1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass{emulateapj}
3: \shorttitle{Temperature of Starless Cores}
4: \shortauthors{Schnee, S. et al.}
5:
6: \begin{document}
7: %\setlength\textheight{9in}
8:
9: \newcommand{\myemail}{schnee@astro.caltech.edu}
10: \newcommand{\amm}{NH$_3$}
11: \newcommand{\kms}{km s$^{-1}$}
12: \newcommand{\nh}{N$_{\rm H}$}
13: \newcommand{\td}{T$_{\rm d}$}
14: \newcommand{\tc}{T$_{\rm kin}$}
15: \newcommand{\tco}{T$_{\rm CO}$}
16: \newcommand{\sigamm}{$\sigma_{\rm NT}$}
17: \newcommand{\sigco}{$\sigma_{\rm CO}$}
18: \newcommand{\AV}{$A_V$}
19: \newcommand{\ccm}{cm$^{-3}$}
20:
21: \title{The Gas Temperature of Starless Cores in Perseus}
22: \author{S. Schnee$^1$, E. Rosolowsky$^{2,3}$, J. Foster$^2$, M. Enoch$^4$, A.
23: Sargent$^1$}
24: \affil{$^1$Division of Physics, Mathematics and Astronomy,
25: California Institute of Technology, 770 South Wilson Avenue,
26: Pasadena, CA 91125 \\
27: $^2$Harvard-Smithsonian Center for Astrophysics, 60 Garden
28: Street, Cambridge, MA 02138 \\
29: $^3$University of British Columbia Okanagan, 3333 University
30: Way, Kelowna, BC V1V 1V7, Canada \\
31: $^4$Department of Astronomy, University of California, Berkeley,
32: CA 94720}
33: \email{\myemail}
34:
35: \begin{abstract}
36:
37: In this paper we study the determinants of starless core temperatures
38: in the Perseus molecular cloud. We use \amm\ (1,1) and (2,2)
39: observations to derive core temperatures (\tc) and data from the
40: COMPLETE Survey of Star Forming Regions and the c2d Spitzer Legacy
41: Survey for observations of the other core and molecular cloud
42: properties. The kinetic temperature distribution probed by \amm\ is
43: in the fairly narrow range of $\sim$ 9 - 15 K. We find that cores
44: within the clusters IC348 and NGC1333 are significantly warmer than
45: ``field'' starless cores, and \tc\ is higher within regions of larger
46: extinction-derived column density. Starless cores in the field are
47: warmer when they are closer to class O/I protostars, but this effect
48: is not seen for those cores in clusters. For field starless cores,
49: \tc\ is higher in regions in which the $^{13}$CO linewidth and the
50: 1.1mm flux from the core are larger, and \tc\ is lower when the the
51: peak column density within the core and average volume density of the
52: core are larger. There is no correlation between \tc\ and $^{13}$CO
53: linewidth, 1.1mm flux, density or peak column density for those cores
54: in clusters. The temperature of the cloud material along the line of
55: sight to the core, as measured by CO or far-infrared emission from
56: dust, is positively correlated with core temperature when considering
57: the collection of cores in the field and in clusters, but this effect
58: is not apparent when the two subsamples of cores are considered
59: separately.
60:
61: \end{abstract}
62:
63: \keywords{stars: formation --- ISM: molecules}
64:
65: \section{Introduction}
66:
67: Starless cores are the link between molecular clouds and Class 0
68: protostars. They are smaller and denser (r $\sim$ 0.1 pc, n $\gtrsim
69: 10^4$ cm$^{-3}$) than, and kinematically distinct from, the molecular
70: clouds in which they are found \citep{Goodman98}. Because starless
71: cores have no source of internal heating, the temperature of a
72: starless core is determined by external illumination and
73: self-shielding. The temperatures of starless cores are set by the
74: interstellar radiation field, and in some cases luminous nearby stars,
75: attenuated by the molecular cloud. The temperature profile within a
76: starless core is determined by the heating of its outer layers and the
77: internal extinction, making the centers of starless cores colder than
78: their outer layers \citep{Evans01, Stamatellos07, Zucconi01}.
79:
80: The temperatures of starless cores can be inferred from thermal
81: emission from dust or from molecular lines. Deriving core
82: temperatures from continuum emission is complicated by uncertainties
83: in the dust emission spectrum. For instance, in cold and dense
84: regions dust grains are expected to coagulate and become covered in
85: icy mantles, creating variation in the emissivity spectral index
86: \citep{Ossenkopf94}. In addition, below a density of $\sim3 \times
87: 10^4$ the dust temperature is not coupled to the gas temperature
88: \citep{Goldsmith01, Galli02}, which is the more important quantity
89: because the gas makes up $\sim$99\% of the mass. One of the best
90: tracers of gas temperature is the rotation-inversion lines of \amm.
91: Ammonia is particularly well suited for this type of study because 1)
92: \amm\ does not seem to deplete onto dust grains in starless cores, and
93: in fact may have an enhanced abundance where CO is depleted
94: \citep{Tafalla02} 2) \amm\ has hyperfine splitting of its inversion
95: lines, making fits to the temperature, velocity and optical depth more
96: certain 3) the \amm\ (1,1) and (2,2) transitions are separated by a
97: small difference in frequency, so it is possible to simultaneously
98: observe both sets of lines. In this paper we use ammonia (1,1) and
99: (2,2) observations \citep{Rosolowsky08} to derive the temperature of
100: starless cores. We examine whether these data support the several
101: models of the relationship between core properties and the molecular
102: cloud where they are found \citep[e.g.,][]{Evans01,
103: Bethell04,Stamatellos04}
104:
105: Perseus is one of the Gould's belt molecular clouds included in the
106: Spitzer legacy project ``From Molecular Cores to Planet-Forming
107: Disks'' (c2d) \citep{Evans03} and the COMPLETE survey of star-forming
108: regions \citep{Ridge06}, at a distance of 250 pc \citep{Cernis03,
109: Belikov02}. Because it is so well observed, there is a wealth of data
110: on the cores and the cloud that enables us to study the influences of
111: various internal and environmental factors on the temperature of 50
112: starless cores in Perseus.
113:
114: \section{Data}
115:
116: \subsection{Far-Infrared} \label{OBSFIR}
117:
118: As described in \citet{Schnee08}, we used far-infrared emission maps
119: at 70, 100 and 160 \micron\ to derive the dust temperature and column
120: density of the Perseus molecular cloud. The 70 and 160 \micron\ maps
121: were obtained as part of the c2d survey \citep{Rebull07}, and the 100
122: \micron\ map was taken from the IRIS reprocessing of the IRAS all-sky
123: maps \citep{Miville05}. The resolution of the emission-derived column
124: density (\nh) and dust temperature (\td) maps with which we will
125: compare the \amm\ temperature (\tc) is 40\arcsec, as set by the 160
126: \micron\ data.
127:
128: The dust temperature is calculated from the ratio of the observed flux
129: at 70 and 160 \micron, taking into account the color correction at
130: both wavelengths and the emission from transiently heated very small
131: dust grains at 70 \micron. The column density is derived from the
132: flux at 160 \micron\ and the dust temperature, allowing for a variable
133: dust emissivity and normalized to the extinction measured from NIR
134: reddening. Maps of the emission-derived dust temperature and column
135: density, and the procedure used to make the maps, are presented in
136: detail in \citet{Schnee08}.
137:
138: \subsection{Near-Infrared} \label{OBSNIR}
139:
140: Data from the final release of the Two Micron All Sky Survey (2MASS)
141: point source catalog was used to create a large-scale extinction map
142: of Perseus as part of the COMPLETE survey \citep{Ridge06}. The NICER
143: algorithm estimates the column density towards background stars by
144: comparing their observed colors with the (assumed) instrinsic colors
145: \citep{Lombardi01}. Although we expect the extinction map to be our
146: least-biased estimate of the cloud column density \citep{Schnee06,
147: Goodman08}, its 5\arcmin\ resolution is roughly ten times coarser than
148: our other maps, which smooths out density gradients and makes it
149: somewhat harder to compare the cloud and core properties.
150:
151: \subsection{\amm\ (1,1) and (2,2)} \label{OBSNH3}
152:
153: We used observations of 193 dense cores and core candidates in the
154: Perseus Molecular Cloud using the 100-m Robert F. Byrd Green Bank
155: Telescope (GBT) as described in \citet{Rosolowsky08}. For each target,
156: single-pointing (31\arcsec\ FWHM) observations were made of the \amm\
157: (1,1) and (2,2) lines with 0.024 \kms\ spectral resolution. The
158: targets for the ammonia survey were drawn, in order of precedence,
159: from 1) the locations of millimeter cores in the Bolocam survey of the
160: region \citep{Enoch06} 2) the locations of submillimeter cores in the
161: SCUBA survey of the region \citep{Kirk06}, 3) sources in the
162: literature survey of \citet{Jijina99}, and 4) cold,
163: high-column-density objects in the dust map produced by
164: \citet{Schnee08}, and 5) sources identified by eye that appear in both
165: the Bolocam and SCUBA maps but were not included in the published
166: catalogs. The starless cores discussed in this paper were drawn from
167: from only the first two categories.
168:
169: The kinetic temperatures of the starless cores (Section \ref{IDENT}
170: describes how we identify which cores are starless) are determined by
171: a simultaneous fit to the (1,1) and (2,2) emission profiles,
172: accounting for collisional deexcitation of the two states. The same
173: fit also yields the width of the \amm\ line, which is decomposed into
174: its thermal and non-thermal components based on the temperature
175: determination. Although there are likely gas temperature and density
176: gradients within each starless core on both theoretical and
177: observational grouds \citep{Crapsi07}, the fitting of the
178: ammonia-derived temperature is essentially an average over the core
179: properties. However, a single-temperature model with uniform
180: excitation conditions is a reasonably good fit to nearly all of the
181: starless core in this sample. The NH$_3$ observations can be used to
182: accurately derive the kinetic temperatures of starless cores with
183: $8\mbox{ K}\le T_{\mathrm{kin}} \le 40\mbox{ K}$ where the lower and
184: upper limits are set by sensitivity and transition thermalization
185: respectively \citep{Walmsley83}. Further details about the
186: observations and data reduction are described in \citet{Rosolowsky08}.
187:
188: \subsection{$^{12,13}$CO (1--0)} \label{OBSCO}
189:
190: Observations of the $^{12}$CO (1--0) and $^{13}$CO (1--0) transitions
191: at 115 and 110 GHz were carried at the 14-m Five College Radio
192: Astronomy Observatory (FCRAO) telescope as part of the COMPLETE
193: survey. The data have 0.07 \kms\ spectral resolution and 40\arcsec\
194: spatial resolution. The data reduction and mapping techniques are
195: described in \citet{Ridge06}. The CO maps cover approximately the same
196: area in Perseus as the c2d survey, giving us a good estimate of the
197: kinematics of the ambient, less dense, material in the molecular
198: cloud. \citet{Pineda08} use this data set to determine the gas
199: temperature and column density in regions in which the $^{13}$CO
200: (1--0) transition is optically thin (the $^{12}$CO line is assumed to
201: be optically thick everywhere) and the gas is not subthermally
202: excited.
203:
204: \subsection{1.1mm Continuum Maps} \label{OBSBOLO}
205:
206: As part of a survey that is complemetary to c2d, \citet{Enoch06}
207: mapped 7.5 deg$^2$ of the Perseus molecular cloud at 1.1mm with
208: Bolocam at the Caltech Submillimeter Observatory. The resolution is
209: 31\arcsec, which matches very well with our \amm\ observations, and
210: all 122 1.1mm-detected cores were observed with the GBT. The Bolocam
211: observations are most sensitive to dense cores ($n \gtrsim 2 \times
212: 10^4$ cm$^{-3}$) with density contrasts at least 30 times greater than
213: the background cloud, which matches the description of a starless core
214: embedded in a molecular cloud quite well \citep{Bergin07}. In addition
215: to using the 1.1mm data for target selection, we use the 1.1mm derived
216: fluxes and sizes to compare the \tc\ with core properties such as
217: mass, column density and volume density.
218:
219: \section{Identification of Starless Cores} \label{IDENT}
220:
221: To ensure that we only study the starless core population in Perseus,
222: we compare the positions of our \amm\ observations with the
223: coordinates of known protostars, as found through the analysis of
224: \citet{Enoch08}, \citet{Jorgensen07} and \citet{Hatchell07}.
225: \citet{Jorgensen07} identify 49 deeply embedded Young Stellar Objects
226: (YSOs) in Perseus using Spitzer/IRAC color cuts, Spitzer/MIPS and 850
227: \micron\ SCUBA correlations, and SCUBA morphologies.
228: \citet{Hatchell07} uses NIR/MIR (2MASS/IRAC) detections and/or
229: detections of protostellar outflows to separate starless cores from
230: protostars, and find 56 protostars. \citet{Enoch08} identifies 55
231: protostellar cores out of the total 122 millimeter cores in Perseus
232: using the spatial coincidence between 1.1mm continuum cores and
233: sources identified in the c2d Delivery Document \citep{Evans07,
234: Harvey07} as ``YSOc'' (YSO candidate), ``red'' or that are 70 \micron\
235: point sources and are not classified as ``Galc'' (Galaxy candidate).
236: In this paper we only consider those 50 ammonia pointings that are
237: called ``starless'' by the various definitions of all three papers, or
238: that are more than 30\arcsec\ from any identified protostar in any of
239: the three papers. None of the cores we classify as ``starless'' are
240: identified as a low-luminosity embedded protostar by \citet{Dunham08}.
241: We also require that the starless cores have solid \amm-derived
242: temperatures and linewidths, meaning that \tc\ is greater than 2.73 K,
243: the \amm\ (2,2) detection is $> 5\sigma$ and the model and observed
244: spectra fit well by eye. Of the 193 observations in the \amm\ survey,
245: 137 of the pointings are not coincident with an identified protostar,
246: and 50 of these satisfy the signal to noise and goodness of fit
247: requirements.
248:
249: \section{Analysis} \label{ANALYSIS}
250:
251: Here we compare the ammonia-derived temperature with several
252: properties of the starless cores and their environment. We considered
253: only those 50 starless cores with well-defined \tc\ and linewidth, so
254: we are able to derive their non-thermal linewidths. All were detected
255: in the 1.1mm survey of \citet{Enoch06}, so for each we have the total
256: flux, peak flux (from which we derive the peak column density), mass
257: (from the total flux and the \amm\ temperature) and mean particle
258: density. The FIR-derived dust temperature and column density are
259: considered to be believable as long as they fall within the range $12
260: \le T_d \le 20$ and $0 \le A_V \le 24.5$, which is true except in
261: those regions where the 160 \micron\ map is saturated or corrupted by
262: a nearby bright source. There are 46 starless cores with believable
263: FIR-derived \td\ and \AV, and all 50 starless cores fall within the
264: area of our NIR-derived extinction map. The gaussian fitting algorithm
265: failed to converge when calculating the $^{13}$CO linewidth of \tco\
266: for 6 cores. The internal properties of our sample of starless cores
267: are listed in Table \ref{COREPROPTAB}, and the properties of Perseus
268: cloud along the line of sight to the cores are given in Table
269: \ref{COREENVIRTAB}. Although the distance from each starless core to
270: the nearest protostar found in the surveys of \citet{Enoch08,
271: Jorgensen07} and \citet{Hatchell07} is known, in our discussion of
272: possible correlations we consider only those with separations of 0.25
273: pc or smaller.
274:
275: \subsection{\tc} \label{TC}
276:
277: The kinetic temperature distribution probed by \amm\ is quite narrow,
278: from $\sim$ 9 - 15 K, as seen in Fig.~\ref{TKINHIST}. This may be due
279: to the relatively low angular resolution of the GBT pointings, whose
280: 31\arcsec\ beam corresponds to probing the inner 0.019 pc (3.9$\times
281: 10^3$ AU) of each core. Models of externally heated prestellar cores
282: show that they are only significantly colder than 9 K within the inner
283: $\sim 10^3$ AU or warmer than 14 K outside a radius of $\sim 10^4$ AU,
284: with variation depending on the strength of the ISRF incident upon the
285: core \citep{Evans01, Stamatellos04}. We are unlikely to be biased by
286: chemical effects, given that \amm\ is not seen to deplete in starless
287: cores \citep[e.g.,][]{Tafalla02}. The critical density of \amm, $\sim
288: 10^4$ \ccm\ \citep{Swade89}, is low compared the typical gas densities
289: observed in the cores in our sample ($10^4$ \ccm\ $\le n \le 10^6$
290: \ccm), so we expect the \amm\ observations to sample the full volume
291: of our cores.
292:
293: The range of temperatures seen in the Perseus starless cores is
294: typical of those seen in other molecular clouds. For instance,
295: \citet{Kirk07} derived dust temperatures of pre-stellar cores in
296: Taurus and Ophiuchus and found that most cores fall within the 9-13K
297: range. The peak of the \tc\ distribution shown in Figure
298: \ref{TKINHIST} is between 10-11K, which agrees well with the median
299: temperature (11 K) derived from \amm\ observations of the 36 starless
300: as observed and identified by \citet{Jijina99} across many nearby star
301: forming regions. The lower and upper quartiles of the temperatures of
302: the \citet{Jijina99} starless core sample (9.7 K and 14.0 K) are also
303: a close match to that shown in Figure \ref{TKINHIST}.
304:
305: Numerical simulations of starless cores have been used to quantify how
306: their temperature profile is affected by variations in many
307: parameters, such as the strength of the ISRF, the density profile, and
308: the amount of extinction provided by the molecular cloud in which the
309: cores are embedded \citep[e.g.][]{Evans01, Galli02, Stamatellos03,
310: Stamatellos04}. \citet{Evans01} calculate that for Bonnor-Ebert
311: spheres with central densities varying by about a factor of 100 (from
312: 10$^5$ to 10$^7$ cm$^{-3}$), the central temperature varies by a
313: factor of $\sim$2 (6.5 - 10 K), but the temperature at a radius of
314: 10$^{3.5}$ AU (the radius of the GBT beam at the distance of Perseus)
315: only varies by $\sim$1 K (9 - 10 K). A doubling of the strength of
316: the ISRF results in a 1 K increase of the core temperature file over
317: the range of radii probed by the \amm\ observations in this study
318: \citep{Evans01}, while a factor of 100 increase in the ISRF would
319: result in an increase in the core temperature by a factor of $\sim$2
320: \citep{Galli02}. Variations in the dust opacity also affect the
321: temperature of starless core, but \citet{Goncalves04} show that this
322: effect is quite small ($< 1$ K) for the range of dust properties that
323: one expects in these cores. Given that the temperature of a
324: prestellar core (at a radius of 10$^{3.5}$ AU) is not especially
325: sensitive to its central density, the dust opacity, the strength of
326: the ISRF or the degree of extinction for the range of values that we
327: expect in Perseus, the narrow width of the temperature distribution
328: shown in Figure \ref{TKINHIST} and its peak at 10 K are roughly what
329: one would expect.
330:
331: \subsection{\tc\ vs.~Location} \label{TGDIST}
332:
333: The presence of local star formation appears to affect \tc.
334: Fig.~\ref{TKINHIST} shows the temperature distribution of all the
335: starless cores in Perseus (filled grey histogram), as well as that for
336: starless cores in the clusters NGC1333 and IC348 (open solid
337: histogram) and field starless cores (open dashed histogram). The
338: boundaries for NGC1333 and IC348 are taken from \citet{Jorgensen07}.
339: It is clear that those cores in clusters are warmer than those outside
340: clusters. The probability that the temperature distributions of
341: starless cores inside and outside clusters have the same mean is
342: $\simeq 10^{-5}$, as calculated by a Wilcoxon Rank-Sum Test, Student's
343: T-statistic or two-sided Kolmogorov-Smirnov test.
344:
345: There are reasons to expect that starless cores in clusters would be
346: warmer than those outside clusters. For instance, a cluster of
347: protostars may directly heat their environment. In addition,
348: protostars inject turbulence into the surrounding medium, making it
349: more porous and less effective as a shield from the ISRF. In
350: addition, \citet{Jijina99} find that dense cores that are not
351: associated with {\it IRAS} sources are warmer when found inside
352: clusters than outside.
353:
354: \citet{Stamatellos07} predict that the presence of a nearby Class 0
355: protostar should not affect the temperature of a nearby starless core
356: at all, and that a nearby Class I protostar will only heat the outer
357: layers of a core, and only by $\sim1-2$K. We test this by comparing
358: the \amm-derived temperatures of starless cores in Perseus to the
359: distance to the nearest Class 0 and I protostars (as determined by
360: \citet{Enoch08}). As shown in Fig.~\ref{TKINENVIRONMENT} and Table
361: \ref{CORRELTAB}, we confirm that there is no significant correlation
362: between \tc\ and the distance to the nearest protostar for those
363: starless cores in the clusters NGC1333 and IC348. However, there is a
364: significant negative correlation between \tc\ and the distance to the
365: nearest class 0 and/or I protostar for those starless cores in the
366: field (i.e. starless cores further from protostars are colder). It is
367: not at all certain that this result contradicts the prediction of
368: \citet{Stamatellos07}, because we cannot say that nearby protostars
369: heat starless cores, just that warmer cores are found near protostars.
370:
371: \subsection{\tc\ vs. Cloud Temperature} \label{TGTDTCO}
372:
373: Here we compare the temperatures of starless cores with the
374: temperature of the material in which it is embedded. We estimate the
375: line-of-sight averaged temperature of the dust (\td) \citep{Schnee08}
376: and the gas (\tco) \citep{Pineda08} in the Perseus molecular cloud and
377: compare it with the temperature derived from the \amm\ pointings. The
378: resolution of the \td\ and \tco\ maps are both $\sim$40\arcsec, which
379: is roughly comparable to the 31\arcsec\ resolution of the ammonia
380: data. However, both \td\ and \tco\ are imperfect measures of cloud
381: temperature. Because $^{12}$CO is optically thick (and perhaps
382: depleted) it is not sensitive to the colder and denser cloud material
383: in which the starless cores are embedded. The dust temperature
384: determination is emission-weighted, and at the FIR wavelengths (70 and
385: 160 \micron) of the Spitzer maps the emission is dominated by warm
386: dust along the line of sight, and therefore is less sensitive to the
387: colder and denser clumps in which cores form.
388:
389: Nevertheless, one might expect that the temperature of a starless core
390: could be affected, or predicted, by the temperature of the medium in
391: which is it embedded. Starless cores are, by definition, externally
392: heated, so whatever is heating the cloud should also be heating the
393: embedded cores. This analysis can be biased because the gas and dust
394: derived temperatures are warmer within the clusters than outside,
395: which is also true for \tc, so any derived correlations must take this
396: into account. Figure \ref{TKINENVIRONMENT} and Tab.~\ref{CORRELTAB}
397: show that although both \tco\ and \td\ are significantly and
398: positively correlated with \tc\ when considering the entire population
399: of starless cores, this effect is entirely explained by the
400: cluster/field bias, and within each sub-population there is no
401: significant correlation.
402:
403: \subsection{\tc\ vs.~Column Density} \label{TGNH}
404:
405: The heating of starless cores is governed by the strength of the local
406: interstellar radiation field (ISRF). Given that there are no O or B
407: stars within the Perseus molecular cloud, the ISRF seen by the cores
408: should depend largely on dust shielding but not on the distance to any
409: special source of external radiation. Because we are unable to
410: measure the volume-averaged shielding around each starless core, we
411: use as a proxy the column density along the line of sight. Based on
412: this simple analysis, one would predict that \tc\ would be
413: anti-correlated with column density.
414:
415: On the other hand, for starless cores that are in pressure
416: equilibrium, the inward pressures due to the weight of the Perseus
417: molecular cloud and the self-gravity of the cores will be counter
418: balanced by the internal gas pressure from thermal and non-thermal
419: motion. The external pressure on the starless cores from the
420: molecular cloud is proportional to the square of the column density
421: ($P_{\rm cloud} \propto A_V^2$) \citep{Bertoldi92, Lada08} and the
422: internal gas pressure from thermal motions is proportial to \tc\
423: ($P_{\rm thermal} \propto \sigma_{\rm T}^2 \propto$ \tc). From a
424: balance of pressure analysis, one would expect a positive correlation
425: between \tc\ and column density, though this would not necessarily be
426: a strong effect because outward pressure is also supplied by
427: non-thermal motions and magnetic fields. We calculate that for the
428: starless cores in our sample the inward pressure of the weight of the
429: molecular cloud is roughly a factor of 2 weaker than their
430: self-gravity.
431:
432: Here we compare the temperatures of starless cores with the column
433: density of molecular cloud material along the line of sight to each
434: core. We estimate the cloud column density using far-infrared dust
435: emission \citep{Schnee08} and near-infrared extinction
436: \citep{Ridge06}. Although extinction may be the most secure estimator
437: of column density in a molecular cloud \citep{Goodman08}, the
438: 40\arcsec\ resolution of the emission-derived column density map
439: compares favorably with the 5\arcmin\ resolution of the near-infrared
440: extinction map. As with the cloud temperature, there is an inherent
441: bias in that the column density in the clusters is higher and the core
442: temperatures are higher, so we consider the field and cluster starless
443: cores separately. Figure \ref{TKINENVIRONMENT} and Table
444: \ref{CORRELTAB} show that there is a significant positive correlation
445: between extinction-derived column density and \tc\ when considering
446: just the starless cores in the field and just the starless cores in
447: the clusters, as well as for the field+cluster ensemble. There is no
448: significant trend between \tc\ and the dust emission-derived column
449: density for field or cluster cores.
450:
451: One should keep in mind that the emission-derived column
452: density derived from dust emission loses accuracy at values of $A_V >$
453: a few, so it is not so surprising to get different results depending
454: on how the column density is estimated \citep{Schnee06}. The dust
455: temperature derived from FIR maps is emission-weighted, and the column
456: density derivation depends on the dust temperature, so the column
457: density along lines of sight with significant variations in
458: temperature and density is both systemically underestimated and
459: subject to large scatter. This effect could wash out any true
460: correlations between the column density of the cloud material and the
461: temperature of the core embedded in it.
462:
463: In a numerical simulation of embedded starless cores,
464: \citet{Stamatellos03} show that the temperature profile of a core
465: shielded by $A_V$ = 5 material (a typical value for $A_V$ seen towards
466: the starless cores in Perseus) is not significantly different from a
467: naked core, while deeply embedded cores ($A_V$ = 20) are noticably
468: colder than naked cores. In addition, we expect that the column
469: density along the line of sight is a poor indicator of volumetric
470: shielding due to the porosity of molecular clouds like Perseus. The
471: visual impression given by two-dimensional maps and
472: position-position-velocity cubes of nearby molecular clouds is one of
473: non-relaxed structures with many filaments, voids and shells,
474: supporting this view. The observational evidence supports the idea
475: that the shielding provided by higher \AV\ material in Perseus does
476: not lead to lower temperatures of the embedded starless cores. The
477: opposite effect is observed, which may be explained by the greater
478: thermal support needed to balance the increased pressure on the cores
479: by the weight of the molecular cloud.
480:
481: \subsection{\tc\ vs.~\sigamm\ and \sigco}
482: \label{TGSIG}
483:
484: Clumpiness formed by turbulence in a molecular cloud is known to
485: significantly affect the temperature of the cloud material, with
486: higher turbulence leading to greater porosity and penetration by the
487: ISRF, and therefore higher temperatures \citep{Bethell04}. However,
488: \citet{Stamatellos07} find that the densest portions of the cloud
489: (where starless cores are found) are primarily heated by long
490: wavelength radiation that is not effectively shielded by the clumpy
491: structure. It is therefore not obvious if one would expect starless
492: core temperatures to be correlated with turbulence.
493:
494: Here we examine whether turbulence affects the temperature of the
495: starless cores in the Perseus molecular cloud. We use the non-thermal
496: linewidth of \amm\ as a proxy for the turbulence within the cores, and
497: the total linewidth of $^{13}$CO as an estimate of the turbulence in
498: the surrounding medium (the measured $^{13}$CO linewidth is roughly a
499: factor of 10 higher than the thermal linewidth). As before, we
500: consider field and cluster starless cores separately to avoid
501: confusing a correlation between \tc\ and turbulence with a correlation
502: between \tc\ and the field vs.~cluster environment.
503:
504: Figure \ref{TKININTERNAL} and Table \ref{CORRELTAB} show that there is
505: no correlation between the non-thermal linewidth of \amm\ and \tc\
506: when considering the field and cluster popoluations of starless cores
507: separately, but there is a significant positive correlation for the
508: ensemble. This implies that those starless cores in Perseus are more
509: turbulent when found in clusters than in the field, but the turbulence
510: itself does not necessarily lead to higher temperatures. Our
511: expectation of starless core geometry is that they are in some sense
512: ``relaxed'' structures and therefore not porous, regardless of the
513: level of internal turbulence. CO linewidth does not correlate with
514: \tc\ for those cores in clusters, but it is significantly positively
515: correlated with \tc\ for field starless cores, as shown in Figure
516: \ref{TKINENVIRONMENT}. One interpretation of of the field/cluster
517: difference is that within clusters the CO linewidth is likely to be
518: affected strongly by the ongoing star formation, while the CO
519: linewidth in the field better traces the cloud geometry, so only in
520: the field does a larger CO linewidth imply better penetration by the
521: ISRF.
522:
523: \subsection{\tc\ vs.~1.1mm Continuum Derived Properties}
524:
525: It might be supposed that the temperatures of starless cores are
526: partly determined by the internal properties of the cores themselves,
527: and not just by their environment. Here we compare \tc\ with 1.1mm
528: Bolocam flux, the peak column density (expressed in $A_V$), the mass
529: and the average volume density derived from the 1.1mm map and
530: \amm-derived temperature of each core.
531:
532: The peak \AV\ is derived from the peak 1.1mm flux ($S_{\nu}^{beam}$)
533: of each core and the \amm-derived temperature.
534: \begin{equation}
535: N(\mathrm{H_2}) = \frac{S_{\nu}^{beam}}{\Omega_{beam} \mu_{H_2} m_H \kappa_{\nu} B_{\nu}(T_{\rm kin})}, \label{aveq}
536: \end{equation}
537: with $N($H$_2)/A_V = 0.94 \times 10^{21}$ cm$^{2}$~mag$^{-1}$
538: \citep{Frerking82}. Here $\Omega_{beam}$ is the beam solid angle,
539: $m_H$ is the mass of hydrogen, $\kappa_{1.1mm} = 0.0114$ cm$^2$
540: g$^{-1}$ is the dust opacity per gram of gas interpolated from Table~1
541: column~5 of \citet{Ossenkopf94} for dust grains with thin ice mantles,
542: $B_{\nu}$ is the Planck function, and $\mu_{H_2}=2.8$ is the mean
543: molecular weight per H$_2$ molecule. A gas to dust mass ratio of 100
544: is included in $\kappa_{1.1mm}$.
545:
546: The core mass is derived from the total 1mm flux ($S_{\nu}$) and \tc\
547: using the equation,
548: \begin{equation}
549: M = \frac{d^2 S_{\nu}}{B_{\nu}(T_{\rm kin}) \kappa_{\nu}}, \label{masseq}
550: \end{equation}
551: where $\kappa_{\nu}$ is the dust opacity, $d$ is the distance to
552: Perseus (250 pc). The density is derived from the core mass and the
553: deconvolved average FWHM size ($D$) of the core ($n = 6M/\pi D^3
554: \mu_p$), where $\mu_p = 2.33$ is the mean molecular weight per
555: particle, as explained in \citet{Enoch06}.
556:
557: When deriving the core mass, peak column density and average volume
558: density from the 1.1mm continuum emission, we use the gas temperature
559: (\tc) in the calculation, under the assumption that the gas and dust
560: temperatures are coupled. Although this assumption is valid for the
561: densities ($n > 2 \times 10^4$ cm$^{-3}$) derived for these cores
562: \citep{Goldsmith01, Galli02}, we are introducing a potential bias in
563: our analysis because \tc\ and the derived mass are not independent.
564: It would be preferable to use the T$_{\rm dust}$ rather than \tc\ to
565: determine the 1.1mm-derived core properties, but this requires mapping
566: the dust emission from each core at several wavelengths, sampling both
567: the peak of the spectral energy distribution (SED) as well as the
568: Rayleigh-Jeans tail to derive the dust temperature, column desnity and
569: emissivity spectral index profiles. Existing surveys of Perseus
570: \citep{Enoch06, Kirk06} have only sampled the Rayleigh-Jeans portion
571: of the SED, so we do not have accurate estimates of the dust
572: temperatures within the starless cores, and we are forced to use the
573: gas temperature in our calculations.
574:
575: Note that some of the Bolocam cores, whose numbers are given in Table
576: \ref{COREPROPTAB} (Per-Bolo 24, 75, 100 and 122), have been observed
577: in multiple GBT pointings with each position corresponding to
578: different sub-condensations within the core. Each of these pointings
579: is assigned the same 1.1mm total and peak flux and dimensions as that
580: of the entire Bolocam core, and the peak $A_V$, mass and density are
581: all derived from the overall core Bolocam flux measurements and the
582: \tc\ derived for the sub-core. An alternative analysis would have
583: been to break up each Bolocam core with more than one \amm\ pointing
584: into sub-cores and assign different fractions of the total 1.1mm flux
585: to each portion, but the procedure to do the division would be
586: somewhat arbitrary, just as the decision of which cores should be
587: observed only once and which in multiple positions was not based on an
588: absolute prescription. Since neither analysis is obviously the
589: correct one, we decided to go with the simpler of the two and
590: double-count the 1.1mm properties of the four Bolocam cores with more
591: than one \amm\ observation.
592:
593: In Fig.~\ref{TKININTERNAL} and Table \ref{CORRELTAB} we compare \tc\
594: with the 1.1mm flux and properties derived from the \citet{Enoch06}
595: Bolocam map of Perseus. We find no correlation between \tc\ and the
596: 1.1mm flux, core mass, average volume density or peak \AV\ for those
597: cores in IC348 and NGC1333. On the other hand, for starless cores in
598: the field there are significant negative correlations between \tc\ and
599: the peak \AV\ and average volume density, and a positive correlation
600: between \tc\ and the Bolocam flux. If we assume that starless cores
601: are effectively self-shielded from the ISRF because they are not
602: porous, then we would expect higher peak column densities and volume
603: densities to correspond to higher volumetric shielding and lower
604: internal temperatures. It is not obvious why the self-shielding
605: argument does not hold for those starless cores in NGC1333 and IC348.
606: The connection between 1.1mm flux and \tc\ may be simply that for a
607: given mass a warmer core will produce more emission, but this argument
608: does not explain why the positive correlation is not seen in cluster
609: starless cores.
610:
611: \section{Conclusions} \label{CONCLUSIONS}
612:
613: In this paper we investigate the temperature distribution of starless
614: cores in Perseus. Although, the range of temperatures derived from
615: \amm\ (1,1) and (2,2) pointing is quite narrow, from $\sim 9-15$ K,
616: there are several interesting correlations and lack of correlations.
617:
618: The location of a starless core does affect its temperature. We find
619: that those cores inside the two main star-forming regions in Perseus
620: (IC348 and NGC1333) are warmer than those not in clusters. Although
621: there is no correlation between \tc\ and the distance to the nearest
622: class 0/I protostar for those starless cores in clusters, starless
623: cores in the field are colder the further away they are from
624: protostars.
625:
626: Certain properties of the molecular cloud are related to the
627: temperatures of the starless cores embedded in it. \tc\ is higher
628: when the extinction-derived column density of the molecular cloud is
629: larger, despite the additional shielding from the ISRF that the higher
630: column density implies, which may be due to the increased pressure on
631: the cores from the molecular cloud in higher column density regions.
632: This correlation is not seen when column density along the line of
633: sight in Perseus is measured from far-infrared dust emission. After
634: controlling for the cluster/field temperature difference, we find that
635: \tc\ does not depend on the temperature of the molecular cloud, as
636: measured by CO or dust emission. The clusters have higher values of
637: \tco\ and \td\ on average, so presumably the same processes that are
638: heating the clusters are also heating the cores within them. Those
639: starless cores outside NGC1333 and IC348 are warmer in regions with
640: higher $^{13}$CO linewidth, suggesting the the turbulence causing the
641: high linewidths makes the cloud more porous to the ISRF, which in turn
642: heats the embedded cores. There is no significant correlation between
643: $\sigma_{CO}$ and \tc\ for those cores in the clusters.
644:
645: Some internal properties of starless cores are correlated with their
646: temperature. For those starless cores in the field, total 1.1mm flux
647: is higher in cores that are warmer, and peak column density and
648: average volume density are higher in cores that are colder. A higher
649: temperature naturally leads to more 1.1mm flux for a given core mass,
650: and the correlations between \tc\ and peak core column density and
651: volume density can be explained by the amount of shielding from the
652: ISRF that the additional material provides. However, it is surprising
653: that none of these correlations exist for starless cores inside the
654: clusters. Support was provided to MLE by NASA through the Spitzer
655: Space Telescope Fellowship Program.
656:
657: \acknowledgements
658:
659: We thank an anonymous referee for comments that have improved this
660: paper. SS acknowledges support from the Owens Valley Radio
661: Observatory, which is supported by the National Science Foundation
662: through grant AST 05-40399. ER was supported by an NSF Astronomy and
663: Astrophysics Postdoctoral Fellowship (AST-0502605) and a Discovery
664: Grant from NSERC of Canada. JBF is supported by a grant from the NRAO
665: Student Observing Support Program (GSSP06-0015)
666:
667: \begin{thebibliography}{}
668: \bibitem[Belikov et al.(2002)]{Belikov02} Belikov, A.~N., Kharchenko,
669: N.~V., Piskunov, A.~E., Schilbach, E., \& Scholz, R.-D.\ 2002, \aap,
670: 387, 117
671: \bibitem[Bergin \& Tafalla(2007)]{Bergin07} Bergin, E.~A., \& Tafalla,
672: M.\ 2007, \araa, 45, 339
673: \bibitem[Bertoldi \& McKee(1992)]{Bertoldi92} Bertoldi, F., \& McKee,
674: C.~F.\ 1992, \apj, 395, 140
675: \bibitem[Bethell et al.(2004)]{Bethell04} Bethell, T.~J., Zweibel,
676: E.~G., Heitsch, F., \& Mathis, J.~S.\ 2004, \apj, 610, 801
677: \bibitem[{\v C}ernis \& Strai{\v z}ys(2003)]{Cernis03} {\v C}ernis,
678: K., \& Strai{\v z}ys, V.\ 2003, Baltic Astronomy, 12, 301
679: \bibitem[Crapsi et al.(2007)]{Crapsi07} Crapsi, A., Caselli, P.,
680: Walmsley, M.~C., \& Tafalla, M.\ 2007, \aap, 470, 221
681: \bibitem[Dunham et al.(2008)]{Dunham08} Dunham, M.~M., Crapsi, A.,
682: Evans, N.~J., II, Bourke, T.~L., Huard, T.~L., Myers, P.~C., \&
683: Kauffmann, J.\ 2008, ArXiv e-prints, 806, arXiv:0806.1754
684: \bibitem[Enoch et al.(2008b)]{Enoch08b} Enoch, M.~L. et al.\ 2008b, in
685: prep.
686: \bibitem[Enoch et al.(2008)]{Enoch08} Enoch, M.~L., Evans, N.~J., II,
687: Sargent, A.~I., Glenn, J., Rosolowsky, E., \& Myers, P.\ 2008, ArXiv
688: e-prints, 805, arXiv:0805.1075
689: \bibitem[Enoch et al.(2006)]{Enoch06} Enoch, M.~L., et al.\ 2006,
690: \apj, 638, 293
691: \bibitem[Evans et al.(2001)]{Evans01} Evans, N.~J., II, Rawlings,
692: J.~M.~C., Shirley, Y.~L., \& Mundy, L.~G.\ 2001, \apj, 557, 193
693: \bibitem[Evans et al.(2007)]{Evans07} Evans, N.~J., II, et al.\ 2007,
694: Final Delivery of Data from the c2d Legacy Project: IRAC and MIPS
695: \bibitem[Evans et al.(2003)]{Evans03} Evans, N.~J., II, et al.\ 2003,
696: \pasp, 115, 965
697: \bibitem[Frerking et al.(1982)Frerking, Langer, \& Wilson]{Frerking82}
698: Frerking, M. A., Langer, W. D., \& Wilson, R. W. 1982, \apj, 262, 590
699: \bibitem[Galli et al.(2002)]{Galli02} Galli, D., Walmsley, M., \&
700: Gon{\c c}alves, J.\ 2002, \aap, 394, 275
701: \bibitem[Goldsmith(2001)]{Goldsmith01} Goldsmith, P.~F.\ 2001, \apj,
702: 557, 736
703: \bibitem[Gon{\c c}alves et al.(2004)]{Goncalves04} Gon{\c c}alves, J.,
704: Galli, D., \& Walmsley, M.\ 2004, \aap, 415, 617
705: \bibitem[Goodman et al.(1998)]{Goodman98} Goodman, A.~A., Barranco,
706: J.~A., Wilner, D.~J., \& Heyer, M.~H.\ 1998, \apj, 504, 223
707: \bibitem[Goodman et al.(2008)]{Goodman08} Goodman, A.~A., Pineda,
708: J. \& Schnee, S.\ \apj, submitted
709: \bibitem[Harvey et al.(2007)]{Harvey07} Harvey, P., Mer{\'{\i}}n, B.,
710: Huard, T.~L., Rebull, L.~M., Chapman, N., Evans, N.~J., II, \&
711: Myers, P.~C.\ 2007, \apj, 663, 1149
712: \bibitem[Hatchell et al.(2007)]{Hatchell07} Hatchell, J., Fuller,
713: G.~A., Richer, J.~S., Harries, T.~J., \& Ladd, E.~F.\ 2007, \aap,
714: 468, 1009
715: \bibitem[Jijina et al.(1999)]{Jijina99} Jijina, J., Myers, P.~C., \&
716: Adams, F.~C.\ 1999, \apjs, 125, 161
717: \bibitem[J{\o}rgensen et al.(2007)]{Jorgensen07} J{\o}rgensen, J.~K.,
718: Johnstone, D., Kirk, H., \& Myers, P.~C.\ 2007, \apj, 656, 293
719: \bibitem[Kirk et al.(2006)]{Kirk06} Kirk, H., Johnstone, D., \& Di
720: Francesco, J.\ 2006, \apj, 646, 1009
721: \bibitem[Kirk et al.(2007)]{Kirk07} Kirk, J.~M., Ward-Thompson, D., \&
722: Andr{\'e}, P.\ 2007, \mnras, 375, 843
723: \bibitem[Lada et al.(2008)]{Lada08} Lada, C.~J., Muench, A.~A.,
724: Rathborne, J., Alves, J.~F., \& Lombardi, M.\ 2008, \apj, 672, 410
725: \bibitem[Lombardi \& Alves(2001)]{Lombardi01} Lombardi, M., \& Alves,
726: J.\ 2001, \aap, 377, 1023
727: \bibitem[Miville-Desch{\^e}nes \& Lagache(2005)]{Miville05}
728: Miville-Desch{\^e}nes, M.-A., \& Lagache, G.\ 2005, \apjs, 157, 302
729: \bibitem[Ossenkopf \& Henning(1994)]{Ossenkopf94} Ossenkopf, V., \&
730: Henning, T.\ 1994, \aap, 291, 943
731: \bibitem[Pineda et al.(2008)]{Pineda08} Pineda, J., Caselli, P., \&
732: Goodman, A.\ 2008 \apj, accepted
733: \bibitem[Rebull et al.(2007)]{Rebull07} Rebull, L.~M., et al.\ 2007,
734: \apjs, 171, 447
735: \bibitem[Ridge et al.(2006)]{Ridge06} Ridge, N.~A., et al.\ 2006, \aj,
736: 131, 2921
737: \bibitem[Rosolowsky et al.(2008)]{Rosolowsky08} Rosolowsky, E.~W.,
738: Pineda, J.~E., Foster, J.~B., Borkin, M.~A., Kauffmann, J., Caselli,
739: P., Myers, P.~C., \& Goodman, A.~A.\ 2008, \apjs, 175, 509
740: \bibitem[Schnee et al.(2006)]{Schnee06} Schnee, S., Bethell, T., \&
741: Goodman, A.\ 2006, \apjl, 640, L47
742: \bibitem[Schnee et al.(2008)]{Schnee08} Schnee, S., Li, J., Goodman,
743: A.~A., \& Sargent, A.~I.\ 2008, ArXiv e-prints, 805, arXiv:0805.4215
744: \bibitem[Stamatellos \& Whitworth(2003)]{Stamatellos03} Stamatellos,
745: D., \& Whitworth, A.~P.\ 2003, \aap, 407, 941
746: \bibitem[Stamatellos et al.(2004)]{Stamatellos04} Stamatellos, D.,
747: Whitworth, A.~P., Andr{\'e}, P., \& Ward-Thompson, D.\ 2004, \aap,
748: 420, 1009
749: \bibitem[Stamatellos et al.(2007)]{Stamatellos07} Stamatellos, D.,
750: Whitworth, A.~P., \& Ward-Thompson, D.\ 2007, \mnras, 379, 1390
751: \bibitem[Swade(1989)]{Swade89} Swade, D.~A.\ 1989, \apj, 345, 828
752: \bibitem[Tafalla et al.(2002)]{Tafalla02} Tafalla, M., Myers, P.~C.,
753: Caselli, P., Walmsley, C.~M., \& Comito, C.\ 2002, \apj, 569, 815
754: \bibitem[Walmsley \& Ungerechts(1983)]{Walmsley83} Walmsley, C.~M., \&
755: Ungerechts, H.\ 1983, \aap, 122, 164
756: \bibitem[Zucconi et al.(2001)]{Zucconi01} Zucconi, A., Walmsley,
757: C.~M., \& Galli, D.\ 2001, \aap, 376, 650
758: \end{thebibliography}
759:
760: \clearpage
761:
762: \begin{deluxetable}{ccc|cccccc}
763: \tablewidth{0pt}
764: \tabletypesize{\scriptsize}
765: \tablecaption{Internal Properties of Starless Cores \label{COREPROPTAB}}
766: \tablehead{
767: \colhead{RA\tablenotemark{1}} &
768: \colhead{Dec\tablenotemark{2}} &
769: \colhead{\tc\tablenotemark{3}} &
770: \colhead{\sigamm\tablenotemark{4}} &
771: \colhead{1.1mm Flux\tablenotemark{5}} &
772: \colhead{Peak \AV\tablenotemark{6}} &
773: \colhead{Mass\tablenotemark{7}} &
774: \colhead{Density\tablenotemark{8}} &
775: \colhead{Number\tablenotemark{9}} \\
776: \colhead{J2000} & \colhead{J2000} &
777: \colhead{K} & \colhead{\kms} &
778: \colhead{Jy} & \colhead{mag} &
779: \colhead{M$_{solar}$} & \colhead{\ccm} &
780: \colhead{}}
781: \startdata
782: 3:25:07.8 & 30:24:22 & 9.2 & 0.09 & 0.1 & 9.1 & 0.3 & 1.4E+05 & 1 \\
783: 3:25:09.7 & 30:23:53 & 9.2 & 0.14 & 0.1 & 10.1 & 0.4 & 3.3E+05 & 2 \\
784: 3:25:10.1 & 30:44:41 & 10.0 & 0.12 & 0.4 & 9.0 & 0.9 & 2.5E+04 & 3 \\
785: 3:25:17.1 & 30:18:53 & 10.3 & 0.11 & 1.2 & 10.1 & 2.8 & 4.5E+04 & 4 \\
786: 3:25:26.9 & 30:21:53 & 9.1 & 0.10 & 1.2 & 12.3 & 3.4 & 5.3E+04 & 6 \\
787: 3:25:46.1 & 30:44:11 & 11.2 & 0.18 & 0.9 & 14.2 & 1.8 & 8.5E+04 & 11 \\
788: 3:25:48.8 & 30:42:24 & 9.1 & 0.13 & 0.5 & 34.6 & 1.3 & 3.8E+05 & 13 \\
789: 3:25:50.6 & 30:42:02 & 9.0 & 0.14 & 0.4 & 29.6 & 1.2 & 3.2E+05 & 14 \\
790: 3:27:02.1 & 30:15:08 & 10.4 & 0.13 & 0.6 & 7.8 & 1.3 & 4.2E+04 & 19 \\
791: 3:27:28.9 & 30:15:04 & 10.7 & 0.10 & 0.7 & 7.3 & 1.5 & 5.1E+04 & 20 \\
792: 3:27:40.0 & 30:12:13 & 10.5 & 0.17 & 0.4 & 18.6 & 0.9 & 1.6E+05 & 23 \\
793: 3:27:55.9 & 30:06:18 & 9.1 & 0.05 & 0.7 & 18.9 & 1.9 & 1.9E+05 & 24 \\
794: 3:28:05.5 & 30:06:19 & 9.4 & 0.11 & 0.7 & 17.7 & 1.8 & 1.8E+05 & 24 \\
795: 3:28:32.4 & 31:04:43 & 10.6 & 0.13 & 0.5 & 7.6 & 1.1 & 7.1E+04 & 26 \\
796: 3:28:33.4 & 30:19:35 & 10.7 & 0.12 & 1.0 & 10.0 & 2.2 & 3.7E+04 & 27 \\
797: 3:28:39.5 & 31:18:35 & 11.7 & 0.17 & 1.9 & 21.2 & 3.6 & 2.3E+05 & 31 \\
798: 3:28:48.5 & 31:16:03 & 11.3 & 0.10 & 0.2 & 7.6 & 0.4 & 1.8E+05 & 35 \\
799: 3:28:52.2 & 31:18:08 & 13.8 & 0.17 & 0.4 & 6.8 & 0.5 & 1.4E+05 & 37 \\
800: 3:29:18.5 & 31:25:13 & 11.9 & 0.12 & 1.4 & 18.1 & 2.5 & 1.4E+05 & 53 \\
801: 3:29:22.5 & 31:36:24 & 9.8 & 0.12 & 0.6 & 7.6 & 1.5 & 4.6E+04 & 56 \\
802: 3:29:25.8 & 31:28:17 & 10.3 & 0.11 & 0.3 & 18.8 & 0.8 & 6.3E+05 & 58 \\
803: 3:30:24.1 & 30:27:39 & 10.5 & 0.07 & 0.4 & 7.0 & 0.8 & 4.2E+04 & 61 \\
804: 3:30:45.6 & 30:52:36 & 10.5 & 0.13 & 0.7 & 9.4 & 1.5 & 5.5E+04 & 63 \\
805: 3:30:50.5 & 30:49:17 & 9.8 & 0.08 & 0.2 & 6.4 & 0.5 & 9.9E+04 & 64 \\
806: 3:32:26.9 & 30:59:11 & 10.5 & 0.13 & 0.9 & 11.0 & 1.9 & 1.0E+05 & 67 \\
807: 3:32:44.1 & 31:00:01 & 10.2 & 0.11 & 2.0 & 16.7 & 4.6 & 9.2E+04 & 70 \\
808: 3:32:51.3 & 31:01:48 & 11.2 & 0.22 & 1.1 & 10.7 & 2.3 & 7.2E+04 & 71 \\
809: 3:32:57.0 & 31:03:21 & 10.1 & 0.12 & 0.9 & 16.8 & 2.2 & 1.6E+05 & 72 \\
810: 3:33:02.0 & 31:04:33 & 10.0 & 0.14 & 0.4 & 18.8 & 0.9 & 1.8E+05 & 74 \\
811: 3:33:04.3 & 31:04:57 & 9.9 & 0.18 & 0.4 & 19.2 & 0.9 & 1.8E+05 & 75 \\
812: 3:33:06.3 & 31:06:26 & 10.3 & 0.15 & 0.4 & 17.7 & 0.9 & 1.6E+05 & 75 \\
813: 3:33:25.2 & 31:05:35 & 9.7 & 0.10 & 0.4 & 11.1 & 0.9 & 1.5E+05 & 82 \\
814: 3:40:49.5 & 31:48:35 & 12.4 & 0.08 & 1.0 & 7.6 & 1.7 & 3.5E+04 & 89 \\
815: 3:41:40.2 & 31:58:05 & 9.5 & 0.07 & 0.2 & 7.5 & 0.6 & 8.7E+04 & 92 \\
816: 3:41:46.0 & 31:57:22 & 9.6 & 0.07 & 0.3 & 10.2 & 0.8 & 1.0E+05 & 94 \\
817: 3:43:44.0 & 32:02:52 & 12.9 & 0.13 & 0.8 & 11.7 & 1.3 & 1.5E+05 & 100 \\
818: 3:43:45.5 & 32:01:44 & 14.0 & 0.10 & 0.3 & 5.3 & 0.4 & 8.0E+04 & 101 \\
819: 3:43:45.8 & 32:03:11 & 11.7 & 0.16 & 0.8 & 13.8 & 1.5 & 1.7E+05 & 100 \\
820: 3:43:57.8 & 32:04:06 & 12.9 & 0.16 & 0.6 & 13.3 & 1.0 & 1.6E+05 & 105 \\
821: 3:44:05.1 & 32:00:28 & 10.7 & 0.20 & 0.4 & 8.2 & 0.8 & 9.4E+04 & 109 \\
822: 3:44:05.3 & 32:02:05 & 10.9 & 0.14 & 0.4 & 16.0 & 0.9 & 1.7E+05 & 110 \\
823: 3:44:14.6 & 31:57:59 & 10.6 & 0.12 & 0.8 & 11.1 & 1.8 & 6.9E+04 & 111 \\
824: 3:44:36.4 & 31:58:40 & 10.8 & 0.15 & 0.5 & 10.5 & 1.1 & 1.1E+05 & 115 \\
825: 3:44:48.8 & 32:00:29 & 10.8 & 0.10 & 0.4 & 9.4 & 0.8 & 1.6E+05 & 117 \\
826: 3:44:56.1 & 32:00:32 & 10.9 & 0.12 & 0.3 & 6.8 & 0.7 & 7.4E+04 & 118 \\
827: 3:45:15.9 & 32:04:49 & 10.7 & 0.08 & 1.3 & 11.6 & 2.8 & 4.9E+04 & 119 \\
828: 3:47:33.5 & 32:50:55 & 10.0 & 0.08 & 0.7 & 9.9 & 1.6 & 5.0E+04 & 121 \\
829: 3:47:38.6 & 32:52:19 & 10.1 & 0.17 & 1.3 & 18.5 & 3.2 & 1.8E+05 & 122 \\
830: 3:47:39.7 & 32:53:57 & 10.1 & 0.10 & 1.3 & 18.5 & 3.2 & 1.8E+05 & 122 \\
831: 3:47:39.8 & 32:53:34 & 10.0 & 0.11 & 1.3 & 18.8 & 3.2 & 1.8E+05 & 122
832: \enddata
833: \tablenotetext{1}{J2000 position of \amm\ pointing, $^2$J2000 position
834: of \amm\ pointing, $^3$Kinetic temperature of core derived from \amm\
835: (1,1) and (2,2) spectra, $^4$Non-thermal 1-D velocity dispersion of
836: \amm\ spectra, $^5$Total 1.1mm flux of the Bolocam core, $^6$Peak
837: column density of the core derived from the peak 1.1mm flux and \tc,
838: $^7$Total mass of Bolocam core derived from the integrated 1.1mm flux
839: and \tc, $^8$Average density of free particles of Bolocam core derived
840: from its mass and size, $^9$Per-Bolo number taken from Table 4 in
841: \citet{Enoch08}}
842: \end{deluxetable}
843:
844: \begin{deluxetable}{ccc|cccccccc}
845: \tablewidth{0pt}
846: \tabletypesize{\scriptsize}
847: \tablecaption{Environment of Starless Cores \label{COREENVIRTAB}}
848: \tablehead{
849: \colhead{RA\tablenotemark{1}} &
850: \colhead{Dec\tablenotemark{2}} &
851: \colhead{\tc\tablenotemark{3}} &
852: \colhead{\td\tablenotemark{4}} &
853: \colhead{\tco\tablenotemark{5}} &
854: \colhead{NIR \AV\tablenotemark{6}} &
855: \colhead{FIR \AV\tablenotemark{7}} &
856: \colhead{\sigco\tablenotemark{8}} &
857: \colhead{Distance Class 0/I\tablenotemark{9}} &
858: \colhead{In a Cluster?\tablenotemark{10}} \\
859: \colhead{J2000} & \colhead{J2000} &
860: \colhead{K} & \colhead{K} &
861: \colhead{K} & \colhead{mag} &
862: \colhead{mag} & \colhead{\kms} &
863: \colhead{pc} & \colhead{}}
864: \startdata
865: 3:25:07.8 & 30:24:22 & 9.2 & 14.6 & 11.6 & 5.0 & 2.6 & 0.5 & 1.51 & N \\
866: 3:25:09.7 & 30:23:53 & 9.2 & 14.6 & 11.2 & 5.0 & 2.6 & 0.6 & 1.50 & N \\
867: 3:25:10.1 & 30:44:41 & 10.0 & 15.2 & 12.9 & 5.1 & 4.0 & 0.6 & 0.19 & N \\
868: 3:25:17.1 & 30:18:53 & 10.3 & 12.9 & 12.0 & 5.0 & 7.2 & 0.7 & 1.28 & N \\
869: 3:25:26.9 & 30:21:53 & 9.1 & 12.9 & 12.7 & 3.7 & 6.7 & 0.5 & 1.20 & N \\
870: 3:25:46.1 & 30:44:11 & 11.2 & 15.5 & 11.0 & 5.5 & 5.1 & 0.7 & 0.12 & N \\
871: 3:25:48.8 & 30:42:24 & 9.1 & 14.3 & 11.2 & 4.8 & 5.4 & 0.7 & 0.20 & N \\
872: 3:25:50.6 & 30:42:02 & 9.0 & 12.5 & 10.6 & 4.8 & 10.4 & 0.7 & 0.24 & N \\
873: 3:27:02.1 & 30:15:08 & 10.4 & 13.5 & 13.1 & 4.8 & 3.8 & 0.6 & 0.39 & N \\
874: 3:27:28.9 & 30:15:04 & 10.7 & 14.9 & NaN & 9.5 & 2.9 & NaN & 0.16 & N \\
875: 3:27:40.0 & 30:12:13 & 10.5 & 17.4 & NaN & 8.8 & 2.8 & NaN & 0.04 & N \\
876: 3:27:55.9 & 30:06:18 & 9.1 & 13.7 & NaN & 3.1 & 3.1 & NaN & 0.17 & N \\
877: 3:28:05.5 & 30:06:19 & 9.4 & 13.7 & NaN & 2.9 & 3.1 & NaN & 0.16 & N \\
878: 3:28:32.4 & 31:04:43 & 10.6 & 12.8 & 18.8 & 5.4 & 8.3 & 1.1 & 0.14 & Y \\
879: 3:28:33.4 & 30:19:35 & 10.7 & 13.0 & NaN & 4.5 & 6.0 & NaN & 0.90 & N \\
880: 3:28:39.5 & 31:18:35 & 11.7 & 12.3 & 18.6 & 7.8 & 19.1 & 0.8 & 0.05 & Y \\
881: 3:28:48.5 & 31:16:03 & 11.3 & 16.0 & 16.2 & 8.5 & 0.0 & 0.9 & 0.15 & Y \\
882: 3:28:52.2 & 31:18:08 & 13.8 & 15.0 & 19.4 & 9.2 & 10.7 & 1.3 & 0.10 & Y \\
883: 3:29:18.5 & 31:25:13 & 11.9 & NaN & 49.0 & 6.5 & NaN & 0.7 & 0.13 & Y \\
884: 3:29:22.5 & 31:36:24 & 9.8 & 15.2 & 15.5 & 5.9 & 2.9 & 0.5 & 0.23 & N \\
885: 3:29:25.8 & 31:28:17 & 10.3 & 15.6 & 23.3 & 5.5 & 0.0 & 1.2 & 0.14 & Y \\
886: 3:30:24.1 & 30:27:39 & 10.5 & 14.2 & NaN & 7.0 & 5.5 & NaN & 0.16 & N \\
887: 3:30:45.6 & 30:52:36 & 10.5 & 13.3 & 12.4 & 2.7 & 4.8 & 0.9 & 0.74 & N \\
888: 3:30:50.5 & 30:49:17 & 9.8 & 13.5 & 11.6 & 2.1 & 2.8 & 0.8 & 0.54 & N \\
889: 3:32:26.9 & 30:59:11 & 10.5 & 13.4 & 13.5 & 10.3 & 9.0 & 0.7 & 0.23 & N \\
890: 3:32:44.1 & 31:00:01 & 10.2 & 12.2 & 12.8 & 10.0 & 21.7 & 0.9 & 0.30 & N \\
891: 3:32:51.3 & 31:01:48 & 11.2 & 12.3 & 13.7 & 8.8 & 20.3 & 0.9 & 0.36 & N \\
892: 3:32:57.0 & 31:03:21 & 10.1 & 12.4 & 14.4 & 8.5 & 15.4 & 0.8 & 0.39 & N \\
893: 3:33:02.0 & 31:04:33 & 10.0 & 12.5 & 13.9 & 8.5 & 16.7 & 0.9 & 0.27 & N \\
894: 3:33:04.3 & 31:04:57 & 9.9 & 12.5 & 14.0 & 8.5 & 15.3 & 0.8 & 0.23 & N \\
895: 3:33:06.3 & 31:06:26 & 10.3 & 12.7 & 13.3 & 8.5 & 13.3 & 0.9 & 0.14 & N \\
896: 3:33:25.2 & 31:05:35 & 9.7 & 12.5 & 15.8 & 8.6 & 19.1 & 0.9 & 0.10 & N \\
897: 3:40:49.5 & 31:48:35 & 12.4 & 16.1 & 15.4 & 7.7 & 9.1 & 0.9 & 0.42 & N \\
898: 3:41:40.2 & 31:58:05 & 9.5 & 16.3 & 21.3 & 8.4 & 7.5 & 0.6 & 1.09 & N \\
899: 3:41:46.0 & 31:57:22 & 9.6 & 15.6 & 19.7 & 8.4 & 10.3 & 0.7 & 1.09 & N \\
900: 3:43:44.0 & 32:02:52 & 12.9 & NaN & 21.8 & 9.5 & NaN & 0.6 & 0.11 & Y \\
901: 3:43:45.5 & 32:01:44 & 14.0 & 14.6 & 20.7 & 10.1 & 15.8 & 0.5 & 0.14 & Y \\
902: 3:43:45.8 & 32:03:11 & 11.7 & 15.1 & 21.4 & 8.7 & 21.3 & 0.6 & 0.07 & Y \\
903: 3:43:57.8 & 32:04:06 & 12.9 & 15.4 & 27.0 & 7.4 & 19.0 & 0.8 & 0.08 & Y \\
904: 3:44:05.1 & 32:00:28 & 10.7 & 14.7 & 21.9 & 9.0 & 14.8 & 0.8 & 0.13 & Y \\
905: 3:44:05.3 & 32:02:05 & 10.9 & NaN & 23.5 & 9.0 & NaN & 0.6 & 0.06 & Y \\
906: 3:44:14.6 & 31:57:59 & 10.6 & NaN & 17.0 & 7.9 & NaN & 0.7 & 0.16 & Y \\
907: 3:44:36.4 & 31:58:40 & 10.8 & 15.3 & 16.8 & 7.2 & 9.1 & 0.8 & 0.22 & Y \\
908: 3:44:48.8 & 32:00:29 & 10.8 & 14.8 & 16.9 & 5.9 & 12.5 & 0.6 & 0.10 & Y \\
909: 3:44:56.1 & 32:00:32 & 10.9 & 14.9 & 18.6 & 5.4 & 10.0 & 0.5 & 0.20 & Y \\
910: 3:45:15.9 & 32:04:49 & 10.7 & 14.3 & 24.5 & 5.2 & 12.3 & 0.5 & 0.55 & Y \\
911: 3:47:33.5 & 32:50:55 & 10.0 & 14.0 & 16.5 & 5.8 & 6.5 & 0.5 & 0.14 & N \\
912: 3:47:38.6 & 32:52:19 & 10.1 & 13.4 & 16.0 & 7.6 & 10.7 & 0.7 & 0.06 & N \\
913: 3:47:39.7 & 32:53:57 & 10.1 & 12.7 & 14.2 & 7.6 & 11.9 & 0.6 & 0.16 & N \\
914: 3:47:39.8 & 32:53:34 & 10.0 & 12.4 & 14.8 & 7.6 & 14.8 & 0.6 & 0.13 & N \\
915: \enddata
916: \tablenotetext{1}{J2000 position of \amm\ pointing, $^2$J2000 position
917: of \amm\ pointing, $^3$Kinetic temperature of core derived from \amm\
918: (1,1) and (2,2) spectra, $^4$Dust temperature of the Perseus molecular
919: cloud derived from FIR emission, $^5$Gas temperature of the Perseus
920: molecular cloud derived from $^{12}$CO (1--0) emission, $^6$Column
921: density of the Perseus molecular cloud derived from NIR extinction,
922: $^7$Column density of the Perseus molecular cloud derived from FIR
923: emission, $^8$ 1-D velocity dispersion of $^{13}$CO (1--0),
924: $^9$Distance to nearest Class 0/I protostar, $^{10}$Is the core within
925: a cluster (NGC1333 or IC348)?}
926: \end{deluxetable}
927:
928: \begin{deluxetable}{lccccc}
929: \tablewidth{0pt}
930: \tabletypesize{\scriptsize}
931: \tablecaption{\tc\ Correlations\label{CORRELTAB}}
932: \tablehead{
933: \colhead{Property} &
934: \colhead{Location} &
935: \colhead{r value\tablenotemark{1}} &
936: \colhead{t value\tablenotemark{2}} &
937: \colhead{confidence\tablenotemark{3}} &
938: \colhead{DOF\tablenotemark{4}}}
939: \startdata
940: \td & field+cluster & 0.27 & 1.83 & 96.3 & 44 \\
941: \tco & field+cluster & 0.45 & 3.26 & $>$99.9 & 42 \\
942: NIR \AV & field+cluster & 0.47 & 3.66 & $>$99.9 & 48 \\
943: FIR \AV & field+cluster & 0.34 & 2.43 & 99.0 & 44 \\
944: \sigco & field+cluster & 0.20 & 1.31 & 90.1 & 42 \\
945: \sigamm & field+cluster & 0.28 & 2.06 & 97.8 & 48 \\
946: 1.1mm Flux & field+cluster & 0.11 & 0.73 & 76.6 & 48 \\
947: Peak \AV & field+cluster & -0.33 & 2.39 & 99.0 & 48 \\
948: Mass & field+cluster & -0.11 & 0.78 & 78.0 & 48 \\
949: Density & field+cluster & -0.18 & 1.25 & 89.1 & 48 \\
950: dist 0\tablenotemark{5} & field+cluster & -0.45 & 2.29 & 98.3 & 21 \\
951: dist I\tablenotemark{5} & field+cluster & -0.21 & 1.14 & 86.8 & 29 \\
952: dist 0/I\tablenotemark{5} & field+cluster & -0.29 & 1.71 & 95.1 & 33 \\[3pt]
953: \hline \\
954: \td & field only & 0.16 & 0.93 & 82.0 & 31 \\
955: \tco & field only & 0.04 & 0.20 & 57.8 & 25 \\
956: NIR \AV & field only & 0.36 & 2.17 & 98.1 & 31 \\
957: FIR \AV & field only & 0.16 & 0.91 & 81.5 & 31 \\
958: \sigco & field only & 0.50 & 2.90 & 99.6 & 25 \\
959: \sigamm & field only & 0.27 & 1.55 & 93.5 & 31 \\
960: 1.1mm Flux & field only & 0.35 & 2.06 & 97.6 & 31 \\
961: Peak \AV & field only & -0.36 & 2.17 & 98.1 & 31 \\
962: Mass & field only & 0.15 & 0.86 & 80.1 & 31 \\
963: Density & field only & -0.54 & 3.58 & $>$99.1 & 31 \\
964: dist 0\tablenotemark{5} & field only & -0.61 & 2.32 & 97.7 & 9 \\
965: dist I\tablenotemark{5} & field only & -0.54 & 2.43 & 98.6 & 14 \\
966: dist 0/I\tablenotemark{5} & field only & -0.51 & 2.42 & 98.7 & 17 \\[3pt]
967: \hline \\
968: \td & cluster only & 0.06 & 0.19 & 57.4 & 11 \\
969: \tco & cluster only & 0.12 & 0.46 & 67.4 & 15 \\
970: NIR \AV & cluster only & 0.63 & 3.13 & 99.7 & 15 \\
971: FIR \AV & cluster only & 0.39 & 1.41 & 90.7 & 11 \\
972: \sigco & cluster only & -0.09 & 0.36 & 63.8 & 15 \\
973: \sigamm & cluster only & 0.14 & 0.56 & 70.8 & 15 \\
974: 1.1mm Flux & cluster only & -0.03 & 0.12 & 54.7 & 15 \\
975: Peak \AV & cluster only & -0.25 & 0.99 & 83.1 & 15 \\
976: Mass & cluster only & -0.19 & 0.76 & 77.0 & 15 \\
977: Density & cluster only & -0.19 & 0.75 & 76.8 & 15 \\
978: dist 0\tablenotemark{5} & cluster only & -0.26 & 0.86 & 79.5 & 10 \\
979: dist I\tablenotemark{5} & cluster only & -0.15 & 0.55 & 70.4 & 13 \\
980: dist 0/I\tablenotemark{5} & cluster only & -0.18 & 0.69 & 74.9 & 14 \\
981: \enddata
982: \tablenotetext{1}{coefficient of correlation ($r$), $^2$t statistic:
983: $r/{\rm sqrt}[(1-r^2)/{\rm DOF}]$, $^3$\% confidence of correlation,
984: $^4$degrees of freedom, $^5$distance to a Class 0/I source, as
985: identified in \citet{Enoch08b} and plotted in Figure
986: \ref{TKINENVIRONMENT}}
987: \end{deluxetable}
988:
989: \clearpage
990:
991: \begin{figure}
992: \epsscale{1.0}
993: \plotone{f1.eps}
994: \caption{Histogram of \tc. The filled (grey) histrogram is the
995: temperature distribution of all 50 of the starless cores in Perseus
996: for which we have well-fit \amm\ spectra. The dashed (red) and
997: solid (blue) histograms are the temperature distribution of the
998: starless cores in the field and inside the clusters, respectively.
999: \label{TKINHIST}}
1000: \end{figure}
1001:
1002: \begin{figure}
1003: \epsscale{1.0}
1004: \plotone{f2.eps}
1005: \caption{\tc\ plotted against properties of the enviroment around the
1006: starless cores: (top left) distance to nearest Class 0 protostar,
1007: (top middle) distance to nearest Class I protostar, (top right)
1008: distance to nearest Class 0 or Class I protostar (middle left) dust
1009: temperature, (center) gas temperature, (middle right) column density
1010: of the molecular cloud derived from NIR extinction, (bottom left)
1011: column density of the cloud derived from FIR emission and (bottom
1012: middle) $^{13}$CO line width (1-D velocity dispersion). In this
1013: plot Class 0 cores and Class I cores are identified as in
1014: \citet{Enoch08b}. Starless cores in NGC1333 and IC348 are plotted
1015: in black, and field starless cores are plotted in grey. The
1016: 1$\sigma$ errors in the fit of \tc\ are plotted for each
1017: core.\label{TKINENVIRONMENT}}
1018: \end{figure}
1019:
1020: \begin{figure}
1021: \epsscale{1.0}
1022: \plotone{f3.eps}
1023: \caption{\tc\ plotted against properties of the starless cores: (top
1024: left) non-thermal line width (1-D velocity dispersion) of \amm, (top
1025: middle) Bolocam Flux, (top right) peak \AV, (bottom left) core mass
1026: and (center) average volume density. Starless cores in NGC1333 and
1027: IC348 are plotted in black, and field starless cores are plotted in
1028: grey. The 1$\sigma$ errors in the fit of \tc\ and non-thermal
1029: linewidth are plotted for each core. \label{TKININTERNAL}}
1030: \end{figure}
1031:
1032:
1033:
1034: \end{document}
1035: