{jets}.
200: For the final precision the $Z\to ee$ sample corresponding to $\approx$2\invfb of data
201: is needed to reach the specified target of 0.5$\%$.
202:
203: \begin{figure}[h!]
204: \centering
205: {\includegraphics[width=0.40\linewidth]{cms-jet.eps}}
206: {\includegraphics[width=0.40\linewidth]{cms-jet2.eps}}
207: \caption{CMS inter-calibration precision in barrel (left) and endcap (right) which can be
208: obtained with 11 million Level-1 jet trigger events and the limit on the inter-calibration
209: precision due to tracker material inhomogeneity as a function of $\eta$ (from Ref.~\cite{cmstdr}).} \label{jets}
210: \end{figure}
211:
212: \begin{figure}[ht!]
213: \centering
214: {\includegraphics[width=0.40\linewidth]{2se5-track-log80.eps}}
215: {\includegraphics[width=0.40\linewidth]{et_oleg.eps}}
216: \caption{(Left) Expected differential cross-section for low-mass electron pairs using
217: the dedicated ATLAS low-\pt dielectron trigger. Shown is the invariant di-electron mass
218: distribution reconstructed using tracks for $J/\psi \to e e$ decays (dotted histogram),
219: $\Upsilon \to e e$ decays (dashed histogram) and Drell-Yan production (full histogram).
220: Also shown is the expected background (full circles).
221: (Right) Expected differential rates in ATLAS as a function of electron \et for
222: 100\invpb of data and for electrons satisfying tight single-electron selection.
223: Shown are the isolated electrons from $W$ and $Z$ decays (solid histogram),
224: the non-isolated electrons from $b,c\to e$ decays (light gray/red histogram), and the
225: estimated QCD background (dark gray/blue histogram).} \label{electron}
226: \end{figure}
227:
228: \begin{figure}[ht!]
229: \centering
230: {\includegraphics[width=0.40\linewidth]{cms-w.eps}}
231: \caption{CMS inter-calibration precision as a function of HLT events per crystal for different $\eta$
232: regions. Upper curve: last 10 crystals in the EM calorimeter barrel ($1.305<\eta<1.479$),
233: middle curve: 10 crystals in the middle EM calorimeter barre ($0.783<|\eta|<0.957$), lower curve:
234: the first 15 crystals in the EM calorimeter barrel ($0.0<|\eta|<0.261$). The third point on each line
235: gives the precision for 5\invfb of integrated luminosity (from Ref.~\cite{cmstdr}).} \label{ew}
236: \end{figure}
237:
238: In ATLAS, specific triggers have been implemented for electron pairs from direct $J/\psi$ and $\Upsilon$ production,
239: which are expected to provide about 200k and 50k events respectively for 100\invpb of data (compared to
240: 50k of events on the $Z$ resonance). The expected signal and background dielectron mass distributions after tight
241: electron selection are shown in Fig.~\ref{electron} (left).
242: The obtained $J/\psi$ sample can also be used for the calorimeter inter-calibration and
243: preliminary studies~\cite{fred} show that a precision of $\approx 0.6\%$ may be achieved.
244: In addition to a cross-check on the inter-calibration result obtained using the $Z$ resonance,
245: the $J/\psi$ and $\Upsilon$ samples will also provide an
246: in-situ check of the linearity of the EM calorimeter response as a function
247: of electron energy. The additional electron identification
248: capabilities provided by the ATLAS transition radiation detector may turn out to be
249: crucial to identify low-energy electrons
250: from $J/\psi$ and $\Upsilon$ decays, since they are usually not as well isolated as those
251: from $Z$ decay and since the charged-pion combinatorial background is very large.
