1: %%\documentclass[12pt,manuscript,draft]{aastex}
2:
3: %% manuscript produces a one-column, double-spaced document:
4:
5: %%\documentclass[manuscript]{aastex}
6: \documentclass[12pt,preprint]{aastex}
7:
8:
9: %% preprint2 produces a double-column, single-spaced document:
10: %%\documentclass[apj]{emulateapj}
11: %%\usepackage{apjfonts}
12: %\input psfig.sty
13: %\usepackage[light]{draftcopy}
14:
15: %\usepackage{epsfig}
16: %% \documentclass[preprint2]{aastex}
17:
18: \shorttitle{Hercules Thick Disk Cloud}
19: \shortauthors{Larsen et al.}
20:
21: \begin{document}
22:
23: \title{Mapping the Asymmetric Thick Disk: The Hercules Thick Disk Cloud}
24:
25: \author{Jeffrey A. Larsen}
26: \affil{Physics Department, United States Naval Academy,
27: Annapolis, MD 21402}
28: \email{larsen@usna.edu}
29:
30: \author{Roberta M. Humphreys}
31: \affil{Astronomy Department, University of Minnesota, Minneapolis MN, 55455}
32: \email{roberta@umn.edu}
33:
34: \and
35:
36: \author{Juan E. Cabanela}
37: \affil{Department of Physics and Astronomy, Minnesota State University Moorhead, Moorhead MN, 56563}
38: \email{cabanela@mnstate.edu}
39:
40: \begin{abstract}
41: The stellar asymmetry of faint thick disk/inner halo stars in the first
42: quadrant ({\it l} = 20 --45$\arcdeg$) first
43: reported by \cite{lar96} and investigated further by \cite{par03,par04} has
44: recently been
45: confirmed by the SDSS \citep{jur08}. Their interpretation of the
46: excess in the star counts as a
47: ringlike structure, however, is not supported by
48: critical complementary data in the fourth quadrant,
49: not covered by the SDSS. We present stellar
50: density maps from the Minnesota Automated Plate
51: Scanner (MAPS) Catalog of the POSS I showing that
52: the overdensity does not extend into the fourth
53: quadrant. The overdensity is most probably not a ring. It
54: could be due to interaction with the disk bar,
55: evidence for a triaxial thick disk,
56: or a merger remnant/stream.
57: We call this feature the Hercules Thick Disk Cloud.
58:
59: \end{abstract}
60:
61: \keywords{Galaxy: structure, Galaxy: kinematics and dynamics}
62:
63: \section{Introduction}
64:
65: \cite{lar96} initially reported a substantial asymmetry of faint blue stars
66: in the first quadrant (Q1)
67: of the inner Galaxy, $l = 20\arcdeg - 45\arcdeg$ compared with
68: complementary fields in the fourth
69: quadrant (Q4) based on star counts from MAPS\footnote{The Minnesota Automated Plate
70: Scanner Catalog of the POSS I is online
71: at: http://aps.umn.edu} \citep{cab03}.
72: \cite{par03} made a more in-depth survey to map the extent
73: of the asymmetry using
74: 40 contiguous fields in each of three regions: Q1 above and
75: below the plane and Q4 above the plane. Q4 below the plane
76: is not covered in the POSS I. They found a
77: 25\% excess in the number of probable thick disk stars in Q1
78: above \it{and }\rm below the plane when compared to the
79: complementary Q4 fields. The region was irregular in shape
80: and covered several hundred square degrees, but with a completeness
81: limit at $\approx$ 18--18.5 mag, the stars showing the excess in Q1 were
82: relatively nearby, $\sim$ 1 -- 2 kpc from the Sun. \cite{par04} also found an
83: associated kinematic signature, a significant lag of ~80 to 90 km/sec in the
84: direction of Galactic rotation for the associated thick disk stars in Q1.
85:
86: The recent release of the SDSS Data Release 5 (DR5) photometry in
87: the direction of the observed asymmetry in
88: Q1 led to the discovery of a feature at much fainter magnitudes,
89: the distant Hercules-Aquila
90: cloud \citep{bel07}
91: and the photometric parallax study by \cite{jur08} confirmed our nearer asymmetry
92: in the inner Galaxy as an overdensity at a galactocentric radius of 6.5 kpc situated
93: 1.5 kpc above the plane.
94: The SDSS survey however is not
95: well designed for a good study of the thick disk inside the Solar orbit. It
96: extends below $b = 30\arcdeg$ in only a few directions in Q1 and has only
97: limited coverage in Q4.
98:
99: We are continuing our program of photometric and spectroscopic
100: observations to map the size and extent of the asymmetry along our line of sight
101: and to determine the degree of spatial and kinematic asymmetry above and below
102: the plane. We are using the SMARTS Consortium CTIO 1-meter Y4KCam and the
103: Steward Observatory 90" Bok telescope 90Prime Mosaic imager to obtain
104: wide-field multi-color CCD imaging to fainter completeness limits than
105: the POSS I. Spectra of the candidate thick disk stars for radial velocities
106: and metallicity estimates have
107: been observed using the Hydra multi-object spectrometer on the
108: CTIO Blanco 4-meter telescopes and the Hectospec on the MMTO 6.5 meter.
109:
110: In this Letter we present a stellar number
111: density map we created from the MAPS POSS I scans to provide a more global
112: reference for our current deeper but more spatially restricted photometric and
113: spectroscopic study of the thick disk in the inner Galaxy. Other
114: works \citep{xu07} have used plate data to supplement gaps in the SDSS
115: at more southern declinations with good success.
116: Our map of the stellar density in Q1 and Q4 covers much of the sky
117: unavailable to SDSS and demonstrates that the nearby asymmetry in Q1 does
118: not represent a ring above the Galactic plane,
119: but instead is a significant substructure or cloud extending over many
120: square degrees in galactic longitude and latitude.
121: We call this feature the Hercules Thick Disk cloud.
122:
123: \section{The Stellar Density Maps}
124:
125: The figures presented in this Letter were originally created to provide a
126: reference frame for the interpretation of our ongoing observations
127: of the thick disk using narrower yet deep multicolor CCD imaging.
128: They were made
129: using the same set of POSS I fields selected by \cite{par03} above the
130: galactic plane
131: and the fields are fully described in that publication. The 80
132: POSS I plates cover 2900 square
133: degrees on the sky and all are complete to fainter than 18th magnitude in the
134: O band (blue) and have colors (O--E) available from the paired observation in
135: the red (E band). Their placement on the sky is shown in Figure~\ref{fig1} for
136: Q1 and for Figure~\ref{fig2} for Q4. The B-V extinction in each field is
137: plotted from \cite{sch98} and is substantially less than $E(B-V) < 0.2$ in
138: all fields. We defer a discussion of the Q1 data below the plane to a
139: later paper because it is not relevant to the present Letter. The MAPS
140: Catalog has superior stellar photometry
141: when compared with most other digitized plate catalogs of the POSS
142: because each
143: plate has its own independent photometric calibration derived from our own
144: CCD observations as described in \cite{lar03} or from photoelectric photometry
145: \citep{las88} and uses an isophotal diameter-to-magnitude relation for the
146: stars.
147:
148: The individual plate catalogs were merged, duplicates in the plate
149: overlap regions
150: removed, and interstellar extinction from \cite{sch98} was applied on
151: a star by star
152: basis using the standard interstellar extinction law. We assume
153: that the bulk of the extinction comes from relatively near the Sun
154: and that our objects
155: of interest are much farther away, so that the extinction is
156: a zero point correction. A global geometric vignetting correction
157: was applied to all of the plates
158: in the MAPS Catalog. However, we found that some plates had larger
159: vignetting problems (probably due to moonlight).
160: As a result we applied an additional radial magnitude
161: correction to the MAPS magnitudes for 5 of the 80 plates to bring
162: the number densities at the plate edges into line with the well behaved and
163: well calibrated plate center. Star-galaxy classification uses a neural
164: network \citep{1992AJ....103..318O,1993PASP..105.1354O} and is described
165: in those papers and in \cite{cab03}. Any uncertainty in object
166: classification is not
167: a significant factor in the creation of these maps.
168:
169: We then selected stars in the magnitude range $14 < O < 18$ (or
170: approximately $13.5 < V < 17.5$) above the completeness limits.
171: Since the plates still have some magnitude zero point differences and cosmic
172: scatter, we used the ``blue ridgeline" at
173: $O-E \approx 1.0$ or $B-V \approx 0.6$ as a fiducial reference
174: to select a sample of stars with blue and intermediate
175: colors \citep[see][Fig.5]{par03}. Stars bluer than the ridgeline
176: are representative of the stars
177: which show the asymmetry \citep{par03}. While the exact location
178: of the blue ridge in O-E color varies somewhat from plate to plate and with
179: galactic longitude and latitude, due to the relative contributions of
180: stars from the halo, disk and
181: thick disk, it provides
182: a strongly identifiable feature on each color-magnitude
183: diagram (see Figure~\ref{fig4}). The color variations between
184: adjacent plates will thus be small compared to the variations across the
185: full 180 degrees of sky covered. Our magnitude and color search limits
186: are
187: illustrated on one of our color magnitude diagrams shown in Figure ~\ref{fig4}.
188:
189: \section{Discussion and Conclusions}
190:
191: After the reduction steps outlined above, we then
192: binned the stars $0.25 \arcdeg \times 0.25 \arcdeg$
193: in {\it l} and {\it b} to create maps of the stellar
194: density distribution of the faint blue
195: and intermediate color stars shown in
196: Figures~\ref{fig5} and ~\ref{fig6}. The figures are color-coded with respect to
197: number density per 0.0625 square degree.
198:
199: \cite{jur08} described the excess
200: in Q1 as due to a ``ringlike"
201: structure because the overdensity
202: appeared to be radially constant and circular in cross
203: section in their Figure 27.
204: The center of the overdensity region
205: ranges from $(X,Y,Z)=(6.5 kpc, -2.2 kpc, 1.5 kpc)$ to
206: $(6.5 kpc, 0.3 kpc, 1.5 kpc)$ and can easily be converted into galactic
207: coordinates. This is shown as the purple line on Figure~\ref{fig5}.
208: If the feature were symmetric about $l=0\arcdeg$ it would project
209: into Q4 to $(6.5 kpc, 2.2 kpc, 1.5 kpc)$. This projection is also shown
210: as a purple line on Figure~\ref{fig6}.
211: Comparison of Figures~\ref{fig5} and ~\ref{fig6}, however show that
212: the density of these stars is not symmetric with
213: respect to the Sun-Center line. There is a clear
214: excess of stars in Q1 over Q4 in the range
215: {\it l} $= 25$ -- $45\arcdeg$ and {\it b} $= 30$ -- $40\arcdeg$.
216: Furthermore, we emphasize that the excess would not have been initially
217: discovered if
218: it had been a symmetric ring since we \citep{lar96} were
219: comparing star counts for
220: complementary fields in Q1 and Q4.
221:
222: Could the ``ringlike" structure be inclined to the plane and therefore
223: not visible in Q4? Most probably not. The full height of the overdensity
224: in Z over an azimuthal distance of 2500 parsecs is only 500 pc (Juric{\c'} et al.).
225: Even for the pathological case of
226: a paper thin inclined distribution in Z the maximum inclination could only be
227: 11 degrees and given its width in X it should have been visible in Figure~\ref{fig6}.
228: Additionally, there is strong evidence from Juri{\'c}
229: et al.'s Figure 27 (left panel) that for X = 7250-7750 parsecs the
230: overdensity is exclusively above $Z = 1500$ pc. Examination of the same figure's
231: middle panel shows that all significant contributors to the overdensity in
232: this same X range have $Y < 1000$ pc. This would be
233: the opposite of what should be happening if the overdensity were falling into
234: the plane in Q4. Finally, \cite{par04} studied the velocities of
235: samples of stars taken from overdensity regions in Q1 compared with a control
236: sample Q4 and found in a somewhat weak result that the Z component of
237: velocity was less negative for Q1 than it was for Q4.
238: If a coherent population of stars were moving together towards the disk from
239: above the plane as they entered Q4, the opposite should be true.
240:
241: The cloud of stars detected by Juri{\'c} et al. is
242: not symmetric about the $l=0\arcdeg$ line and almost certainly is
243: not a ring.
244: This does not change their other possible
245: explanation for the feature, however. Given the broad extent of the cloud
246: (Figures~\ref{fig5} and ~\ref{fig6})
247: together with its apparent small ranges in radial distance
248: from the Sun and its distance above the galactic plane
249: (their Figure 27), the Hercules Thick Disk
250: cloud may be a debris stream consistent with
251: the disk formation scenario described by \cite{abd03}.
252: While the Hercules Thick Disk Cloud is relatively nearby on the sky it
253: is not related to the more distant (10 - 20 kpc) Hercules-Aquila cloud of
254: \citep{bel07}. The northern
255: extent of the Hercules-Aquila cloud (Figure 2 in \citet{bel07}) is confined to
256: galactic latitudes less than $30\arcdeg$ and
257: even in those regions the bulk of the stars are much fainter than our
258: magnitude limit. The contamination of our sample by Hercules-Aquila
259: cloud stars would be less than 5 objects per square degree (0.5 stars/bin)
260: in any case given our relatively bright magnitude limits.
261:
262: Other explanations are still possible such as a triaxial thick disk
263: and an overdensity due to an interaction with the disk bar.
264: Analysis of our CCD photometry
265: and spectroscopy for fainter stars in Q1 and Q4 will be used to address
266: this question.
267:
268: \acknowledgments
269:
270: This work was supported by the National Science Foundation grants
271: AST0507309 to Larsen and AST0507170 to Humphreys.
272: Larsen is pleased
273: to thank Debora Katz and C. Elise Albert for useful conversations.
274:
275: This research has made use of the MAPS Catalog of POSS I supported by the University of Minnesota. The APS databases can be accessed at http://aps.umn.edu/.
276:
277: \begin{thebibliography}{}
278:
279: \bibitem[Abadi et al.(2003)]{abd03} Abadi, M.~G., Navarro,
280: J.~F., Steinmetz, M., \& Eke, V.~R.\ 2003, \apj, 597, 21
281: \bibitem[Belokurov et al.(2007)]{bel07} Belokurov, V., et
282: al.\ 2007, \apjl, 657, L89
283: \bibitem[Cabanela et al.(2003)]{cab03} Cabanela, J.~E., Humphreys, R.~M., Aldering, G., Larsen, J.~A., Odewahn, S.~C., Thurmes, P.~M., \& Cornuelle, C.~S.\ 2003, \pasp, 115, 837
284: \bibitem[Juri{\'c} et al.(2008)]{jur08} Juri{\'c}, M., et
285: al.\ 2008, \apj, 673, 864
286: \bibitem[Larsen \& Humphreys(1996)]{lar96} Larsen, J.~A., \& Humphreys, R.~M.\ 1996, \apjl, 468, L99
287: \bibitem[Larsen \& Humphreys(2003)]{lar03} Larsen, J.~A., \& Humphreys, R.~M.\ 2003, \aj, 125, 1958
288: \bibitem[Lasker et al.(1988)]{las88} Lasker, B.~M., et al.\
289: 1988, \apjs, 68, 1
290: \bibitem[Odewahn et al.(1992)]{1992AJ....103..318O} Odewahn, S.~C.,
291: Stockwell, E.~B., Pennington, R.~L., Humphreys, R.~M.,
292: \& Zumach, W.~A.\ 1992, \aj, 103, 318
293: \bibitem[Odewahn et al.(1993)]{1993PASP..105.1354O} Odewahn, S.~C.,
294: Humphreys, R.~M., Aldering, G., \& Thurmes, P.\ 1993, \pasp, 105, 1354
295: \bibitem[Parker et al.(2003)]{par03} Parker, J.~E., Humphreys, R.~M., \& Larsen, J.~A.\ 2003, \aj, 126, 1346
296: \bibitem[Parker et al.(2004)]{par04} Parker, J.~E., Humphreys, R.~M., \& Beers, T.~C.\ 2004, \aj, 127, 1567
297: \bibitem[Schlegel et al.(1998)]{sch98} Schlegel, D.~J.,
298: Finkbeiner, D.~P., \& Davis, M.\ 1998, \apj, 500, 525
299: \bibitem[Xu et al.(2007)]{xu07} Xu, Y., Deng, L.~C.,
300: \& Hu, J.~Y.\ 2007, \mnras, 379, 1373
301:
302: \end{thebibliography}
303:
304: \clearpage
305:
306: \begin{figure*}
307: \begin{center}
308: \leavevmode
309: % \psfig{file=leftside.ps,width=\textwidth,angle=-90}
310: \epsscale{0.6}
311: \plotone{f1.eps}
312: \caption{The fields from the MAPS Catalog of POSS I used to create the star count areal density image for Q1 ($l = 0\arcdeg - 90\arcdeg$). The centers are further described in \cite{par03} and are superimposed on the extinction plots of \cite{sch98}. While the interstellar extinction is clumpy it is not high enough to influence our primary result.}
313: \label{fig1}
314: \end{center}
315: \end{figure*}
316:
317: \begin{figure*}
318: \begin{center}
319: \leavevmode
320: %% \psfig{file=rightside.ps,width=\textwidth,angle=-90}
321: \epsscale{0.6}
322: \plotone{f2.eps}
323: \caption{The fields from the MAPS Catalog of POSS I used to create the star count areal density image for Q4 ($l = 270\arcdeg - 360\arcdeg$). The centers are further described in \cite{par03} and are superimposed on the extinction plots of \cite{sch98}. While the interstellar extinction is clumpy it is not high enough to influence our primary result.\label{fig2}}
324: \end{center}
325: \end{figure*}
326:
327: \begin{figure*}
328: \begin{center}
329: \leavevmode
330: \epsscale{0.7}
331: \plotone{f3.eps}
332: %% \psfig{file=cmd.ps,width=\textwidth,height=\textheight}
333: \caption{An example Color Magnitude diagram from the MAPS Catalog indicating the region of color and magnitude used to generate these images.\label{fig4}}
334: \end{center}
335: \end{figure*}
336:
337: \begin{figure*}
338: \begin{center}
339: \leavevmode
340: \epsscale{0.4}
341: %% \psfig{file=excess.ps,width=\textwidth,angle=-90}
342: \plotone{f4.eps}
343: \caption{The density image created for the stars in Q1 with the POSS I plate boundaries from Figure ~\ref{fig1} overlayed in green. Notice how there is a distinct overdensity on the order of about 30\% between $l=25\arcdeg - 45\arcdeg$ and $b = 30\arcdeg - 35\arcdeg$ when compared with the corresponding Q4 stars in Figure ~\ref{fig6}. Comparison with Figure ~\ref{fig1} shows that extinction cannot be causing the feature we see. The location of Juri{\c'} et al.'s overdensity is shown by the purple line. \label{fig5}}
344: \end{center}
345: \end{figure*}
346:
347: \begin{figure*}
348: \begin{center}
349: \leavevmode
350: %% \psfig{file=comparison.ps,width=\textwidth,angle=-90}
351: \epsscale{0.4}
352: \plotone{f5.eps}
353: \caption{The density image created for the stars in Q4 with the POSS I plate boundaries from Figure ~\ref{fig1} overlayed in green. There is no enhancement of faint blue star counts between $l=335\arcdeg - 315\arcdeg$ to match the excess shown in Figure ~\ref{fig5}. Comparison with Figure ~\ref{fig2} shows that extinction cannot be hiding this lack of stars. The location of the Q4 projection of Juri{\c'} et al.'s overdensity is indicated by the purple line. \label{fig6}}
354: \end{center}
355: \end{figure*}
356:
357: \end{document}
358: