Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.naic.edu/~gibson/hisa/poster/2000/poster.ps.gz Äàòà èçìåíåíèÿ: Fri Jan 21 05:12:07 2000 Äàòà èíäåêñèðîâàíèÿ: Sun Dec 23 01:26:38 2007 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ï ï ï ï ï ð ï c 1999 s4 linear

Motivation
Though a major constituent of the interstellar medium, cold atomic gas,
with T Ÿ 100 K, is elusive. Maps of 21cm emission are dominated by
warm H i, and most observations of H i absorption against continuum
sources are limited to discrete points. However, H i self­absorption
(HISA) against warmer background H i can give a better view of the
structure and distribution of cold H i clouds in the Galaxy.
A systematic HISA study of cold Galactic H i requires broad angular cov­
erage to remain unbiased, as well as high angular resolution to detect small­
scale features which might otherwise be washed out. Our investigation is
the first to employ wide­field synthesis imaging to these ends. We are using
Canadian Galactic Plane Survey (Taylor et al. 2000; English et al.
1998) maps taken with the DRAO Synthesis Telescope (Landecker et al.
1999). Our CGPS images have ¸ 1 0 resolution with 0:8 km s \Gamma1 velocity
channels over the region [147:3 ffi ? ` ? 74:2 ffi ; \Gamma3:6 ffi ! b ! +5:6 ffi ].

Current Results
At right are maps of H i emission, HISA, and 12 CO emission in the eastern
25% of the survey, given as projections along the velocity (Figure 1) and
latitude (Figure 2) axes. We find:
ffl A rich and diverse population of HISA is revealed, with a wide variety
of shapes, sizes, and contrasts against the background H i. Some of
these features appear unresolved, and many would be invisible to a
single­dish survey.
ffl Despite sharing some of the compact, filamentary qualities of molecu­
lar clouds, few of these HISA features appear associated with detected
CO emission (far­IR dust emission is found in some HISA features
lacking CO, but not all). Either the HISA is not probing the small
fraction of H i in H 2 clouds, as has been presumed in the past, or CO
is not a good tracer of H 2 under some circumstaces.
ffl The bulk of the HISA occurs at velocities placing it in the Perseus
spiral arm (\Gamma35 to \Gamma60 km s \Gamma1 ); Local gas contains a some faint
HISA features, but no strong absorption.
ffl The fainter HISA seems widely distributed in space and velocity, while
the darker HISA is organized into discrete complexes. One question
we wish to address is whether these dark complexes represent a par­
ticular aspect of arm structure. For example, they might trace cold
H i downstream of the Perseus density wave shock prior to H 2 con­
densation and star formation. Figure 2 compares HISA velocities
against Roberts' (1972) spiral shock model (green line).
ffl The total mass of cold H i we detect as HISA is at least 7:5 \Theta 10 7 M fi .
Since this value presumes all H i emission to lie behind the HISA and
a gas spin temperature of 30 K (likely an overestimate), the actual
HISA mass may be much greater.

REFERENCES
English, J., et al., 1998, Pub. Ast. Soc. Aust., 15, 56
Heyer, M. H., Brunt, C., Snell, R. L., Howe, J. E., Schoerb, F. P., & Carpenter,
J. M., 1998, Ap. J. Supp., 115, 241
Landecker, T. L., et al. 1999, A. & A., submitted
Roberts, W. W., 1972, Ap. J., 173, 259
Taylor, A. R., et al., 2000, A. J., submitted

Figure 1: H i Survey Map (`; b): H i 21cm brightness integrated over all velocity channels; white
is bright and black is faint. Blue contours show HISA column density (8 \Theta 10 19 cm \Gamma2 ), computed for
an assumed HISA spin temperature of 30 K. Red contours show FCRAO 12 CO brightness integrated
over the Perseus arm, from the Heyer et al. (1998) survey.
Figure 2: H i Survey Map (`; v LSR
): As above, except integrated over latitude instead of velocity;
HISA column contours are for 0.5 and 1:0 \Theta 10 21 cm \Gamma2 . Local gas occurs near 0 km s \Gamma1 , and Perseus
arm gas near \Gamma40 km s \Gamma1 . The green line marks the position of the Roberts (1972) model H i spiral
shock ridge in the Perseus arm.
Figure 3: H i Brightness: Close view of a ``raw'' H i velocity channel with self­absorption.
Brightness ranges from 40 K (black) to 130 K (white). ON and OFF H i spectra, from the cross
and boxes respectively, are given below for one strong but compact HISA feature. The upper panel
compares ON (solid) with OFF (dashed), and the lower panel shows the ON­OFF temperature
difference.
Figure 4: Velocity Search: We are developing automated methods of locating HISA in velocity
cubes by removing large­scale emission in space and velocity and flagging small­scale dark residuals.
This image shows (negative) temperatures of the dark residuals found with our spectral search,
ranging from 0 to \Gamma40 K.
Figure 5: Spatial Search: Here are residual (negative) temperatures from our spatial search. We
combine these results with those of the spectral search to flag suspected HISA voxels (3­D pixels).
Figure 6: HISA Amplitude: Final ON­OFF temperature differences for all identified HISA voxels.
The voxels have been assembled into contiguous 3­D groups whose non­HISA spatial and velocity
edges are used to obtain the best estimate of H i brightness behind the HISA feature. This is
essential for determining its absorption properties and mass.