252:
253: Isolated single electrons from $W\to e\nu$ decays will provide further information for
254: tracker-EM calorimeter inter-calibration by comparing the energy deposited in the calorimeter to
255: the track momentum measurement. As shown in Fig.~\ref{ew}, an inter-calibration precision of
256: $0.5-1.5\%$ may be achieved in CMS with 5\invfb of data. Non-isolated electrons from $b,c\to e$ decays could
257: also be used for this purpose during the early data-taking while the single-electron trigger
258: thresholds are low. As one can see from Fig.~\ref{electron} (right), these decays are
259: dominant for \et$<$35\gev and they should provide ten times more electrons with \et$>$10\gev
260: than those from $W$ decay.
261:
262: Since the trigger efficiency represents a fundamental ingredient of any physics analysis,
263: it must be verified as independently as possible from simulations and as a function of the LHC physics and detector
264: operation conditions. One of the most widely spread techniques under study is the
265: ``tag-and-probe'' method, where an event is triggered by one of the electrons in a $Z\to ee$ decay and
266: the efficiency to trigger the other electron is measured using offline information.
267: With 100\invpb of data, one will measure the trigger efficiency for electrons of $\et>20\gev$
268: with a statistical accuracy of approximately 0.2$\%$~\cite{atldet,cmstag}.
269:
270: \section{Muons}
271:
272: Muons are identified through matching reconstructed tracks in
273: the inner tracker and the muon spectrometer.
274: ATLAS and CMS follow complementary concepts of muon detection. ATLAS has an
275: instrumented air-toroid magnetic system serving as a stand-alone high-precision muon spectrometer. CMS relies
276: on high bending power and momentum resolution in the inner tracker, and on the iron yoke used for the return of
277: its solenoidal magnetic field. The iron yoke is instrumented with chambers used for muon identification and triggering.
278: Stand-alone muon tracks can be reconstructed, but in order to obtain
279: a precise momentum measurement it is essential to combine the
280: inner-tracker information with that from the muon chambers.
281:
282: \begin{figure}
283: \centering
284: {\includegraphics[width=0.60\linewidth]{m2.eps}}
285: \caption{Relative momentum resolution as a function of the muon transverse momentum
286: showing the stand-alone resolution of muon systems, the stand-alone resolution of the inner tracker and
287: the combined resolution: (a) and (c) in ATLAS for $|\eta|<1.5$ and $|\eta|>1.5$, respectively; (b) and
288: (d) in CMS for $0<|\eta|<0.2$ and $1.8<|\eta|<2.0$, respectively. From Ref.~\cite{gigi}.} \label{muonres}
289: \end{figure}
290:
291: \begin{figure}
292: \centering
293: {\includegraphics[width=0.50\linewidth]{muoneff.eps}}{\includegraphics[trim=11cm 0cm 0cm 0cm,clip,width=0.4\linewidth]{m1.eps}}
294: \caption{(left) For muons with \pt$=$100\gev, expected fractional stand-alone momentum resolution in ATLAS
295: as a function of $\phi$ and $|\eta|$ (from Ref.~\cite{atldet}). No momentum measurement is possible at $|\eta|<$0.1 over
296: most of the azimuth, nor at $|\eta|=1.3$ because of holes in the acceptance
297: of the muon spectrometer.
298: (right) Relative momentum resolution for combined muon reconstruction
299: in CMS as a function of $\eta$ for two
300: different values of transverse momentum (from Ref.~\cite{gigi}).} \label{muoneff}
301: \end{figure}
302:
303: The momentum resolution of the ATLAS and CMS detectors are shown in Fig.~\ref{muonres}.
304: Whereas the stand-alone muon resolution is fairly uniform over most of the
305: $\eta-\phi$ plane in ATLAS as is shown in Fig.~\ref{muoneff}(left) %~\footnote{No stand-alone momentum measurement
306: %is possible at $|\eta|<0.1$ and $|\eta|=$1.3 because of holes in acceptance of the muon
307: %spectrometer.}
308: with optimal resolution achieved at 100\gev, the resolution degrades at
309: high $\eta$ in CMS because of the limited coverage of the solenoid (Fig.~\ref{muoneff}(right)).
310: The resolution of the combined measurement in the barrel region is slightly
311: better in CMS owing to the higher precision of the measurement in the tracking system,
312: whereas the reverse is true in the end-cap region owing to the better coverage of the ATLAS
313: muon spectrometer at higher rapidities.
314:
315: The muon reconstruction efficiencies are high (above $96\%$) for both experiments with fake rates of
316: about $1\%$ from neutron background at the LHC design luminosity, see Ref.~\cite{kortner} for details.
317: The combination of inner tracker, calorimeter and muon spectrometer measurements is required in ATLAS
318: to minimize inefficiencies due to holes in the coverage of the muon spectrometer
319: system as is shown in Fig.~\ref{muoneff}.
320:
321:
322: \begin{figure}
323: \centering
324: {\includegraphics[width=0.40\linewidth]{muon1.eps}}
325: %{\includegraphics[width=0.40\linewidth,height=6cm]{muon1.eps}}
326: %{\includegraphics[width=0.40\linewidth,height=5.5cm]{CMS_Zmisal.eps}}
327: %{\includegraphics[width=0.40\linewidth,height=6cm]{SW_TagProbe_Luminosity.eps}}
328: \caption{Difference between reconstructed and true dimuon mass from $Z\to\mu\mu$ decays, as obtained from
329: an aligned and a misaligned muon spectrometer in ATLAS.
330: %(right) Average statistical error expected for the measured muon reconstruction efficiency in ATLAS for 320 regions
331: %of the muon spectrometer as a function of integrated luminosity.
332: } \label{muon1}
333: \end{figure}
334:
335: \begin{figure}
336: \centering
337: %{\includegraphics[width=0.40\linewidth,height=6cm]{muon1.eps}}
338: {\includegraphics[width=0.40\linewidth]{CMS_Zmisal.eps}}
339: %{\includegraphics[width=0.40\linewidth,height=6cm]{SW_TagProbe_Luminosity.eps}}
340: \caption{Effect on the reconstructed $Z$ boson mass in CMS
341: of the misalignment and of the uncertainties on the
342: magnetic field after the muon scale calibration. Fits are sums of a Lorentzian and a
343: linear function. See Ref.~\cite{cmsmu} for details.
344: } \label{muon2}
345: \end{figure}
346:
347:
348: \begin{figure}
349: \centering
350: %{\includegraphics[width=0.40\linewidth,height=6cm]{muon1.eps}}
351: %{\includegraphics[width=0.40\linewidth,height=5.5cm]{CMS_Zmisal.eps}}
352: {\includegraphics[width=0.40\linewidth,height=6cm]{SW_TagProbe_Luminosity.eps}}
353: \caption{%(left) Difference between reconstructed and true dimuon mass from $Z\to\mu\mu$ decays, as obtained from
354: %an aligned and a misaligned muon spectrometer in ATLAS. (right)
355: Average statistical error expected for the measured muon reconstruction efficiency in ATLAS for 320 regions
356: of the muon spectrometer as a function of integrated luminosity.} \label{muon3}
357: \end{figure}
358:
359: The large expected rates of $Z\to\mu\mu$ decays provide an excellent tool to untangle various effects
360: which might lead to distortions of the measured dimuon invariant mass spectrum. One example is shown
361: in Fig.~\ref{muon1} for ATLAS stand-alone muon measurements, where the performance obtained with a misaligned
362: layout (with random displacements of 1\mm and a random rotations of 1\mrad of each chamber in the muon spectrometer)
363: is compared to that expected from a perfectly aligned layout. The large assumed misalignments lead to an increase in the
364: fitted Gaussian resolution of the dimuon reconstructed invariant mass from 2.5\gev to 8\gev.
365: Similar distortions including a potential shift in the mean of the peak
366: could be caused by muon momentum scale biases and uncertainties in the
367: initial knowledge of the magnetic field as is shown in Fig.~\ref{muon2}.
368:
369: The muon reconstruction and identification efficiency will also be measured from data using
370: $Z\to\mu\mu$ decays and the tag-and-probe method described in Section~\ref{el}, with similar results
371: expected in terms of the accuracy of the measurement. This type of analysis can be extended to
372: study efficiency with increasing granularity as the integrated luminosity increases,
373: in particular as a function of $\eta$ and $\phi$.
374: An example of the expected statistical uncertainty on the reconstruction efficiency
375: in the muon spectrometer,
376: averaged over 320 regions in the $\eta-\phi$ plane, is shown in Fig.~\ref{muon3}.
377: Other resonances, such as $J/\psi$ and $\Upsilon$, could also be used for this analysis
378: at lower initial luminosities. The expected rates for 100\invpb of data in ATLAS/CMS are 1.6$\,$M $J/\psi$,
379: 300k $\Upsilon$ and 60k $Z$-decays.
380:
381: \section{$\tau-$leptons}
382:
383: The $\tau-$lepton is the heaviest lepton observed to-date and it decays to electrons, muons and hadrons.
384: This section concentrates on the identification on $\tau-$leptons decaying hadronically. They
385: %For the $35\%$ of taus decaying leptonically the electron and muon identifications described above
386: %are used. Hadronically decaying $\tau-$leptons (the remaining $65\%$)
387: are reconstructed by matching
388: narrow calorimeter clusters with a small number of tracks reconstructed in the inner tracker.
389: %For detailed description please see ...
390:
391: %As $\tau$ reconstruction is not trivial, {\it a priori}
392: One will need to demonstrate that the observed signals from $W\to\tau\nu$ and $Z\to\tau\tau$ decays are in
393: agreement with fundamental properties of the $\tau-$lepton (e.g. visible mass and decay length
394: measurement for three-prong decays).
395: One of the distinctive features of hadronic $\tau$ decays is the track multiplicity spectrum shown in Fig.~\ref{tau}
396: for $Z\to\tau\tau$ decays (left) and for jets (right). The distributions are shown after the reconstruction
397: step, after a cut-based identification algorithm and finally after applying a multi-variate
398: discrimination technique using a neural network. Figure~\ref{tau} shows that the signal track
399: multiplicity distributions do not depend strongly on the reconstruction and identification cuts used. It
400: also shows that the track multiplicity in the
401: QCD jet sample is quite different from that in the signal sample
402: and that the candidates with track multiplicity above three may be used to
403: normalize the background. This would allow a reasonably precise calibration of the performance
404: of the $\tau$-identification algorithms using real data, provided the rejection against QCD jets
405: is proven to be sufficient to extract clean signals in the single-prong and three-prong categories.
406:
407: \begin{figure}[ht!]
408: \centering
409: {\includegraphics[width=0.40\linewidth]{tau1.eps}}
410: {\includegraphics[width=0.40\linewidth]{tau2.eps}}
411: \caption{Track multiplicity distributions expected in ATLAS for hadronic $\tau$-decays (left) and
412: for the background from hadronic jets (right), for $\tau-$candidates
413: with visible transverse energy above 20\gev reconstructed using a track-based $\tau$-identification
414: algorithm. The distributions are shown after reconstruction, after cut-based
415: identification, and finally after applying a neural network discrimination
416: technique, resulting in an efficiency of 30$\%$ for the signal.
417: } \label{tau}
418: \end{figure}
419:
420: Triggering on hadronic $\tau-$decays is extremely challenging because of the huge backgrounds from
421: hadronic jets and of the requirement to bias minimally the track multiplicity spectrum.
422: The $\tau-$triggers increase the discovery potential for many physics channels and
423: in rare cases provide the only trigger, e.g. $W\to\tau\nu$ decays.
424:
425: $W\to\tau\nu$ decays provide the most abundant source of $\tau-$leptons in SM processes.
426: Due to trigger rate limitations, they are only expected to be accessible at initial low luminosities.
427: The dominant background for it arises from hardonic jets and the signal-to-background ratio is expected to be
428: a factor of ten worse than that
429: observed at the TeVatron. Preliminary studies show that it is possible to reach a signal-to-background
430: ratio of about three in this channel with $\approx 1500$ $W\to\tau\nu$ signal events for 100\invpb of data.
431: The goal of such an analysis would be to show track-multiplicity spectrum of identified $\tau-$leptons as a proof
432: of $\tau-$lepton observability.
433:
434: The most interesting SM sample is that from $Z\to\tau\tau$ decays. Although the expected rate is
435: ten times less than $W\to\tau\nu$ decays, this process
436: exhibits a more interesting topology and is easily triggered by requiring a single lepton.
437: Same-sign events which are nearly signal-free will be used to control the
438: dominant QCD background in the signal-enriched
439: opposite-sign events. The goal of this analysis is to set $\tau$ energy scale from the excess of
440: signal events in the
441: invariant mass of the visible decay products. The complete $Z$ invariant mass can be also reconstructed
442: in the collinear approximation. Preliminary studies from ATLAS expect to see about 520 signal events and
443: 85 background events in 100\invpb. The visible decay mass could be used to determine the $\tau$
444: energy scale with a precision of 3$\%$ (taking into account only statistical errors). Also assuming
445: $\tau$s are well-calibrated and using additional selection ($\approx$ 200 signal and 20 background
446: events remaining) and the collinear approximation, one could determine $E_T^{miss}$ scale with $3\%$
447: statistical precision. One should note that the $E_T^{miss}$ scale could be more easily
448: understood with $W\to e(\mu)\nu$ events.
449:
450: \section{Acknowledgments}
451:
452: The author would like to thank Daniel Froidevaux, Pascal Vanlaer, Oliver Kortner, Martijn Mulders and
453: Elzbieta Richter-Was for the fruitful discussions and exchanges during this work.
454:
455: \begin{thebibliography}{9} % Use for 1-9 references
456: %\begin{thebibliography}{99} % Use for 10-99 references
457: \bibitem{thomas} J. Thomas, Commissioning of ATLAS, Proceedings of 19th Hadron Collider Physics
458: Symposium (HCP 2008), this volume, Galena, IL, USA.
459: \bibitem{atldet} The ATLAS Experiment at the CERN Large Hadron Collider,
460: the ATLAS Collaboration, G Aad {\it et al.}, JINST 3 S08003 (2008).
461: \bibitem{luca} L. Malgeri, Commissioning of CMS, Proceedings of 19th Hadron Collider Physics
462: Symposium (HCP 2008), this volume, Galena, IL, USA; The CMS experiment at the CERN LHC,
463: the CMS Collaboration, S Chatrchyan {\it et al.}, JINST 3 S08004 (2008).
464: \bibitem{charlot} C. Charlot, Electron/Photon Identification in ATLAS and CMS,
465: Proceedings of 17th Hadron Collider Physics Symposium (HCP 2006) [arXhiv:0709.2479].
466: \bibitem{cmstdr} CMS Physics TDR vol. 1, CERN/LHCC 2006-001.
467: \bibitem{fred} F. Derue, A. Kaczmarska, Ph. Schwemling, Reconstruction of DC1 $J/\psi\to ee$ decays
468: and use of the low energy calibration of the ATLAS electromagnetic calorimeter, 2006, ATL-PHYS-PUB-2006-004.
469: \bibitem{cmstag} G.Daskalakis {\it et al.}, Measuring electron efficiencies at CMS with early data, CMS EGM-07-001-PAS.
470: \bibitem{gigi} F. Ragusa and L. Rolandi, Tracking at LHC, New J. Phys. 9 (2007) 336.
471: \bibitem{kortner} O. Kortner, Muon Identification in ATLAS and CMS,
472: Proceedings of 17th Hadron Collider Physics Symposium (HCP 2006) [arXhiv:0707.0905].
473: \bibitem{cmsmu} The CMS Collaboration, Towards a measurement of the inclusive $W\to\mu\nu$ and
474: $Z\to\mu\mu$ cross sections in pp collisions at $\sqrt{s}=14$\tev, CMS PAS 2007/002.
475: \end{thebibliography}
476:
477: \end{document}
478: