1: \documentclass[usegraphicx,usenatbib]{mn2e}
2: \usepackage{times}
3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4: \def\mpc{h^{-1}{\rm{Mpc}}}
5: \def\apj {ApJ}
6: \def\apjl {ApJL}
7: \def\apjs {ApJS}
8: \def\aj {AJ}
9: \def\aap {A\&A}
10: \def\mnras {MNRAS}
11: \def\pasp {PASP}
12: \def\nn {astro-ph/}
13: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
14: \begin{document}
15: \title
16: [Alignments of Luminous Red Galaxies]
17: {Alignment between Luminous Red Galaxies and surrounding structures at $z\sim ~0.5$}
18: \author[Donoso et al.]
19: {Emilio Donoso$^{1,2}$\thanks{E-mail: edonoso@speedy.com.ar}, Ana O'Mill $^{3}$ \&
20: Diego G. Lambas$^{3,4}$ \\
21: $^1$Universidad Nacional de San Juan, Av. Ignacio de la Roza 590(O), 5400 San Juan, Argentina\\
22: $^2$Observatorio Astronomico Felix Aguilar, Av. Benavidez 8175(O), 5413 San Juan, Argentina\\
23: $^3$IATE, Observatorio Astronomico de la Universidad Nacional de Cordoba, Laprida 851, 5000 Cordoba, Argentina\\
24: $^4$Consejo de Investigaciones Cient\'{\i}ficas y T\'ecnicas (CONICET),
25: Avenida Rivadavia 1917, C1033AAJ, Buenos Aires, Argentina\\
26: }
27: \date{\today}
28: \pagerange{\pageref{firstpage}--\pageref{lastpage}}
29: \maketitle
30: \label{firstpage}
31: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32: \begin{abstract}
33: We analyse a high redshift sample ($0.4<z<0.5$) of LRG's extracted from the
34: Sloan Digital Sky Survey Data Release 4 and their
35: surrounding structures to explore the presence of alignment
36: effects of these bright galaxies with neighbor objects.
37: In order to avoid projection effects we compute photometric redshifts
38: for galaxies within $3~\mpc$ in projection of LRGs and calculate
39: the relative angle between the LRG major axis and the direction to
40: neighbors within $1000$ km/s.
41: We find a clear signal of alignment between LRG orientations
42: and the distribution of galaxies within $1.5\mpc$.
43: The alignment effects are present only for the red population
44: of tracers, LRG orientation is uncorrelated
45: to the blue population of neighbor galaxies.
46: These results add evidence to the alignment effects
47: between primaries and satellites detected at low redshifts. We conclude that
48: such alignments were already present at $z\sim 0.5$.
49:
50: \end{abstract}
51:
52: \begin{keywords}
53: cosmology: theory - galaxies: formation -
54: galaxies: alignment - galaxies: Large scale distribution
55: \end{keywords}
56:
57: \section{Introduction}
58:
59: The study of galaxy orientations has the potential to provide important information
60: about the formation and evolution of cosmic structures. However, the interpretation of alignments,
61: and particularly the search of its observational evidence, has a long confusing history.
62: Bingelli (1982) introduced the idea of analyzing the distribution of orientations of clusters
63: or their dominant galaxies relative to neighboring clusters or galaxies. Using a sample of well known
64: Abel clusters he found that the position angle of first ranked galaxies strongly correlates
65: with the orientation of the cluster itself, as determined from the spatial distribution of members.
66: This author found that the orientation of a cluster is related to the distribution of
67: neighboring clusters, i.e. clusters separated less than $\sim 30 \mpc$ tend to point to each other.
68: Later, Struble and Peebles (1985) argued against this last effect claiming that it could be biased by
69: systematic errors. In agreement with Bingelli (1982), Argyres et al. (1986) found that galaxy counts are
70: systematically high along the line defined by the major axis of Abell clusters or its dominant member,
71: up to at least $15 \mpc$ in projected distance.
72:
73: Lambas et al. (1988) found that elliptical galaxies in the Uppsala General Catalog of Bright Galaxies
74: showed a significant alignment signal up to $2~\mpc$ with respect to the surrounding galaxies identified
75: by means of the Lick maps of galaxy counts, an effect likely to be associated to anisotropy of large
76: scales transferred to the elliptical galaxy population through mergers events. Since no similar
77: correlation was found for spiral galaxies, this gave rise to the first detection of a
78: morphology-orientation effect. Again, Rhee \& Katgert (1987) and West (1989) found convincing
79: evidence of Bingelli's results, and with a large sample of rich Abell clusters, Lambas et al. (1990)
80: obtained a $30\%$ excess of brightest cluster galaxies pointing to the nearest-neighbor
81: cluster at scales up to $15\mpc$. Also west (1989) presented evidence suggesting
82: that the orientation of groups of galaxies in superclusters is not random, showing
83: a strong correlation with surrounding groups within $\sim (15-30)\mpc$.
84:
85: More recently, such alignment effects have also been found by Yang et al. (2006),
86: using a large sample of $\sim 53.000$ poor galaxy groups extracted from the New York University
87: Value Added Galaxy Catalogue in the redshift range $0.01<z<0.2$. These authors determined that the
88: distribution of satellite is aligned with the major axis of the brightest galaxies in groups.
89: It is interesting to note that this systematic effect is statistically significant only for the red
90: population of satellites and central galaxies.
91:
92: From the theoretical point of view, numerical studies (West et al. 1991; van Haarlem \& van de
93: Weygaert 1993, Splinter et al.
94: 1998; Faltenbacher et al. 2002) have shown that these cluster-cluster and substructure-cluster alignments
95: occur naturally in hierarchical clustering scenarios for structure formation such as the Cold Dark Matter Model;
96: a fact that can be interpreted as the result of correlations of density fluctuations at different
97: scales.
98: Also, if galaxies formed after the collapse of their parent cluster, then the
99: anisotropic initial conditions could be imprinted in member galaxy orientations, producing alignment effects.
100: Both the tidal field of the parent cluster
101: (Barnes \& Efstathiou 1987, Usami \& Fujimoto 1997), or an anisotropic merger scenario,
102: (where interactions occur along the spatial directions corresponding to the primordial large-scale filaments)
103: could explain the strong alignment between dominant galaxies and surrounding structures.
104:
105: As previously discussed, evidence for the tendency of structures to align each other extends
106: smoothly from the richest clusters of galaxies, to small groups and
107: galaxies, over scales from $Kpc$ up to several $Mpc$,
108: providing a test for models of the
109: origin and evolution of structure in the universe. All these works concern
110: low redshifts, typically $z\la 0.1$, and concentrate in rich and/or poor
111: cluster environments. This limitation is not surprising since only a handful
112: of high redshift clusters are known.
113: Therefore, it is of interest to explore their presence in deeper samples.
114:
115: The Luminous Red Galaxies (LRG) sample corresponds mostly to intermediate redshift
116: ($0.2 \la z \la 0.55$) early type galaxies and, consequently, these objects may be ideal
117: targets to test the alignment hypothesis at larger redshifts. A problem one faces
118: with such a project is the lack of redshifts for the neighbor objects in projection
119: that can dilute the alignment signals due mainly to contamination by distant
120: background galaxies. In this paper, we use photometric redshifts calibrated
121: with the publicly Artificial Neural Network code (Collister \& Lahav (2004), ANNz). The galaxy
122: training set used in the code from galaxy sample randomly selected the SDSS DR4
123: (main galaxy sample and LRGs). In this particular data set the $rms$ redshift
124: error in the range $z_s<0.3$ is $\sigma_{rms}=0.03$ (O'Mill et al. 2006, submitted)
125: to estimate the redshifts of objects close in projection to the
126: LRGs and compute the relative angle between the LRG position angle and the
127: radius vector from the LRG position to that of each neighbor galaxy.
128: The data and statistics are briefly described as well as the main results and
129: conclusions.
130:
131: \section{Data}
132: The spectroscopic sample of the LRG used in this paper comprise the
133: highest redshift objects of the total LRG sample. These galaxies were selected on the
134: basis of color and magnitude to yield a sample of luminous intrinsically red
135: galaxies (Eisenstein et al (2001)) that extends fainter than the SDSS galaxy spectroscopic
136: sample.
137:
138: Together, this sample of luminous red galaxies covers an enormous volume of space,
139: about $1h^{-3}$ Gpc$^3$ when completed, and is expected to trace clusters of galaxies
140: at higher distances while providing a fairly homogeneous population of galaxies
141: suitable to study large-scale structure formation and giant elliptical evolution.
142: The details of selection algorithms as well as its tune-up to remove the passive
143: evolution of an old stellar population are explained in Einsenstein (2001).
144:
145: SDSS photometry is accurate to $\sim 2\%$ $rms$ in $g$,$r$,$i$ bands;
146: and $\sim3\%$ $rms$ in $u$ and $z$. In the $r$ band, a $95\%$
147: completeness is achieved at $m_r\leq 22.2$. All magnitudes used in
148: this work are modified Petrosian magnitudes (Blanton et al. (2001)), which measure galaxy
149: fluxes within a circular aperture whose radius is determined by the azimuthally averaged
150: light profile, and therefore measure a constant fraction of the total light independent
151: of the position and distance. Spectroscopy is taken with two multifiber spectrographs
152: that allow 640 spectra to be simultaneously acquired using pre-drilled aluminum plates.
153: At a spectral resolution of 1800, the typical signal-to-noise per pixel is $>4$ at the
154: peek of the $g$ band. Redshift accuracy is $30$ $km/s$. The fourth data release (DR4)
155: covers a footprint area of 6670 and 4783 square degrees of sky for imaging and
156: spectroscopy, respectively. A complete description of the survey is given
157: by York et al. (2000).
158:
159: Since we are interested in searching for alignment effects at the highest redshifts possible,
160: we adopt the redshift range $0.4 <z <0.5$ for the LRG targets. Position angle $\theta$,
161: major axis $a$ and minor axis $b$ of LRGs were derived by the $photo$ reduction
162: pipeline from the $r$ band isophote at 25 magnitudes per square arcsec. Briefly
163: explained, the radius of a particular isophote is measured as a function of angle
164: and then expanded in Fourier series. Coefficients for this expansion are
165: straightforward to calculate and account for the centroid, major and minor axis,
166: and average radius of the isophote in question. After careful examination of LRG
167: images provided by SDSS, we adopted a conservative range of galaxy flattening,
168: $0.35< b/a < 0.85$, for which misidentification, star contamination, and other
169: problems are greatly reduced while keeping at the same time the highest number
170: of objects and the lowest number of round shaped galaxies.
171:
172: We have computed photometric redshifts using the ANNz code for all galaxies in
173: fields centered on each LRG target within
174: $\sim 2.5~\mpc$. We applied k-corrections using the method of Blanton et
175: al (2003), version 4.1.
176:
177: \begin{figure}\includegraphics[width=0.48\textwidth]{fig1.eps}
178: \caption{Petrosian absolute magnitude $M_r$ vs redshift for $\sim25.000$ tracer
179: galaxies with photometrically derived redshifts calculated with the publicity available
180: Artificial Neural Network code (Collister \& Lahav (2004), ANNz). The ten contours enclose
181: $10\% ~- ~100\%$ respectively. The $+$ signs correspond to the LRG sample.}
182: \label{figura1}
183: \end{figure}
184:
185: In Figure \ref{figura1}, we show the distribution of redshifts and Petrosian
186: absolute magnitudes of the spectroscopic LRG target sample, and of $\sim~25.000$
187: tracer galaxies with photometrically derived redshifts. It can be appreciated that
188: LRGs are typically brighter by $\sim$3.5 magnitudes than the most luminous
189: galaxies in the fields.
190:
191:
192: \section{Analysis and Results}
193:
194: To search for alignment signals we calculate the relative angle $\phi$ between the
195: position angle of the central luminous red galaxy and the vector pointing to
196: each tracer galaxy within $1.5\mpc$ and $dv=1000$ km/s; and then count the number
197: of galaxies in angular bins between 0\degr and 90\degr . For the quantification of
198: the strength of these alignments we define the relative fraction distribution $f(\phi)$ as
199: \[
200: f(\phi)=\frac{N(\phi)-\langle N(\phi) \rangle}{\langle N(\phi) \rangle}
201: \]
202: where $N(\phi)$ is the number of galaxies at each angular bin and $\langle N(\phi)
203: \rangle$ its mean value. This allows to effectively measure the fractional excess of
204: tracer galaxy counts as a function of $\phi$ with respect to the mean. With this
205: definition the absence of alignments would be characterized by a flat distribution
206: around the zero mean. On the other hand, an excess of galaxy counts at low values
207: of $\phi$ indicates the presence
208: of an alignment effect.
209:
210: In order to assess the statistical significance of alignments we adopt the
211: following procedure:
212:
213: \begin{table*}
214: \begin{minipage}{170mm}
215: \centering
216: \caption{Definition of samples and alignment statistics}
217: \label{tabla1}
218: \begin{tabular}{@{}lcccccccccccc}
219: \hline
220: Sample & LRG & \multicolumn{2}{c}{Tracer galaxies}\\
221: \cline{3-4}
222: Name & Magnitude & Magnitude & Color & $\sigma$ & $\chi^2$ & $\langle\phi\rangle$ & $P_>(\langle\phi\rangle)$ & $b$ & $P_>(b)$ & $n_{45}$ & $P_>(n_{45})$ & $N_{gal}$\\
223: \hline
224: S & $M_r<-21.5$ & $M_r<-14.0$ & no restriction & 0.26 & 2.94 & 44.5 & 0.05 & 0.02 & 0.081 & 1.019 & 0.174 & 9883\\
225: S0 & $M_r<-21.5$ & $M_r<-14.0$ & $(g-r)>1.0$ & 0.49 & 7.00 & 43.7 & 0.002 & 0.066 & 0.002 & 1.100 & 0.006 & 2834\\
226: S1 & $M_r<-21.5$ & $M_r<-19.8$ & $(g-r)>1.0$ & 0.65 & 10.2 & 42.9 & 0.000 & 0.109 & 0.001 & 1.179 & 0.000 & 1593\\
227: S2 & $M_r<-21.5$ & $M_r>-19.8$ & $(g-r)>1.0$ & 0.73 & 0.14 & 44.7 & 0.36 & 0.011 & 0.400 & 1.008 & 0.462 & 1241\\
228: S3 & $M_r<-21.5$ & $M_r<-19.3$ & $(g-r)<0.2$ & 0.75 & 0.95 & 44.2 & 0.14 & 0.036 & 0.171 & 1.033 & 0.291 & 1210\\
229: S4 & $M_r<-21.5$ & $M_r>-19.3$ & $(g-r)<0.2$ & 1.01 & 0.05 & 45.2 & 0.42 & -0.022 & 0.350 & 0.923 & 0.845 & 656\\
230: S5 & $M_r<-22.8$ & $M_r<-14.0$ & $(g-r)>1.0$ & 0.54 & 6.97 & 43.5 & 0.002 & 0.076 & 0.003 & 1.120 & 0.000 & 2241\\
231: S6 & $M_r>-22.8$ & $M_r<-14.0$ & $(g-r)>1.0$ & 1.06 & 0.42 & 44.3 & 0.26 & 0.031 & 0.293 & 1.031 & 0.365 & 593\\
232: S7 & $M_r<-22.8$ & $M_r<-14.0$ & $(g-r)<0.2$ & 0.69 & 0.21 & 44.6 & 0.29 & 0.008 & 0.380 & 0.959 & 0.769 & 1391\\
233: S8 & $M_r>-22.9$ & $M_r<-14.0$ & $(g-r)<0.2$ & 0.95 & 0.08 & 44.7 & 0.39 & 0.011 & 0.410 & 1.043 & 0.279 & 748\\
234: \hline
235: \end{tabular}
236: \end{minipage}
237: \end{table*}
238:
239: \begin{enumerate}
240: \item We compute the mean and dispersion of the relative angle $\phi$ between the
241: LRG and the tracers. For an isotropic distribution is expected
242: $\langle \phi \rangle \approx 45$\degr while any departure from this value
243: indicates either alignments or anti-alignments.
244: \item We also calculate the distribution of $\delta = \langle \phi \rangle - 45$\degr
245: which should exhibit for the isotropic case a mean value
246: $\langle \delta \rangle \approx 0$ and standard deviation $\sigma = 90 /
247: \sqrt{12N}$, where $N$ is the total number of LBG-galaxy pairs involved (Struble \& Peebles (1985)).
248: The mean and dispersion of these deviations across all galaxies in the sample allows us
249: to construct $\chi ^2 = \delta ^2 / \sigma ^2$, which is another useful
250: measure of the statistical significance of the results.
251: \item We fit the $f(\phi)$ distribution with $f(\phi)=b$ $cos(2\phi)$ and quote $b$
252: values. This coefficient quantifies the anisotropy amplitude
253: ($b>0$ alignment, $b<0$ anti-alignment).
254: \item We calculate the ratio between the number of $\phi$ values lesser (grater)
255: than 45\degr and define $n_{45}= 2 N_{<45}/ N_{>45}$.
256: \end{enumerate}
257: To address the question of how significant are the alignments signals detected we
258: proceed as follows. We calculate the distribution of random occurrence of
259: $\langle \phi \rangle$, $b$, and $n_{45}$ using a Monte-Carlo method
260: assigning random position angles to the central LRGs. By computing
261: the statistics for 1000 random realizations we provide a robust estimate of
262: the reliability of the results independently of a possible non-gaussian behavior. Also,
263: since the galaxies used for each random sample are exactly the same as real
264: data ones, with the same radial dependence, and the same physical properties;
265: we can avoid any bias caused by clustering or other effects, leaving
266: the LRG orientation as the only parameter that varies across the random set.
267: Then, if an observed parameter value is well outside the corresponding
268: distribution of the random realization we assure its reliability since it
269: could hardly been obtained by chance. We compute the
270: probability of obtaining values higher than the observed ones for $\langle \phi\rangle$,
271: $b$ and $n_{45}$, quoted in Table \ref{tabla1} as $P_>(\langle\phi\rangle)$, $P_>(b)$
272: and $P_>(n_{45})$, respectively. Low values of these probabilities indicate a strong
273: confidence of the observed quantities being non-random.
274:
275: \begin{figure}
276: \includegraphics[width=0.45\textwidth]{fig2.eps}
277: \caption{
278: Normalized distribution of the relative angle $\Phi$ between LRGs position
279: angle and the radius vector to neighbors within $rp\la 1.5~\mpc$ and
280: $dv=1000$ km/s. The dashed lines correspond to $M_r<-14$
281: (sample S), the dotted lines to sample S0 ($M_r<-14.0$ and $(g-r)>1$),
282: while the solid lines correspond to $M_r<-19.8$ and $(g-r)>1$ (sample S1). In the
283: case of no alignment we expect $f=0$. It can be appreciated a significant alignment effect
284: between luminous red galaxies and red neighbors (samples S0 and S1). Error bars
285: correspond to Poisson statistical uncertainty.}
286: \label{figura2}
287: \end{figure}
288:
289: \begin{figure*}\includegraphics[width=0.8\textwidth]{fig3.eps}
290: \caption{
291: Alignment signal for sample S1 for different values of relative radial velocity
292: difference $dv$ between the central LRG and the surrounding tracers (left),
293: and for different projected radius $rp$ (right). The panel for $dv$ shows how
294: the relative excess of galaxy counts for $dv=1000$ km/s (solid line), $dv=3000$
295: km/s (dashed line) and $dv=10000$ km/s (dotted line) gradually tending to isotropy.
296: A similar tendency is shown in the right panel for values of $rp=1.5~\mpc$ (solid line),
297: $rp=2.5~\mpc$ (dashed line) and $rp=3~\mpc$ (dotted line).}
298: \label{figura3}
299: \end{figure*}
300:
301: \begin{figure*}\includegraphics[width=0.8\textwidth]{fig4.eps}
302: \caption{
303: Dependence of $\langle\phi\rangle$, $\chi^2$ and $b$ on velocity difference (top row)
304: and projected radius (bottom row) for sample S1. It is clear that for larger separations (in both
305: velocity and projected radius) all parameters exhibit a behavior consistent with a gradual
306: tendency to isotropy.}
307: \label{figura4}
308: \end{figure*}
309:
310: We applied this set of statistical tests to our sample of LRGs and surrounding galaxies,
311: and extracted different subsamples taking into account luminosity and color.
312: We perform this analysis taking into account Yang et al. (2006) results, who found the red population to be
313: more strongly aligned, an effect that is desired to be tested in the present analysis.
314: By considering the tracer galaxy distributions of luminosity, shown in Figure
315: \ref{figura1}, and ($g-r$) colour index, we have constructed samples of luminous (faint)
316: tracers, $M_r < -19.8$ ($M_r > -19.8$); and red (blue) tracers with $(g-r) > 1$
317: ($(g-r) < 0.2$). In addition, we also analyzed complementary samples of bright (faint)
318: LRGs, $M_r < -22.8$ ($M_r > -22.8$); with the same red (blue) population of tracers.
319: All samples are defined in Table \ref{tabla1}, conveniently labeled, and listed along
320: with the obtained parameters of alignment and significance.
321:
322: The results for samples S, S0 and S1 are shown in figure \ref{figura2}. There is a
323: marginal alignment signal for sample S, while sample S0 has a larger and significant
324: alignment amplitude. It is also clear that the most significant alignment signal is found for
325: bright, red tracers (sample S1). This red bright population is strongly aligned as is evident
326: by inspection to table \ref{tabla1}. The magnitude of the effect is of the order
327: of $11\%$ as accounted by the amplitude of the cosine fit ($\langle \phi \rangle =42\degr.9$).
328: Given the strong alignment detected for bright, red tracers, we have also explored for
329: faint red tracers (sample S2) finding no significant alignment signal,
330: as can be appreciated in table \ref{tabla1}. So that, the alignment signal for
331: sample S0 is likely to be due almost entirely to the bright galaxies of sample S1.
332:
333: \begin{figure*}\includegraphics[width=0.7\textwidth]{fig5.eps}
334: \caption{
335: Distribution of statistical parameters $\langle\phi\rangle$, $b$ and $n_{45}$
336: drawn from 1000 samples where the LRG position angle $\theta$ is extracted from
337: homogeneous distribution. Vertical lines indicate the values obtained directly from observed data
338: for the different samples (see Table \ref{tabla1}). The first row corresponds to the
339: bright red population of tracer galaxies around LRGs, showing a significant alignment
340: signal singe by observed values are well outside the random distributions. The second
341: row (sample S5, bright LRGs, red tracers) also show a correlation. The third row is a
342: good example of no alignment effect consistent with isotropy corresponding to blue
343: neighbors surrounding faint LRGs (sample S7).
344: }
345: \label{figura5}
346: \end{figure*}
347:
348: It is natural to expect a dependence between alignment effects and the velocity difference
349: $dv$ between central and surrounding tracer galaxies, so that an
350: additional analysis is shown in figure \ref{figura3}, where the alignment signal and
351: statistical parameters for sample S1 (bright red population) are plotted at several
352: intervals of $dv$ in the range $10^3 - 10^4$ km/s. As expected, the alignment
353: strength declines with increasing $dv$ values as more uncorrelated
354: tracers are included. This tendency is reflected also in the progression of
355: $\langle \phi \rangle$ values towards isotropy at 45\degr and the
356: lower values of $\chi^2$ and alignment amplitude $b$. For $dv<1000$ km/s, the number
357: of objects fall well beyond reasonable limits. Therefore, we adopt this limit value
358: as the optimal one for the purposes of this analysis.
359:
360: In figure \ref{figura4}, we show the dependence of the alignment signal
361: ($\langle \phi \rangle , ~ b,$ and $\chi^2$) of sample S1 on relative velocity
362: $dv$ and projected separation ($r_p$). As it can be appreciated, there is a
363: smooth tendency to isotropy as either $dv$ and $r_p$ increase.
364:
365: In order to explore if the alignment effects comprising the red population extend to other
366: type of galaxies we tested the alignment in two samples with both, bright and faint
367: blue tracers (samples S3 and S4, respectively). As it can be appreciated in table
368: \ref{tabla1}, the results are fairly consistent with isotropy, implying no correlation
369: between LRG orientation and blue late-type neighbors, a fact that is in good
370: agreement with most previous investigations about alignments at low redshift
371: (Yang et al. (2006)).
372:
373: The question whether the luminosity of the central LRG affects the alignment
374: pattern with neighboring galaxies is also worth to be investigated. For this
375: matter, we repeated the analysis for samples of bright (faint) LRGs, with
376: $M_r < -22.8$ ($M_r > -22.8$), and red tracers with $(g-r) > 1$ (samples S5
377: and S6, respectively). This indicates that the most significant alignment signal is
378: obtained for bright LRGs even while considering red tracer galaxies.
379: As the results described above suggest that color is a parameter that determines
380: the presence or lack of correlation with the LRG position angle, we performed the
381: test again, but considering blue tracer $(g-r) < 0.2$, with the same magnitude cut
382: dividing bright and faint central LRGs (samples S7 and S8). The results indicate
383: clearly that these cases are consistent with isotropy.
384: Again, blue galaxy positions do not show correlation with LRG
385: orientations, independently of their luminosity.
386:
387:
388: A visualization of the reliability of the results is given in the
389: panels of figure \ref{figura5} which show the distributions of occurrence of
390: the alignment parameters $\langle \phi \rangle ,~ n_{45}$, and $b$ for the Monte-Carlo
391: realizations with randomly assigned LRG's position angles for samples S1, S5 and S7.
392: It can be appreciated that true value of the parameters for samples S1 and S5 are well beyond
393: the Monte-Carlo simulation results indicating a statistically significant departure
394: from isotropy. On the other hand, in sample S7 the observed values are well within the
395: Monte-Carlo distributions.
396:
397: \section{Conclusions}
398: From the statistical analysis applied to the data we may summarize the
399: following results:
400: \begin{enumerate}
401: \item A study of alignments equivalent to those typically performed at
402: lower redshifts using information spectroscopic Catalogue can
403: also be performed at higher redshifts, using distance estimates
404: obtained via photometric redshift techniques. This allows to probe
405: deeper into the universe, overcoming the natural limitation of wide-area
406: redshift surveys.
407: \item The orientation of objects from the Luminous Red Galaxy sample
408: extracted from the SDSS shows a significant alignment signal with respect
409: to the direction to neighbors within $1.5~\mpc$ involving an excess of $11\%$
410: galaxies respect to an uncorrelated population. This signal stands at over
411: $3\sigma$ level only for the bright red population of tracer galaxies.
412: The probability of obtaining such alignment pattern from 1000 random realizations is
413: lesser than $1$.
414: \item While the above tendency is quite remarkable for red, bright, neighbors,
415: we found no evidence of such an effect for bluer and/or fainter galaxies around LRGs.
416: \end{enumerate}
417: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
418: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
419: \begin{thebibliography}{}
420: \bibitem[Argyres et al.(1986)]{arg}
421: Argyres, P.~C., Groth, E.~J., Peebles, P.~J.~E., \& Struble, M.~F.\ 1986, \aj, 91, 471
422: \bibitem[Barnes \& Efstathiou(1987)]{barnes}
423: Barnes, J., \& Efstathiou, G.\ 1987, \apj, 319, 575
424: \bibitem[Binggeli (1982)]{bing}
425: Binggeli, B.\ 1982, \aap, 107, 338
426: \bibitem[Blanton et al.(2003)]{blanton03}
427: Blanton, M.~R., et al.\ 2003, \aj, 125, 2348
428: \bibitem[Blanton et al.(2001)]{blanton01}
429: Blanton, M.~R., et al.\ 2001, \aj, 121, 2358
430: \bibitem[Collister \& Lahav(2004)]{annz}
431: Collister, A.~A., \& Lahav, O.\ 2004, \pasp, 116, 345
432: \bibitem[Eisenstein et al.(2001)]{lrgs}
433: Eisenstein, D.~J., et al.\ 2001, \aj, 122, 2267
434: \bibitem[Faltenbacher et al.(2002)]{faltenb}
435: Faltenbacher, A., Kerscher, M., Gottloeber,S., \& Mueller, M.\ 2002, \aap, 395, 1
436: \bibitem[Lambas et al.(1988)]{dgl1}
437: Lambas, D.~G., Groth, E.~J., \& Peebles, P.~J.~E.\ 1988, \aj, 95, 996
438: \bibitem[Lambas et al.(1990)]{dgl2}
439: Lambas, D.~G., Nicotra, M., Muriel, H., \& Ruiz, L.\ 1990, \aj, 100, 1006
440: \bibitem[O'Mill, Valotto \& Lambas (2006)]{ana} O'Mill, Valotto \& Lambas G.\ 2006,
441: \mnras,submitted
442: \bibitem[Rhee \& Katgert(1987)]{reek}
443: Rhee, G., \& Katgert, P.\ 1987, \aap, 183, 217
444: \bibitem[Splinter et al.(1997)]{splinter}
445: Splinter, R.J., Melott, A.L., Linn, A.M., Buck, C., \& Tinker, J.\ 1997, \aj, 479, 632
446: \bibitem[Struble \& Peebles(1985)]{strub}
447: Struble, M.~F., \& Peebles, P.~J.~E.\ 1985, \aj, 90, 582
448: \bibitem[Usami \& Fujimoto(1997)]{usami}
449: Usami, M., \& Fujimoto, M.\ 1997, \apj, 487, 489
450: \bibitem[van Haarlem \& van de Weygaert(1993)]{haarlem}
451: van Haarlem, M., \& van de Weygaert, R.\ 1993, \apj, 418, 544
452: \bibitem[West(1989)]{west1}
453: West, M.J.\ 1989, \apj, 344, 535
454: \bibitem[West(1991)]{west2}
455: West, M.J., Villumsen, J.V., Dekel, A.\ 1991, \apj, 369, 287
456: \bibitem[Yang et al.2006]{yang}
457: Yang, X., van den Bosch, F.C, Mo, H.J., Mao, S., Kang, X., Weinmman, S.M.,Jing, Y.P., 2006,\nn
458: 0601040
459: \bibitem[York et al.(2000)]{sdss}
460: York, D.~G., et al.\ 2000, \aj, 120, 1579
461: \end{thebibliography}
462:
463: \label{lastpage}
464: \end{document}
465: