1: \documentclass{emulateapj}
2: %\documentclass[12pt,preprint]{aastex}
3: \bibliographystyle{apj}
4: \newcommand{\HI}{\ion{H}{1}}
5: \newcommand{\HII}{\ion{H}{2}}
6: \slugcomment{Accepted for publication in ApJ}
7: \shorttitle{HI CLOUDS IN THE LOWER HALO}
8: \shortauthors{Ford et al.}
9: \begin{document}
10:
11: \title{\HI\ CLOUDS IN THE LOWER HALO: I. THE GALACTIC ALL-SKY SURVEY
12: PILOT REGION}
13: \author{H. Alyson Ford\altaffilmark{1,2},
14: N. M. McClure-Griffiths\altaffilmark{2},
15: Felix J. Lockman\altaffilmark{3},
16: J. Bailin\altaffilmark{4},
17: M. R. Calabretta\altaffilmark{2},
18: P. M. W. Kalberla\altaffilmark{5},
19: T. Murphy\altaffilmark{6,7},
20: D. J. Pisano\altaffilmark{3}}
21: \altaffiltext{1}{Center for Astrophysics and Supercomputing, Swinburne
22: University of Technology, Mail H39, P.O. Box 218, Hawthorn, Victoria
23: 3122, Australia; alyson@astro.swin.edu.au}
24: \altaffiltext{2}{Australia Telescope National Facility, CSIRO,
25: P.O. Box 76, Epping, NSW 1710, Australia.}
26: \altaffiltext{3}{National Radio Astronomy Observatory, P.O. Box 2,
27: Green Bank, WV 24944, USA.}
28: \altaffiltext{4}{Department of Physics \& Astronomy, McMaster
29: University, 1280 Main Street West, Hamilton, ON, L8S 4M1, Canada.}
30: \altaffiltext{5}{Radioastronomisches Institut der Universit\"at Bonn,
31: Auf dem H\"ugel 71, 53121 Bonn, Germany.}
32: \altaffiltext{6}{School of Physics, University of Sydney, NSW 2006, Australia.}
33: \altaffiltext{7}{School of Information Technologies, University of Sydney, NSW 2006, Australia.}
34:
35: \begin{abstract}
36:
37: We have detected over $400$ \HI\ clouds in the lower halo of the
38: Galaxy within the pilot region of the Galactic All-Sky Survey (GASS),
39: a region of the fourth quadrant that spans $18\degr$ in longitude,
40: $40\degr$ in latitude and is centered on the Galactic equator. These
41: clouds have a median peak brightness temperature of $0.6$~K, a median
42: velocity width of $12.8$~km~s$^{-1}$, and angular sizes
43: $\lesssim 1\degr$. The motion of these clouds is dominated by Galactic
44: rotation with a random cloud-to-cloud velocity dispersion of
45: $18$~km~s$^{-1}$. A sample of clouds likely to be near tangent
46: points was analyzed in detail. These clouds have radii on the order of
47: $30$~pc and a median \HI\ mass of $630 M_{\odot}$. The population has a
48: vertical scale height of $400$~pc and is concentrated in
49: Galactocentric radius, peaking at $R=3.8$~kpc. This confined structure
50: suggests that the clouds are linked to spiral features, while
51: morphological evidence that many clouds are aligned with loops and
52: filaments is suggestive of a relationship with star formation. The
53: clouds might result from supernovae and stellar winds in the form of
54: fragmenting shells and gas that has been pushed into the halo rather
55: than from a galactic fountain.
56: \end{abstract}
57:
58: \keywords{galaxies: structure --- Galaxy: halo --- ISM: clouds ---
59: ISM: structure --- radio lines: ISM }
60:
61: \section{INTRODUCTION}
62:
63: Neutral atomic hydrogen (\HI) is ubiquitous throughout the Galaxy with
64: a wide variety of morphologies and kinematics, and exhibits many complex
65: structures, including worms \citep{1992Koo}, sheets and filaments
66: \citep{1967Heiles,1990Dickey}, shells
67: \citep{1979Heiles,2002McClure-Griffiths}, and clouds
68: \citep{2002Lockman}. The Galactic \HI\ disk extends to Galactocentric
69: radii, $R\geq 30$~kpc and its thickness varies from $\leq 100$~pc
70: inside $R=3.5$ kpc to $\sim 3$ kpc in the outer Galaxy, with a roughly
71: uniform thickness of $230$~pc between $3.5$~kpc and the solar circle
72: (see \citealt{2001Ferriere} and references within). \HI\ is also known
73: to extend far beyond the thin \HI\ disk as a layer into the Galactic
74: halo \citep{1984Lockman}. However, recent high angular resolution
75: observations using the Green Bank Telescope (GBT) have revealed that this
76: layer is not smooth but instead is composed of small \HI\ clouds with
77: sizes on the order of a few tens of parsecs and masses of $50
78: M_\odot$ \citep{2002Lockman}; confusion may limit the detectability of
79: such clouds at low heights, in which case they may not be confined to
80: the halo (see, e.g., \citealt{2006Stil}). These clouds follow
81: Galactic rotation and are discrete clumps of \HI\ that are localized
82: in space and velocity. Although they are sometimes related to larger
83: structures, for example, embedded in filaments, each cloud appears to
84: be a distinct object. The gross properties of the Milky Way's thick
85: \HI\ layer may in fact be a consequence of the statistical properties
86: of these \HI\ clouds and it is possible that thick \HI\ disks in other
87: galaxies, once observed with sufficient resolution, would reveal a
88: similar structure.
89:
90: While the origin of the halo clouds is unknown, one possible
91: explanation is a ``galactic fountain'' model, where hot gas produced
92: by supernovae rises into the halo of the Galaxy, cools and condenses
93: into \HI\ clouds, which then fall back to the plane
94: \citep{1976Shapiro,1980Bregman}. \citet{1990Houck} predicted that,
95: from a lower temperature fountain, gas in Galactic rotation could be
96: formed at heights close to the plane. This scenario is supported by
97: observations of intermediate velocity clouds (IVCs), such as those of
98: cloud g1, whose location, kinematics and abundances match those
99: expected \citep{2008Wakker}. The abundances in the
100: IV Arch \citep{2001RichterA} and LLIV Arch \citep{2001RichterB} also
101: suggest a fountain origin; as their abundances are near solar it
102: is likely that they originated from material enriched from the disk. The
103: location of these IVCs also support this, as they are roughly
104: $1$~kpc from the disk.
105:
106: Another possibility is that the halo clouds originate in environments
107: where supernovae and stellar winds disrupt the surrounding
108: medium. Such events can result in the formation of a bubble, and models
109: suggest that with a large enough energy source it is possible for
110: these bubbles to expand beyond the thickness of the Galactic disk
111: (e.g., \citealt{1988Tomisaka,1990Heiles}). These bubbles are
112: encompassed by an \HI\ shell as a result of either radiative cooling,
113: which then accumulates more \HI\ as the bubble continues to expand, or
114: solely the sweeping up of ambient material, depending on the wind
115: speed \citep{1992KooB}. The shell remains as a single entity until in
116: some cases Rayleigh-Taylor instabilities cause it to fragment
117: \citep{1989MacLow}. Once this fragmentation has occurred, gas that has
118: been shock heated is expelled outwards, mixing material from the disk
119: with that in the halo \citep{1989Norman} and the remaining fragments
120: of the \HI\ shell may be the observed halo clouds
121: \citep{2006McClure-Griffiths}. It is also possible for the hot gas
122: that has been expelled to cool and recombine, as seen in models of
123: \citet{2000deAvillez}, or perhaps energy from increased supernova
124: activity in areas of active star formation has simply pushed disk gas
125: into the halo, forming clumps of \HI.
126:
127: Large samples of halo clouds are required to constrain the properties
128: and distribution of the population and provide insight into
129: their origin. As the \citet{2002Lockman} sample only has $38$ clouds
130: in a small region of the Galaxy, the properties of the
131: population that constitute the \HI\ layer are currently not
132: well-determined. In this paper we present a catalog of over $400$ \HI\
133: clouds in the lower halo of the inner Galaxy, which we have detected
134: in the Galactic All-Sky Survey pilot region, along with an analysis
135: and discussion of their properties and distribution. We begin with an
136: overview of the observations and data in \S\ref{observations} and
137: present the observed properties of all clouds in \S \ref{clouds}. In
138: \S \ref{tangentpoints} we determine the physical properties of a
139: subset of these clouds that can be assumed to be located at tangent
140: points, where the observer's line-of-sight is tangent to circles of
141: constant Galactocentric radius. An analysis of the cloud properties is
142: presented in \S \ref{discussion} and implications of these results are
143: discussed in \S \ref{nature}. We summarize the results and discuss
144: future work in \S\ref{conclusions}.
145:
146: \section{OBSERVATIONS AND DATA}
147: \label{observations}
148:
149: The data presented in this paper are from the Galactic All-Sky Survey
150: (GASS), a fully sampled Galactic \HI\ survey that covers the entire
151: sky south of declination $\delta=0\degr$. GASS data were taken with
152: the 21 cm Multibeam receiver \citep{1996StaveleySmith} at the Parkes
153: Radio Telescope and cover $-400 \le V_{\mathrm{LSR}} \le
154: 450$~km~s$^{-1}$, where $V_{\mathrm{LSR}}$ is the velocity with
155: respect to the local standard of rest. The spectral resolution of the
156: data is $0.8$~km~s$^{-1}$ and the half power beam width of the
157: Multibeam is $\sim 15\arcmin$.
158:
159: GASS observations began in 2005 January and were completed by the end
160: of 2006. Data reduction was performed using the Livedata package,
161: which is part of the ATNF subset of the aips++ distribution. The
162: bandpass correction was performed using an algorithm designed
163: specifically for GASS data and the Doppler correction was
164: applied. Fluxes were calibrated from observations of the standards S6,
165: S8 and S9 \citep{1973Williams}. The reduced data were gridded into a
166: 3D data cube with voxel dimensions of $4'\times 4'\times0.8$ km
167: s$^{-1}$ using the Gridzilla package, which is also part of the ATNF
168: subset of the aips++ distribution, and was based on the gridding
169: algorithm described in \citet{2001Barnes}. The rms spectral noise,
170: $\Delta T_b \sim 60$ mK, was determined by measuring brightness
171: temperature fluctuations in a $\sim 200$ square degree region of the
172: survey that spanned 48 velocity channels free of any obvious
173: emission. The final GASS data release will be corrected for stray
174: radiation according to the procedure described in
175: \citet{2005Kalberla}. Further details on GASS can be found in
176: \citet{2006McClure-Griffiths}, while extensive details will be
177: presented in a future paper.
178:
179: In this paper we present results from the GASS pilot region, a region
180: that was chosen for a preliminary study on halo clouds to develop
181: techniques that will be applied to the entire survey in the
182: future. The GASS pilot region is in the fourth quadrant in the inner
183: Galaxy and spans $325\degr \lesssim l \lesssim 343\degr$, $|b|\lesssim
184: 20\degr$, and $-200 \le V_{\mathrm{LSR}} \le -70$ km~s$^{-1}$
185: (see Figure~\ref{tangptdiagram}). Our focus is on
186: spatially discrete \HI\ features with an angular size $\lesssim
187: 1\degr$. These features are not characteristic of stray radiation, so
188: although these data have not been corrected for stray radiation, it is
189: unlikely that any of the clouds discussed here are a spurious result
190: of this effect.
191:
192: \begin{figure}
193: \plotone{f1.eps}
194: \caption{\label{tangptdiagram} The area of the Galaxy discussed in
195: this paper is bound by the longitude limits of the GASS pilot region
196: (dotted arrows) and by $V_{\mathrm{LSR}}= -70$ km~s$^{-1}$, which,
197: for a flat rotation curve, gives the area enclosed by the solid
198: line. Dotted circles mark Galactocentric radii of $2.47$~kpc and
199: $8.5$~kpc and the locus of tangent points is shown by the dashed
200: curve that connects the Sun and the Galactic center.}
201: \end{figure}
202:
203:
204: \section{HI CLOUDS IN THE GASS PILOT REGION}
205: \label{clouds}
206:
207: \begin{figure}
208: \plotone{f2.eps}
209: \caption{\label{primage} GASS pilot region at
210: $V_{\mathrm{LSR}}=-105$~km~s$^{-1}$. Many \HI\ clouds with angular
211: sizes $\lesssim 1\degr$ are observed,
212: both near the Galactic plane and into the lower halo.
213: The curved lines at the top and
214: bottom are the boundaries of the region searched.}
215: \end{figure}
216:
217: \begin{figure}
218: \plotone{f3.eps}
219: \caption{\label{primage2} \HI\
220: clouds are observed at a variety of velocities, as can be seen in this
221: latitude--velocity diagram at $l=334.60\degr$. The contour represents
222: $T_b=3$~K, which is the brightness temperature above which
223: few clouds can be distinguished due to confusion.}
224: \end{figure}
225:
226: We detect numerous discrete \HI\ features in the lower halo of the
227: Galaxy with angular sizes $\lesssim 1\degr$, similar to the population
228: of halo clouds discovered by \citet{2002Lockman}. Samples of these clouds can be seen in Figures
229: \ref{primage} and \ref{primage2}, where we display a longitude-latitude
230: image at $V_{\mathrm{LSR}}=-105$~km~s$^{-1}$ and a latitude-velocity
231: image at $l=334.60\degr$, respectively.
232: The curved lines at the top and bottom of Figure \ref{primage} represent the
233: boundaries of the region studied in this paper. The contour in Figure
234: \ref{primage2} represents $T_b=3$ K. Where the emission is brighter than this it
235: is difficult to distinguish clouds because of confusion, although we
236: do detect some clouds above this threshold. These figures clearly
237: demonstrate the presence of discrete \HI\ clouds at a variety of
238: longitudes, latitudes and velocities, which are seen both close to the
239: disk and into the halo. Some clouds are extended and some compact, and many
240: appear to be related to diffuse and filamentary structures.
241:
242: \subsection{Search Method and Criteria}
243: \label{pilotregion}
244:
245: We chose the following selection criteria to generate a homogeneously
246: selected catalog of halo clouds:\\
247: 1. The cloud must be within the GASS pilot region, i.e., within
248: $324.7\degr \le l \le 343.1\degr$, $|b|\lesssim 20\degr$ (see Figure
249: \ref{primage} for exact boundaries), and $-200 \le V_{\mathrm{LSR}} \le
250: -70$ km~s$^{-1}$.\\
251: 2. The cloud must span four or more pixels and be clearly visible over
252: three or more channels in the spectra. Most cloud detections were
253: made with $T_b\geq 5\Delta T_{b}$, where $\Delta T_{b}=60$~mK.\\
254: 3. The cloud must be distinguishable from unrelated background
255: emission. It was impossible to separate clouds from this emission at
256: low latitudes in the least negative velocity channels, where the
257: emission is particularly complex.
258:
259: We believe that we have identified all obvious clouds that meet the
260: criteria listed above. However, it should be noted that some clouds
261: appeared to have double-peaked velocity profiles. In these
262: circumstances, each peak was cataloged as an individual cloud because
263: confusion may be important. If so, confusion may have resulted in the
264: merging of profiles of multiple clouds at similar locations but with
265: distinct velocities, resulting in a double-peaked profile. We will
266: discuss such effects in detail in a subsequent publication.
267:
268: Most of the clouds are not isolated and spherical, but are instead
269: often part of nebulous, filamentary structures and/or sitting in a
270: fluctuating diffuse background. As a result, all automated cloud
271: finding algorithms that we tested had difficulty differentiating
272: clouds from neighbouring clouds and from background emission. Clouds
273: were therefore identified and their properties measured by visual
274: inspection of the data cubes.
275:
276: \subsection{Properties of the Entire Cloud Sample}
277: \label{observedcolumns}
278:
279: Using the criteria presented in \S \ref{pilotregion} we measured $403$
280: \HI\ clouds in the GASS pilot region. The observed properties of these
281: clouds are presented in Table \ref{catalog}. We provide a description
282: of each property and how it was determined below, along with sample
283: spectra in Figure \ref{spectra}. An integrated intensity map, which
284: has the summed intensities over the velocity range for a given cloud,
285: was used to aid in the determination of some properties. These maps
286: have had a background subtracted that was the mean flux of unrelated
287: emission in three interactively chosen areas surrounding the
288: cloud. It is apparent from Figure
289: \ref{background} that at the low end of the peak brightness temperature
290: distribution ($\lesssim 0.5$~K) there is significant incompleteness in
291: regions of higher background levels. We are confident in our
292: background subtraction because beyond incompleteness there is no
293: trend between the peak brightness temperature and mean background
294: level.
295:
296: \begin{figure}
297: \plotone{f4.eps}
298: \caption{\label{spectra} Spectra from a random sample of \HI\ clouds
299: from Table \ref{catalog}. Many clouds appear to be
300: sitting on broad spectral wings. Arrows mark the
301: velocity where the brightness temperature peaks (after background
302: subtraction) for each profile.}
303: \end{figure}
304:
305: \begin{figure}
306: \plotone{f5.eps}
307: \caption{\label{background} Peak brightness temperature as a function
308: of the mean background subtracted from the integrated intensity
309: map. In regions of high background levels there is a clear decline in
310: the number of clouds with low peak brightness temperatures.}
311: \end{figure}
312:
313: The columns in Table \ref{catalog} are:
314:
315: Columns 1-2: The Galactic longitude, $l$, and latitude, $b$, of the
316: cloud at the position of the peak brightness temperature.
317:
318: Column 3: The cloud velocity with respect to the local standard of
319: rest, $V_{\mathrm{LSR}}$, in km~s$^{-1}$, measured as the velocity of the
320: cloud's peak brightness temperature after background subtraction. The
321: background level was determined by fitting a line between each edge of
322: the cloud velocity profile where it merges with surrounding emission.
323:
324: Column 4: The cloud's peak brightness temperature after background subtraction,
325: $T_{\mathrm{pk}}$, in K. Most clouds that we measured had $T_b < 3$~K
326: prior to background subtraction, as clouds with $T_b$ greater than
327: this tended to be found only in areas of high confusion.
328:
329: Column 5: $\Delta v$ is the full-width at half-maximum (FWHM) of the
330: velocity profile, determined by inspection after background
331: subtraction, in km~s$^{-1}$.
332:
333: Column 6: The \HI\ column density, $N_{HI}$, at the cloud center is
334: $1.94\times 10^{18}T_{\mathrm{pk}}\Delta v$~cm$^{-2}$ in the optically
335: thin limit, an assumption that is reasonable because the emission is
336: faint. $N_{HI}$ was determined after we subtracted a background from
337: the integrated intensity map. The background level was highly variable
338: from cloud to cloud being mainly dependent on the latitude and velocity of
339: the cloud.
340:
341: Column 7: $\theta_{\mathrm{maj}}$ and $\theta_{\mathrm{min}}$
342: are the maximum and minimum extent of the cloud in arcminutes and were
343: determined by inspection from the integrated intensity map of the
344: cloud. Many of the clouds are unresolved in at least one dimension but
345: we have not deconvolved their angular sizes due to the uncertainties
346: associated with the fluctuating background levels of the integrated
347: intensity maps. These values therefore represent upper limits of the
348: angular extent.
349:
350: Column 8: $M_{HI}d^{-2}$ represents the \HI\ mass of the cloud in
351: units of $M_\odot$ kpc$^{-2}$, where $d$ is the distance to the cloud
352: in kpc. A background was subtracted from the integrated intensity map,
353: thereby leaving only the flux of the cloud, which was then summed.
354: The mass also relies on the assumption that the \HI\ is optically thin.
355:
356: Column 9: Details of prior detections in the literature, if any, are
357: noted.
358:
359: \subsection{Uncertainties in Observed Properties}
360: \label{observeduncertainties}
361:
362: $\delta V_{\mathrm{LSR}}$: The difference between $V_{\mathrm{LSR}}$
363: and the velocity where the profile decreases from the peak by $\sqrt{\Delta
364: T_b^2+\delta T_{\mathrm{pk}}^2}$, assuming the profile can be
365: approximated by a Gaussian, where $\delta T_{\mathrm{pk}}$ is the
366: error on $T_{\mathrm{pk}}$ (see below) and $\Delta T_b$ is the rms
367: noise.
368:
369: $\delta T_{\mathrm{pk}}$: This error is assumed to be
370: $\sqrt{\Delta T_b^2+(\Delta T_b/\sqrt{2})^2}$, where the first term
371: represents the error in $T_{\mathrm{pk}}$ prior to the background
372: subtraction and the second term represents the error of the mean of
373: the two points defining the background.
374:
375: $\delta \Delta v$: Each of the two half maximum points on the profile
376: is defined as the point where the profile, with error $\Delta T_b$,
377: crosses $(1/2)T_{\mathrm{pk}}$, with error $(1/2)\delta
378: T_{\mathrm{pk}}$. We translate these $T_b$ errors into velocity errors
379: using the slope of a Gaussian profile. We also add the channel width
380: in quadrature.
381:
382: $\delta N_{HI}$: The error on the \HI\ column density, $\delta
383: N_{HI}=N_{HI}\sqrt{(\delta T_{\mathrm{pk}}/T_{\mathrm{pk}})^2+(\delta
384: \Delta v/\Delta v)^2}$.
385:
386: $\delta \theta_{\mathrm{maj}}$ and $\delta \theta_{\mathrm{min}}$: The
387: uncertainties due to the interactive nature of the angular extent
388: measurements and the fluctuating background levels of the integrated
389: intensity maps dominate over the nominal statistical error. The
390: uncertainties were therefore calibrated by comparing values obtained
391: during different trials for a randomly selected subset of clouds and
392: are estimated to be approximately $25\%$ of the determined angular
393: extent.
394:
395: $\delta M_{HI}d^{-2}$: The errors introduced by the interactive
396: process of determining the cloud area and subtracting the background
397: dominate the mass uncertainties and are estimated to be $40\%$ of the
398: determined mass, based on the comparison of different trials of
399: randomly selected clouds mentioned above.
400:
401: \subsection{Observed Properties}
402: \label{obsprops}
403:
404: The $V_{\mathrm{LSR}}$ as a function of longitude for all clouds is
405: shown in Figure \ref{lv}, along with a solid line denoting the
406: terminal velocity, $V_{\mathrm{t}}$, which, in the fourth quadrant, is
407: the most negative velocity permitted by Galactic rotation. The clouds
408: are abundant at velocities allowed by Galactic rotation and there is
409: clearly a decline in the number of clouds beyond (more negative than)
410: $V_{\mathrm{t}}$, which demonstrates that the motions of this cloud
411: population are dominated by Galactic rotation. Clouds beyond
412: $V_{\mathrm{t}}$ are said to have ``forbidden velocities'' and the amount
413: their $V_{\mathrm{LSR}}$ differs from $V_{\mathrm{t}}$ is the
414: deviation velocity,
415: $V_{\mathrm{dev}}=V_{\mathrm{LSR}}-V_{\mathrm{t}}$, as defined by
416: \citet{1991bWakker}. For $l \leq 339.695$, we used $V_{\mathrm{t}}$
417: determined by \citet{2007McClure-Griffiths} from \HI\ observations
418: from the Southern Galactic Plane Survey
419: \citep{2005McClure-Griffiths}. For all remaining longitudes we used
420: $V_{\mathrm{t}}$ determined by \citet{2006Luna} from the
421: Columbia-Universidad de Chile CO surveys.
422:
423: \begin{figure}
424: \plotone{f6.eps}
425: \caption{\label{lv}$V_{\mathrm{LSR}}$ as a function of longitude for
426: all clouds in Table~\ref{catalog}. Crosses represent a subset of
427: clouds with $V_{\mathrm{dev}} \leq 0.8$ km~s$^{-1}$ (the tangent
428: point sample; \S \ref{tangentpoints}) while circles represent
429: all other GASS pilot region clouds. The solid line is the
430: $V_{\mathrm{t}}$ curve determined by \citet{2007McClure-Griffiths}
431: from \HI\ observations for $l \leq 339.695\degr$ and by
432: \citet{2006Luna} from CO observations for all remaining
433: longitudes. As we have searched for all clouds within $-200 \leq
434: V_{\mathrm{LSR}} \leq -70$ km~s$^{-1}$, the decline in the number of
435: clouds at velocities more negative than the terminal velocity
436: demonstrates that the kinematics of the clouds are dominated by
437: Galactic rotation.}
438: \end{figure}
439:
440: In Figure \ref{bvdev} we display the latitude distribution with
441: deviation velocity. For clouds beyond the terminal velocity (with
442: negative deviation velocities), those at lower latitudes
443: have larger absolute deviation velocities. This is likely an artefact,
444: as the number of clouds would naturally
445: fall off with both more negative deviation velocities and larger
446: latitudes if they were dominated by Galactic rotation. There are, however, some outliers at large positive
447: latitudes with very negative deviation velocities.
448:
449: \begin{figure}
450: \plotone{f7.eps}
451: \caption{\label{bvdev}Latitude distribution with deviation velocity,
452: $V_{\mathrm{dev}}$, for all clouds in Table \ref{catalog}, where
453: $V_{\mathrm{dev}}=V_{\mathrm{LSR}}-V_{\mathrm{t}}$, as defined by
454: \citet{1991bWakker}. We used $V_{\mathrm{t}}$ determined by
455: \citet{2007McClure-Griffiths} from \HI\ observations for $l \leq
456: 339.695\degr$ and by \citet{2006Luna} from CO observations for all
457: remaining longitudes. Crosses denote a subset of clouds with
458: $V_{\mathrm{dev}}\leq 0.8$~km~s$^{-1}$ (the tangent
459: point sample, \S\ref{tangentpoints}), while circles denote all other
460: clouds in the GASS pilot region. For clouds that are observed at
461: forbidden velocities ($V_{\mathrm{dev}}<0$~km~s$^{-1}$, denoted by
462: the dashed line), those further from the plane have smaller absolute
463: deviation velocities. This trend is expected of a population of
464: clouds whose kinematics are dominated by Galactic rotation, as the
465: number of clouds would fall off at more negative velocities and
466: larger latitudes, but there are some outliers at large positive
467: latitudes with more negative deviation velocities.}
468: \end{figure}
469:
470: Histograms of $T_{\mathrm{pk}}$, $\Delta v$ and angular size are
471: presented in Figures \ref{pTbhist}a--c, respectively. The majority of
472: clouds have $T_{\mathrm{pk}}\leq 1$~K, with the median
473: $T_{\mathrm{pk}}=0.6$~K. The number of clouds decreases sharply below
474: $T_{\mathrm{pk}}\simeq 5\Delta T_b$ and suggests that the sample is
475: sensitivity limited. The median FWHM of the
476: velocity profiles is $12.8$~km~s$^{-1}$ and very few clouds have
477: linewidths larger than $30$~km~s$^{-1}$ (Figure \ref{fwhmhist}b). All
478: but one of the linewidths are greater than $3.4$~km~s$^{-1}$; as the
479: velocity resolution of the survey is $0.8$~km~s$^{-1}$, most lines are
480: therefore well resolved. The median angular diameter of the clouds is $29'$,
481: which is approximately twice the beam size. This value and
482: the steep cutoff at small angular sizes seen in Figure
483: \ref{angularsizehist}c are most likely due to the spatial resolution limit
484: of the data and suggest that many of the clouds are unresolved.
485:
486: \begin{figure*}
487: \includegraphics[scale=0.36]{f8a.eps}\qquad\includegraphics[scale=0.36]{f8b.eps}\qquad\includegraphics[scale=0.36]{f8c.eps}
488: \caption{\label{pTbhist}(a) Histogram of the peak brightness
489: temperature of the clouds, $T_{\mathrm{pk}}$. The median is
490: $T_{\mathrm{pk}}= 0.6$~K and the cutoff at small $T_{\mathrm{pk}}$
491: is likely due to the sensitivity limit. The arrow represents the
492: $5\Delta T_b$ detection level. We assume $\sqrt{N}$
493: errors. (b)\label{fwhmhist} Histogram of the FWHM of the velocity
494: profile of the clouds, where the median is $12.8$~km~s$^{-1}$. As
495: the spectral resolution is $0.8$~km~s$^{-1}$ and all but one
496: linewidth is greater than $3.4$~km~s$^{-1}$, the linewidths are well
497: resolved. (c)\label{angularsizehist} Histogram of the angular size
498: of clouds,
499: $\sqrt{\theta_{\mathrm{maj}}\theta_{\mathrm{min}}}$, where
500: $\theta_{\mathrm{maj}}$ and $\theta_{\mathrm{min}}$ are from Table
501: \ref{catalog}. The median angular diameter of the clouds is $29'$,
502: which is approximately twice the beam size. This median, along with
503: the apparent lack of clouds with small angular sizes, is most likely
504: due to the spatial resolution limit ($15'$, denoted by an arrow) and
505: suggests that many of the clouds are unresolved.}
506: \end{figure*}
507:
508: \section{Selection of a set of ``Tangent Point'' Clouds}
509: \label{tangentpoints}
510:
511: The largest magnitude velocity from Galactic rotation in the inner
512: Galaxy occurs at the tangent point,
513: defined as the location where the line-of-sight is
514: perpendicular to a circle of constant Galactocentric radius (see Figure
515: \ref{tangptdiagram}). Here $R_{\mathrm{t}}= R_0|\sin{l}|$ and the LSR
516: velocity from Galactic rotation is
517: $V_{\mathrm{t}}=R_0\left[\Theta/R_{\mathrm{t}}-\Theta_0/R_0\right]\sin{l}\cos{b}\mbox{,}$
518: where $R_0$ is the radius of the solar circle, $\Theta$ is the
519: circular velocity and $\Theta_0$ is the circular velocity at the solar
520: circle. We adopt $R_0\equiv 8.5$~kpc and $\Theta_0\equiv
521: 220$~km~s$^{-1}$, as recommended by the IAU \citep{1986Kerr}. Clouds
522: in pure Galactic rotation cannot have a circular velocity beyond
523: $V_{\mathrm{t}}$. However, the random motion of a cloud near the
524: tangent point might increase the cloud's $|V_{\mathrm{LSR}}|$ beyond
525: $|V_{\mathrm{t}}|$. Clouds in
526: Galactic rotation with $|V_{\mathrm{LSR}}|\geq|V_{\mathrm{t}}|$ must
527: therefore lie near the tangent point and thus at a known distance,
528: $d_{\mathrm{t}}=R_0\cos{l}/\cos{b}$. While an assumption, it is
529: reasonable to adopt a distance of $d_{\mathrm{t}}$ for clouds with
530: $|V_{\mathrm{LSR}}|\geq|V_{\mathrm{t}}|$ given the rapid decline
531: in the number of clouds beyond the terminal velocity, as shown in
532: Figure \ref{lv}. Tangent point clouds constitute a sample uniquely
533: suited for investigating the population's distribution and properties,
534: such as physical size and mass.
535:
536: From the population of clouds detected in the GASS pilot region, we
537: define the tangent point sample as all clouds with
538: $V_{\mathrm{LSR}}\leq V_{\mathrm{t}}+0.8$~km~s$^{-1}$, where
539: $0.8$~km~s$^{-1}$ is one channel width, and assume that they are at
540: the tangent point; we assess the effect of this assumption in \S
541: \ref{disterrs}. We also assume that Galactic rotation is constant with
542: distance from the plane, i.e., $\Theta(z)=\Theta(0)$; deviations from
543: cylindrical rotation will be discussed in a subsequent work. As the
544: tangent point clouds in the GASS pilot region are nearly all at the
545: same distance and the latitude boundary of the region corresponds to a
546: constant height of $2.5$~kpc (at tangent points), they provide a
547: uniformly selected sample.
548:
549: \subsection{Derived Properties}
550: \label{derivedprops}
551:
552: The physical properties and positions of the tangent point clouds are
553: presented in Table~\ref{catalogtangent}, while descriptions of the
554: derived quantities are presented below.
555:
556: Columns 1-3: As in Table \ref{catalog}.
557:
558: Column 4: The deviation velocity,
559: $V_{\mathrm{dev}}=V_{\mathrm{LSR}}-V_{\mathrm{t}}$
560: \citep{1991bWakker}, where $V_{\mathrm{t}}$ is the most negative
561: velocity expected from Galactic rotation in the fourth quadrant, in
562: km~s$^{-1}$, and was determined by \citet{2007McClure-Griffiths}
563: from \HI\ observations for $l \leq 339.695\degr$ and by
564: \citet{2006Luna} from CO observations for all remaining longitudes.
565:
566: Column 5: The distance, $d$, along the line-of-sight from the
567: Sun to the cloud determined by assuming the cloud is at the tangent
568: point: $d=R_0\cos{l}/\cos{b}$, in kpc.
569:
570: Column 6: The Galactocentric radius, $R=R_0|\sin{l}|$, of the tangent
571: point at the cloud's location, in kpc.
572:
573: Column 7: The height, $z$, of the cloud from the plane of the Galaxy,
574: determined geometrically to be $z=d\sin{b}$, in kpc.
575:
576: Column 8: The radius, $r$, of the cloud in pc, determined by
577: $\sqrt{r_{\mathrm{maj}}r_{\mathrm{min}}}$, where
578: $r_{\mathrm{maj}}=(1/2)d\theta_{\mathrm{maj}}$,
579: $r_{\mathrm{min}}=(1/2)d\theta_{\mathrm{min}}$, and
580: $\theta_{\mathrm{maj}}$ and $\theta_{\mathrm{min}}$ are from
581: Table~\ref{catalog}.
582:
583: Column 9: The physical mass of \HI\ in the cloud, $M_{HI}$, determined
584: as in Column 8 of \S \ref{observedcolumns} but with the tangent point
585: distance assumed, in $M_\odot$.
586:
587: \subsection{Uncertainties in Derived Properties}
588: \label{deriveduncertainties}
589:
590: Here we present error estimates for the derived properties in Table
591: \ref{catalogtangent}.
592:
593: $\delta V_{\mathrm{LSR}}$: As in Table \ref{catalog}.
594:
595: $\delta V_{\mathrm{dev}}$: This error is $\sqrt{{\delta
596: V_{\mathrm{LSR}}}^2+{\delta V_{\mathrm{t}}}^2}$, where $\delta
597: V_{\mathrm{t}}=3$~km~s$^{-1}$ for clouds located at longitudes where
598: $V_{\mathrm{t}}$ was determined using \HI\ ($l \leq 339.695\degr$;
599: \citealt{2007McClure-Griffiths}). For all other longitudes, where
600: the terminal velocity was determined using CO observations
601: \citep{2006Luna}, the error is assumed to be $9$~km~s$^{-1}$, as
602: suggested from the scatter on Figure 7 of
603: \citet{2007McClure-Griffiths}.
604:
605: $\delta d$: Distance errors, which are inherent to the assumption that
606: the clouds are located at tangent points, were estimated using a
607: simulated population of clouds (see \S \ref{disterrs}).
608:
609: $\delta R$: The error on the Galactocentric radius was determined
610: analogously to $\delta d$. Because the closest point to the Galactic
611: center along a given line-of-sight is the tangent point, the adopted
612: $R$ is always a lower limit and the error can only be positive.
613:
614: $\delta z$: The error on the height is estimated to be $\delta d|\sin{b}|$.
615:
616: $\delta r$: The error on the radius of the cloud depends on the
617: uncertainties in the distance and angular size estimates and is
618: $\delta r=r\sqrt{{(\delta \theta/\theta)}^2+{(\delta d/d)}^2}$, where
619: $\theta=\sqrt{\theta_{\mathrm{maj}}\theta_{\mathrm{min}}}$ \\
620: and $\delta \theta=(1/2)\sqrt{{\delta \theta_{\mathrm{min}}}^2\left(\theta_{\mathrm{maj}}/\theta_{\mathrm{min}}\right)+{\delta \theta_{\mathrm{maj}}}^2\left(\theta_{\mathrm{min}}/\theta_{\mathrm{maj}}\right)}$.
621:
622: $\delta M_{HI}$: The error on the mass is
623: $M_{HI}\sqrt{(\delta M_{HI}d^{-2}/M_{HI}d^{-2})^2+(2\delta
624: d/d)^2)}$, where $M_{HI}d^{-2}$ is given in Table \ref{catalog}.
625:
626: \section{ANALYSIS OF THE TANGENT POINT CLOUD POPULATION}
627: \label{discussion}
628:
629: \subsection{Simulated Halo Cloud Population}
630: \label{simulations}
631:
632: To constrain the spatial and kinematic properties of the observed
633: tangent point cloud population we simulated a population of
634: clouds to which were applied the same $l$, $b$ and $V_{\mathrm{LSR}}$
635: selection criteria as for the GASS pilot region clouds.
636: The simulated clouds were randomly sampled from the following
637: distribution:
638: \begin{equation}
639: n(R,z) = \Sigma(R)\exp\left[{-\frac{|z|}{h}}\right],
640: \end{equation}
641: where $\Sigma(R)$ is the radial surface density distribution, $h$ is
642: the exponential scale height and $R$ and $z$ are the cylindrical
643: coordinates. $\Sigma(R)$ is composed of $12$ independent radial bins
644: of width $0.25$~kpc, spanning $R=2.5$ to $5.5$~kpc. The amplitude
645: of each bin was optimized to best fit the observed longitude
646: distribution of the tangent point clouds by minimizing the Kolmogorov-Smirnov (K-S) $D$ statistic
647: (the maximum deviation between the cumulative distributions) using
648: Powell's algorithm \citep{1992Press}. We optimized the fits using
649: three different initial estimates on $\Sigma(R)$. All converged on a
650: similar solution and we adopted the mean of the three as the best fit
651: to the data, which is shown in Figure \ref{Rhist}.
652:
653: \begin{figure}
654: \plotone{f9.eps}
655: \caption{ \label{Rhist} Best estimate of the radial surface density,
656: $\Sigma(R)$, of clouds in the GASS pilot region. Error bars denote the
657: range in amplitude within each bin for the three radial solutions.
658: Clouds were detected at all longitudes within the GASS pilot region, where
659: corresponding tangent point boundaries in $R$ are indicated by arrows. The
660: distribution suggests that the clouds are concentrated in radius,
661: peaking at $R=3.8$~kpc.}
662: \end{figure}
663:
664: Velocities of simulated clouds were based on a
665: flat rotation curve where $\Theta=\Theta_0$ with a random velocity
666: component drawn from a Gaussian of dispersion $\sigma_{cc}=18$~km~s$^{-1}$, which is
667: discussed in detail in \S\ref{veldispersion}. We generated
668: $5\times 10^4$ clouds in a half-galaxy (third and fourth quadrants),
669: of which $2475$ were within
670: the Galactic coordinates of the GASS pilot region and also within the
671: defined velocity range of the tangent point cloud sample. We then normalized
672: this distribution to compare directly with the observed
673: distribution. We performed K-S tests to estimate
674: the quality of the fit between the observed and simulated
675: distributions. Based on these tests, we find that the parameters of
676: the simulated population reproduce the distribution of observed
677: tangent point clouds well and are therefore good estimates for those
678: of the intrinsic population. Results of the fits to the
679: distributions are presented in \S \ref{veldispersion}--\ref{verthist}
680: along with tests of other functional forms.
681:
682: \subsection{Cloud-to-Cloud Velocity Dispersion}
683: \label{veldispersion}
684:
685: We assume that a random cloud-to-cloud velocity dispersion
686: ($\sigma_{cc}$) is responsible for the presence of clouds at forbidden
687: velocities within the GASS pilot region. Simulations of the global
688: cloud population were required to model the effects of these motions,
689: which can cause clouds that are not located at tangent points
690: to have velocities that are near or beyond $V_{\mathrm{t}}$. The
691: cloud-to-cloud velocity dispersion can provide information on the
692: expected scale height of the distribution and a better understanding
693: of the formation mechanisms of the clouds. We find the
694: $V_{\mathrm{dev}}$ distribution of the tangent point sample of clouds
695: to be consistent with that derived from a Gaussian distribution of
696: random velocities whose dispersion is $\sigma_{cc}=18$~km~s$^{-1}$, with a K-S test probability greater than $97\%$%
697: \footnote{For reference, a $97\%$ probability is roughly equivalent in
698: confidence to a $0.04\sigma$ detection of a difference, i.e.,
699: no detectable difference in the distributions, $15\%$ corresponds to
700: $1.4\sigma$, and $1\%$ corresponds to $2.6\sigma$.}.
701: These distributions are presented in Figure
702: \ref{simvdev}, where the simulated $V_{\mathrm{dev}}$ distribution is
703: represented by a dashed line and the observed distribution by a solid
704: line. Velocity dispersions of $16$ to $22$~km~s$^{-1}$ also provide
705: acceptable fits (K-S test probabilities greater than or equal to
706: $15\%$), as do fits where the random velocity component is drawn
707: from an exponential distribution rather than a Gaussian distribution
708: (with a K-S test probability of $89\%$ for a scale velocity of
709: $13$~km~s$^{-1}$).
710: The implied kinematics of the cloud population based on this result are
711: discussed in \S \ref{kinematics}.
712:
713:
714: \begin{figure}
715: \plotone{f10.eps}
716: \caption{ \label{simvdev} Distribution of deviation velocities of
717: the observed tangent point and simulated
718: population of clouds, where the simulated clouds are normalized to
719: the same total number of observed tangent point clouds. Observed
720: clouds are represented by a solid line, while simulated clouds are
721: represented by a dashed line. To
722: reduce the effects of small number statistics, simulations were run
723: with $5\times 10^4$ clouds. We assume $\sqrt{N}$ errors.}
724: \end{figure}
725:
726: Estimating the uncertainties on our determined $\sigma_{cc}$ is
727: extremely difficult. We expect these uncertainties to be coupled to
728: errors introduced by the measurements of $V_{\mathrm{t}}$, which are
729: difficult to untangle due to the possibilty that random motions can
730: systematically offset the measured $V_{\mathrm{t}}$. Another possible
731: source of uncertainty is streaming motions associated with spiral
732: features. However, because $V_{\mathrm{t}}$ was determined directly
733: from \HI\ and CO measurements rather than from a fitted rotation
734: curve, streaming motions should already be reflected in the adopted
735: $V_{\mathrm{t}}$.
736:
737: \subsection{Distances Errors}
738: \label{disterrs}
739:
740: The simulations allow us to estimate the error in our assumption that
741: clouds with $V_{\mathrm{dev}}\leq 0.8$~km~s$^{-1}$ are at the tangent
742: point. We have calculated the fractional distance error of the
743: simulated clouds as a function of deviation velocity within a
744: $5$~km~s$^{-1}$ wide $V_{\mathrm{dev}}$ bin (Figure
745: \ref{disterrors}). Clouds at increasingly forbidden velocities have
746: smaller distance errors, i.e., are more likely to be near the tangent
747: point; the degree to which this is true depends on the magnitude of
748: $\sigma_{cc}$. We also calculated the fractional distance error as a
749: function of longitude and found that clouds with longitudes
750: corresponding to the Galactocentric radii where few clouds are
751: detected ($l \sim 328\degr$) have larger distance errors (this is
752: expected because few clouds are intrinsically at these radii and
753: therefore a larger fraction of the forbidden velocity clouds at these
754: longitudes are likely interlopers from larger radii). We assume that
755: the errors due to $V_{\mathrm{dev}}$ and longitude are
756: independent. The adopted fractional distance error is the product of
757: the fractional distance error due to $V_{\mathrm{dev}}$ (relative to
758: the typical error of $0.12$), due to longitude (relative to the
759: typical error) and the typical error itself. The fractional distance
760: errors due to $V_{\mathrm{dev}}$ and longitude are taken to be the rms
761: error for the simulated clouds within the same bin. We have confirmed
762: that the fractional distance error of the simulated clouds does not
763: significantly depend on their latitude. Based on the relative distance
764: errors, we believe that our assumption that clouds with
765: $V_{\mathrm{dev}} \leq 0.8$~km~s$^{-1}$ are located near their tangent
766: point is reasonable.
767:
768: \begin{figure}
769: \plotone{f11.eps}
770: \caption{\label{disterrors} Relative distance error as a function of
771: deviation velocity (crosses) for a population of simulated
772: clouds. Clouds with a large forbidden velocity (i.e., large negative value
773: of $V_{\mathrm{dev}}$) are more likely to be at the tangent point
774: and thus have smaller errors in their estimated distances.
775: These points have been placed into bins of width
776: $5$~km~s$^{-1}$, denoted by the horizontal bars, and the
777: vertical error bars denote the mean and standard
778: deviation of the distribution within each bin. Approximately one
779: fifth of simulated points are displayed here but the error bars have
780: taken all simulated data into account. The left most error bar is
781: large due to small number statistics.}
782: \end{figure}
783:
784: \subsection{Radial Distribution}
785: \label{radhist}
786:
787: The adopted radial distribution of the tangent point clouds along with
788: that of the simulated population of clouds is shown in Figure
789: \ref{Rerrors}. The apparent offset in the distributions results from
790: the assumption that the forbidden velocity clouds are located at
791: tangent points, which are always at the smallest value of $R$ along the
792: line of sight.
793:
794: Another useful quantity that can be extracted from the simulations is
795: the mean radial surface density distribution of clouds within the GASS
796: pilot region (Figure \ref{Rhist}). Although clouds are observed at all
797: longitudes within the GASS pilot region, the distribution is
798: concentrated in Galactocentric radius and peaks at $R=3.8$~kpc. The
799: error bars at $R\gtrsim 5$~kpc are significantly larger because clouds
800: at these radii must have large random velocities in order to meet the
801: sample criteria and therefore represent a small tail of the
802: population. With a K-S test probability of $95\%$, the simulated
803: longitude distribution fits that of the observed distribution of the
804: tangent point sample well (Figure \ref{lhist}a).
805:
806: The number of clouds
807: in a uniformly distributed population of tangent point clouds should
808: monotonically decrease by a factor of $\sim 2$ between $l=325\degr$
809: and $l=343\degr$, as demonstrated by Equation (A2) in
810: \citet{2006Stil}, where they show that the line-of-sight distance
811: effectively surveyed over forbidden velocities is $\Delta
812: d=\sqrt{8}R_0|\sin{l}|\sqrt{\sigma_{cc}/\Theta_0}\propto|\sin{l}|$. We
813: have overlayed the distribution of a uniform surface density
814: population in Figure \ref{lhist}a and it is clear that it is in
815: stark contrast to the centrally peaked longitude distribution we observe. We
816: therefore conclude that the peaked radial distribution is real. In Figure
817: \ref{lhist}b we present the longitude distributions of the simulated
818: and observed clouds within the entire pilot region, i.e., at all
819: $-200 \le V_{\mathrm{LSR}} \le -70$ km~s$^{-1}$.
820: Even though the vast majority of these clouds were not used to
821: constrain the simulated radial distribution, their longitude
822: distribution is well reproduced. The tangent point sample therefore
823: appears to be a fair representation of the entire GASS pilot region.
824:
825: \begin{figure}
826: \plotone{f12.eps}
827: \caption{\label{Rerrors} Histogram showing the fraction of the sample
828: located at each Galactocentric radius. The solid line denotes the
829: clouds assumed to be located at tangent points and the dashed line
830: denotes the corresponding
831: simulated radial distribution (dashed line). The shift between the
832: simulated and adopted radial distributions results from systematic
833: errors that stem from the assumption that the forbidden velocity
834: clouds are located at tangent points, i.e., at the minimum possible
835: value of $R$. Arrows represent the radial boundaries of tangent
836: points within the GASS pilot region. We assume $\sqrt{N}$ errors.}
837: \end{figure}
838:
839: \begin{figure*}
840: \plottwo{f13a.eps}{f13b.eps}
841: \caption{
842: \label{lhist}(a) Histogram presenting the longitude distribution of
843: the observed tangent point clouds (solid line) and corresponding
844: simulated distribution (dashed line), along with the distribution
845: expected from a uniform
846: population (dotted line). There is a peak in the observed and
847: simulated distribution, which is clearly different from that
848: expected of a uniform population. (b) Histogram presenting the longitudes of all
849: clouds detected in the GASS pilot region (solid line) along with
850: the corresponding simulated population (dashed line). The
851: distributions are similar, suggesting that the tangent point sample
852: represents the pilot region fairly well. We assume $\sqrt{N}$ errors.}
853: \end{figure*}
854:
855: It is worth investigating whether our assumptions regarding the
856: functional form of the velocity distribution affects the inferred
857: radial distribution, i.e., could a distribution with more velocity
858: outliers and a less strongly peaked radial distribution also fit the
859: data? To test this we have performed the same optimization of the
860: radial distribution while using an exponential velocity
861: distribution. We find that the resulting radial profile is identical
862: to within the errors, confirming the robustness of this result.
863:
864: \subsection{Vertical Distribution}
865: \label{verthist}
866:
867: The vertical distribution of both the tangent point and simulated
868: cloud population is presented in Figure \ref{zhist}. Clouds have been detected
869: throughout the entire range of latitudes covered by the GASS pilot
870: region, up to corresponding heights of $|z|=2.5$ kpc. However, there
871: are very few clouds at $|b|\leq 2.5\degr$ because identification of
872: clouds was extremely difficult close to and within the Galactic plane,
873: except in cases where clouds were observed at large forbidden
874: velocities.
875: Because of this incompleteness at low latitudes, we compare
876: the latitude distributions of simulated and observed clouds with
877: $|b|>2\degr$ and find that they are consistent for vertical scale
878: heights between $400-500$~pc, with K-S test probabilities of at least
879: $25\%$. As incompleteness may still be a problem at $|b| \approx
880: 2\degr$, we also compare the height distributions for only clouds at
881: $|b|>3\degr$ and find acceptable fits for scale heights between $300$
882: and $400$~pc, both with probabilities greater than $50\%$. Based on
883: these comparisons we conclude that the population is best represented
884: with a exponential scale height of $h=400$~pc. A $\mathrm{sech}^2(z/z_0)$
885: distribution with $z_0=700$~pc provides an equally good fit the the
886: data, which is not surprising given that the exponential and $\mathrm{sech}^2$
887: distributions differ primarily near $z=0$~pc where we cannot
888: constrain the fit.
889:
890:
891: \begin{figure}
892: \plotone{f14.eps}
893: \caption{ \label{zhist} Histogram of the fraction of the observed (solid
894: line) and simulated (dashed line) clouds as a function of
895: height. The lack of observed clouds at low $|z|$ is a selection
896: effect due to confusion. This effect was taken into account during
897: comparisons with the simulated population by omitting all simulated
898: clouds with $|b|\leq 2\degr$. The exponential scale height of the
899: distribution is $h=400$~pc. We assume $\sqrt{N}$ errors.}
900: \end{figure}
901:
902: Although incomplete, the combination of surface density, mean mass and
903: scale height of the clouds gives a rough estimate of the vertical
904: distribution of \HI\ contributed by the cloud population. The
905: mid-plane \HI\ number density can be estimated by
906: $n(0)=\Sigma(R)\left<M\right>/(2hM_{\mathrm{H}})$, where
907: $\left<M\right>$ is the mean cloud \HI\ mass and $M_{\mathrm{H}}$
908: is the \HI\ atom mass. We find that the clouds are responsible for
909: $\sim 5\%$, by \HI\ number density, of the exponential component of
910: the \HI\ layer in \citet{1990Dickey} and have a very similar scale height.
911:
912: We note that there is an asymmetry in the number of observed clouds
913: above and below the disk at $|z|\leq750$~pc and an excess of clouds
914: at large positive latitudes. Possible explanations for these are discussed
915: in \S \ref{nature}.
916:
917: \subsection{Physical Size and Mass}
918: \label{mass}
919:
920: Cloud radii, $r$, vary from $\sim 15$ pc to $65$ pc, with a median
921: radius of $32$ pc, as can be seen in Figure \ref{radiushist}a. The
922: angular resolution of the telescope sets a lower limit on the observed
923: cloud size. The maximum angular extent of the detected clouds suggests
924: that roughly $80\%$ of the entire sample is unresolved in at least
925: one dimension. Derived radii should therefore be thought of as
926: upper limits. A histogram of the \HI\ mass of the clouds at tangent
927: points is presented in Figure \ref{masshist}b, demonstrating that the
928: clouds range in size from hundreds to thousands of solar masses, with
929: a median \HI\ mass of $630~M_\odot$. These values may be overestimates
930: if confusion is significant.
931:
932: \begin{figure*}
933: \plottwo{f15a.eps}{f15b.eps}
934: \caption{ (a) \label{radiushist} Histogram of the
935: physical radius of clouds at tangent points. The median
936: $r=32$~pc. The cutoff at small radii is a result of the spatial
937: resolution limit. (b) \label{masshist} Histogram of log $M_{HI}$ of
938: clouds at tangent points. The $\sim 5\Delta T_b$ limit is
939: represented by an arrow for a cloud with the median $r=32$~pc. We
940: assume $\sqrt{N}$ errors.}
941: \end{figure*}
942:
943: \subsection{Comparison of GASS Clouds to Lockman Clouds}
944: \label{comparisons}
945:
946: Based on their angular sizes and location in the lower halo, the
947: clouds detected in the GASS pilot region appear to be similar to those
948: observed by \citet{2002Lockman}. We investigate this possibility further by
949: comparing the properties of each distribution summarized in
950: Table~\ref{comptable}. The median $\Delta v$ is strikingly similar in
951: both sets of data, which is not surprising if the clouds
952: are part of the same
953: population. The derived $\sigma_{cc}$ are also in agreement. The
954: median $|z|$ of this sample is smaller than that of
955: \citet{2002Lockman} but this is most likely due to a selection effect,
956: as the areas searched by \citet{2002Lockman} tended to be further from
957: the Galactic plane to avoid areas of confusion. The most obvious
958: differences in the sample stem from the difference in the angular
959: resolutions of each survey: $15\arcmin$ for GASS data versus
960: $9\arcmin$ for the GBT data. This affects $T_{\mathrm{pk}}$, $N_{HI}$, and $r$, where $T_{\mathrm{pk}}$ and $N_{HI}$ would naturally
961: be lower for unresolved clouds, and $r$ would be
962: larger. If confusion is important, $M_{HI}$ would also be larger. \citet{2002Lockman} estimated that $25\%$ of the clouds were
963: unresolved while we have estimated $\sim 80\%$ here. The clouds
964: observed by \citet{2002Lockman} are much less massive, with
965: approximately one third having $M_{HI} \leq 30 M_\odot$, while none of
966: the observed GASS clouds have masses that low. As the only differences
967: in the observed properties are due to differences in the observations,
968: these comparisons reveal that the clouds belong to the same population
969: of clouds as those detected by \citet{2002Lockman}.
970:
971: \subsection{Observed Trends}
972:
973: There does not appear to be any correlation between the height of the
974: clouds and $V_{\mathrm{LSR}}$, $R$, or $r$. However, as the data in
975: Figure \ref{zfwhm} suggest, there may be a trend between $\Delta v$
976: and $|z|$, where clouds near the plane ($|z|\leq 1$~kpc) have a median
977: FWHM of $10$~km~s$^{-1}$ and a large dispersion, where as those at
978: $|z|> 1$~kpc have a median FWHM of $17$~km~s$^{-1}$ and a smaller
979: dispersion. One possible explanation for such a trend could be that
980: the clouds at larger heights belong to a different population of
981: clouds than those at lower heights.
982: Another possibility is that if the clouds are in pressure
983: equilibrium, the trend is reflecting pressure variations throughout the
984: halo. Similar to our results, \citet{2002Lockman} found that
985: $V_{\mathrm{LSR}}$ and $r$ are independent of $z$, and also found
986: evidence that clouds with more narrow linewidths lay closer to the plane.
987:
988: \begin{figure}
989: \plotone{f16.eps}
990: \caption{\label{zfwhm} FWHM of the velocity profile, $\Delta v$, of the tangent
991: point clouds as a function of height. Clouds at larger heights
992: may have larger linewidths.}
993: \end{figure}
994:
995: As one might expect if a population of clouds has a narrow range of
996: densities, the larger the radius of the cloud, the more massive it
997: is. This trend is evident in Figure \ref{mradius} and does not appear
998: to be solely due to the \HI\ mass sensitivity limit of the
999: data. This limit is denoted by the curved line and is based on the
1000: minimum observable \HI\ column density, $N_{HI_\mathrm{min}}=1.94\times
1001: 10^{18}T_{b_\mathrm{min}}\Delta v_{\mathrm{med}}$ in cm$^{-2}$, where
1002: $T_{b_\mathrm{min}}$ is the minimum observable $T_b$ (assumed to be
1003: $5\Delta T_b$) and $\Delta v_{\mathrm{med}}$ is the
1004: observed median $\Delta v$.
1005:
1006: \begin{figure}
1007: \plotone{f17.eps}
1008: \caption{\label{mradius} Derived \HI\ mass of the clouds as a function
1009: of cloud radius (crosses). The curved line
1010: represents the lower \HI\ mass limit and is based on the minimum
1011: observable $N_{HI}$. The minimum brightness temperature is assumed
1012: to be $5\Delta T_{\mathrm{b}}$. The larger clouds are more massive.}
1013: \end{figure}
1014:
1015: Another apparent trend, as seen in Figure \ref{zNHI}, is that the clouds
1016: with higher \HI\ column densities appear to be at lower
1017: heights. This could be due to inclusion of unrelated diffuse emission
1018: with the clouds at lower heights if the
1019: background subtraction was not effective. It could also be explained
1020: by a scenario where each cloud was given the same kinetic energy from
1021: a formation process or via equipartition in the subsequent evolution
1022: of the cloud population. The higher mass clouds, which have higher
1023: $N_{HI}$, would then have preferentially lower velocities and would
1024: not reach heights as large as those reached by clouds with higher
1025: velocities. Figure \ref{mz} provides tentative support for this
1026: hypothesis, revealing that at $|z|\geq 1$~kpc there are very few
1027: clouds with $M_{HI}\geq 10^3M_{\odot}$ (note, however, that at these
1028: heights our statistics are poor). \citet{2002Lockman} did not find a correlation between
1029: the \HI\ column density or mass with height, but this could be due to
1030: the limited vertical range of his data. Similarly, a correlation between the \HI\ mass of a cloud and its
1031: deviation velocity is suggested in Figure \ref{mvdev}; in particular,
1032: the more massive clouds may have lower deviation velocities. If the
1033: more massive clouds have lower random motions, they could have lower
1034: typical deviation velocities. This would be consistent with a scenario
1035: in which clouds evolve to an equipartition of energy or in which each
1036: cloud is given a similar initial kick, for example, from similar
1037: supernovae explosions. This could also explain why all of the most
1038: massive clouds are seen closer to the plane (Figure \ref{mz}): if they
1039: have smaller initial velocities, they would not move as far into the
1040: halo before falling back towards the plane. At this stage, however, we
1041: can not exclude the possibility that at lower heights confusion is
1042: affecting the determined \HI\ column density and mass of the clouds.
1043: We discuss the kinematics of the cloud population further in \S \ref{kinematics}.
1044:
1045: \begin{figure}
1046: \plotone{f18.eps}
1047: \caption{\label{zNHI} \HI\ column density as a function of
1048: height. All clouds with large column densities are located at lower
1049: heights, i.e., closer to the Galactic disk.}
1050: \end{figure}
1051:
1052: \begin{figure}
1053: \plotone{f19.eps}
1054: \caption{\label{mz} \HI\ mass of the clouds as a function of
1055: height (crosses). Horizontal lines represent the median $M_{HI}$
1056: per $18$ clouds and error bars represent the $25^{\mathrm{th}}$ to
1057: $75^{\mathrm{th}}$ percentile range. There is no evidence for a
1058: trend until $|z|\sim 1$~kpc, where there may be a lack of massive
1059: clouds.}
1060: \end{figure}
1061:
1062: \begin{figure}
1063: \plotone{f20.eps}
1064: \caption{\label{mvdev} \HI\ mass of the clouds as a function of
1065: deviation velocity (crosses). Horizontal lines represent the median $M_{HI}$
1066: per $18$ clouds and error bars represent the $25^{\mathrm{th}}$ to
1067: $75^{\mathrm{th}}$ percentile range. The more massive clouds have
1068: lower deviation velocities.}
1069: \end{figure}
1070:
1071: \section{The Origin and Nature of Halo Clouds}
1072: \label{nature}
1073:
1074: \subsection{Kinematics of Halo Cloud Population}
1075: \label{kinematics}
1076:
1077: The velocity dispersion of the cloud population is likely a remnant of
1078: a common formation process, such as violent supernovae explosions. If
1079: the population is in equilibrium, the vertical distribution and
1080: velocity dispersion are linked via the potential. By using the
1081: simulated population of clouds, we found that the vertical
1082: distribution is best fit with an exponential scale height of $400$~pc
1083: (\S\ref{verthist}) and the $V_{\mathrm{dev}}$ distribution is best fit
1084: with a Gaussian dispersion of $\sigma_{cc}=18$~km~s$^{-1}$
1085: (\S\ref{veldispersion}). However, assuming a vertical force using the
1086: mass model of \citet{2007Kalberla} at $R=3.8$~kpc, we derive an
1087: exponential scale height of $90$~pc for an isothermal population of
1088: clouds with $\sigma_{cc}=18$~km~s$^{-1}$. To produce our observed
1089: scale height of $400$~pc within this potential, the cloud-to-cloud
1090: velocity dispersion would have to be $60$~km~s$^{-1}$. The difference
1091: between the $\sigma_{cc}$ required and that observed may be due to the
1092: lack of clouds observable in the disk; if a large number of clouds
1093: within the disk have gone undetected, the scale height could be lower
1094: than $400$ pc, and could therefore be explained by the observed
1095: velocity dispersion. It is also possible that the distribution cannot
1096: be explained by a single component, but this is not obvious in the
1097: current data (we will address this possibility in a subsequent
1098: paper). However, the magnitude of this difference suggests that the
1099: clouds do not belong to an equilibrium population and their heights
1100: must, in part, result from processes that do not increase the velocity
1101: dispersion, such as uniformly expanding \HI\ shells, instead of bursts
1102: of energy from areas of active star formation generating random
1103: ``kicks.'' The clouds could have also originated above the disk, as in
1104: a galactic fountain model.
1105:
1106: Similar clouds have been detected at forbidden velocities within the
1107: Galactic disk using data from the VLA Galactic Plane Survey,
1108: suggesting that the clouds are not restricted to the halo
1109: \citep{2006Stil}. Those cloud diameters are much smaller than the
1110: ones derived here (they have diameters $\sim 10$~pc),
1111: likely due to the higher spatial resolution of the VLA data. Based on
1112: their models, \citet{2006Stil} find that a vertical Gaussian half-width at
1113: half-maximum (HWHM) larger than $1$ kpc (equivalent to an exponential
1114: scale height of $1.2$~kpc) best fits their data, but given that they
1115: only surveyed within $|b|\leq 1.3\degr$, this is likely to be strongly
1116: affected by the lack of coverage at high latitudes, especially because
1117: we observe similar clouds up to $|z|=2.5$ kpc and \citet{2002Lockman}
1118: has observed clouds up to $\sim1.5$ kpc. \citet{2006Stil} find a lower limit
1119: to the HWHM of the clouds to be $180$~pc (exponential scale height of
1120: $216$~pc), which is consistent with the value we derive and also
1121: inconsistent with the value expected for the derived velocity
1122: dispersion.
1123:
1124: It is worth noting that there is evidence that similar clouds may
1125: also be abundant in the outer Galaxy \citep{2006Stanimirovic}. It is
1126: not yet certain whether they belong to the same population of clouds,
1127: but with larger surveys of \HI\ clouds in the lower halo of the
1128: Galaxy, such as those presented here and in the entire inner Galaxy
1129: within GASS, along with those underway by the Galactic Arecibo L-Band
1130: Focal Plane Array Consortium and other groups, properties
1131: such as $\sigma_{cc}$ and the spatial distribution can be determined
1132: more accurately and will therefore help constrain the kinematics and
1133: formation mechanisms of the clouds.
1134:
1135: \subsection{Halo Clouds and Spiral Structure}
1136: \label{spiralarms}
1137:
1138: As discussed in \S \ref{radhist}, the surface density of halo clouds
1139: is not uniformly distributed but instead peaks at a Galactocentric
1140: radius of $R=3.8$ kpc and the population is concentrated in radius. This confined nature
1141: suggests that the halo clouds are related to the spiral structure of the
1142: Galaxy. Although distances to spiral arms are currently not well
1143: constrained, the location of the peaked radial distribution of the
1144: clouds indicates that they may be related to the expanding ``$3$~kpc''
1145: arm \citep{1957vanWoerden,1960Rougoor}, which is tangential to an
1146: observer's line-of-sight at roughly $l=336\degr$
1147: (\citealt{2008Bronfman}; see also \citealt{2008Vallee} who argues that
1148: this feature lies at $l\sim 339\degr$ and is the start of the Perseus
1149: arm, distinct from the $3$~kpc--Norma arm), corresponding to
1150: $R=3.5$~kpc. Also, it has been suggested that the $3$~kpc arm must be
1151: confined to an annulus of less than $1$ kpc in extent
1152: \citep{1981Lockman}, which corresponds well with the radial
1153: concentration of clouds. At this time, however, the
1154: possibility that these clouds are instead related to other Galactic
1155: structures such as the ``$5$~kpc molecular ring'' \citep{2004Jackson}
1156: cannot be ruled out. This ``ring'' is likely not a coherent structure
1157: but instead a complex region where multiple spiral arms originate
1158: \citep{2008Vallee}.
1159:
1160: The apparent association between many of the \HI\ clouds and
1161: filamentary structures within the GASS pilot region suggests that the
1162: clouds are related to star formation because such structures are
1163: common in areas of significant supernova activity or stellar winds
1164: \citep{1990Dickey}. Figure \ref{newshell} displays an integrated
1165: intensity map with many clouds that are clearly aligned along
1166: filaments and loops, reminiscent of the clouds observed to be
1167: associated with the cap of a superbubble by
1168: \citet{2006McClure-Griffiths}. Recently, proper motion measurements of
1169: the molecular cloud NGC 281 West, which is associated with an \HI\
1170: loop, revealed that the cloud is moving away from the Galactic plane at
1171: $20-30$~km~s$^{-1}$ \citep{2007Sato}. This velocity is
1172: similar to our observed velocity dispersion and further supports the
1173: scenario that the clouds are related to expanding shells. In this scenario, violent
1174: supernovae and stellar winds may have pushed \HI\ from the disk up
1175: into the halo or the clouds may be fragments of \HI\ shells
1176: \citep{1989MacLow,2006McClure-Griffiths}, rather than a result of
1177: a standard galactic fountain. In the galactic fountain model hot gas
1178: rises from the disk, cools and condenses, then falls back to the
1179: plane \citep{1976Shapiro,1980Bregman}. This would result in a more
1180: uniform radial distribution of clouds \citep{1980Bregman},
1181: while we clearly observe a peak in the radial distribution of the
1182: clouds that may be associated with the $3$~kpc arm. If the halo
1183: clouds are related to star formation, the asymmetry in the number of
1184: detected clouds at low heights ($|z|\leq 750$~pc) could be a result of
1185: this, as any asymmetry in the structure of the ISM and the location of
1186: star forming regions may be reflected in the distribution of clouds,
1187: and there appears to be more filaments below than above the Galactic
1188: plane in the GASS pilot region.
1189:
1190: \begin{figure}
1191: \plotone{f21.eps}
1192: \caption{\label{newshell} Filamentary structures are plentiful within
1193: the GASS pilot region, as demonstrated in this longitude--latitude
1194: integrated intensity map over $ -135 \leq V_{\mathrm{LSR}}\leq
1195: -120$~km~s$^{-1}$. Many \HI\ clouds in the GASS pilot region are
1196: aligned with these loops and filaments, suggesting that they are
1197: related to expanding superbubbles or other structures common in areas
1198: of star formation. The diagonal striations are instrumental artifacts.}
1199: \end{figure}
1200:
1201: If the clouds are related to spiral structure and star formation then
1202: we would expect to see a correlation between the radial surface
1203: density distribution
1204: of the \HI\ clouds and that of Galactic \HII\ regions. We compare
1205: these distributions, along with the mass surface densities of \HI\ and
1206: $\mathrm{H}_2$, in Figure \ref{H2}. The mass surface densities have been
1207: averaged over the entire Galaxy and were derived by \citet{1993Dame}
1208: using \HI\ data from \citet{1990Dickey} and \citet{1978Burton}, and
1209: $\mathrm{H}_2$ data from \citet{1988Bronfman}, whereas the halo cloud
1210: distribution from this study includes only
1211: the GASS pilot region. The \HII\ regions are taken
1212: from \citet{2004Paladini}. There is no obvious relationship between
1213: the \HI\ cloud and \HII\ region distributions or between the \HI\
1214: clouds and the \HI\ and $\mathrm{H}_2$ surface densities, providing
1215: conflicting evidence for the relationship between the \HI\ clouds and
1216: current star formation. As the evaporation timescale for a cold cloud
1217: in a hot medium is expected to be much longer than other timescales
1218: associated with cloud evolution \citep{1977Cowie,2006Nagashima},
1219: and if evaporation is the main disruptive mechanism, the clouds
1220: are long-lived. It is therefore possible that the clouds are tracing
1221: past rather than current star formation. With the present data we are
1222: only able to make a preliminary study of the relationship between the
1223: halo cloud distribution and spiral structure and star formation in the
1224: Galaxy. Future analysis of clouds over a larger range of longitudes
1225: will allow us to test the assertion that the clouds are related to
1226: star formation in greater detail.
1227:
1228: \begin{figure}
1229: \plotone{f22.eps}
1230: \caption{\label{H2} Radial surface density distribution of the
1231: simulated GASS pilot region \HI\ clouds (solid histogram) and \HII\
1232: regions (scaled up by a factor of $7$ for ease of comparison and
1233: shown by the
1234: dash-dot histogram; \citealt{2004Paladini}). Mass surface densities of the
1235: average Galactic \HI\ (dotted line) and Galactic $\mathrm{H}_2$
1236: (dashed line) are overlaid and were both determined by
1237: \citet{1993Dame} using data from \citet{1990Dickey} and
1238: \citet{1978Burton} for \HI\, and data from \citet{1988Bronfman} for
1239: $\mathrm{H}_2$. Arrows indicate limits of $R$ for tangent points in the GASS pilot region.
1240: The peaked distribution of halo clouds does not appear to be similar
1241: to that of the \HII\ regions nor the \HI\ and $\mathrm{H}_2$
1242: mass surface densities, suggesting that the halo clouds are not
1243: directly related to current star formation.}
1244: \end{figure}
1245:
1246: \subsection{Possible Association with High Velocity Cloud Complex L}
1247: \label{complexLsection}
1248:
1249: An excess of clouds at large positive heights can be seen in the
1250: vertical distribution of the clouds (Figure \ref{zhist}); $9\%$ of the
1251: clouds at positive heights lie at $z\geq 2$~kpc while none of the
1252: clouds below the plane are seen at such heights, and many of them have
1253: unusually large deviation velocities (Figure \ref{bvdev}). This excess
1254: could have several possible origins including infalling gas, increased
1255: disk activity on one side of the disk that has resulted in outflowing
1256: gas reaching larger heights, or small number statistics. Given the
1257: proximity of high velocity cloud complex L to the halo clouds in the
1258: upper portion of the GASS pilot region, we compared the \HI\
1259: associated with each population to determine whether or not the
1260: presence of complex L could be responsible for the observed excess.
1261:
1262: Complex L was first described by \citet{1991Wakker} to have velocities
1263: ranging from $-190 \lesssim V_{\mathrm{LSR}} \lesssim -85$ km
1264: s$^{-1}$, longitudes ranging from $341\degr \leq l \leq 348\degr$ and
1265: latitudes ranging from $31\degr\leq b \leq 41\degr$, and they
1266: speculated that the clouds were part of
1267: a population that was related to a galactic fountain. Since then,
1268: H$\alpha$ distance limits have been determined for complex L, placing
1269: it within the Galactic halo at heights of $4\leq z\leq 12$~kpc from
1270: the plane and heliocentric distances of $8$ to $22$~kpc
1271: \citep{2001Weiner,2003PutmanB}.
1272:
1273: In Figure \ref{complexL} we display a region of the GASS data that
1274: encompasses both the upper GASS pilot region, outlined by the solid
1275: black lines, and the lower velocity gas of complex L. We have overlaid
1276: circles on high velocity clouds associated with complex L as
1277: catalogued by \citet{1991Wakker}, regardless of their velocity.
1278: The spatial morphology of the \HI\ at velocities where gas is detected
1279: in complex L suggests that there is a connection between the clouds in
1280: the GASS pilot region and those in complex L, in the form of a
1281: filamentary structure that is contiguous in velocity and connects
1282: complex L with the disk. Also, with the assumption that
1283: the halo clouds at large positive latitudes are located at tangent
1284: points, their heights are between $1.5\leq z \leq 2.5$ kpc with
1285: distances $\sim 8$~kpc, which place them in the vicinity of the lower
1286: height estimates of complex L. These correlations suggest that complex
1287: L may have similar origins as the clouds presented here and it may
1288: be responsible for the observed excess of clouds at large positive
1289: latitudes.
1290:
1291: \begin{figure}
1292: \plotone{f23.eps}
1293: \caption{\label{complexL} GASS data encompassing the upper portion of
1294: the GASS pilot region, outlined by the solid black lines, along with
1295: the high velocity cloud complex L. Circles have been placed at the
1296: positions of high velocity clouds within complex L, regardless of
1297: their $V_{\mathrm{LSR}}$, as determined by \citet{1991Wakker}. The
1298: morphology of the \HI\ in the upper GASS pilot region, in complex L, and
1299: between them, suggests that complex L may have similar origins as
1300: the observed halo clouds and may be responsible for the observed
1301: excess of clouds at large positive latitudes. }
1302: \end{figure}
1303:
1304: \subsection{Stability of Halo Clouds}
1305: \label{stability}
1306:
1307: The amount of mass required for a spherical cloud to be
1308: gravitationally bound is $M\approx r\Delta v^2/G$, where $r$ is the
1309: radius of the cloud, $\Delta v$ is the linewidth of its velocity
1310: profile and $G$ is the gravitational constant. For a cloud with
1311: $r=32$~pc and $\Delta v=12.8$~km~s$^{-1}$, the median observed values,
1312: this would require a mass on the order of $10^6 M_{\odot}$. As none of
1313: the detected clouds have masses this large, they are either pressure
1314: confined or transitory.
1315:
1316: Because turbulence also contributes to the linewidths, an upper limit
1317: for the thermal pressure within the clouds is $P=nT \leq
1318: 22\left<n\right>\Delta v^2$, in units of K cm$^{-3}$, where
1319: $\left<n\right>$ is the average number density and $T$ is the thermal
1320: temperature. However, as $\left<n\right>$ is a lower limit and $T$ is
1321: an upper limit, and $\left<n\right>$ and $T$ are uncorrelated, we are
1322: unable to put any constraints on the pressure of these clouds with the
1323: present data. It is possible that pressure changes are responsible for
1324: the observed trend in Figure \ref{zfwhm}, which tentatively shows that clouds
1325: further from the disk of the Galaxy have larger linewidths. If the
1326: clouds are in pressure equilibrium, their pressures could provide us
1327: with insight into the pressure structure of the halo, which is
1328: currently not well understood.
1329:
1330: \section{SUMMARY}
1331: \label{conclusions}
1332:
1333: We have detected over $400$ \HI\ clouds in the lower halo of the
1334: Galaxy in the Galactic All-Sky Survey pilot region. These clouds have
1335: a median peak brightness temperature of $0.6$~K, a median velocity
1336: width of $12.8$~km~s$^{-1}$, and angular sizes
1337: $\lesssim 1\degr$. As these clouds follow Galactic rotation, a
1338: subset was selected that is likely to be located at tangent points of the inner
1339: Galaxy, allowing us to determine their distances and therefore their
1340: sizes and masses. The tangent point clouds have radii on the order of
1341: $30$~pc and a median \HI\ mass of $630 M_{\odot}$. The properties of
1342: these clouds suggest that they belong to the same population of clouds
1343: discovered by \citet{2002Lockman}.
1344:
1345: We simulated the population of clouds to constrain their random
1346: cloud-to-cloud velocity dispersion, $\sigma_{cc}$, and spatial
1347: distribution. We found that $\sigma_{cc}=18$~km~s$^{-1}$, but if the
1348: clouds were left to evolve in the Galactic potential without any
1349: disruptions, these random motions would produce
1350: a scale height of $90$~pc, which is inconsistent with our derived
1351: scale height of $400$~pc. This suggests that the clouds do not belong to
1352: an equilibrium population. We detected clouds throughout the entire
1353: GASS pilot region, up to the latitude boundaries ($|b|\lesssim
1354: 20\degr$). Few clouds were observed at low latitudes due to confusion,
1355: which may have resulted in an underestimate of the number of clouds at
1356: low heights, and therefore an overestimate in the derived scale
1357: height.
1358:
1359: Our large, homogeneously-selected sample has allowed us to determine
1360: the spatial distribution of these halo clouds for the first time and
1361: has revealed that although clouds were observed at all longitudes
1362: within the GASS pilot region, they do not appear to be uniformly
1363: distributed but instead are concentrated in radius, peaking at
1364: $R=3.8$~kpc. We analyzed this distribution
1365: and suggest that the clouds are related to the spiral structure of the Galaxy.
1366: In particular, the peak in the radial distribution is suggestive of a
1367: relation to the $3$~kpc arm. This relation to a specific spiral
1368: feature remains speculative until further analysis using a larger
1369: sample of clouds and better constrained spiral structure models can be
1370: performed. It is therefore unlikely that the halo clouds are a
1371: result of a standard galactic fountain, as radial enhancements
1372: would not be expected in this scenario \citep{1980Bregman}. Instead, it
1373: appears that the clouds may be directly related to areas of active
1374: star formation, in the form of fragmenting \HI\ shells and \HI\ gas
1375: that has been pushed into the halo. This is further supported by the
1376: appearance of numerous clouds related to filaments and loops, whose
1377: structures may have resulted from stellar winds and supernovae
1378: \citep{1990Dickey}. However, a comparison between the radial surface
1379: density distribution
1380: of \HI\ clouds and \HII\ regions provides conflicting evidence: if
1381: clouds are related to areas of star formation, a relationship between
1382: the clouds and \HII\ regions would be expected but is not observed.
1383:
1384: The morphology of \HI\ at large positive latitudes in the GASS pilot
1385: region suggests that some of the clouds may be related to high
1386: velocity cloud complex L, whose lower height estimates of $4$~kpc and
1387: distance estimates of $8$~kpc place it in the vicinity of the clouds
1388: presented here. If they are related, it is possible that whatever
1389: process is responsible for the halo clouds, e.g., star formation, is
1390: also responsible for complex L. This is further supported by recent
1391: observations of
1392: IVCs, which suggest that the IVCs are
1393: related to energetic events in the Galactic disk and that they are
1394: likely linked to spiral structure \citep{2006Kerton}. The majority of
1395: those IVCs have $V_{\mathrm{dev}}\leq 20$~km~s$^{-1}$, implying a
1396: Galactic origin, and we suggest that they are similar to the cloud
1397: population presented here.
1398:
1399: A future study of halo clouds within the entire inner Galaxy observed
1400: by the Galactic All-Sky Survey will provide a much larger sample of
1401: tangent point clouds, enabling more complete statistics for
1402: distribution studies and a better understanding of Galactic
1403: structure. These studies will include comparisons between the
1404: distribution of halo clouds and that of tracers of star formation,
1405: such as \ion{H}{2} and $\mathrm{H}_2$, and will allow us to better
1406: constrain the origin of these clouds.
1407:
1408: \acknowledgements
1409: The Parkes Observatory is operated by the Australia Telescope National
1410: Facility, a division of the Commonwealth Scientific and Industrial
1411: Research Organisation.
1412:
1413: {\it Facilities:} \facility{Parkes ()}
1414:
1415: \bibliography{ms.bib}
1416:
1417: \LongTables
1418:
1419: \begin{deluxetable}{rrrrrrccr}
1420: \tabletypesize{\footnotesize}
1421: \tablewidth{0pt}
1422: \tablecaption{HI Clouds - Observed Properties\label{catalog}}
1423: \tablehead{\colhead{$l$}&\colhead{$b$}&\colhead{$V_{\mathrm{LSR}}$}&\colhead{$T_{\mathrm{pk}}$\tablenotemark{a}}&\colhead{$\Delta v$}&\colhead{$N_{HI}$}&\colhead{$\theta_{\mathrm{min}}\times\theta_{\mathrm{maj}}$\tablenotemark{b}}&\colhead{$M_{HI}d^{-2}$\,\,\,\tablenotemark{c}}&\colhead{Notes}\\
1424: \colhead{(deg)}&\colhead{(deg)}&\colhead{(km~s$^{-1}$)}&\colhead{(K)}&\colhead{(km~s$^{-1}$)}&\colhead{($\times
1425: 10^{19}$~cm$^{-2}$)}&\colhead{(\,\,$\arcmin$\, $\times$\, $\arcmin$\,\,)}&\colhead{($M_{\odot}$ kpc$^{-2}$)}}
1426: \startdata
1427: $324.78$ & $13.39$ & $-91.1 \pm 4.2$ & $0.20$ & $8.8 \pm 3.3$ & $0.34 \pm 0.18$ & $24 \times 38$ & $6.3 $ & \\
1428: $324.87$ & $-8.30$ & $-73.2 \pm 2.2$ & $1.16$ & $12.7 \pm 1.1$ & $2.87 \pm 0.31$ & $17 \times 36$ & $18.6 $ & \\
1429: $325.11$ & $-8.56$ & $-80.5 \pm 6.3$ & $0.34$ & $18.2 \pm 3.9$ & $1.20 \pm 0.37$ & $28 \times 36$ & $24.0 $ & \\
1430: $325.34$ & $-4.61$ & $-94.2 \pm 1.8$ & $0.44$ & $6.1 \pm 1.3$ & $0.52 \pm 0.14$ & $27 \times 29$ & $20.5 $ & \\
1431: $325.38$ & $5.14$ & $-88.4 \pm 2.2$ & $0.78$ & $10.3 \pm 1.2$ & $1.56 \pm 0.24$ & $26 \times 40$ & $29.3 $ & \\
1432: \enddata
1433: \tablecomments{Observed properties of \HI\ clouds in the GASS pilot
1434: region. The complete version of Table \ref{catalog} is available in
1435: the electronic edition of the Astrophysical Journal. We provide this
1436: sample as a guide to its content. Descriptions of each property are
1437: presented in \S \ref{observedcolumns}. Clouds that have been
1438: cataloged elsewhere in the literature are noted by the following labels:\\
1439: 1: detected by \citet{2002Putman} and\\
1440: 2: detected by \citet{1991Wakker}.\\
1441: Although the clouds detected elsewhere do not necessarily have the
1442: exact Galactic coordinates and $V_{\mathrm{LSR}}$ as listed here,
1443: it is likely that they are the same cloud and that the
1444: differences are a result of observational constraints. Also, we note
1445: that \citet{2000Morras} detected \HI\ in some areas of these
1446: clouds but such detections were not identified as individual
1447: objects.}
1448: \tablenotetext{a}{Uncertainties in $T_{\mathrm{pk}}$ are $0.07$~K.}
1449: \tablenotetext{b}{Uncertainties in the maximum angular extents are
1450: dominated by background levels surrounding the cloud and are assumed
1451: to be $25\%$ of the estimated values.}
1452: \tablenotetext{c}{Mass uncertainties are dominated by the interactive process
1453: used in mass determination and are assumed to be $40\%$ of the
1454: estimated values.}
1455: \end{deluxetable}
1456:
1457: \begin{deluxetable}{rrrrrrrrr}
1458: \tabletypesize{\footnotesize}
1459: \tablewidth{0pt}
1460: \tablecaption{Tangent Point \HI\ Clouds - Derived Properties \label{catalogtangent}}
1461: \tablehead{
1462: \colhead{$l$}&\colhead{$b$}&\colhead{$V_{\mathrm{LSR}}$}&\colhead{$V_{\mathrm{dev}}$}&\colhead{$d$}&\colhead{$R$\,\,\tablenotemark{a}}&\colhead{$z$}&\colhead{$r$}&\colhead{$M_{HI}$}\\
1463: \colhead{(deg)}&\colhead{(deg)}&\colhead{(km~s$^{-1}$)}&\colhead{(km~s$^{-1}$)}&\colhead{(kpc)}&\colhead{(kpc)}&\colhead{(kpc)}&\colhead{(pc)}&\colhead{($M_{\odot}$)}}
1464: \startdata
1465: $325.34$ & $-4.61$ & $-94.2\pm 1.8$ & $ -1.7 \pm 3.5$ & $ 7.0 \pm 1.5$ & $ 4.8^{+0.4}$ & $ -0.56 \pm 0.12$ & $ 29 \pm 8$ & $ 1000 \pm 600$ \\
1466: $325.69$ & $-2.87$ & $-98.8\pm 0.7$ & $ -3.9 \pm 3.1$ & $ 7.0 \pm 1.5$ & $ 4.8^{+0.4}$ & $ -0.35 \pm 0.07$ & $ 25 \pm 7$ & $ 470 \pm 270$ \\
1467: $325.95$ & $-3.33$ & $-103.2\pm 0.9$ & $ -6.1 \pm 3.1$ & $ 7.0 \pm 1.4$ & $ 4.8^{+0.4}$ & $ -0.41 \pm 0.08$ & $ 34 \pm 9$ & $ 700 \pm 400$ \\
1468: $326.34$ & $-4.47$ & $-118.0\pm 3.6$ & $ -19.7 \pm 4.7$ & $ 7.1 \pm 1.3$ & $ 4.7^{+0.3}$ & $ -0.55 \pm 0.10$ & $ 27 \pm 7$ & $ 270 \pm 140$ \\
1469: $327.65$ & $13.25$ & $-108.3\pm 6.6$ & $ -4.5 \pm 7.3$ & $ 7.4 \pm 1.9$ & $ 4.5^{+0.6}$ & $ 1.69 \pm 0.44$ & $ 38 \pm 12$ & $ 600 \pm 400$ \\
1470: $328.24$ & $2.67$ & $-135.0\pm 2.5$ & $ -24.7 \pm 3.9$ & $ 7.2 \pm 1.4$ & $ 4.5^{+0.4}$ & $ 0.34 \pm 0.06$ & $ 24 \pm 6$ & $ 1200 \pm 700$ \\
1471: $329.79$ & $3.74$ & $-112.0\pm 1.5$ & $ -0.8 \pm 3.4$ & $ 7.4 \pm 1.9$ & $ 4.3^{+0.6}$ & $ 0.48 \pm 0.12$ & $ 30 \pm 9$ & $ 1800 \pm 1200$ \\
1472: $330.12$ & $-3.40$ & $-129.3\pm 2.0$ & $ -18.8 \pm 3.6$ & $ 7.4 \pm 0.8$ & $ 4.2^{+0.2}$ & $ -0.44 \pm 0.05$ & $ 30 \pm 6$ & $ 120 \pm 50$ \\
1473: $330.79$ & $4.34$ & $-114.1\pm 1.9$ & $ -5.5 \pm 3.6$ & $ 7.4 \pm 0.9$ & $ 4.2^{+0.3}$ & $ 0.56 \pm 0.07$ & $ 24 \pm 5$ & $ 1100 \pm 500$ \\
1474: $330.85$ & $13.32$ & $-138.3\pm 10.1$ & $ -31.6 \pm 10.6$ & $ 7.6 \pm 0.6$ & $ 4.1^{+0.1}$ & $ 1.76 \pm 0.14$ & $ 36 \pm 7$ & $ 240 \pm 100$ \\
1475: $330.90$ & $6.35$ & $-120.3\pm 7.8$ & $ -11.1 \pm 8.4$ & $ 7.5 \pm 0.9$ & $ 4.1^{+0.2}$ & $ 0.83 \pm 0.10$ & $ 25 \pm 5$ & $ 440 \pm 200$ \\
1476: $330.91$ & $5.74$ & $-115.0\pm 4.5$ & $ -5.7 \pm 5.4$ & $ 7.5 \pm 0.9$ & $ 4.1^{+0.3}$ & $ 0.75 \pm 0.09$ & $ 30 \pm 7$ & $ 110 \pm 50$ \\
1477: $331.04$ & $12.43$ & $-114.6\pm 6.3$ & $ -6.5 \pm 7.0$ & $ 7.6 \pm 0.9$ & $ 4.1^{+0.3}$ & $ 1.64 \pm 0.20$ & $ 48 \pm 10$ & $ 1000 \pm 500$ \\
1478: $331.06$ & $3.87$ & $-119.7\pm 2.1$ & $ -9.2 \pm 3.6$ & $ 7.5 \pm 0.9$ & $ 4.1^{+0.3}$ & $ 0.50 \pm 0.06$ & $ 27 \pm 6$ & $ 2400 \pm 1100$ \\
1479: $331.20$ & $10.53$ & $-109.0\pm 4.5$ & $ -1.3 \pm 5.4$ & $ 7.6 \pm 1.0$ & $ 4.1^{+0.3}$ & $ 1.38 \pm 0.18$ & $ 26 \pm 6$ & $ 280 \pm 140$ \\
1480: $331.25$ & $-5.41$ & $-113.2\pm 3.3$ & $ -4.5 \pm 4.4$ & $ 7.5 \pm 1.0$ & $ 4.1^{+0.3}$ & $ -0.71 \pm 0.09$ & $ 48 \pm 10$ & $ 5700 \pm 2700$ \\
1481: $331.33$ & $-3.40$ & $-135.5\pm 3.0$ & $ -26.8 \pm 4.2$ & $ 7.5 \pm 0.7$ & $ 4.1^{+0.2}$ & $ -0.44 \pm 0.04$ & $ 46 \pm 9$ & $ 300 \pm 130$ \\
1482: $331.46$ & $3.40$ & $-112.4\pm 3.6$ & $ -3.4 \pm 4.7$ & $ 7.5 \pm 1.0$ & $ 4.1^{+0.3}$ & $ 0.44 \pm 0.06$ & $ 41 \pm 9$ & $ 4900 \pm 2300$ \\
1483: $332.64$ & $8.16$ & $-115.0\pm 2.4$ & $ -4.9 \pm 3.9$ & $ 7.6 \pm 1.1$ & $ 3.9^{+0.3}$ & $ 1.08 \pm 0.15$ & $ 28 \pm 6$ & $ 120 \pm 60$ \\
1484: $332.66$ & $-5.34$ & $-115.7\pm 1.1$ & $ -4.6 \pm 3.2$ & $ 7.6 \pm 1.1$ & $ 3.9^{+0.3}$ & $ -0.71 \pm 0.10$ & $ 31 \pm 7$ & $ 2000 \pm 1000$ \\
1485: $332.73$ & $-3.47$ & $-112.7\pm 2.0$ & $ -0.7 \pm 3.6$ & $ 7.6 \pm 1.1$ & $ 3.9^{+0.3}$ & $ -0.46 \pm 0.06$ & $ 20 \pm 5$ & $ 800 \pm 400$ \\
1486: $332.79$ & $-6.41$ & $-132.2\pm 5.5$ & $ -20.4 \pm 6.2$ & $ 7.6 \pm 0.8$ & $ 3.9^{+0.2}$ & $ -0.85 \pm 0.09$ & $ 33 \pm 7$ & $ 800 \pm 400$ \\
1487: $332.86$ & $-2.73$ & $-143.7\pm 2.7$ & $ -29.7 \pm 4.0$ & $ 7.6 \pm 0.8$ & $ 3.9^{+0.2}$ & $ -0.36 \pm 0.04$ & $ 25 \pm 5$ & $ 340 \pm 150$ \\
1488: $332.87$ & $-2.07$ & $-151.4\pm 5.7$ & $ -37.4 \pm 6.4$ & $ 7.6 \pm 0.8$ & $ 3.9^{+0.2}$ & $ -0.27 \pm 0.03$ & $ 24 \pm 5$ & $ 230 \pm 100$ \\
1489: $333.18$ & $-9.24$ & $-114.4\pm 1.5$ & $ -1.0 \pm 3.4$ & $ 7.7 \pm 1.1$ & $ 3.8^{+0.3}$ & $ -1.23 \pm 0.17$ & $ 37 \pm 8$ & $ 460 \pm 220$ \\
1490: $333.19$ & $-5.61$ & $-128.8\pm 2.9$ & $ -14.6 \pm 4.2$ & $ 7.6 \pm 1.0$ & $ 3.8^{+0.2}$ & $ -0.75 \pm 0.10$ & $ 33 \pm 7$ & $ 900 \pm 400$ \\
1491: $333.20$ & $-3.87$ & $-116.2\pm 2.7$ & $ -1.7 \pm 4.1$ & $ 7.6 \pm 1.1$ & $ 3.8^{+0.3}$ & $ -0.51 \pm 0.07$ & $ 30 \pm 7$ & $ 900 \pm 400$ \\
1492: $333.26$ & $-5.21$ & $-115.4\pm 1.2$ & $ -1.0 \pm 3.2$ & $ 7.6 \pm 1.1$ & $ 3.8^{+0.3}$ & $ -0.69 \pm 0.10$ & $ 46 \pm 10$ & $ 5100 \pm 2500$ \\
1493: $333.51$ & $-9.44$ & $-133.4\pm 8.0$ & $ -21.4 \pm 8.5$ & $ 7.7 \pm 0.8$ & $ 3.8^{+0.2}$ & $ -1.27 \pm 0.13$ & $ 29 \pm 6$ & $ 340 \pm 150$ \\
1494: $333.53$ & $-2.67$ & $-126.2\pm 6.1$ & $ -12.5 \pm 6.8$ & $ 7.6 \pm 1.0$ & $ 3.8^{+0.2}$ & $ -0.35 \pm 0.05$ & $ 20 \pm 4$ & $ 270 \pm 130$ \\
1495: $333.53$ & $3.27$ & $-130.4\pm 0.8$ & $ -16.8 \pm 3.1$ & $ 7.6 \pm 0.9$ & $ 3.8^{+0.2}$ & $ 0.43 \pm 0.05$ & $ 33 \pm 7$ & $ 550 \pm 260$ \\
1496: $333.67$ & $-3.20$ & $-122.6\pm 4.3$ & $ -7.4 \pm 5.3$ & $ 7.6 \pm 1.0$ & $ 3.8^{+0.3}$ & $ -0.43 \pm 0.06$ & $ 23 \pm 5$ & $ 550 \pm 270$ \\
1497: $333.80$ & $-4.67$ & $-132.5\pm 1.7$ & $ -16.0 \pm 3.4$ & $ 7.7 \pm 0.9$ & $ 3.8^{+0.2}$ & $ -0.62 \pm 0.07$ & $ 19 \pm 4$ & $ 280 \pm 130$ \\
1498: $334.13$ & $-2.60$ & $-124.8\pm 1.8$ & $ -8.2 \pm 3.5$ & $ 7.7 \pm 1.0$ & $ 3.7^{+0.3}$ & $ -0.35 \pm 0.05$ & $ 23 \pm 5$ & $ 230 \pm 110$ \\
1499: $334.33$ & $3.80$ & $-124.4\pm 2.2$ & $ -5.2 \pm 3.7$ & $ 7.7 \pm 1.0$ & $ 3.7^{+0.3}$ & $ 0.51 \pm 0.07$ & $ 29 \pm 6$ & $ 400 \pm 190$ \\
1500: $334.46$ & $-4.47$ & $-123.0\pm 2.3$ & $ -1.7 \pm 3.8$ & $ 7.7 \pm 1.1$ & $ 3.7^{+0.3}$ & $ -0.60 \pm 0.08$ & $ 38 \pm 9$ & $ 2000 \pm 1000$ \\
1501: $334.47$ & $3.00$ & $-127.3\pm 1.8$ & $ -5.8 \pm 3.5$ & $ 7.7 \pm 1.0$ & $ 3.7^{+0.3}$ & $ 0.40 \pm 0.05$ & $ 31 \pm 7$ & $ 900 \pm 400$ \\
1502: $334.60$ & $-3.00$ & $-134.3\pm 4.4$ & $ -12.5 \pm 5.3$ & $ 7.7 \pm 1.0$ & $ 3.6^{+0.2}$ & $ -0.40 \pm 0.05$ & $ 26 \pm 6$ & $ 570 \pm 270$ \\
1503: $334.80$ & $-7.82$ & $-136.3\pm 2.2$ & $ -15.0 \pm 3.7$ & $ 7.8 \pm 1.0$ & $ 3.6^{+0.2}$ & $ -1.06 \pm 0.14$ & $ 42 \pm 9$ & $ 560 \pm 270$ \\
1504: $334.93$ & $-11.95$ & $-131.5\pm 8.0$ & $ -12.0 \pm 8.6$ & $ 7.9 \pm 1.0$ & $ 3.6^{+0.2}$ & $ -1.63 \pm 0.21$ & $ 35 \pm 8$ & $ 570 \pm 270$ \\
1505: $335.00$ & $0.80$ & $-136.5\pm 1.9$ & $ -14.4 \pm 3.5$ & $ 7.7 \pm 1.0$ & $ 3.6^{+0.2}$ & $ 0.11 \pm 0.01$ & $ 28 \pm 6$ & $ 220 \pm 100$ \\
1506: $335.01$ & $-7.22$ & $-123.6\pm 2.2$ & $ -2.5 \pm 3.7$ & $ 7.8 \pm 1.1$ & $ 3.6^{+0.3}$ & $ -0.98 \pm 0.14$ & $ 30 \pm 7$ & $ 220 \pm 110$ \\
1507: $335.07$ & $-6.62$ & $-128.2\pm 1.5$ & $ -7.0 \pm 3.3$ & $ 7.8 \pm 1.1$ & $ 3.6^{+0.3}$ & $ -0.89 \pm 0.12$ & $ 25 \pm 6$ & $ 200 \pm 90$ \\
1508: $335.07$ & $-4.88$ & $-123.4\pm 2.8$ & $ -1.9 \pm 4.1$ & $ 7.7 \pm 1.1$ & $ 3.6^{+0.3}$ & $ -0.66 \pm 0.09$ & $ 31 \pm 7$ & $ 610 \pm 300$ \\
1509: $335.13$ & $-4.40$ & $-142.7\pm 4.8$ & $ -20.7 \pm 5.7$ & $ 7.7 \pm 0.8$ & $ 3.6^{+0.2}$ & $ -0.59 \pm 0.06$ & $ 38 \pm 8$ & $ 800 \pm 400$ \\
1510: $335.13$ & $3.00$ & $-139.4\pm 2.3$ & $ -17.2 \pm 3.8$ & $ 7.7 \pm 0.9$ & $ 3.6^{+0.2}$ & $ 0.40 \pm 0.05$ & $ 39 \pm 8$ & $ 900 \pm 400$ \\
1511: $335.20$ & $-5.88$ & $-124.5\pm 3.0$ & $ -1.9 \pm 4.2$ & $ 7.8 \pm 1.1$ & $ 3.6^{+0.3}$ & $ -0.79 \pm 0.11$ & $ 24 \pm 5$ & $ 180 \pm 90$ \\
1512: $335.20$ & $4.54$ & $-133.5\pm 2.6$ & $ -10.7 \pm 4.0$ & $ 7.7 \pm 1.0$ & $ 3.6^{+0.3}$ & $ 0.61 \pm 0.08$ & $ 28 \pm 6$ & $ 350 \pm 170$ \\
1513: $335.27$ & $-2.47$ & $-126.6\pm 2.2$ & $ -2.2 \pm 3.7$ & $ 7.7 \pm 1.1$ & $ 3.6^{+0.3}$ & $ -0.33 \pm 0.05$ & $ 32 \pm 7$ & $ 3200 \pm 1600$ \\
1514: $335.27$ & $2.13$ & $-127.0\pm 2.0$ & $ -2.5 \pm 3.6$ & $ 7.7 \pm 1.1$ & $ 3.6^{+0.3}$ & $ 0.29 \pm 0.04$ & $ 25 \pm 6$ & $ 600 \pm 300$ \\
1515: $335.33$ & $4.07$ & $-128.8\pm 1.5$ & $ -4.3 \pm 3.4$ & $ 7.7 \pm 1.1$ & $ 3.5^{+0.3}$ & $ 0.55 \pm 0.08$ & $ 53 \pm 12$ & $ 2300 \pm 1100$ \\
1516: $335.74$ & $-5.07$ & $-151.0\pm 1.6$ & $ -22.4 \pm 3.4$ & $ 7.8 \pm 0.8$ & $ 3.5^{+0.2}$ & $ -0.69 \pm 0.07$ & $ 58 \pm 12$ & $ 3200 \pm 1400$ \\
1517: $335.94$ & $-4.07$ & $-141.2\pm 2.3$ & $ -11.5 \pm 3.8$ & $ 7.8 \pm 1.0$ & $ 3.5^{+0.2}$ & $ -0.55 \pm 0.07$ & $ 32 \pm 7$ & $ 1100 \pm 500$ \\
1518: $336.00$ & $2.40$ & $-132.6\pm 0.8$ & $ -3.5 \pm 3.1$ & $ 7.8 \pm 1.1$ & $ 3.5^{+0.3}$ & $ 0.33 \pm 0.05$ & $ 17 \pm 4$ & $ 70 \pm 30$ \\
1519: $336.20$ & $3.13$ & $-129.8\pm 1.0$ & $ -0.2 \pm 3.2$ & $ 7.8 \pm 1.1$ & $ 3.4^{+0.3}$ & $ 0.43 \pm 0.06$ & $ 27 \pm 6$ & $ 1200 \pm 600$ \\
1520: $336.87$ & $-5.01$ & $-141.3\pm 5.0$ & $ -13.2 \pm 5.8$ & $ 7.8 \pm 1.0$ & $ 3.3^{+0.2}$ & $ -0.68 \pm 0.09$ & $ 49 \pm 11$ & $ 5100 \pm 2400$ \\
1521: $336.94$ & $6.08$ & $-139.6\pm 4.1$ & $ -12.3 \pm 5.1$ & $ 7.9 \pm 1.0$ & $ 3.3^{+0.2}$ & $ 0.83 \pm 0.11$ & $ 30 \pm 7$ & $ 490 \pm 240$ \\
1522: $337.01$ & $-5.14$ & $-132.3\pm 3.2$ & $ -4.7 \pm 4.4$ & $ 7.9 \pm 1.1$ & $ 3.3^{+0.3}$ & $ -0.70 \pm 0.10$ & $ 44 \pm 10$ & $ 4000 \pm 1900$ \\
1523: $337.27$ & $-2.94$ & $-127.5\pm 2.1$ & $ -0.3 \pm 3.7$ & $ 7.8 \pm 1.1$ & $ 3.3^{+0.3}$ & $ -0.40 \pm 0.06$ & $ 41 \pm 9$ & $ 12000 \pm 6000$ \\
1524: $337.35$ & $-4.47$ & $-145.0\pm 1.3$ & $ -18.9 \pm 3.3$ & $ 7.9 \pm 0.9$ & $ 3.3^{+0.2}$ & $ -0.61 \pm 0.07$ & $ 26 \pm 5$ & $ 100 \pm 50$ \\
1525: $337.55$ & $5.41$ & $-146.4\pm 3.3$ & $ -20.6 \pm 4.5$ & $ 7.9 \pm 0.8$ & $ 3.2^{+0.2}$ & $ 0.74 \pm 0.08$ & $ 39 \pm 8$ & $ 190 \pm 90$ \\
1526: $337.61$ & $16.01$ & $-131.1\pm 18.9$ & $ -8.6 \pm 19.1$ & $ 8.2 \pm 1.1$ & $ 3.2^{+0.3}$ & $ 2.25 \pm 0.30$ & $ 32 \pm 7$ & $ 170 \pm 80$ \\
1527: $337.87$ & $-2.47$ & $-145.4\pm 1.8$ & $ -16.6 \pm 3.5$ & $ 7.9 \pm 0.9$ & $ 3.2^{+0.3}$ & $ -0.34 \pm 0.04$ & $ 44 \pm 9$ & $ 3600 \pm 1700$ \\
1528: $337.87$ & $2.60$ & $-137.4\pm 1.2$ & $ -8.6 \pm 3.2$ & $ 7.9 \pm 1.0$ & $ 3.2^{+0.3}$ & $ 0.36 \pm 0.05$ & $ 29 \pm 6$ & $ 230 \pm 110$ \\
1529: $337.88$ & $-5.07$ & $-155.1\pm 5.4$ & $ -26.6 \pm 6.2$ & $ 7.9 \pm 0.8$ & $ 3.2^{+0.2}$ & $ -0.70 \pm 0.07$ & $ 52 \pm 11$ & $ 1100 \pm 500$ \\
1530: $337.88$ & $-3.80$ & $-152.2\pm 2.0$ & $ -23.5 \pm 3.6$ & $ 7.9 \pm 0.8$ & $ 3.2^{+0.2}$ & $ -0.52 \pm 0.05$ & $ 31 \pm 6$ & $ 380 \pm 170$ \\
1531: $337.94$ & $-3.13$ & $-142.5\pm 2.0$ & $ -13.9 \pm 3.6$ & $ 7.9 \pm 1.0$ & $ 3.2^{+0.3}$ & $ -0.43 \pm 0.05$ & $ 30 \pm 6$ & $ 1100 \pm 500$ \\
1532: $337.94$ & $4.14$ & $-147.4\pm 4.5$ & $ -18.9 \pm 5.4$ & $ 7.9 \pm 0.9$ & $ 3.2^{+0.3}$ & $ 0.57 \pm 0.07$ & $ 40 \pm 8$ & $ 600 \pm 280$ \\
1533: $337.99$ & $17.47$ & $-176.0\pm 4.1$ & $ -53.4 \pm 5.1$ & $ 8.3 \pm 0.4$ & $ 3.2^{+0.1}$ & $ 2.48 \pm 0.11$ & $ 34 \pm 6$ & $ 240 \pm 100$ \\
1534: $338.40$ & $-2.33$ & $-152.8\pm 2.7$ & $ -25.1 \pm 4.0$ & $ 7.9 \pm 0.8$ & $ 3.1^{+0.2}$ & $ -0.32 \pm 0.03$ & $ 33 \pm 7$ & $ 800 \pm 400$ \\
1535: $338.54$ & $-3.74$ & $-159.8\pm 4.1$ & $ -30.0 \pm 5.1$ & $ 7.9 \pm 0.7$ & $ 3.1^{+0.1}$ & $ -0.52 \pm 0.04$ & $ 43 \pm 8$ & $ 550 \pm 240$ \\
1536: $339.08$ & $-3.13$ & $-134.2\pm 1.9$ & $ -0.4 \pm 3.6$ & $ 8.0 \pm 1.1$ & $ 3.0^{+0.3}$ & $ -0.43 \pm 0.06$ & $ 37 \pm 8$ & $ 2700 \pm 1300$ \\
1537: $339.29$ & $-5.07$ & $-141.4\pm 2.3$ & $ -5.5 \pm 3.8$ & $ 8.0 \pm 1.0$ & $ 3.0^{+0.3}$ & $ -0.71 \pm 0.09$ & $ 40 \pm 9$ & $ 1000 \pm 500$ \\
1538: $339.65$ & $8.36$ & $-140.0\pm 4.9$ & $ -6.3 \pm 5.8$ & $ 8.1 \pm 1.1$ & $ 3.0^{+0.3}$ & $ 1.17 \pm 0.15$ & $ 50 \pm 11$ & $ 800 \pm 400$ \\
1539: $339.88$ & $2.13$ & $-153.9\pm 5.0$ & $ -18.1 \pm 10.3$ & $ 8.0 \pm 0.9$ & $ 2.9^{+0.2}$ & $ 0.30 \pm 0.03$ & $ 31 \pm 7$ & $ 900 \pm 400$ \\
1540: $340.25$ & $-5.94$ & $-144.6\pm 5.4$ & $ -9.5 \pm 10.5$ & $ 8.0 \pm 1.0$ & $ 2.9^{+0.3}$ & $ -0.83 \pm 0.10$ & $ 30 \pm 6$ & $ 1000 \pm 500$ \\
1541: $340.35$ & $-2.87$ & $-142.3\pm 2.5$ & $ -7.6 \pm 9.3$ & $ 8.0 \pm 1.0$ & $ 2.9^{+0.3}$ & $ -0.40 \pm 0.05$ & $ 43 \pm 9$ & $ 2100 \pm 1000$ \\
1542: $340.41$ & $2.40$ & $-164.8\pm 3.0$ & $ -29.5 \pm 9.5$ & $ 8.0 \pm 0.8$ & $ 2.9^{+0.2}$ & $ 0.34 \pm 0.03$ & $ 35 \pm 7$ & $ 1300 \pm 600$ \\
1543: $341.65$ & $-4.47$ & $-150.5\pm 4.9$ & $ -6.8 \pm 10.2$ & $ 8.1 \pm 1.0$ & $ 2.7^{+0.3}$ & $ -0.63 \pm 0.08$ & $ 33 \pm 7$ & $ 370 \pm 170$ \\
1544: $342.68$ & $16.98$ & $-177.4\pm 12.4$ & $ -35.7 \pm 15.3$ & $ 8.5 \pm 0.6$ & $ 2.5^{+0.1}$ & $ 2.48 \pm 0.17$ & $ 27 \pm 5$ & $ 200 \pm 80$ \\
1545: $342.91$ & $3.54$ & $-149.2\pm 3.9$ & $ 0.4 \pm 9.8$ & $ 8.1 \pm 0.8$ & $ 2.5^{+0.2}$ & $ 0.50 \pm 0.05$ & $ 31 \pm 6$ & $ 340 \pm 150$ \\
1546: \enddata
1547: \tablecomments{Derived properties of a subset of the clouds assumed to
1548: be located at tangent points (those with $V_{\mathrm{dev}}\leq
1549: 0.8$~km~s$^{-1}$).}
1550: \tablenotetext{a}{Along a given line-of-sight, the smallest
1551: Galactocentric radius possible is at the tangent point. If the cloud
1552: is not located at the tangent point it must be further away from the
1553: center and the error on $R$ must be positive.}
1554: \end{deluxetable}
1555:
1556: \begin{deluxetable}{lcccc}
1557: \tablecaption{Property Summary and Comparison to Lockman Clouds\label{comptable}}
1558: \tablehead{ & \multicolumn{2}{c}{This Sample} & \multicolumn{2}{c}{Lockman (2002)}\\
1559: \colhead{Parameter} & \colhead{Median} & \colhead{90\%\ Range} & \colhead{Median} & \colhead{90\%\ Range}}
1560: \startdata
1561: $T_{\mathrm{pk}}$ (K) & $0.6$ & $0.2 \rightarrow 2.2$ & $1.0$ &$0.4 \rightarrow 2.7$\\
1562: $\Delta v$ (km~s$^{-1}$) & $12.8$ & $5.8 \rightarrow 26.2$ & $12.2$ & $5.4\rightarrow 26.3$\\
1563: $N_{HI}$ ($\times 10^{19}$ cm$^{-2}$) & $1.4$ & $0.2 \rightarrow 2.2$ & $2$ &
1564: $0.7\rightarrow 6.3$\\
1565: $r$ (pc) & $32$ & $< 23 \rightarrow 50$ & $12$ &$<9.5\rightarrow 17.5$\\
1566: $M_{HI}$ ($M_{\odot}$) & $630$ & $120 \rightarrow 4850 $ & $50$ &$12\rightarrow 290$\\
1567: $|z|$ (pc) & $560$ & $320 \rightarrow 1690 $ & $940$ & $640\rightarrow 1210$ \\
1568: $\sigma_{cc}$ (km~s$^{-1}$) & $16-22$ & & $15-20$ &\\
1569: \enddata
1570: \tablecomments{Median values of the observed halo cloud properties in
1571: this sample, where the number of clouds $n=403$ for $T_{\mathrm{pk}}$,
1572: $\Delta v$, and $N_{HI}$ and $n=81$ for remaining properties
1573: (of tangent point clouds) and the \cite{2002Lockman} sample, where
1574: $n=38$. Most properties have a large scatter about the median in
1575: both samples, as demonstrated by the $90\%$ range.}
1576: \end{deluxetable}
1577:
1578: \end{document}