nucl-ex0111004/piAnalysis.tex
1: %
2: % piAnalysis.tex
3: %
4: \chapter{Results and Discussion: Identified Pions}
5: \label{ch:PiResults} The next step in the analysis of charged
6: particles is to use the particle identification capabilities of
7: the TPC.  This allows us to select a specific particle for further
8: study. The first natural candidate is the $\pi$ meson, the
9: lightest of the hadrons and the most copiously produced particle
10: in a high energy heavy ion collision.
11: 
12: \section{Raw Yields}
13: The starting point to obtain the raw yields for the identified
14: particle analyses is Fig.~\ref{fig:dedxvsp}.  One can obtain the
15: raw yields of pions, kaons and protons (and their antiparticles)
16: selecting events according to their centrality and fitting the
17: \dedx\ distribution in a given y-\pt\ phase space cell.  The
18: procedure adopted here is a variation from this which offers a few
19: advantages.   Tracks were selected according to quality criteria
20: based on number of points on the track and on the pointing
21: accuracy to the primary event vertex.  The cuts used in the pion
22: analysis were more stringent than the ones used for the \hminus\
23: analysis.  This was done mainly to select tracks with a good
24: \dedx\ resolution as it depends mainly on the number of valid
25: ionization samples, see Fig.\ref{fig:dedxResolution}.
26: 
27: For this analysis, we followed a procedure similar to the one
28: outlined in Ref.~\cite{aguilar-benitez:91} for ionization
29: measurements.  We use the truncated mean to estimate the
30: ionization of a given track ($I_m$).  The expected ionization for
31: a given momentum and a given mass hypothesis ($I_h$)is known to an
32: accuracy of better than 0.1\%, as represented by the curves in
33: Fig.\ref{fig:dedxvsp}.  We then construct the $z$ variable defined
34: as
35: \begin{equation}\label{eq:zvariable}
36:   z = \ln\left(\frac{I_m}{I_h}\right)
37: \end{equation}
38: The $z$ variable follows a Gaussian distribution with mean zero
39: for a given particle population and with standard deviation given
40: by the measured \dedx\ resolution as a function of the number of
41: \dedx\ samples, Fig.\ref{fig:dedxResolution}.  The mean will be
42: zero independent of momentum for the given particle species under
43: consideration, since the momentum dependence is contained in
44: $I_h$. By incorporating $I_h$ and the measured resolution
45: $\sigma_{dE/dx}$, we can obtain a new distribution, $Z_\pi$, that
46: has unit width and zero mean for the particle of interest:
47: $Z_{\pi}=z_{\pi}/\sigma_{dE/dx}$.
48: 
49: Fig.~\ref{fig:pionzfits} shows the result of fitting to the $Z$
50: distributions in different phase-space bins.  Shown is the region
51: 0.2 < y < 0.3 for 6 different \pt\ bins in steps of 50 \mevc\
52: starting from \pt = 0.3 \gevc.
53: \begin{figure}[htb]
54: \begin{center}
55: \includegraphics[width=.6\textwidth]{plots/piMinusZFitsc05Y13.eps}
56: \caption[Fits to $Z_{\pi}$ distributions for raw \piminus\
57: yields.]{Fits to the $Z_{\pi}$ distribution to extract the yield
58: of $\pi^-$. Shown is the region $0.2 < y < 0.3$ and $0.3 < \pt <
59: 0.6\ \gevc$.} \label{fig:pionzfits}
60: \end{center}
61: \end{figure}
62: The abscissa is in log scale, and the ordinate is in linear scale
63: (note that $Z_\pi$ is already the log of the ionization).
64: Strictly, only the pion population should follow a Gaussian in the
65: $z_{\pi}$ variable. One can of course construct a $z$ variable for
66: the other species ($z_{K}$, $z_{p}$ and $z_{e}$) in order to
67: extract their yields. The analysis of the other particle species
68: is being carried out in STAR also. Here we focus on the pions
69: which are the bulk of the produced particles as can be seen from
70: the relative heights of the different Gaussians in the plot. The
71: figure shows the result of the different fits using a Gaussian for
72: each of the particle species. The pion yield is extracted from the
73: Gaussian centered at zero. In the lower right panel, the electron
74: yield is small enough and close enough to the kaon population that
75: one can use a single Gaussian function.  The fit also stops at
76: $Z_{\pi}=5$ in the bottom right panel, thus the anti-proton
77: Gaussian is not shown.
78: 
79: \section{Corrections}
80: The corrections applied to the raw data follow the same procedure
81: as for the \hminus\ distribution with a few important differences.
82: First of all, we chose to do them in cells of ($y,\pt$) instead of
83: ($\eta,\pt$).  The acceptance and efficiency use the embedding
84: procedure as in the \hminus\ analysis.  Embedding was done in the
85: region .1 < \pt\ < 2 \gevc, and |$y$|<1.  For the lowest \pt\ bin,
86: 50 < \pt\ < 100 \mevc, we used a full \Hijing\ simulation. The
87: efficiency obtained from embedding is shown in
88: Fig.\ref{fig:piEfficiency}.
89: \begin{figure}[htb]
90:   \centering
91:   \includegraphics[width=.7\textwidth]{plots/pionEfficiency.eps}
92:   \caption[\piminus\ reconstruction efficiency \vs\ \pt.]
93:   {The track reconstruction efficiency
94:   as a function of \pt\  for \piminus\ mesons.  The efficiency
95:   is lower than in Fig.~\ref{fig:efficiency} as the
96:   requirements for tracks in the \dedx\ analysis are more
97:   stringent than for the \hminus\ analysis.}
98:   \label{fig:piEfficiency}
99: \end{figure}
100: The stringent requirement on 24 or more fit points on the track
101: effectively removed the need to undertake a split track
102: correction. For the momentum range $\pt<0.6\ \gevc$ the distortion
103: of the spectrum due to resolution effects was found to be
104: negligible. This fact and the good statistics of the sample
105: allowed for a much finer bin size in the \pt\ spectrum.  This was
106: also desirable since this narrowed the phase space where the fits
107: were made to obtain the yield. Event though the use of the $z$
108: variable fixes the region where the pions are found, the other
109: particles will still vary with momentum.  Therefore, a narrower
110: \pt\ bin will reduce this effect making the fits more stable.  The
111: final spectrum was also corrected for the products of weak decays
112: and for the expected contribution of pions resulting from
113: secondary interactions. Since for this analysis the requirement on
114: the distance of closest approach to the primary interaction vertex
115: was set at 1 cm, the background correction was smaller than for
116: the \hminus\ analysis, and was found to be $\lesssim 5\%$. We
117: present now the resulting distributions for identified \piminus\
118: mesons.
119: 
120: \section{\pt\ and \mt\ Distributions}
121: 
122: \subsection{Results}
123: We first look at the spectra for central events.
124: Figure~\ref{fig:pionspectraptcentral} shows the \pt\ distribution
125: for \piminus\ for the 5\% most central collisions (selected by a
126: cut on raw \hminus\ multiplicity).  The data are grouped for
127: different bins in rapidity, and scaled by successive powers of 2
128: for the figure.
129: \begin{figure}[htb]
130: \begin{center}
131: \includegraphics[width=0.5\textwidth]{plots/piPtCentY.eps}
132: \caption[\piminus\ \pt\ spectrum for central collisions]{\piminus\
133:  \pt\ distribution for central collisions.  The different bins in rapidity
134:  are scaled by successive powers of 2 for display purpose.} \label{fig:pionspectraptcentral}
135: \end{center}
136: \end{figure}
137: 
138: The more common way to show such a distribution is in transverse
139: mass, \mt, and this is shown in Figure
140: ~\ref{fig:pionspectracentral}.
141: \begin{figure}[htb]
142: \begin{center}
143: \includegraphics[width=0.6\textwidth]{plots/piMtCentY.eps}
144: \caption[\piminus\ \mt\ spectrum for central collisions]{\piminus\
145: $d^2N/(2\pi\ dy\ \mt d\mt)$ distribution for central collisions.
146: The rapidity bins are scaled by successive powers of 2 for display
147: purpose.} \label{fig:pionspectracentral}
148: \end{center}
149: \end{figure}
150: The main corrections applied to the raw data are acceptance,
151: efficiency and contributions from weak decay background and
152: secondary interactions.  The effect of energy loss in the
153: determination of the \piminus\ momenta was found to be less than
154: 2\% at 100 \mevc, decreasing rapidly for larger momenta.  We also
155: studied the effect of momentum resolution as for the \hminus\ data
156: and found that the main effect was the feeding of the lowest \pt\
157: bin from its nearest neighbour. The change in the slope for $\pt\
158: > 100\ \mevc$ was found to be $\ll 1\%$.  The data for the central
159: rapidity bins, $|y|<0.4$, was obtained in the range $0.05 < \pt <
160: 0.75\ \gevc$, which accounts for $\sim 85\%$ of the total yield.
161: As we go towards the forward and backward regions, the \pt\ range
162: where the identification via \dedx\ in the TPC can be done
163: shrinks, and we can only measure in the range 0.2 < \pt\ < 0.6
164: \gevc.
165: 
166: \subsection{Discussion}
167: Pions are the bulk of the produced particles, they are the
168: lightest of the hadrons and have large interaction cross sections
169: in nuclear matter; they are therefore expected to thermalize
170: easily.  In addition, their spectra are the least affected by a
171: given collective transverse flow velocity, and are thus good
172: probes to study the kinetic freeze-out properties of the system.
173: The pion data are fit with a Bose-Einstein distribution of the
174: form $A/(\exp(\mt/T) - 1)$. Typically, these distributions are fit
175: with a Boltzmann distribution \cite{schukraft:91}.  The data at
176: low-\pt\ fit with a Boltzmann distribution always seemed to show
177: an enhancement.  This partly was attributed to contributions
178: coming from resonance decays. There are also enhancements in the
179: high-\pt\ region which can arise from contributions of pions
180: coming from high momentum transfer of partons early in the
181: collision.  A consequence of all these structures was that simple
182: Boltzmann fits to the same spectrum would lead to different slope
183: parameters depending upon the \pt\ region included in the fit.
184: Since we focus here on the low \pt\ part of the spectrum, and
185: since pions are bosons, we chose to fit the data with a
186: Bose-Einstein distribution.  We find that this naturally gives an
187: enhancement in the very low \pt\ region of the spectrum.  Also, we
188: have already estimated the contribution from weak decays to the
189: distribution and corrected for this effect.  We have not estimated
190: the contribution from resonance decays however. We fit the data
191: over the measured range, instead of excluding the low momentum
192: data from the fit, and find no significant enhancement at low \pt.
193: The absence of a noticeable contribution from resonances might be
194: due to reinteraction of the daughter decay particles. A fit using
195: a Boltzmann distribution for the range above $\pt = 0.3\ \gevc$
196: causes a variation in the slope parameters of $\sim 10\ \mev$.
197: Since the data at forward and backward rapidity have a smaller
198: range in \pt\, as a systematic check we performed a fit restricted
199: to the \pt\ range which was measured for all the rapidity bins.
200: This resulted in a change in the slope parameters of 5-10 \mev\
201: for the $|y|<0.4$ spectra.  We find the slope parameter for the
202: central collisions to be $\Teff = 210\ \pm 20\ \mev$, where the
203: systematic uncertainty comes from studying the variation in the
204: slope parameter with several changes, including variations in the
205: fit range, choice of fit function, centrality selection, and
206: analyzing the pions in $\eta$ instead of $y$.  The error on
207: $\Teff$ from any particular fit is on the order of $5\ \mev$. The
208: rapidity dependence of the \piminus\ yield and of the slope
209: parameters is discussed in the following section.
210: 
211: The shape of the low \pt\ pion spectra, in contrast to the
212: \hminus\ distribution, shows very little change when going from
213: SPS energies to RHIC.  The slope parameters reported by NA49
214: \cite{Jones:1996xc} for \piminus\ near midrapidity are $\Teff =
215: 188 \pm 6\ \mev$ for \piplus\ and $\Teff = 192 \pm 3\ \mev$ using
216: an exponential fit.  WA98 data for $\pi^0$ spectra for central
217: collisions yields also similar values of $\Teff \simeq 204$
218: (although fitting the data with a Bessel function, $\mt
219: K_1(\mt/\Teff)$ yields a smaller value of $\Teff=155$, so one must
220: be careful when comparing). The pion spectra from NA45
221: \cite{ceretto:98} were analyzed in terms of a local inverse slope
222: changing with \mt\ to take into account the difference at low and
223: high \pt.  For the low \mt\ range ($\mt-m_{\pi} < 0.8\ \gevcc$)
224: the local inverse slope is found to be in the range $180-200\
225: \mev$. The slopes at the AGS have smaller values, E895 finds
226: \cite{Liu:1998rt} $\Teff \simeq 110\ \mev$.  While the initial
227: temperature reached at RHIC is expected to be larger than at the
228: SPS and the AGS, the \piminus\ spectrum probes the final state.
229: The saturation of the \piminus\ slope parameter would indicate
230: that the freeze-out temperature, assuming that thermal equilibrium
231: is reached, is the same at SPS and RHIC. This does not mean that
232: the final state is the same, for there is evidence of an increased
233: collective radial flow velocity at RHIC based on the preliminary
234: \pt\ distributions of heavier particles from STAR
235: \cite{harris:01,calderon:01}. Since the collective velocity
236: affects the heavier particles the most, the pion inverse slope is
237: the one numerically closer to the actual freeze-out temperature,
238: \Tfo.  The slopes of the \kminus\ and \pbar\ will be much larger
239: than for the \piminus\ even if they all freeze out at the same
240: temperature in the presence of large radial flow.
241: 
242: In order to gain further insight into the dynamics of the system,
243: a study of the heavier particle spectra can be made to address the
244: question of radial flow.  In a different approach, the combination
245: of the 2-particle correlation results\cite{Adler:2001zd} along
246: with the pion spectra is also useful: they determine the
247: 6-dimensional pion \textit{phase space density}, \ie\ the
248: dimensionless average number of pions per 6-dimensional phase
249: space cell $\hbar^3$ (see \eg\ \cite{Ferenc:1999ku}).  Such
250: analyses can help to disentangle the intrinsic freeze-out
251: temperature and the transverse flow velocity.  For example, for
252: the 5\% most central data, preliminary analysis of the phase space
253: density yield values of $\Tfo \sim 120\ \mev$ and $\langle
254: \beta_\perp \rangle \sim 0.53$.  This is a rather large flow
255: velocity, yet since the pions are so light, the slope parameter
256: $\Teff$ is only increased to $\sim 200$ \mev, while the slope
257: parameter for the heavier \kminus\ increases to $\sim 300\ \mev$
258: and is even larger for the \pbar\ \cite{harris:01}. The
259: systematics of the centrality dependence of the pion phase space
260: density is a currently active area of study in STAR which will
261: certainly help to shed light on the determination of the kinetic
262: freeze-out conditions of the system.
263: 
264: \section{Rapidity Distribution}
265: \subsection{Results}
266: By integrating the \pt\ spectra in
267: Figure~\ref{fig:pionspectraptcentral} for each of the rapidity
268: bins, we are able to obtain the $dN/dy$ \vs\ $y$ distribution for
269: \piminus.  The result is shown in Figure~\ref{fig:piondndy}.
270: Several data points are shown.  The black circles are obtained by
271: integrating the \pt\ spectrum in the limited \pt\ region $0.2 -
272: 0.6\ \gevc$ where we have data for $|y|<0.8$.  This shows that the
273: measured yield in this region is relatively independent of
274: rapidity. The measured yield in this region is, however, only
275: about half of the total yield.  To illustrate the behaviour of the
276: yield in the full available $\pt$ region ($0.05 - 0.75\ \gevc$)
277: for a narrower $y$ region, we plot the data shown as the black
278: squares.  Finally, the open circles are the extrapolated yield to
279: all $\pt$ based on the fits to the data using the Bose-Einstein
280: distribution.  We find the pion yield to be $\dndy|_{|y|<0.1} =
281: 286 \pm 10$ for the 5\% most central events.
282: \begin{figure}[htb]
283: \begin{center}
284: \includegraphics[width=.6\textwidth]{plots/pidndy.eps}
285: \caption[\piminus\ rapidity distribution, $dN/dy$]{\piminus\
286: rapidity distribution. The black circles (squares) are obtained
287: summing the yield between \pt\ 0.2 - 0.6 \gevc\ (0.05 - 0.75) from
288: Fig.~\ref{fig:pionspectraptcentral}. The hollow data points are
289: the yields obtained using the fit function extrapolated to all
290: \pt.} \label{fig:piondndy}
291: \end{center}
292: \end{figure}
293: Not only is it important to study the yields as a function of
294: rapidity, but it is equally important to examine the rapidity
295: dependence of the slope parameters.  At the SPS and at the AGS,
296: they have been found to decrease with increasing $|y|$.  The
297: $\Teff$ slope parameter extracted from the Bose-Einstein fit to
298: the data is shown in Figure~\ref{fig:pionTeffY}.
299: \begin{figure}[htb]
300: \begin{center}
301: \includegraphics[width=.7\textwidth]{plots/piTeffY.eps}
302: \caption[\piminus\ slope parameter $T_{\mathrm{eff}}$ \vs\ $y$.
303: ]{\piminus\ slope parameter for the different rapidity bins from
304: Fig.~\ref{fig:pionspectracentral}. The flat line at $\Teff = 210$
305: is the average for $|y|<0.1$} \label{fig:pionTeffY}
306: \end{center}
307: \end{figure}
308: The error bars in the figure are the uncorrelated point-to-point
309: systematic uncertainties on $\Teff$, the overall correlated
310: systematic uncertainty is $\pm 20\ \mev$.  The rapidity dependence
311: of the yield, the slope parameter \Teff\ and \meanpt\ are
312: collected in
313: \begin{table}[htb]
314:     \centering
315:     \begin{tabular}{cccc}
316:         \hline \hline
317:         $y$   &      \dndy\            &  \Teff\ [\mev]   &  \meanpt\ [\mevc] \\
318:         -0.75 & 273 $\pm$ 13  &  185.0 $\pm$ 5.0 &  356 $\pm$ 9.9  \\
319:         -0.65 & 274 $\pm$ 11  &  198.1 $\pm$ 4.2 &  376 $\pm$ 8.4  \\
320:         -0.55 & 273 $\pm$ 10  &  201.7 $\pm$ 4.6 &  385 $\pm$ 9.2  \\
321:         -0.45 & 280 $\pm$ 10  &  207.5 $\pm$ 4.0 &  391 $\pm$ 7.9  \\
322:         -0.35 & 283 $\pm$ 10  &  215.2 $\pm$ 3.6 &  397 $\pm$ 7.2  \\
323:         -0.25 & 288 $\pm$ 10  &  213.4 $\pm$ 3.1 &  402 $\pm$ 6.2  \\
324:         -0.15 & 284 $\pm$ 10  &  209.5 $\pm$ 3.2 &  396 $\pm$ 6.4  \\
325:         -0.05 & 287 $\pm$ 10  &  207.7 $\pm$ 3.9 &  394 $\pm$ 7.8  \\
326:          0.05 & 285 $\pm$ 10  &  213.3 $\pm$ 3.4 &  400 $\pm$ 6.8  \\
327:          0.15 & 286 $\pm$ 10  &  208.9 $\pm$ 3.4 &  392 $\pm$ 6.9  \\
328:          0.25 & 284 $\pm$ 10  &  216.5 $\pm$ 3.8 &  404 $\pm$ 7.6  \\
329:          0.35 & 285 $\pm$ 10  &  211.9 $\pm$ 3.1 &  400 $\pm$ 6.2  \\
330:          0.45 & 272 $\pm$ 10  &  203.4 $\pm$ 3.2 &  383 $\pm$ 6.4  \\
331:          0.55 & 275 $\pm$ 10  &  206.5 $\pm$ 3.4 &  388 $\pm$ 6.8  \\
332:          0.65 & 275 $\pm$ 11  &  191.5 $\pm$ 4.0 &  367 $\pm$ 8.0  \\
333:          0.75 & 269 $\pm$ 14  &  182.4 $\pm$ 5.6 &  350 $\pm$ 11.2 \\
334:          \hline \hline
335: \end{tabular}
336: \caption[\piminus\ \dndy, \Teff\ and mean \pt\ \vs\ $y$.]{\piminus\ \dndy, \Teff\ and \meanpt\ \vs\ $y$.}
337: \label{tab:piRapidity}
338: \end{table}
339: Table~\ref{tab:piRapidity}. The \pt\ distributions for the most
340: central data for the various $y$ bins are collected in
341: Table~\ref{tab:piPtRapidity}.
342: 
343: \subsection{Discussion}
344: The rapidity distribution has nearly a plateau shape at
345: mid-rapidity. However, there remains a slight rapidity dependence
346: of the slope parameter $\Teff$.  This behaviour is indicative that
347: the idealized boost-invariant mid-rapidity region is not yet
348: reached in \AuAu\ collisions at \sqrtsNN = 130 \gev.  A possible
349: cause for the observed effect can be a change in the baryon
350: content of the system with rapidity. In the presence of radial
351: flow driven by the dominant \piminus\ mesons, a larger fraction of
352: baryons would take away some of the pion's kinetic energy,
353: reducing the observed slope parameter.  The anti-proton to proton
354: ratios reported by STAR~\cite{Adler:2001bp} and
355: BRAHMS~\cite{Bearden:2001kt} show a drop in the $\pbar/p$ ratio
356: from 0.65 at mid-rapidity to 0.41 for the region $y\approx 2$,
357: which already hints at such effects, but a better understanding of
358: the dynamics must wait for the measured baryon yields and the
359: baryon rapidity distributions.
360: 
361: \section{Centrality Dependence}
362: \subsection{Centrality Selection}
363: For the \piminus\ centrality dependence, we chose to do a
364: selection based on the event multiplicity.  This is in contrast to
365: the centrality selection for the \hminus\ analysis which was based
366: on ZDC and CTB trigger signals.  The reason for this choice was
367: simply to use a common set of cuts which would be useful not just
368: for this analysis, but also for other studies such as particle
369: correlations (Hanbury-Brown Twiss) and in particular for the
370: studies of the pion phase space density. Having a standard set of
371: values based on the (uncorrected) \hminus\ yield facilitates
372: comparison and combination of different observables.  There was a
373: worry that since the pions are the main component of the charged
374: hadron spectra, a centrality selection based on charged
375: multiplicity would introduce a sizable auto-correlation.
376: Certainly, the shape of the \hminus\ multiplicity distribution for
377: 5\% most central events (the shaded region in
378: Fig.~\ref{fig:hminus}) would look different by selecting events
379: with a straight cut on uncorrected multiplicity.  The mean of this
380: distribution, \ie\ the value of $dN/d\eta$ for the region
381: $|\eta|<0.5$, will also shift to higher values by doing such a
382: selection.  The question is how significant is this shift.  We
383: studied this by selecting the 5\% most central events based on
384: uncorrected multiplicity and comparing the mean of the resulting
385: distribution to the mean of the histogram in the shaded region of
386: Figure ~\ref{fig:hminus}.  The mean of the histogram using the
387: centrality definition based on multiplicity was 4.5\% higher than
388: than the one using the ZDC as centrality definition.  The
389: difference in the mean of the histograms using these two
390: centrality definitions will also decrease for more peripheral
391: collisions.  We therefore conclude that the differences are not
392: problematic, and when encountered, they can be reconciled by at
393: most a 4.5\% effect.
394: 
395: \subsection{\mt\ Distributions \vs\ Multiplicity}
396: Figure~\ref{fig:pionPtCent} shows the \mt\ distribution for 10
397: different centrality selections.  The data are taken in the
398: rapidity interval $0 < y < 0.1$.  Again, the data are fit by a
399: Bose-Einstein distribution which agrees well with the spectra even
400: at the lowest \mt.  The slope parameters are very similar for the
401: different centralities.
402: \begin{figure}[htb]
403: \begin{center}
404: \includegraphics[width=.6\textwidth]{plots/piMtCentrality.eps}
405: \caption[\piminus\ \mt\ distribution for different
406: centralities.]{\piminus\ \mt\ distribution for different
407: multiplicity selections.  The distributions are plotted in the
408: rapidity slice 0 < $y$ < 0.1 units. The data are fit to
409: Bose-Einstein distributions (curves) from which we extract the
410: slope parameter \Teff.} \label{fig:pionPtCent}
411: \end{center}
412: \end{figure}
413: They are collected and plotted \vs\ the \hminus\ mean multiplicity
414: of the corresponding centrality bin in
415: Figure~\ref{fig:pionTeffCent}.
416: \begin{figure}[htb]
417: \begin{center}
418: \includegraphics[width=.6\textwidth]{plots/piTeffCentrality.eps}
419: \caption[Centrality dependence of the \piminus\ slope
420: parameter.]{Centrality dependence of the slope parameter obtained
421: by a fit to the \piminus\ \mt\ distribution.}
422: \label{fig:pionTeffCent}
423: \end{center}
424: \end{figure}
425: The horizontal error bars are the systematic uncertainty in the
426: determination of the mean multiplicity of each centrality bin. The
427: vertical error bars are the uncorrelated systematic uncertainty in
428: the measurement of the slope parameter.  For the most peripheral
429: bins, the statistical error is also important, and the error bar
430: is therefore larger.
431: 
432: We see that there is a only a slight dependence of the slope
433: parameter with centrality.  The main difference happens for the
434: 2nd and 3d most peripheral bins (the most peripheral bin has a
435: larger error, so it is hard to see a systematic effect).  The 2nd
436: most peripheral bin has a slope $\Teff = 177 \pm 8\ \mev$, while
437: the most central bin has a slope $\Teff = 210 \pm 4\ \mev$ (where
438: the errors are the uncorrelated systematic uncertainties for the
439: comparison). This yields an increase of $19 \pm 5 \%$. The
440: increase occurs rapidly, as the 60\% of \sigmahad\ bin already has
441: a slope parameter of $\Teff = 196 \pm 9 \mev$, and from then on
442: the data are consistent with having no further centrality
443: dependence.  For comparison, the increase in \hminus\ \meanpt\
444: from the $70-80\%$ bin to the most central bin is $14 \pm 3\%$. In
445: addition, the \hminus\ distribution includes a \pbar\ component,
446: and the \pbar\ slope has a stronger centrality dependence than the
447: \piminus\ slope. At SPS energies, the centrality dependence of the
448: pion slopes was also found to be rather constant. This finding is
449: also in agreement with the argument that the slope of the
450: \piminus\ spectrum is not changed significantly by the collective
451: expansion.
452: 
453: 
454: \begin{table}[htb]
455:   \centering
456:   \begin{tabular}{c|r|r|r|r|r} %\hline \hline
457:   %\multicolumn{6}{c}{\ }\\
458: %    % after \\: \hline or \cline{col1-col2} \cline{col3-col4} ...
459:     %$\mt - m_\pi$ & \multicolumn{5}{c}{Fraction of $\sigma_{\mathrm{hadronic}}$} \\
460:     $\mt - m_\pi$ {\tiny(\gevcc)}\ & \multicolumn{1}{c}{0-5\%}  & \multicolumn{1}{c}{5-10\%} & \multicolumn{1}{c}{10-20\%} & \multicolumn{1}{c}{20-30\%} & \multicolumn{1}{c}{30-40\%} \\ \hline
461:  0.0189 & 871  $\pm$ 70  & 707  $\pm$ 57   & 550  $\pm$ 43   &  377  $\pm$ 31   & 254  $\pm$ 21    \\
462:  0.0478 & 679  $\pm$ 30  & 544  $\pm$ 25   & 446  $\pm$ 19   &  312  $\pm$ 14   & 214  $\pm$ 9.5   \\
463:  0.0842 & 473  $\pm$ 25  & 388  $\pm$ 20   & 312  $\pm$ 13   &  224  $\pm$ 9.7  & 164  $\pm$ 7.0   \\
464:  0.125  & 357  $\pm$ 12  & 289  $\pm$ 11   & 226  $\pm$ 6.9  &  169  $\pm$ 5.9  & 111  $\pm$ 3.9   \\
465:  0.169  & 281  $\pm$ 11  & 231  $\pm$ 11   & 180  $\pm$ 6.4  &  120  $\pm$ 5.2  & 87.1 $\pm$ 2.8   \\
466:  0.214  & 206  $\pm$ 7.2 & 173  $\pm$ 6.7  & 138  $\pm$ 4.1  &  100  $\pm$ 2.8  & 65.7 $\pm$ 2.4   \\
467:  0.261  & 165  $\pm$ 5.8 & 134  $\pm$ 5.8  & 108  $\pm$ 3.2  &  78.1 $\pm$ 2.1  & 50.8 $\pm$ 1.5   \\
468:  0.308  & 127  $\pm$ 5.6 & 104  $\pm$ 6.9  & 84.3 $\pm$ 2.9  &  61.6 $\pm$ 1.8  & 37.6 $\pm$ 1.0   \\
469:  0.356  & 105  $\pm$ 3.2 & 85.4 $\pm$ 3.1  & 65.1 $\pm$ 2.1  &  46.6 $\pm$ 1.6  & 29.1 $\pm$ 1.2   \\
470:  0.404  & 79.6 $\pm$ 2.9 & 65.6 $\pm$ 2.6  & 50.9 $\pm$ 1.5  &  35.6 $\pm$ 1.6  & 23.8 $\pm$ 0.74  \\
471:  0.452  & 64.0 $\pm$ 1.9 & 47.5 $\pm$ 1.7  & 37.2 $\pm$ 1.2  &  26.2 $\pm$ 0.90 & 18.8 $\pm$ 0.60  \\
472:  0.501  & 47.6 $\pm$ 1.6 & 37.4 $\pm$ 1.4  & 27.7 $\pm$ 0.95 &  20.3 $\pm$ 0.77 & 14.2 $\pm$ 0.50  \\
473:  0.550  & 36.4 $\pm$ 1.3 & 29.7 $\pm$ 1.1  & 22.4 $\pm$ 0.77 &  15.9 $\pm$ 0.64 & 11.1 $\pm$ 0.42  \\
474:  0.599  & 27.7 $\pm$ 1.3 & 24.5 $\pm$ 0.98 & 17.8 $\pm$ 0.63 &  12.5 $\pm$ 0.56 & 8.82 $\pm$ 0.35  \\
475: %
476: \multicolumn{6}{c}{\ }\\ %\multicolumn{5}{c}{\ }\\ %\hline
477: %$\mt - m_\pi$    & \multicolumn{5}{c}{Fraction of $\sigma_{\mathrm{hadronic}}$} \\
478: $\mt - m_\pi$ {\tiny(\gevcc)}\ & \multicolumn{1}{c}{40-50\%} & \multicolumn{1}{c}{50-60\%} & \multicolumn{1}{c}{60-70\%} & \multicolumn{1}{c}{70-80\%} & \multicolumn{1}{c}{80-100\%} \\
479: \hline
480:  0.0189 & 165  $\pm$ 14.4 & 95.1 $\pm$ 9.0  & 47.7 $\pm$ 5.3  & 23.2  $\pm$ 3.2   & 6.67  $\pm$ 2.2   \\
481:  0.0478 & 148  $\pm$ 7.0  & 80.2 $\pm$ 4.6  & 41.4 $\pm$ 2.7  & 16.0  $\pm$ 1.8   & 4.07  $\pm$ 0.81  \\
482:  0.0842 & 103  $\pm$ 4.5  & 62.8 $\pm$ 3.7  & 34.4 $\pm$ 2.2  & 15.7  $\pm$ 1.5   & 3.00  $\pm$ 0.52  \\
483:  0.125  & 71.3 $\pm$ 2.8  & 43.0 $\pm$ 2.3  & 23.5 $\pm$ 1.5  & 10.0  $\pm$ 1.0   & 1.86  $\pm$ 0.34  \\
484:  0.169  & 52.6 $\pm$ 3.0  & 32.2 $\pm$ 1.2  & 16.4 $\pm$ 1.3  & 7.71  $\pm$ 0.41  & 1.79  $\pm$ 0.17  \\
485:  0.214  & 40.3 $\pm$ 2.2  & 23.7 $\pm$ 1.2  & 13.7 $\pm$ 0.56 & 5.97  $\pm$ 0.34  & 1.26  $\pm$ 0.14  \\
486:  0.261  & 31.1 $\pm$ 1.2  & 18.7 $\pm$ 0.71 & 9.79 $\pm$ 0.48 & 4.55  $\pm$ 0.27  & 0.956 $\pm$ 0.16  \\
487:  0.308  & 22.3 $\pm$ 1.1  & 13.7 $\pm$ 0.55 & 7.38 $\pm$ 0.35 & 3.21  $\pm$ 0.21  & 0.656 $\pm$ 0.13  \\
488:  0.356  & 18.4 $\pm$ 0.66 & 11.2 $\pm$ 0.45 & 5.67 $\pm$ 0.28 & 2.34  $\pm$ 0.37  & 0.588 $\pm$ 0.12  \\
489:  0.404  & 13.6 $\pm$ 0.50 & 8.65 $\pm$ 0.37 & 4.46 $\pm$ 0.33 & 1.68  $\pm$ 0.13  & 0.480 $\pm$ 0.066 \\
490:  0.452  & 10.8 $\pm$ 0.42 & 6.78 $\pm$ 0.30 & 3.31 $\pm$ 0.19 & 1.35  $\pm$ 0.11  & 0.351 $\pm$ 0.053 \\
491:  0.501  & 8.76 $\pm$ 0.40 & 4.75 $\pm$ 0.27 & 2.51 $\pm$ 0.18 & 0.996 $\pm$ 0.10  & 0.228 $\pm$ 0.045 \\
492:  0.550  & 6.54 $\pm$ 0.31 & 3.98 $\pm$ 0.23 & 1.76 $\pm$ 0.14 & 0.740 $\pm$ 0.076 & 0.185 $\pm$ 0.034 \\
493:  0.599  & 5.59 $\pm$ 0.30 & 3.14 $\pm$ 0.20 & 1.47 $\pm$ 0.18 & 0.645 $\pm$ 0.087 & 0.152 $\pm$ 0.040 \\
494:   \end{tabular}
495:   \caption[\piminus\ $d^2N/(2\pi \mt d\mt dy)$ \vs\ multiplicity.]
496:   {\piminus\ $d^2N/(2\pi \mt d\mt dy)$ at $y=0.05$ for various \hminus\ multiplicity bins, Fig.~\ref{fig:pionPtCent}}
497:   \label{tab:piMtMultiplicity}
498: \end{table}
499: 
500: 
501: 
502: \begin{sidewaystable}[htb]
503:     \centering
504:     \begin{tabular}{c|c|c|c|c|c|c|c|c}
505: 
506: \pt\ [\gevc] & $y = -0.75$& $y = -0.65$    & $y = -0.55$    &  $y = -0.45$   & $y = -0.35$     &   $y = -0.25$  &  $y = -0.15$   &  $y = -0.05$  \\
507: 0.075     &               &                &                &                & 5583 $\pm$ 415  & 5480 $\pm$ 437 & 5475 $\pm$ 437 & 5453 $\pm$ 435\\
508: 0.125     &               &                & 4242 $\pm$ 428 & 4332 $\pm$ 190 & 4225 $\pm$ 205  & 4265 $\pm$ 195 & 4039 $\pm$ 214 & 4297 $\pm$ 181\\
509: 0.175     &               & 3238 $\pm$ 108 & 3332 $\pm$  95 & 3347 $\pm$  99 & 3213 $\pm$ 102  & 3145 $\pm$ 112 & 3355 $\pm$ 118 & 3320 $\pm$ 115\\
510: 0.225     & 2613 $\pm$ 77 & 2428 $\pm$  60 & 2139 $\pm$  85 & 2244 $\pm$  64 & 2183 $\pm$  60  & 2361 $\pm$  71 & 2401 $\pm$  77 & 2329 $\pm$ 167\\
511: 0.275     & 1829 $\pm$ 47 & 1767 $\pm$  41 & 1699 $\pm$  45 & 1667 $\pm$  59 & 1652 $\pm$  47  & 1587 $\pm$  85 & 1590 $\pm$  71 & 1668 $\pm$  77\\
512: 0.325     & 1320 $\pm$ 32 & 1253 $\pm$  39 & 1317 $\pm$  33 & 1300 $\pm$  44 & 1244 $\pm$  40  & 1329 $\pm$  52 & 1340 $\pm$  47 & 1327 $\pm$  53\\
513: 0.375     &  963 $\pm$ 23 &  988 $\pm$  22 &  924 $\pm$ 128 & 1016 $\pm$  44 & 1028 $\pm$  29  & 1067 $\pm$  31 & 1038 $\pm$  30 & 1065 $\pm$  30\\
514: 0.425     &  705 $\pm$ 30 &  750 $\pm$  21 &  774 $\pm$  24 &  748 $\pm$  27 &  758 $\pm$  25  &  796 $\pm$  28 &  762 $\pm$  25 &  839 $\pm$  53\\
515: 0.475     &  557 $\pm$ 27 &  582 $\pm$  18 &  590 $\pm$  22 &  621 $\pm$  24 &  633 $\pm$  25  &  674 $\pm$  21 &  639 $\pm$  19 &  642 $\pm$  32\\
516: 0.525     &  457 $\pm$ 12 &  466 $\pm$  13 &  456 $\pm$  15 &  490 $\pm$  15 &  515 $\pm$  18  &  493 $\pm$  15 &  490 $\pm$  13 &  476 $\pm$  56\\
517: 0.575     &  354 $\pm$ 21 &  352 $\pm$  21 &  362 $\pm$  12 &  361 $\pm$  14 &  381 $\pm$  12  &  409 $\pm$  14 &  393 $\pm$  12 &  390 $\pm$  12\\
518: 0.625     &               &  290 $\pm$  14 &                &  293 $\pm$   9 &  296 $\pm$  10  &  297 $\pm$  10 &  310 $\pm$  10 &  306 $\pm$  15\\
519: 0.675     &               &                &                &                &  223 $\pm$   9  &  236 $\pm$   7 &  227 $\pm$  11 &  228 $\pm$  18\\
520: 0.725     &               &                &                &                &  185 $\pm$  18  &  186 $\pm$   7 &  170 $\pm$   7 &  175 $\pm$   9\\ \\
521: 
522: \pt\      &  $y = 0.05$    & $y = 0.15$     & $y = 0.25$     &  $y = 0.35$    &  $y = 0.45$    &  $y = 0.55$   &   $y = 0.65$   &   $y = 0.75$  \\
523: 0.075     & 5472 $\pm$ 437 & 5490 $\pm$ 438 & 5295 $\pm$ 424 & 5112 $\pm$ 410 &                &               &                &               \\
524: 0.125     & 4266 $\pm$ 186 & 4260 $\pm$ 215 & 4060 $\pm$ 395 & 4047 $\pm$ 219 & 4093 $\pm$ 190 &               &                &               \\
525: 0.175     & 2975 $\pm$ 159 & 3160 $\pm$ 126 & 3103 $\pm$  97 & 3109 $\pm$ 146 & 3274 $\pm$ 107 & 3262 $\pm$ 95 & 3444 $\pm$ 102 &               \\
526: 0.225     & 2244 $\pm$  75 & 2401 $\pm$  75 & 2299 $\pm$  78 & 2289 $\pm$  68 & 2237 $\pm$  59 & 2310 $\pm$ 61 & 2415 $\pm$  62 &  2597 $\pm$ 83\\
527: 0.275     & 1765 $\pm$  71 & 1739 $\pm$  69 & 1723 $\pm$  53 & 1738 $\pm$  48 & 1712 $\pm$  44 & 1721 $\pm$ 43 & 1741 $\pm$  43 &  1768 $\pm$ 52\\
528: 0.325     & 1291 $\pm$  45 & 1286 $\pm$  36 & 1299 $\pm$  40 & 1307 $\pm$  40 & 1268 $\pm$  34 & 1298 $\pm$ 31 & 1312 $\pm$  34 &  1276 $\pm$ 36\\
529: 0.375     & 1040 $\pm$  36 & 1045 $\pm$  30 &  991 $\pm$  31 & 1047 $\pm$  29 &  990 $\pm$  26 &  990 $\pm$ 24 & 1002 $\pm$  31 &   964 $\pm$ 26\\
530: 0.425     &  798 $\pm$  35 &  839 $\pm$  31 &  784 $\pm$  29 &  842 $\pm$  24 &  749 $\pm$  22 &  714 $\pm$ 21 &  777 $\pm$  19 &   737 $\pm$ 18\\
531: 0.475     &  660 $\pm$  20 &  657 $\pm$  20 &  613 $\pm$  21 &  677 $\pm$  21 &  594 $\pm$  18 &  589 $\pm$ 16 &  605 $\pm$  15 &   574 $\pm$ 15\\
532: 0.525     &  500 $\pm$  18 &  494 $\pm$  14 &  484 $\pm$  15 &  517 $\pm$  16 &  457 $\pm$  16 &  454 $\pm$ 22 &  457 $\pm$  14 &   452 $\pm$ 37\\
533: 0.575     &  402 $\pm$  12 &  364 $\pm$  13 &  389 $\pm$  14 &  404 $\pm$  13 &  357 $\pm$  22 &  368 $\pm$ 13 &  331 $\pm$  14 &   328 $\pm$ 24\\
534: 0.625     &  299 $\pm$  10 &  299 $\pm$  10 &  315 $\pm$  10 &  299 $\pm$  10 &  273 $\pm$  11 &  294 $\pm$ 11 &  262 $\pm$  14 &               \\
535: 0.675     &  229 $\pm$   8 &  227 $\pm$   9 &  231 $\pm$   9 &  231 $\pm$  10 &  221 $\pm$  11 &  224 $\pm$ 10 &                &               \\
536: 0.725     &  174 $\pm$   8 &  171 $\pm$  16 &  174 $\pm$  22 &  163 $\pm$   7 &  166 $\pm$   7 &  179 $\pm$  9 &                &               \\
537: 
538: \end{tabular}
539: \caption{\piminus\ $d^2N/(\pt d\pt dy)$ for the 5\% most central
540: events.}
541: \label{tab:piPtRapidity}
542: \end{sidewaystable}
543: