H{}_{2}CO Absorption toward Extragalactic Radio Sources

Study of Interstellar Molecular Clouds using Formaldehyde Absorption toward Extragalactic Radio Sources

E. D. Araya11affiliation: Physics Department, Western Illinois University, 1 University Circle, Macomb, IL 61455, USA , N. Dieter-Conklin22affiliation: Emeritus, Radio Astronomy Laboratory, University of California, Berkeley, USA , W. M. Goss33affiliation: National Radio Astronomy Observatory, P.O. Box 0, Socorro, NM 87801, USA , N. Andreev11affiliation: Physics Department, Western Illinois University, 1 University Circle, Macomb, IL 61455, USA
Abstract

We present new Very Large Array 6cm HCO observations toward four extragalactic radio continuum sources (B0212+735, 3C111, NRAO150, BLLac) to explore the structure of foreground Galactic clouds as revealed by absorption variability. This project adds a new epoch in the monitoring observations of the sources reported by Marscher and collaborators in the mid 1990’s. Our new observations confirm the monotonic increase in HCO absorption strength toward NRAO150. We do not detect significant variability of our 2009 spectra with respect to the 1994 spectra of 3C111, B0212+735 and BLLac; however we find significant variability of the 3C111 2009 spectrum with respect to archive observations conducted in 1991 and 1992. Our analysis supports that changes in absorption lines could be caused by chemical and/or geometrical gradients in the foreground clouds, and not necessarily by small scale (AU) high density molecular clumps within the clouds.

slugcomment: Accepted for Publication in The Astrophysical Journal

1 Introduction

The 6cm line of formaldehyde (HCO) is a powerful probe to study molecular gas in the Milky Way and in other galaxies (e.g., Mangum et al. 2008; Dickel et al. 2001). Soon after its discovery (Snyder et al. 1969), the HCO 6cm transition was detected in absorption toward radio continuum sources as well as against the Cosmic Microwave Background (e.g., Palmer et al. 1969; Dieter 1973). Now, more than 40 years of observations have shown that the 6cm HCO line can trace large (pc) molecular clouds (e.g., Araya et al. 2007). Such large structures are not homogeneous, for example, Dieter-Conklin (2010) concluded that molecular clouds traced by HCO absorption show different velocity dispersion when observed at effective resolutions greater or smaller than pc, which gives direct evidence for sub-parsec structure traced by HCO.

Even smaller size-scales can be probed by HCO observations of absorption against compact extragalactic radio sources. In particular, Marscher et al. (1993) and Moore & Marscher (1995) used the VLA to monitor 6cm HCO absorption toward four extragalactic radio sources during a period of years, and found evidence for HCO line variability toward three continuum sources (3C111, NRAO150, BL Lac; they found no significant change in the B0212+735 line profile based on observations conducted less than 2 years apart). As recognized by Marscher et al. (1993) and Moore & Marscher (1995), the variability most likely traces changes in our line of sight that occurs because we move with respect to the distant source and intervening cloud, i.e., the combined effect of Earth’s revolution around the Sun and the cloud’s internal kinematics and proper motion (see Dieter-Conklin 2009, for a discussion of this “searchlight effect”). Marscher et al. (1993) estimated that the change in line of sight through the clouds enables probing physical scales of the order of AU per year, i.e., AU for sources with 3.4years monitoring data.

The use of 6cm HCO absorption as probe for substructure (as revealed by spectral line variability) has clear advantages, in particular, the supercooling effect of the lowest K-doublet transition due to H - HCO collisions (Townes & Cheung 1969; Thaddeus 1972; Garrison et al. 1975; Troscompt et al. 2009a) enhances the optical depth against continuum sources. Also, the separation between the main group of five hyperfine lines (centered near the component) and the component is approximately 18.2kHz (equivalent to 1.13km s; see Tucker et al. 1970) and therefore the line does not significantly overlaps with the main hyperfine components in diffuse/translucent regions and dark clouds (which have linewidths of km s; see for example Araya et al. 2006). Thus, if variability is detected in the main absorption feature, variability of the hyperfine component is also expected, which is a useful test to assess whether artifacts are present in the data.

The monitoring of the HCO line profiles started by Marscher and collaborators did not continue. In this paper we report new observations of the sources observed by Moore & Marscher (1995) to investigate longer term variability in the absorption profiles and the implication of the variability on the structure of the clouds.

2 Observations

We used the Very Large Array (VLA) of the National Radio Astronomy Observatory (NRAO) to observe the 6cm line of HCO (, = 4829.6594MHz) in absorption toward the extragalactic radio sources B0212+735 (J0217+738), 3C111 (J0418+380), BLLac (J2202+422), and NRAO150 (J0359+509). The observations were conducted on 2009 March 2 (see Table 1)111VLA project number AA327.. The quasar J0319+415 was used as baseline/bandpass calibrator and 3C48 (J0137+331) as flux density calibrator (5.58Jy assumed at 4.8GHz). We measured a flux density of 13.8Jy for J0319+415. We observed with 511 channels (channel width of 3.05kHz, 0.19km s), single circular polarization and a bandwidth of 1.56MHz (96.8km s), however, due to aliasing effects in the VLA-JVLA transition, only an effective bandwidth of 34km s was useful. The data reduction was done in the NRAO software package AIPS, and the final spectra were exported to CLASS222CLASS is part of the GILDAS software package developed by IRAM. for analysis. Natural weighting was used to create the images to maximize signal-to-noise. All sources were self-calibrated. Details of the observation setup and synthesized beam are given in Table 1.

In the case of 3C111, we conducted a short (10 min) observation in standard VLA C-Band continuum mode (4IF; 50 MHz per IF). J0414+343 was used as complex gain calibrator. We measured a flux density of 1.4Jy for J0414+343. The radio continuum image of 3C111 was used as initial model for the self-calibration of the 3C111 line data.

We downloaded from the VLA archive the raw-data of previous HCO observations of all our sources, which have been already published by Marscher et al. (1993) and Moore & Marscher (1995). We reduced the archive data in AIPS following standard procedures. The spectra were Hanning smoothed to a final channel width of 3.05kHz (0.19km s). Table 1 also lists the details of the archive data discussed in this paper.

3 Results

We detected radio continuum and HCO absorption toward the four sources. Table 2 lists the line parameters of the HCO detections obtained from single-Gaussian fits. No significant variability was detected toward B0212+735 and BLLac (Figures 1 and 2, respectively). In the upper left panels of Figures 1 and 2, we show the radio continuum images (contours) and the peak-pixel absorption images (grey scale). In the upper right panels of Figures 1 and 2, we show the March 2009 spectra, including a fit to the six hyperfine components (blue dashed lines; Table 3). The lower left sequence of panels show the March 2009 spectrum (in black) and the spectra from the VLA archive obtained at previous epochs (in red). The lower right sequence of panels show the difference between the previous observations and the 2009 data. In contrast to B0212+735 and BLLac, significant variability was detected toward NRAO150 (Figure 3).

These three sources (B0212+735, BLLac, NRAO150) show compact radio emission. In contrast, 3C111 was resolved in a central compact source (Core), and two extended radio lobes to the NE and SW (see Figure 4). The HCO spectra and hyperfine fits of the NE and SW components are also shown in Figure 4. The low signal-to-noise toward the NE and SW regions precludes meaningful analysis of time variability, thus, only the 2009 spectra are shown. The signal-to-noise toward the 3C111 Core is as good as that of B0212+735, NRAO150 and BLLac, and variability is detected (Figure 5).

4 Analysis and Discussion

4.1 Reproducibility of Spectral Line Results

To check the effect of different calibration procedures, we reduced all archive observations of the sources which have been already published by Marscher et al. (1993) and Moore & Marscher (1995). We applied completely independent flagging, calibration and imaging procedures. In our analysis of the physical parameters of the clouds, the peak optical depths and linewidths are input parameters (see section 4.2), while Moore & Marscher (1995) used equivalent widths defined as . To compare our results to those of Moore & Marscher (1995), we measured the equivalent widths of our reduction of the M94 data. Our equivalent widths agree with Moore & Marscher (1995) values within statistical uncertainties from the fit in all sources, which shows that differences in calibration procedures are not a significant source of error in determining variability. For example, in the case of the M94 observations of B0212+735, Moore & Marscher (1995) reported = 0.0893, km s, and km s, while we measured , km s, and km s.

We detect a strong absorption line at km s toward NRAO150 (see Table 3, Figure 3). This line was included in the bandpass (different IFs) of the J91 and D90 epochs, however, the line was not included in the bandpass of the M94 and D92 epochs. As clearly seen in Figure 3, the line profile of the km s component is perfectly well reproduced by our 2009 observations with respect to the previous epochs, showing that the variability of the km s component in NRAO150 is not an artifact. Likewise, the km s velocity component in 3C111 shows no variability (compare red and black spectra in Figure 5) while the main absorption feature at km s does.

Careful inspection of the B0212+735 and NRAO150 spectra (Figures 1 and 3) shows that the bandpasses are not perfectly flat; specifically, the B0212+735 spectrum has a weak absorption feature at 0km s and NRAO150 presents weak absorption features at 14 and 4.5km s (highlighted with arrows). Absorption features at these velocities have been seen before in other molecular transitions. The CO, OH, and HCO spectra of B0212+735 reported by Liszt & Pety (2012) show a velocity component at km s in addition to the main component at km s. In the case of NRAO150, molecular absorption features at 14 and 4.5km s (in addition to the components at and and km s) have been reported by Pety et al. (2008) in CO, HCO, CH, and HNC; of all, the HNC spectrum is the most similar to the HCO spectrum. The same weak absorption features are seen at the other epochs (see Figure 3, left spectra). The remarkable reproducibility of these weak features demonstrates the superb stability of the VLA for long-term variability studies of spectral lines.

4.2 Physical Parameters

We used the Molpop-CEP code (Elitzur & Asensio Ramos 2006) to explore the physical conditions of the foreground clouds based on the observed HCO absorption lines. We used the HCO – para-H collision rates of Troscompt et al. (2009b), and included He collisions using the rates by Green (1991)333Electron collisions could be an important factor in the HCO excitation in translucent/diffuse clouds (e.g., Turner 1993), however development of a model including electron collisions is beyond the scope of this work.. Diffuse/translucent clouds have densities between and cm (e.g., van Dishoeck & Black 1988), and according to Moore & Marscher (1995) may contain even higher density clumps [cm; however see Liszt & Lucas (2000) and our discussion below]. The HCO abundance relative to hydrogen may be somewhere between and ; for example, Zhou et al. (1990) estimated an abundance of toward B335; Turner (1993) report an abundance of ; Troscompt et al. (2009a) obtained an abundance of . We ran grids of models at two kinetic temperatures (10 and 30K), in a range of densities between 10 to 10cm and HCO abundance ratios between and , however, for the analysis, we further constrain the parameters as follows:

Temperature: The extinction toward the extragalactic sources discussed here is mag or greater (e.g., Marscher et al. 1993; Schlegel et al. 1998; Pety et al. 2008; Schlafly & Finkbeiner 2011; Liszt & Pety 2012). Hence, the foreground clouds are in the diffuse/translucent boundary; which are expected to be hotter than 10K at their centers and as hot as K at their boundaries (e.g., van Dishoeck & Black 1988). Given that the HCO molecules are not expected at the boundaries of the clouds but at higher depths where they may be shielded from the interstellar radiation field, a temperature of K is assumed.

Density: observations of 6cm HCO absorption profiles in a number of environments including dark nebulae, pre-protostellar cores and low mass star forming regions with absorption against the CMB (e.g., Palmer et al. 1969; Heiles 1973; Dieter 1973; Young et al. 2004; Araya et al. 2006) have resulted in excitation temperature measurements of the 6cm K-doublet of K, which according to our model imply densities below cm. This is the well known result that the 6cm HCO K-doublet thermalizes at densities of cm (e.g., Troscompt et al. 2009a; Mundy et al. 1987). Thus, most of the HCO gas likely traces regions of densities below cm. A lower limit of the density can be inferred from our data as well: as highlighted in Figures 1 and 3 upper-right panels and discussed above, we detected some weak HCO absorption features. These velocity components are quite prominent in CO and HCO (e.g., Pety et al. 2008), and may trace molecular gas at densities of cm (e.g., Liszt & Lucas 2000; Pety et al. 2008). HCO is likely weaker because higher density conditions would be needed for stronger HCO absorption features (Pety et al. 2008). Thus, the HCO gas likely traces material at densities between to cm, i.e., higher than those traced by HCO and lower than termalization densities; we therefore assume a density of cm for the analysis.

Abundance: we simply assume an abundance of , i.e., similar to the abundance reported by Turner (1993) for high-latitude cirrus clouds.

Based on these assumptions, we can calculate the HCO and H column densities, and the thickness () of the absorbing region (assuming a slab geometry), i.e., the depth of the cloud that would result in the observed absorption spectrum for a given density, temperature and abundance. We note that the slab thickness should be considered an effective thickness, more precisely, a lower limit of the physical path which would differ from the true thickness of the cloud depending on the number of molecular clumps and their separation along the line of sight (see Marscher et al. 1993).

In the case of the HCO absorption region toward the core of 3C111 (see Figure 4), we obtained the following parameters and uncertainty ranges: AU ([, ] AU), cm ([, ] cm), and cm ([, ] cm), where the ranges shown in square brackets were obtained by assuming densities between and cm and formaldehyde abundances with respect to hydrogen between and . We note that HCO column densities of cm have been reported in diffuse clouds, including the ones discussed here (see Liszt et al. 2006).

Even though the value of is not well constrained within a density range of to cm, we can assess how reasonable the AU value is. As pointed out by Moore & Marscher (1995), the extended brightness distribution of 3C111 enables a study of spatial changes of the HCO absorption by comparing spectral lines toward the NE, Core, and SW continuum sources. Detection of absorption along the three sight-lines shows that the overall cloud is more extended than the projected separation of the NE and SW components, i.e., AU (0.3pc) assuming a distance of 350pc (Moore & Marscher 1995; obtained from a photometric/extinction analysis by Ungerechts & Thaddeus 1987)444For consistency with Moore & Marscher (1995) we use the same distance they assumed for the foreground cloud, however, the possible association with the California Molecular Cloud would imply a distance of 450pc (Harvey et al. 2013).. A AU would imply that the depth of the cloud is somewhat less than the extent of the cloud in the plane of the sky, which is consistent with having slightly different line parameters toward the different continuum sources (Table 3)555We note that velocity components at 1km s and km s were detected toward the Core and the NE regions, whereas only a single broad line was detected in our lower signal-to-noise spectrum of the SW 3C111 region. Thus, the significantly different HCO SW 3C111 profile could be caused in part due to blended lines in the low signal-to-noise spectrum..

Assuming the same values of kinetic temperature (30K), density (cm), HCO abundance (), and abundance/density domains as above, we obtain a total molecular column density and slab thickness equal to cm ([, ] cm) and AU ([, ] AU) for B0212+735. In the case of NRAO150 we obtain cm ([, ] cm) and AU ([, ] AU), while for BLLac we obtain cm ([, ] cm) and AU ([, ] AU).

4.3 Physical Scales Probed by the Monitoring Observations

The most important aspect to relate absorption variability to cloud structure is to be able to convert time-lapses to physical displacement of the clouds perpendicular to our sight line, i.e., the proper motion due to the combined effect of Earth’s revolution and the motion of the Solar System and the cloud around the Galaxy. Marscher et al. (1993) and Moore & Marscher (1995) estimated that the magnitude of transverse displacement for the foreground clouds toward NRAO150, 3C111 and BL Lac is of the order of AU per year. Specifically, Moore & Marscher (1995) report that after approximately 3.4 years of monitoring observations, the foreground clouds toward NRAO150 and BL Lac would have had transverse displacements of 13.9AU and 12.4AU, respectively. These estimates depend on the precise knowledge of the transverse velocity of the foreground clouds with respect to Earth, which has not been directly measured. Thus, we recalculated the transverse displacement of the foreground clouds including uncertainty in the cloud’s peculiar velocity around the Galaxy, in addition to the line-of-sight displacements due to Earth’s revolution, LSR reference frame correction, peculiar velocity of the Solar System and estimated Galactic circular rotation of the foreground clouds. To calculate the Galactic circular rotation, we assumed the distances reported by Moore & Marscher (1995), i.e., 700pc (kinematic distance, Marscher et al. 1993), 350pc (estimated from a photometric/extinction analysis by Ungerechts & Thaddeus 1987), and 330pc (estimated from the average vertical height of Galactic molecular clouds in the solar neighborhood and the galactic latitude of the foreground cloud, Lucas & Liszt 1993) for the NRAO150, 3C111, and BLLac foreground clouds, respectively. We used the rotation curve of Clemens (1985) for consistency with Marscher et al. (1993). To estimate the range of possible transverse displacements, we assume a peculiar velocity of the foreground clouds of km s in the three galactocentric cylindrical coordinates. In a time lapse of 748 days [the time lapse of the Marscher et al. (1993) observations], we obtain a maximum displacement of 10AU and a minimum of AU depending on the peculiar velocity of the clouds toward NRAO150, 3C111 and BLLac. This range encompasses the displacements quoted by Marscher et al. (1993), but shows that the peculiar velocity of the foreground clouds has a significant effect on the displacement, and hence, on our ability to “translate” time-lapses into transverse displacements.

The new 2009 observations reported here enable us to study variability in a time lapse of almost two-decades. From 1990 to 2009, the maximum estimated displacement is AU and as small as its parallax motion (AU; i.e., the case when the peculiar velocity of the cloud results in no proper motion) for NRAO150, 3C111, and BLLac. An estimate of the transverse displacement for B0212+735 is uncertain because there is no distance measurement to the foreground cloud reported in the literature, and kinematic distances (e.g., Watson et al. 2003) or statistical distances (e.g., Magnani et al. 1985) are unreliable given the Galactic latitude and the small V velocity of the absorption. However, a similar displacement range is expected for the foreground cloud in B0212+735 because the magnitude of its is within that of 3C111 and NRAO150 (Table 3), and the magnitude of its galactic latitude () is similar to that of BLLac () and 3C111 ().

4.4 Variability

Figures 3 and 5 bottom right panels show the difference between the 2009 spectra and the previous observations for all sources (i.e, “red” minus “black” spectra of the corresponding left panel), which will be called residual spectra hereafter. As mentioned above, clear variability is seen toward NRAO and 3C111 Core, however the variability behavior observed toward both sources is quite different.

In the case of 3C111 Core, we found no long-term variability trend. The 2009 spectrum is the same within the noise to the September 1991 and December 1990 spectra, and only one channel from the 1994 spectrum differs by 3.5 with respect to the 2009 spectrum. Moore & Marscher (1995) reported variability of the HCO line from the July 1991 and December 1992 epochs with respect to their May 1994 spectrum. We also find variability above when comparing our 2009 spectrum with the July 1991 and December 1992 data (Figure 5). Comparing the March 2009 spectrum to the December 1992 spectrum (D92M09 residual spectrum in Figure 5), we see a line that resembles an inverse-P Cygni profile, which may be caused by a small velocity difference (km s; see Table 2) between the two different sight-lines sampled on March 2009 and December 1992. Evidence for such velocity shift was pointed out by Moore & Marscher (1995) when comparing their 1992 and 1994 spectra. The velocity difference is less than the FWHM of the line, thus, velocity gradients within the cloud could be responsible for the observed variability. We note that the M94-M09 residual spectrum shows a similar profile than the D92-M09 residual spectrum, albeit with significantly less signal to noise. We do not detect evidence for such residual line profile in the other epochs, nor toward the other sources, and at this point it is unclear why only the M92-M09 residual spectrum shows such a pronounced profile. Follow up observations of 3C111 are needed to detect more occurrences of velocity shifts and detailed analysis of synthetic spectra from computer simulations of turbulent clouds is needed to estimate the expected occurrence rate of such events.

Figure 5 also shows that the peak of the July 1991 spectrum is less deep than the March 2009 spectrum by ; Moore & Marscher (1995) also report less absorption in the July 1991 spectrum with respect to the May 1994 spectrum by (see their Figure 5), however the line in our residual spectrum is narrower than that from Moore & Marscher (1995). The 3C111 residual spectra do not show evidence of hyperfine structure due to low signal-to-noise, which as mentioned in the introduction, would strengthen the case for variability.

The difference between the July 1991 and March 2009 3C111 spectra could be explained by a change in the line-of-sight overall characteristics of the foreground cloud. Recalling that the main absorption feature in 3C111 Core is consistent with cm, K, HCO abundance, AU, the difference between the July 1991 and March 2009 could be caused by an 8% variability in overall HCO abundance without changing any other parameter. Smaller abundance gradients would be needed to explain the variability if the depth () of the HCO absorption region would also have changed. Significant HCO abundance gradients in molecular clouds are know to occur, e.g., Young et al. (2004) found clear evidence of HCO depletion in the interior of pre-protostellar cores.

In the case of NRAO150 (Figure 3), our new 2009 spectrum confirms the variability trend reported by Moore & Marscher (1995), i.e., the main absorption feature has been monotonically increasing since 1990. The case of NRAO150 shows the advantages of using HCO absorption as a probe for variability against point sources, in particular, the residual spectra show detection of the hyperfine component and no variability of the km s line, which, as mentioned above, demonstrate that the residual lines are true variability and not artifacts. Figure 6 shows the variability as a function of time. The new data point shows a change in the slope of the variability, suggesting that the 1990’s observations sampled a region with a greater column density/abundance gradient than the region sampled by the 2009 data.

Assuming cm, K, HCO abundance, AU and cm, the variability observed between 1990 and 2009 can be explained by an increase of AU in the effective thickness of the molecular cloud in the line of sight of the quasar () without any change in molecular density, temperature or abundance. As mentioned above, the transverse displacement of the line of sight across the cloud between 1990 and 2009 is AU. Assuming a transverse displacement of 50AU (half the maximum expected displacement), and that the cloud has a cylindrical filamentary structure (circular cross section) with the sight line to NRAO150 crossing the filament perpendicular to its length (Figure 7), we can calculate the radius of curvature of the circular cross section. We obtain a filament radius of AU, and find that the line-of-sight would be crossing the filament very close to its edge (see Figure 7). We stress that the value of the radius is representative of the physical scales and not a precise measurement because of the large uncertainties involved in and transverse displacement, and the filament is unlikely to have a perfectly circular cross-section. Nevertheless, the molecular observations by Pety et al. (2008) indeed show that the line of sight toward NRAO150 is crossing a filamentary structure, which has projected width of  (see their Figure 4, km s  range, 5.8 resolution), which at a distance of 700pc would correspond to a radius of AU. This value is similar to the radius of AU calculated above. Thus, without changing any other parameter of the overall physical conditions of the gas (e.g., temperature, abundance, density) the variability observed toward NRAO150 could be explained by a simple geometrical effect caused by the line-of-sight crossing the edge of a molecular filament.

In order to explain the variability in HCO absorption profiles, Marscher et al. (1993) and Moore & Marscher (1995) argue for the presence of high density (cm) molecular clumps at scales of AU. In contrast, here we have found that minor chemical (abundance) gradients and/or geometrical effects can account for the observed variability without the need for high density clumps. Also, high density clumps are not necessarily present in the case of foreground clouds that show no variability, such as toward B0212+735 or the km s component toward NRAO150. Moreover, we note that at such high densities (cm) the HCO would be detected in emission instead of absorption against the CMB, which has never been observed in diffuse clouds.666Thermal emission of the 6cm HCO line has only been seen toward the Orion BN/KL region (Zuckerman et al. 1975, see also Araya et al. 2006) and non-thermal emission has been detected only toward young massive stellar objects (e.g., Araya et al. 2008). Hence, we concur with the conclusions Liszt & Lucas (2000), that presented independent arguments against explaining absorption variability as caused by AU-sized high density inclusions in interstellar clouds. In particular, Liszt & Lucas (2000) found that the high-density clump/inclusion assumption is inconsistent with the weak HCO emission in diffuse clouds. As discussed in their paper, in order to explain the weak HCO emission, a low ratio of total (high-density) clump to cloud area would be needed, but this would be inconsistent with the frequent optical depth variations observed in HCO profiles.

5 Summary

We used the Very Large Array (VLA) in 2009 to obtain a new epoch in the monitoring of 6cm HCO lines from Galactic molecular clouds detected in absorption toward extragalactic radio sources. We observed Galactic absorption from diffuse/translucent molecular clouds toward B0212+735, NRAO150, 3C111, and BLLac. The goal of the project was to check for variability in the absorption lines due to transverse displacement of foreground clouds with respect to our line of sight to the extragalactic continuum sources; a displacement that is caused by the combined motion of the Solar System and the foreground clouds around the Galaxy.

We detected 6cm HCO absorption toward all sources. To check for the effect of different data reduction (calibration and imaging) procedures in observed spectra, we reduced the archived observations reported by Marscher et al. (1993) and Moore & Marscher (1995). Our new spectra, together with the observations reduced from the archive, demonstrate the superb stability of the VLA for long term variability studies of absorption lines. The data reported here, in particular the spectra obtained toward NRAO150, show the advantages of using 6cm HCO absorption for this type of variability studies given the detection of multiple absorption lines in the same line of sight and the presence of the hyperfine component that can be used to rule out spurious variability in the main line.

We did not detect significant variability toward B0212+735 and BLLac. Variability in the archive observations of 3C111 is seen which can be explained by relatively minor chemical (abundance), velocity and/or cloud-thickness gradients in the foreground cloud. In the case of NRAO150, we confirm the monotonic increase in HCO absorption reported by Moore & Marscher (1995). We found that the monotonic variability can be explained by a filamentary (cylindrically)-shaped cloud crossing the line of sight to the quasar, i.e., the variability could be caused by simple geometrical effects without the need of high density molecular substructure (clumps or inclusions) in the medium, in agreement with Liszt & Lucas (2000). If the line of sight toward NRAO150 is probing the edge of a foreground molecular filament as suggested by our analysis and as seen in CO maps of the region (Pety et al. 2008), then, continuing monitoring of the HCO absorption toward this radio source is important to further study the structure of the filament and to explore other effects such as HCO depletion at higher extinctions across the cloud.

The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. We acknowledge the anonymous referee for a careful review of the manuscript and insightful suggestions. This work has made use of the computational facilities donated by Frank Rodeffer to the Astrophysics Research Laboratory of Western Illinois University. E.D.A. acknowledges support from the Provost Travel Award and the WIU Physics Department to present some of these results at an AAS meeting. This research has made use of NASA’s Astrophysics Data System, and the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.

References

• Araya et al. (2007) Araya, E., Hofner, P., Goss, W. M., et al. 2007, ApJS, 170, 152
• Araya et al. (2008) Araya, E. D., Hofner, P., Goss, W. M., et al. 2008, ApJS, 178, 330
• Araya et al. (2006) Araya, E., Hofner, P., Olmi, L., Kurtz, S., & Linz, H. 2006, AJ, 132, 1851
• Clemens (1985) Clemens, D. P. 1985, ApJ, 295, 422
• Dickel et al. (2001) Dickel, H. R., Goss, W. M., & De Pree, C. G. 2001, AJ, 121, 391
• Dieter (1973) Dieter, N. H. 1973, ApJ, 183, 449
• Dieter-Conklin (2009) Dieter-Conklin, N. 2009, AJ, 137, 3920
• Dieter-Conklin (2010) Dieter-Conklin, N. 2010, AJ, 140, 628
• Elitzur & Asensio Ramos (2006) Elitzur, M., & Asensio Ramos, A. 2006, MNRAS, 365, 779
• Garrison et al. (1975) Garrison, B. J., Lester, W. A., Jr., Miller, W. H., & Green, S. 1975, ApJ, 200, L175
• Green (1991) Green, S. 1991, ApJS, 76, 979
• Harvey et al. (2013) Harvey, P. M., Fallscheer, C., Ginsburg, A., et al. 2013, ApJ, 764, 133
• Heiles (1973) Heiles, C. 1973, ApJ, 183, 441
• Liszt & Lucas (2000) Liszt, H., & Lucas, R. 2000, A&A, 355, 333
• Liszt et al. (2006) Liszt, H. S., Lucas, R., & Pety, J. 2006, A&A, 448, 253
• Liszt & Pety (2012) Liszt, H. S., & Pety, J. 2012, A&A, 541, A58
• Lucas & Liszt (1993) Lucas, R., & Liszt, H. S. 1993, A&A, 276, L33
• Magnani et al. (1985) Magnani, L., Blitz, L., Mundy, L. 1985, ApJ, 295, 402
• Mangum et al. (2008) Mangum, J. G., Darling, J., Menten, K. M., & Henkel, C. 2008, ApJ, 673, 832
• Marscher et al. (1993) Marscher, A. P., Moore, E. M. & Bania, T. M. 1993, ApJ, 419, L101
• Moore & Marscher (1995) Moore, E. M., & Marscher, A. P. 1995, ApJ, 452, 671
• Mundy et al. (1987) Mundy, L. G., Evans, N. J., II, Snell, R. L., & Goldsmith, P. F. 1987, ApJ, 318, 392
• Palmer et al. (1969) Palmer, P., Zuckerman, B., Buhl, D., & Snyder, L. E. 1969, ApJ, 156, L147
• Pety et al. (2008) Pety, J., Lucas, R., & Liszt, H. S. 2008, A&A, 489, 217
• Schlafly & Finkbeiner (2011) Schlafly, E. F., & Finkbeiner, D. P. 2011, ApJ, 737, 103
• Schlegel et al. (1998) Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525
• Snyder et al. (1969) Snyder, L. E., Buhl, D., Zuckerman, B., & Palmer, P. 1969, Phys. Rev. Lett., 22, 679
• Townes & Cheung (1969) Townes, C. H., & Cheung, A. C. 1969, ApJ, 157, L103
• Troscompt et al. (2009a) Troscompt, N., Faure, A., Maret, S., et al. 2009a, A&A, 506, 1243
• Troscompt et al. (2009b) Troscompt, N., Faure, A., Wiesenfeld, L., Ceccarelli, C., & Valiron, P. 2009b, A&A, 493, 687
• Tucker et al. (1970) Tucker, K. D., Tomasevich, G. R. & Thaddeus, P. 1970, ApJ, 161, L153
• Turner (1993) Turner, B. E. 1993, ApJ, 410, 140
• Ungerechts & Thaddeus (1987) Ungerechts, H., & Thaddeus, P. 1987, ApJS, 63, 645
• van Dishoeck & Black (1988) van Dishoeck, E. F., & Black, J. H. 1988, ApJ, 334, 771
• Watson et al. (2003) Watson, C., Araya, E., Sewilo, M., Churchwell, E., Hofner, P., & Kurtz, S. 2003, ApJ, 587, 714
• Young et al. (2004) Young, K. E., Lee, J.-E., Evans, N. J., II, Goldsmith, P. F., & Doty, S. D. 2004, ApJ, 614, 252
• Zhou et al. (1990) Zhou, S., Evans, N. J., II, Butner, H. M., et al. 1990, ApJ, 363, 168
• Zuckerman et al. (1975) Zuckerman, B., Palmer, P., & Rickard, L. J. 1975, ApJ, 197, 571
You are adding the first comment!
How to quickly get a good reply:
• Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
• Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
• Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters