1: %
2: %
3: %
4: \section{Introduction}
5:
6: The S-Matrix ansatz provides a coherent way of describing
7: LEP measurements of the cross-section and forward-backward
8: asymmetries in $s$-channel $\eeff$ processes at centre-of-mass energies
9: around the $\Zzero$ resonance, from the
10: {\LEPI} program, and the measurements at centre-of-mass
11: energies from 130 -- 207~GeV from the {\LEPII} program.
12:
13: Compared with the standard 5 and 9 parameter descriptions of the
14: measurements at the $\Zzero$~\cite{smat:ref:5and9}, the S-Matrix formalism
15: includes an extra 3 parameters (assuming lepton universality) or 7 parameters
16: (without lepton universality) which explicitly determine the
17: contributions to the cross-sections and forward-backward asymmetries
18: of the interference between the exchange of a $\Zzero$ and a photon.
19: The {\LEPI} data alone cannot tightly constrain these interference
20: terms, in particular the interference term for hadronic cross-sections,
21: since their contributions are small around the $\Zzero$ resonance and change
22: sign at the pole.
23: Due to strong correlations between the size of the hadronic interference
24: term and the mass of the $\Zzero$, this leads to a larger error on the
25: fitted mass of the $\Zzero$ compared to the standard 5 and 9 parameter fits,
26: where the hadronic interference term is fixed to the value predicted in the
27: Standard Model. Including the {\LEPII} data
28: leads to a significant improvement in the constraints on the interference terms
29: and a corresponding reduction in the uncertainty on the mass of the
30: $\Zzero$. This results in a measurement of {$\MZ$} which is almost as
31: sensitive as the standard results,
32: but without constraining the interference to the Standard Model prediction.
33: %
34: %
35:
36: %
37: This chapter describes the first, preliminary, combination of data from the
38: full data sets of the 4 LEP experiments, to obtain a LEP combined results
39: on the parameters of the S-Matrix ansatz. These results update those
40: of a previous combination~\cite{smat:ref:smat1997} which was based on
41: preliminary {\LEPI} data and only partial statistics from the full {\LEPII}
42: data set.
43:
44: Different strategies are used to combine the {\LEPI} and {\LEPII} data.
45: For {\LEPI} data, an average of the individual experiment's results
46: on the S-Matrix parameters is made. This approach is rather similar to
47: the method used to combine the results of the 5 and 9 parameter fits.
48: To include {\LEPII} data, a fit is made to LEP combined
49: measurements of cross-sections and asymmetries above the $\Zzero$,
50: taking into account the results of the {\LEPI} combination of S-Matrix
51: parameters.
52:
53: In Section~\ref{smat:sec:ansatz} the parameters of the
54: S-Matrix ansatz are explained. In Sections~\ref{smat:sec:lep1}
55: and~\ref{smat:sec:lep2} the average of the {\LEPI} data and the
56: inclusion of the {\LEPII} data are described. The results are discussed in
57: Section~\ref{smat:sec:discuss} and conclusions are drawn in
58: Section~\ref{smat:sec:conc}.
59:
60: %
61: %
62: %
63: \section{The S-Matrix Ansatz}
64: \label{smat:sec:ansatz}
65:
66: The S-matrix
67: ansatz~\cite{smat:ref:bcms,*smat:ref:rgs,*smat:ref:arr,*smat:ref:tr}
68: is a rigorous approach to describe the cross-sections and
69: forward-backward asymmetries in the $s$-channel $\ee$ annihilations
70: under the assumption that the processes can be parameterised as the
71: exchange of a massless and a massive vector boson, in which the
72: couplings of the bosons including their interference are treated as
73: free parameters.
74:
75:
76: In this model, the cross-sections can be parametrised
77: as follows:
78: %
79: %
80: %
81: %
82: %
83: %
84: %
85: %
86: %
87: %
88: %
89: %
90: %
91: %
92: %
93: %
94: %
95: %
96: %
97: %
98: \begin{equation}
99: \sigma^0_{tot, f}(s)=\frac{4}{3}\pi\alpha^2
100: \left[
101: \frac{\gf^{tot}}{s}
102: +\frac{\jf^{tot} (s-\MZbar^2) + \rf^{tot} \, s}
103: {(s-\MZbar^2)^2 + \MZbar^2 \GZbar^2}
104: \right]
105: \,\,\,\mathrm{with}\,\,\mathrm{f=had,e,\mu,\tau}\,,
106: \label{smat:eqn:eq1}
107: \end{equation}
108: while the forward-backward asymmetries are given by:
109: \begin{equation}
110: A^0_\mathrm{fb, f}(s)=\pi\alpha^2
111: \left[
112: \frac{\gf^{fb}}{s} +
113: \frac{\jf^{fb} (s-\MZbar^2) + \rf^{fb} \, s}
114: {(s-\MZbar^2)^2 + \MZbar^2 \GZbar^2}
115: \right]
116: / {\sigma^0_{\mathrm{tot, f}}(s)}\,,
117: \label{smat:eqn:eq2}
118: \end{equation}
119: where $\roots$ is the centre-of-mass energy.
120: The parameters $\rf$ and $\jf$ scale the $\Zzero$ exchange and the
121: \mbox{$\Zzero-\gamma$}
122: interference contributions to the total cross-section and forward-backward
123: asymmetries. The contribution $\gf$ of the pure $\gamma$ exchange was fixed to
124: the value predicted by QED in all fits. Neither the hadronic charge
125: asymmetry, nor the flavour tagged quark forward-backward asymmetries are
126: considered here,
127: %
128: which leaves 16 free parameters to described the LEP data: 14 $\rf$ and
129: $\jf$ parameters and the mass and width of the massive $\Zzero$ resonance.
130: Applying the constraint of lepton universality reduces this to 8
131: parameters.
132:
133:
134:
135: In the Standard Model the $\Zzero$ exchange term, the $\Zzero-\gamma$
136: interference term and the photon exchange term are given in terms of the
137: fermion charges and their effective vector and axial couplings to the
138: $\Zzero$ by:
139: \begin{equation}
140: \begin{array}{l@{}l@{}l@{}l}
141: \displaystyle
142: \rtotf & = & \kappa^2
143: \left[\gae^2+\gve^2\rule{0mm}{4mm}\right]
144: \left[\gaf^2+\gvf^2\right]
145: -2\kappa\,\gve\,\gvf C_{Im} \\[5mm]
146: \jtotf & = & 2\kappa\,\gve\,\gvf \left(C_{Re}+C_{Im}\right) \\[5mm]
147: \gtotf & = & Q^2_{\mathrm{e}}Q^2_{\mathrm{f}}\left|F_A(\MZ)\right|^2 \\[5mm]
148: \rfbf & = & 4\kappa^2\gae\,\gve\,\gaf\,\gvf
149: -2\kappa\,\gae\,\gaf C_{Im} \\[5mm]
150: \jfbf & = & 2\kappa\,\gae\,\gaf \left(C_{Re}+C_{Im}\right) \\[5mm]
151: \gfbf & = & 0 \,,
152: \end{array}
153: \end{equation}
154: with the following definitions:
155: \begin{equation}
156: \begin{array}{l}
157: \displaystyle
158: \kappa = \dfrac{G_F\MZ^2}{2\sqrt{2\,}\pi\alpha} \approx 1.50\\[5mm]
159: C_{Im} = \dfrac{\GZ}{\MZ} \left.Q_{\mathrm{e}}Q_{\mathrm{f}}\right.
160: \mathrm{Im} \left\{F_A(\MZ)\right\} \\[5mm]
161: C_{Re} = \left.Q_{\mathrm{e}}Q_{\mathrm{f}}\right.
162: \mathrm{Re} \left\{F_A(\MZ)\right\} \\[5mm]
163: F_A(\MZ) = \dfrac{\alpha(\MZ)}{\alpha} \,,
164: \end{array}
165: \end{equation}
166: where $\alpha(\MZ)$ is the complex fine-structure constant, and
167: $\alpha\equiv\alpha(0)$.
168: The photonic virtual and bremsstrahlung corrections are included through
169: the convolution of Equations~\ref{smat:eqn:eq1} and~\ref{smat:eqn:eq2}
170: with radiator functions as in the 5 and 9 parameter fits.
171: The expressions of the S-Matrix parameters in terms of the
172: effective vector and axial-vector couplings given above
173: neglect the imaginary parts of the effective couplings.
174:
175:
176: The usual definitions of the mass $\MZ$ and width $\GZ$ of a Breit-Wigner
177: resonance are used, the width being $s$-dependent, such that:
178: \begin{equation}
179: \begin{array}{lllll@{}c@{}r}
180: \MZ & \equiv & \MZbar\sqrt{1+\GZbar^2/\MZbar^2} & \approx & \MZbar & + & 34.20~\MeV/c^2\phantom{\,,} \\[3mm]
181: \GZ & \equiv & \GZbar\sqrt{1+\GZbar^2/\MZbar^2} & \approx & \GZbar & + & 0.94~\MeV\,.\phantom{/c^2}
182: \end{array}
183: \label{smat:eqn:smat-ls}
184: \end{equation}
185:
186: In the following fits, the predictions from the S-Matrix ansatz and
187: the QED convolution for cross-sections and asymmetries are made using
188: SMATASY~\cite{smat:ref:smatasy}, which in turn uses
189: ZFITTER~\cite{smat:ref:lep2xsafbave} to calculate the QED convolution
190: of the electroweak kernel. In case of the $\ee$ final state,
191: $t$-channel and $s/t$ interference contributions are added to the
192: $s$-channel ansatz.
193:
194: %
195: %
196: %
197: \section{LEP combination}
198: \label{smat:sec:comb}
199:
200: In the following sections the combinations of the results from the
201: individual LEP experiments are described: firstly the {\LEPI}
202: combination, then the combination of both {\LEPI} and {\LEPII} data.
203: The results from these combinations are compared in
204: Section~\ref{smat:sec:discuss}. Although all 16 parameters are
205: averaged during the combination, only results for the
206: parameters $\MZ$ and $\jtoth$ are reported here. Systematic studies
207: specific to the other parameters are ongoing.
208:
209:
210: %
211: %
212: %
213: \subsection{\LEPI\ combination}
214: \label{smat:sec:lep1}
215:
216: Individual LEP experiments have their own determinations of the
217: 16 S-Matrix parameters~\cite{smat:ref:aleph,smat:ref:delphi,smat:ref:l3lep1,smat:ref:opallep1} from {\LEPI} data alone, using the full {\LEPI} data sets.
218:
219: These results are averaged using a multi-parameter BLUE technique
220: based on an extension of Reference~\citen{common_bib:BLUE}. Sources of
221: systematic uncertainty correlated between the experiments have been
222: investigated, using techniques described in~\cite{smat:ref:5and9} and are
223: accounted for in the averaging procedure and benefiting from the experience
224: gained in those combinations.
225:
226: The parameters $\MZ$ and $\jtoth$ are the most sensitive of all 16
227: S-matrix parameters to the inclusion of the {\LEPII} data, and are
228: also the
229: most interesting ones in the context of the 5 and 9 parameter fits. For
230: these parameters the most significant source of systematic error which
231: is correlated between experiments comes from the uncertainty on the
232: $\ee$ collision energy as determined by models of the LEP RF system
233: and calibrations using the resonant depolarisation technique. These
234: errors amount to $\pm 3$ MeV on $\MZ$ and $\pm 0.16$ on $\jtoth$ with
235: a correlation coefficient of $-0.86$. The LEP averaged values of
236: $\MZ$ and $\jtoth$ are given in Table~\ref{smat:tab:lepres}, together
237: with their correlation coefficient. The $\chi^2/$D.O.F. for the
238: average of all 16 parameters is 62.0/48, corresponding to a
239: probability of $8\%$, which is acceptable.
240:
241: %
242: \begin{table}[tpb]
243: \begin{center}
244: \renewcommand{\arraystretch}{1.2}
245: \begin{tabular}{|c|r@{$\pm$}l|r@{$\pm$}l|c|}
246: \hline
247: & \multicolumn{2}{c|}{ $\MZ$ [GeV] }& \multicolumn{2}{c|}{ $\jtoth$}
248: & correlation \\ \hline\hline
249: {\LEPI} only & 91.1925 & 0.0059 & -0.084 & 0.324 & -0.935 \\
250: {\LEPI} \& {\LEPII} & 91.1869 & 0.0023 & 0.277 & 0.065 & -0.461 \\ \hline
251: \end{tabular}
252: \caption{Averaged {\LEPI} and {\LEPII} S-Matrix results for $\MZ$ and $\jtoth$.}
253: \label{smat:tab:lepres}
254: \end{center}
255: \end{table}
256: %
257:
258: %
259: %
260: %
261: %
262: %
263: %
264: %
265: %
266: %
267: %
268: %
269: %
270: %
271: %
272: %
273: %
274: %
275: %
276: %
277:
278: %
279: %
280: %
281: \subsection{{\LEPI} and {\LEPII} combination}
282: \label{smat:sec:lep2}
283:
284: Some experiments have determined S-Matrix parameters using
285: their {\LEPI} and {\LEPII} measured cross-sections and forward-backward
286: asymmetries~\cite{smat:ref:aleph,smat:ref:delphi,smat:ref:l3lep2,smat:ref:opallep2}. To do a full LEP combination
287: would require each experiment to provide S-Matrix results and would require
288: an analysis of the correlated systematic errors on each measured parameter.
289:
290: However, preliminary combinations of the measurements of forward-backward
291: asymmetries and cross-sections from all 4 LEP experiments, for the
292: full~{\LEPII} period, have already been made~\cite{smat:ref:lep2xsafbave} and
293: correlations between these measurements have been estimated. The combination
294: procedure averages measurements of cross-sections and asymmetry for those
295: events with reduced centre-of-mass energies, $\rootsp$, close to the actual
296: centre-of-mass energy of the $\ee$ beams, $\roots$, removing those
297: events which are less sensitive to the $\Zzero-\gamma$ interference where,
298: predominantly, initial state radiation reduces the centre-of-mass
299: energy to close to the mass of the $\Zzero$. The only significant correlations
300: are those between hadronic cross-section measurements at different energies,
301: which are around 20--40\%, depending on energies.
302:
303:
304: The predictions from SMATASY are fitted to the combined
305: {\LEPII} cross-section and forward-backward asymmetry
306: measurements~\cite{smat:ref:lep2xsafbave}.
307: The signal definition 1 of Reference~\citen{smat:ref:lep2xsafbave} is
308: used for the data and for the predictions of SMATASY.
309: Theoretical uncertainties on the S-Matrix predictions for the {\LEPII}
310: results and on the corrections of the $\LEPII$ data to the common signal
311: defintion are taken to be the same as for the
312: Standard Model predictions of ZFITTER~\cite{smat:ref:lep2xsafbave} which
313: are dominated by uncertainties in the QED convolution. These
314: amount to a relative uncertainty of $0.26\%$ on the hadronic cross-sections,
315: fully correlated between all {\LEPII} energies.
316:
317: The fit also uses as inputs the averaged {\LEPI} S-Matrix
318: parameters and covariance matrix. These inputs effectively constrain those
319: parameters, such as $\MZ$, which are not accurately determined by
320: {\LEPII} data.
321: There are no significant correlations between the {\LEPI} and {\LEPII} inputs.
322:
323: The LEP averaged values of $\MZ$ and $\jtoth$ for both {\LEPI} and {\LEPII}
324: data are given in Table~\ref{smat:tab:lepres}, together with their
325: correlation coefficient.
326: The $\chi^2/$D.O.F. for the average of all 16 parameters
327: is 64.4/60, corresponding to a probability of $33\%$, which is good.
328:
329: %
330: %
331: %
332: %
333: %
334: %
335: %
336: %
337: %
338: %
339: %
340: %
341: %
342: %
343: %
344: %
345: %
346: %
347: %
348:
349:
350: %
351: %
352: %
353: \subsection{Discussion}
354: \label{smat:sec:discuss}
355:
356: In the {\LEPI} combination the measured values of the Z boson mass
357: $\MZ = 91.1925 \pm 0.0059$~GeV agrees well with the results of the standard
358: 9 parameter fit ($91.1876 \pm 0.0021$~GeV) albeit with a significantly
359: larger error, resulting from the correlation with the large uncertainty
360: on $\jtoth$ which is then the dominant source of uncertainty on $\MZ$ in the
361: S-Matrix fits.
362: The measured value of $\jtoth = -0.084 \pm 0.324 $ , also agrees with the
363: prediction of the Standard Model ($0.2201^{+0.0032}_{-0.0137}$).
364:
365: Including the {\LEPII} data brings a significant improvement in the
366: uncertainty on the size of the interference between $\Zzero$ and photon
367: exchange compared to {\LEPI} data alone. The measured value
368: $\jtoth = 0.277 \pm 0.065$, agrees well with the values predicted from the
369: Standard Model. Correspondingly, the uncertainty
370: on the the mass of the $\Zzero$ in this ansatz, $2.3$ MeV, is close to
371: the precision obtained from {\LEPI} data alone using the standard 9 parameter
372: fit, $2.1$ MeV. The slightly larger error is due to the uncertainty on
373: $\jtoth$ which amounts to 0.9~MeV.
374: The measured value, $\MZ = 91.1869 \pm 0.0023$~GeV, agrees with that
375: obtained from the standard 9 parameter fits.
376: The results are summarised in Figure~\ref{smat:fig:mzjtoth}.
377:
378: The good agreement found between the values of $\MZ$ and $\jtoth$ and their
379: expectations provide a validation of the approach taken in the standard 5
380: and 9 parameter fits, in which the size of the interference between $\Zzero$
381: boson and photon exchange in the hadronic cross-sections was fixed to the
382: Standard Model expectation.
383:
384: The precision on $\jtoth$ is slightly better than that obtained by the VENUS
385: collaboration~\cite{smat:ref:venus} of $\pm 0.08$, which was obtained using
386: preliminary results from {\LEPI} and their own
387: measurements of the hadronic cross-section below the $\Zzero$ resonance.
388: The measurement of the hadronic cross-sections from VENUS~\cite{smat:ref:venus}
389: and TOPAZ~\cite{smat:ref:topaz} could be included in the future to give a
390: further reduction in the uncertainty on $\jtoth$.
391:
392: Work is in progress to understand those sources of systematic error,
393: correlated between experiments, which are significant
394: for the remaining S-Matrix parameter that have not been presented here.
395: In particular, for $\jtote$ and $\jfbe$, it is important to understand the
396: errors resulting from $t$-channel contributions to the $\eeee$ process.
397: %
398: These errors have only limited impact on the standard 5 and 9 parameter fits.
399:
400:
401:
402: %
403: \begin{figure}[p]
404: \begin{center}
405: \mbox{\epsfig{file=smat_contour_col.eps,width=\textwidth}}
406: \caption{Error ellipses for $\MZ$ and $\jtoth$ for {\LEPI} (at 39\% and
407: 68\%) and the combination of {\LEPI} and {\LEPII} (at 68\%).}
408: \label{smat:fig:mzjtoth}
409: \end{center}
410: \end{figure}
411: %
412:
413:
414: %
415: %
416: %
417: \section{Conclusion}
418: \label{smat:sec:conc}
419:
420: Results for the S-Matrix parameter $\MZ$ and $\jtoth$ have been presented
421: for {\LEPI} data alone and for a fit using the full data sets for
422: {\LEPI} and {\LEPII} from all 4 LEP experiments.
423: Inclusion of {\LEPII} data brings a significant improvement
424: in the determination of $\jtoth$, the fitted value $0.277 \pm 0.065$, agrees
425: well with the values predicted from the Standard Model. As a result
426: in the improvement of the uncertainty in $\jtoth$, the uncertainty on the
427: fitted value of $\MZ$ approaches that of the standard 5 and 9 parameter
428: fits and the measured value $\MZ = 91.1869 \pm 0.0023$~GeV is compatible
429: with that from the standard fits.
430:
431:
432:
433:
434:
435:
436:
437: