Split (1)

train · 132k rows

Filename stringlengths 22 64	Paragraph stringlengths 8 5.57k
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The raw data were reduced into calibrated data cubes using mostly standard techniques of bandpass, absolute ﬂux, and complex gain calibration in the MIRIAD package (Sault, Teuben & Wright 1995). To produce continuum-free line data, the continuum was subtracted in the (u,v)-domain and the ﬁtted continuum was output into sep- arate ﬁles. A single continuum image was generated from each of the three observing sessions and deconvolved with the CLEAN al- gorithm. The CLEAN model was input to determine a phase-only self-calibration solution, which was then applied to the line data and the continuum data. A ﬁnal continuum image was generated using the CLEAN algorithm from the 7.5 GHz bandwidth data ob- tained in each observing session.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	As this is among the ﬁrst spectral line surveys completed with a broadband interferometer, we highlight our method for baseline subtraction and challenges for bandpass stability. We performed baseline subtraction using the MIRIAD task UVLIN, generating a linear ﬁt to the baseline of each 2 GHz tuning in the (u,v)-domain. The continuum images were generated from the continuum solu- tions output by UVLIN. Because Sgr B2 has a very high line density and because it is difﬁcult to deﬁne line-free channels in the (u,v)- domain, we only excluded channels covering strongly masing lines from the ﬁt. As very strong lines covered a small fraction of the bandwidth of each image, this had a very small effect on the contin- uum subtraction, resulting in a ∼1 mJy/beam offset that is <0.1% of the continuum level. Such a small effect might be deemed negli- gible in most data sets, however given the sensitivity of this data, it is non-negligible.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	to the very large bandwidth in each tuning. The bandpass shape is quite complex, and while the selected bandpass calibrator was very strong (∼25 Jy), the strong continuum of Sgr B2 (∼1 Jy) made it difﬁcult to model the continuum to the dynamic range (about 3 orders of magnitude) required. Furthermore, the continuum posi- tions of Sgr B2(N) and (M) are the positions that contain most of the molecular line features, so that the residual continuum prefer- entially affected areas that are important for the line study.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	To correct for these effects, we applied a continuum correction to the data in the image domain. Lines were masked at the 3σ level, and each 2 GHz cube was ﬁtted with a high order polynomial to make a baselevel offset spectrum which was subtracted from the data cube. We note further processing of the baseline applied to the published spectra in §3.1.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	We note that the CABB ﬂagged a consistent set of channels in each observation. In the publically released images, in which the ﬁrst and last 50 channels of each 2 GHz wide module are cut, the zero-indexed channels affected include channels 0, 1, 78, 106, 206, 462, 590, 718, 974, 1102, 1126, 1230, 1358, 1486, 1742, 1870, and 1947. All data cubes and continuum images are avail- able at the CDS database(Genova et al. 1995) via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsarc.u- strasbg.fr/viz-bin/qcat?J/MNRAS/. The continuum-subtracted data cubes provided do not have a primary beam correction applied. Models for the ATCA primary beam can be computed from in- formation provided on the ATCA documentation, or by correspon- dence with the authors.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The ATCA line survey contains of order a thousand spectral lines with signiﬁcant spatial and kinematic structure. In order to handle the wealth of information in the dataset, we extracted mean spec- tra from elliptical regions placed over physically distinct regions of Sgr B2(N), transforming a 3-d problem into a set of 1-d problems. We extracted spectra over the entire 29.8 - 50.2 GHz bandwidth from a radio continuum peak along the K6 shell-shaped HII region, the cometary HII region L located 35 arcsec NE of Sgr B2(N), and the LMH hot core. In the higher spatial resolution section of the spectrum (from 29.8 - 44.6 GHz due to the observed conﬁg- urations), we additionally extracted mean spectra towards K4 and “h”. Figure 1 shows the elliptical regions from which the spectra were extracted and Table 3 reports their positions. Note that we did not vary the size of the elliptical regions in different images de- pending on the beam size. The elliptical regions are typically larger than the beam sizes in the high spatial resolution data (from 29.8 - 44.6 GHz), but smaller than the beam sizes in the low spatial res- olution data observed with the compact array conﬁguration (from 44.6 - 50.2 GHz). The spectra are sensitive to the spatial resolution; for example, low spatial resolution data towards the K6 elliptical region includes ﬂux originating on the K5 shell and in the LMH as the regions are poorly resolved. In the case of the spatially iso- lated source L, the elliptical region from which the spectrum was extracted is similar to the source size of L and well matched to the ∼5 arcsec beam size of the high spatial resolution data (Gaume et al. 1995). In the low spatial resolution data, the ﬂux is spread over the larger beam, which has the effect of decreasing the ﬂux in the low spatial resolution section of the spectrum.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	After extracting spectra, we interpolated over bad spectrome- ter channels listed in §2.2 using linear interpolation, and performed additional baseline smoothing by applying a Hodrick-Prescott ﬁl- ter to remove a low frequency baseline wiggle (Hodrick & Prescott 1997). Figure 2 shows the baseline with and without the Hodrick- Prescott ﬁlter applied. A Hodrick-Prescott ﬁlter is a commonly used tool for removing long-term growth trends from economics data and is similar in function to a high-pass ﬁlter. It is simple to implement and proved a useful means of removing residual base- line without degraded results at the band edges. We used a modi- ﬁed version of the HP ﬁlter, as described in Schlickt (2008), that can use datasets with segments of missing data. This was useful in order to extrapolate over spectral lines, particularly over broad spectral lines, such as the absorption feature of CH2CN at 40230- 40270 MHz. Applying the ﬁlter achieved a 10 - 25% reduction in the root mean squared (rms) noise in the spectra extracted from different images.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	We then converted from units of Jy beam−1 to mean ﬂux den- sity per square arcsec (or equivalently, speciﬁc ﬂux density) in or- der to inspect images with distinct beam sizes on the same scale. Finally, we performed a primary beam correction to the extracted spectra. We applied the beam correction to the extracted data in- stead of simply extracting the spectra from beam-corrected images for the following reason. For each image, the standard technique in MIRIAD applies a single primary beam (PB) correction based on the central frequency tuning of the image, and does not include frequency dependence within the image. However, over the 2 GHz bandwidth of each image, frequency-dependence is signiﬁcant, es- pecially towards L. To compute appropriate frequency-dependent beam corrections, we compared the spectra extracted from the re- gions in the non-PB corrected cubes to the spectra extracted from the same regions in PB corrected cubes. We ﬁt a ﬁfth order poly- nomial to the ratio of PB corrected to non-PB corrected data across the 20 GHz of bandwidth. This ﬁt provides the PB corrections. They are substantial (up to a factor of 2.6) for L but are very small (<1.04) for K4, K6, “h”, and the LMH.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The half power beam widths at all frequency tunings and the rms noise levels in the ﬁnal extracted spectra are reported in Table 2. The supplemental material to this work provides the extracted, HP-ﬁltered, and PB-corrected spectra towards all 5 regions, and the line ﬁtting and identiﬁcation PYTHON code described in §3.2. This material is additionally available at https://github.com/jfc2113/MicrowaveLineFitter.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	ber of independent data points remains the same, this triples the number of points used to constrain the ﬁts, improving the ﬁts signif- icantly. This is essential for the ATCA data because in many cases, only 2-3 original channels sample the lines, and a Gaussian ﬁt to so few channels can return a wildly unreasonable ﬁt. For example, if a line proﬁle is sampled by 2 channels and surrounded by a zero- value baseline, one can imagine many different Gaussian proﬁles capable of ﬁtting the data perfectly with highly varying heights in particular.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	lines from various molecules have been shown to be non-thermally excited (Menten 2004; McGuire et al. 2012; Faure, Remijan & Szalewicz 2014). The code is written modularly, enabling adap- tation for different datasets. The PYTHON scripts are further de- scribed in Appendix A and available in the online journal and at https://github.com/jfc2113/MicrowaveLineFitter.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The data contain recombination lines and molecular transitions. Most detected molecular transitions are either observed in absorp- tion preferentially along the K5 and K6 shells, L, and K4, or in emission towards the LMH and “h” hot cores (Figure 3). Figure 4 shows a representative segment of spectra from all ﬁve regions from which spectra were extracted. Detected absorption lines are primar- ily from low energy states (EL (cid:46) 20K), whereas most emission lines from the LMH and “h” have higher energies (EU (cid:38) 80K). The LMH has the most line-dense spectrum as anticipated, and “h” has the second richest spectrum. The excitation and gas phase molecu- lar abundances of “h” appear signiﬁcantly more similar to the LMH than other regions in Sgr B2(N), conﬁrming that it is a second hot core in Sgr B2(N) (Belloche et al. 2013).
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The spectra towards K6, L, and K4 are similar to one another, predominantly showing emission by recombination lines and ab- sorption by molecules. Of all detected features originating at the positions of K6, L, and K4, ∼90% were conﬁdently assigned to recombination line or molecular carriers. The remaining 10% in- cluding unidentiﬁed transitions are discussed in Appendix B. Fig- ure 32 illustrates the raw data spectrum extracted towards K6 with the output of the line ﬁtter overlaid and line identiﬁcations labeled. Figures in this format containing the full spectra towards K6, L, and K4 are provided in Appendix B. Additionally, tables providing line identiﬁcations and Gaussian ﬁt parameters, including veloc- ity center, height, width, and integrated ﬂux towards K6, L, and K4 are available in Appendix C. Table 9 is a sample page of the line parameters towards K6. We do not report errors on individual Gaussian ﬁts in the tables, but discuss errors in §4.2. As mentioned in §3.2, lines with wing broadening are approximated by a primary Gaussian component and additional components with lower am- plitude. For clarity, we only list the primary Gaussian component parameters in the tables, however, the reported integrated ﬂux val- ues include wing components. It follows that the reported ﬂux is not always equivalent to what you would obtain from the primary Gaussian component alone.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	In order to assess the performance of the line ﬁtting and identiﬁca- tion routine, the output of the code was evaluated using hydrogen and helium recombination lines towards K6, L, and K4. Recombi- nation lines, particularly at higher principal quantum numbers (n >90), are not well described by simple Gaussian line shapes but exhibit Voigt proﬁles (von Proch´azka et al. 2010). The line ﬁtting routine applies only Gaussian shapes, which adequately charac- terise the bulk motion of ionized gas via the velocity center, partic- ularly because the α and β transitions have lower quantum numbers with less collisional broadening. However, towards K6 in particu- lar, most H α lines require primary components and weaker com- ponents to recover the ﬂux in the wings.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	the observed band, enabling measurements of recombination line strengths, widths, and HII region kinematic centers. Figures 6 and 7 present H and He α and β transitions towards K6, and Figures 8 and 9 present H and He α transitions towards L and K4 respec- tively. Each panel shows the H and He data with a single quantum number, with the Gaussian ﬁts reported in the online journal over- laid. The ﬁnal panel of each ﬁgure shows a composite averaged spectrum overlaid with a ﬁt to the averaged spectrum. The best ﬁts to the averaged spectra are provided in Table 5. Additionally, re- combination line kinematic measurements generated from H α - γ transitions are provided in Table 6 and a histogram of the line cen- ters is shown in Figure 10.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Towards K6, the automated line ﬁtting routine detected 64 hydrogen recombination transitions including H α - H ζ and 13 helium transitions including He α - He β lines. The ionized gas towards K6 has two primary velocity components, at ∼59 and ∼84 km s−1 (Figure 6). The automated line ﬁtter determined that a 2-component ﬁt is required for all H α transitions and for nearly all H β and He α transitions. Of the 25 unblended H α - γ transi- tions for which the automated ﬁtting routine used a 2-component ﬁt, it obtained a mean velocity of 58.9 ± 0.8 km s−1 and width 28.7 ± 1.1 km s−1 for the low velocity component and a mean ve- locity of 84.7 ± 0.8 km s−1 and width 27.2 ± 1.2 km s−1 for the high velocity component. The best ﬁts to the composite averaged spectra of H α and H β transitions are both consistent with these parameters (Table 5). As is evident in Figures 6 and 7, hydrogen recombination transitions at higher spatial resolution, (namely the H(56) - (53)α and H(70) - (66)β transitions) have more pronounced double-peaked line shapes than the moderate resolution transitions (H(59) - (57)α and H(75) - (71)β), which in turn are more double- peaked than the lowest resolution transitions (H(52) - (51)α and H(65) - (64)β). The line proﬁles are thus highly sensitive to the spatial resolution.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The double-peaked structure towards K6 was ﬁrst reported by de Pree et al. (1995) in an image of the H(66)α transition observed by the VLA with 1.(cid:48)(cid:48)4 resolution. This work reported a 2-component recombination line proﬁle located internal to the K6 shell at a posi- tion roughly 2.5 arcsec north of our selected K6 region. The compo- nents were reported at 50 and 93 km s−1, and de Pree et al. (1995) argued that the H(66)α line proﬁle provides a direct detection of the two sides of a shell with an expansion velocity of ∼21 km s−1. In the ATCA data, a double-peaked proﬁle with components at ∼50 and 90 km s−1 is observed towards the K6-internal position noted in that work. However, the spatial positions of the peak 50 and 90 km s−1 recombination line emission are offset, with the 50 kms component centered on the base of the K6 shell within our K6 ellip- tical region, and the 90 km s−1 gas centered ∼4 arcsec to the NW (Figure 11). If the two components are part of the same structure, the data are consistent with the ionization front expanding into a clumpy molecular cloud and preferentially illuminating at the po- sition of a clumps on the front and back sides of the shell.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	line velocity of 75.8 ± 0.3 km s−1 and width 31.7 ± 0.7 km s−1, so that the line widths in particular somewhat disagree with the ATCA data. This is a result of the differences in spatial resolution. In the ATCA observations, L appears as a single unresolved source; on the other hand, the higher spatial resolution observations in de Pree et al. (1995) treat L and L13.30 located ∼3 arcsec east as sep- arate sources. Towards L13.30, de Pree et al. obtain a velocity of 76.5 ± 1.2 km s−1 and width 20.6 ± 2.8 km s−1. Our line parame- ters are intermediate between those of the two unresolved sources. Recombination lines are weaker towards K4 compared to K6 and L. As a result, primarily H α - γ transitions were detected (24 total) and only 2 He α transitions were detected. As is apparent in Figure 9, the line proﬁles suggest a wing on the high velocity side of the line center, near 100 km s−1. Of the seven H α transitions from 30 - 44.6 GHz, only H(56)α is known to be blended. Of the remaining six H α transitions, four required a 2-component ﬁt to the line emission from 40 - 80 km s−1 (Figures 9 and 10). The other two H α transitions (namely H(57) and H(55)α) did not meet the criteria for a 2-component ﬁt, but a 2-component ﬁt appears more appropriate and the reported ﬁt is adjusted from the output of the line ﬁtter. The line shape appears to vary substantially, suggesting a complex kinematic structure resulting in a line shape that is sensi- tive to the beam size and precise placement of the elliptical region. Indeed Figure 11 shows a steep velocity gradient across K4.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Although most of the H α transitions required a 2-component ﬁt (primary) and a weaker component in the high velocity wing (secondary), the composite line proﬁle is adequately ﬁt by 1- component primary Gaussian ﬁt at 65 km s−1 and a weaker wing component (secondary) on the high velocity side at 100 km s−1, as reported in Table 5. Using ﬁts to H β and H γ lines deter- mined by the line ﬁtter, we ﬁnd that a 1-component treatment of the recombination lines towards K4 has a velocity center of 66.2 ± 0.5 km s−1 and width of 32.1 ± 1.4 km s−1. The values we determine are in mild disagreement with those reported by de Pree et al. (1995), who report a velocity of 63.9 ± 0.8 km s−1 and a width 36.0 ± 1.9 km s−1. This may be because we are looking at different regions of K4. We determined the shape and placement of the elliptical region towards K4 based on the continuum structure as sampled at 40 GHz by the ATCA, which is offset by ∼1 arcsec SE of the continuum peak at 22 GHz observed by de Pree et al. (1995). In the ATCA data, most of the recombination line emission associated with K4 arises 2-3 arcsec southeast of the 22 GHz con- tinuum peak, and higher velocity line emission is observed to the SE (Figure 11). Some of the higher velocity material is included in the elliptical region placed on K4.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	typically of order 0.05 mJy arcsec−2, can increase the number of false detections and affect the best-ﬁt Gaussian shapes. While we minimized the baseline residual using a Hodrick-Prescott ﬁlter, it is impossible to obtain a fully residual-free baseline, and the remaining baseline can af- fect low signal-to-noise transitions signiﬁcantly. This contributes uncertainty in line heights and can artiﬁcially broaden the best-ﬁt Gaussian. As discussed in §3.2, we minimize the latter effect by ﬁtting only the region within 0.8 × FWHM of line center.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Because there are many sources of uncertainty and a large number of lines detected in the survey, the simplest method for evaluating the error is to derive it from the ﬁts themselves. To do so, we assume that line radiation from all thermal molecular line tran- sitions originates at a consistent set of central velocities for each targeted spatial region, setting an upper limit on the typical error of ﬁts. With very few high signal-to-noise exceptions, this assump- tion is supported by the data, as typical differences from the mean center velocity are signiﬁcantly smaller than the channel widths of 6 - 10 km s−1.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	adopt σv = 1.1 km s−1 and σ∆v = 2.3 km s−1 as 1σ errors on the ﬁt parameters of 1-component lines and spectrally resolved 2- component lines with a signal-to-noise ratio >5.4. This includes unblended 1- and 2-component molecular lines and 1-component recombination lines. These values are consistent with the standard deviations of the parameters ﬁt to high signal-to-noise recombina- tion lines towards L (Figure 10 b), further justifying their validity. in which the components determine σv = 3.6 km s−1 and σ∆v = 4.9 km s−1 from 17 H α and β lines towards K6, and line height errors of each component are estimated at ∼25%. Compared to molecular lines, 2-component recombina- tion lines have larger errors because the two velocity components are poorly resolved. As the line widths of the 2-component than the velocity recombination line components are larger separation between the two components (cid:0) ∆v1+∆v2 (cid:1), the components are unresolved. While a 1-component ﬁt does a poor job of matching these proﬁles, a 2-component ﬁt contains signiﬁcant covariance between the six ﬁt parameters. Errors of the velocity center and width are approximately 40% higher for 2-component Hγ and δ lines, and the line heights have a ∼40% estimated error.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Table 7 summarizes the errors of the ﬁt parameters that are adopted in this work and recommended for further use of the re- sults.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Nearly all of the molecular lines detected towards K6, L, and K4 are in absorption against the free-free continuum emission. The follow- ing molecular species were detected towards one or more of these regions. Tentative detections are marked with a “(?)”.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Towards all three regions, molecular lines were primarily de- tected at two velocity components separated by ∼20 km s−1. Fig- ure 12 shows a histogram of the 25th - 90th percentile Gaussian ﬁt parameters of identiﬁed molecular lines, and Table 8 presents the mean ﬁt parameters, characterising the kinematic structure of molecular gas towards these positions. K6 and K4 have ve- locity components centered at 64 and 82 km s−1 and 62 and 82 km s−1 respectively, whereas L has velocity components at 56 and 76 km s−1. While we do not include a full characteri- sation of the spectra towards the LMH and “h”, the spectra of these sources show that line emission from the LMH hot core is at ∼64 km s−1 and emission from “h” is at 73 km s−1, in agree- ment with previous observations (Hollis et al. 2003; Belloche et al. 2013). Towards K6, L, and K4, both primary velocity components have median line widths >10 km s−1, indicating highly supersonic turbulence.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	A larger number of transitions are detected in the low velocity component gas towards all three regions, and the low velocity com- ponent has a higher total integrated line intensity in the detected transitions. Towards K6, the total integrated ﬂux absorbed by the 64 km s−1 component is a factor of 3.5 times greater than the total ﬂux absorbed by the 82 km s−1 component. Towards L and K4, the low and high velocity components typically have more similar line intensities, with total integrated ﬂux ratios of = 1.7 and 1.1 respectively. A larger number of distinct molecules were de- tected towards the low velocity component gas in K6 than towards any other position or velocity component.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	With the exception of ammonia inversion transitions, methanol masing transitions, one transition of H2CO, and three transitions of c-C3H2, all detected molecular transitions have lower state energies below 20 K, with most being below 10 K. For many of the detected molecules, the highest quantum line strength tran- sitions that fall in the 30-50 GHz range are between low energy states; however, a number of stronger lines between higher energy states (40 - 80 K) are missing. The gas is therefore rotationally cold. Given the typical temperatures of molecular gas in the Galactic Center (> 80 K) (Ao et al. 2013, Mart´ın et al. 2008 and references therein) and the observed supersonic turbulence, excitation temper- atures of TEx ∼10 K are highly subthermal. This result agrees with the conclusions of H¨uttemeister et al. (1995), which deter- mined that TEx < 20 K (cid:28) Tkin in the absorbing gas components, and with multiple recent characterisations of species detected in absorption in Sgr B2(N), which have determined that TEx ∼ 9 K (Loomis et al. 2013; Zaleski et al. 2013; McGuire et al. 2013).
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Additionally, ammonia inversion transitions with lower state energies ranging from 1200 - 3500 K are detected in both veloc- ity components towards K6, L, and K4. Most of these transitions have been reported toward Sgr B2(N) previously with single dish observations (H¨uttemeister et al. 1995; Flower, Pineau des Forets & Walmsley 1995). Although ammonia inversion lines are useful diagnostics of gas temperature (Ho & Townes 1983; Walmsley & Ungerechts 1983) the very high energy transitions do not likely in- dicate the kinetic temperature of the molecular gas, as the energy states may be populated due to formation pumping (Lis et al. 2012; E. Bergin, private communication). In work on Sgr B2 using Her- schel data from the HEXOS program, Lis et al. (2012) proposed that H3O+ and NH3, which have homologous structures with inver- sion transitions between metastable energy states, may be excited to very high energy levels by excess energy available from forma- tion in the presence of X-ray radiation. While they favor an X-ray dominated environment for producing observed abundance ratios of H3O+ and H2O, they do not rule out cosmic ray or shock driven chemistry. Regardless of the exact scenario, the high energy ammo- nia transitions can only be explained with signiﬁcant energy avail- able in the gas (via X-rays, CRs, or shocks). As such, the ATCA data are decidedly not probing cold material in the two primary velocity components towards K6, K4, and L, providing additional evidence that the gas is extremely subthermally excited.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The presence of two velocity components at multiple positions across the ﬁeld, the spatial extent of the absorbing gas over the full continuum structure (Figure 3a), the similar excitation conditions across the spatial ﬁeld, and the similarity with the molecular in- ventory observed in more extended material by Jones et al. (2011, 2008) suggests we may be peering through two extended sheets of gas, perhaps with a kinematic gradient across the ﬁeld. Previous work has suggested that absorbing gas at ∼65 and 82 km s−1 is associated with the extended envelope of Sgr B2 (H¨uttemeister et al. 1995) and is not internal to the core. If the components are lo- cated in the envelope, they would be anywhere from 3 - 10 pc from the cores (H¨uttemeister et al. 1995), a distance that is signiﬁcantly larger than the projected distance between the cores, as L and K6 are separated by ∼ 1.3 pc and K4 and K6 are separated by 0.4 pc in the plane of the sky. In this scenario, the extended molecular gas components need not interact directly with the local structure of the core (including expanding HII regions), and it is unclear whether the higher or lower velocity component is closer to the star forming core.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	To determine if the system consists of two extended clouds or if each kinematic component towards each region should be treated as a separate cloud (in which case we would be probing 3 spatial regions × 2 velocity components = 6 clouds), we inspect the line radiation in each component. If the gas is located in the envelope with signiﬁcant distance between the cores and the absorbing gas, we expect the gas phase molecular abundances and excitation to be fairly self-consistent in low velocity gas across the spatial ﬁeld and in high velocity gas across the ﬁeld. The low velocity gas could be quite distinct from the high velocity gas however, as they are clearly separate clouds.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	We refer to the quantity on the left as the normalised line ratio (NLR) of spectral line A with respect to C1 and use the line height generated by the automated line ﬁtter for TL. In the range of 29.8 to 44.6 GHz, we determine the continuum levels using the continuum images with central frequencies of 33.4 and 40.8 GHz and 7.5 GHz bandwidths. From the two images, we derived a power-law ﬁt to the continuum strength towards K6 and linear ﬁts to the continuum ratios for L and K4. We estimate that uncertainty in the continuum ratios contributes 10% error to an NLR.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	for the high velocity gas with respect to the low velocity gas for the two speciﬁc transitions. However, the NLR is clearly much higher, at ∼70%, for the transition of CH2NH and much lower for H2CS. If the line proﬁles of CH3CHO and CH3CH2CN are typical towards K6, we can therefore say that the high velocity compo- nent towards K6 (K6: 82 km s−1) is enhanced in CH2NH 111 – 202 and depleted in H2CS 101 – 000 relative to the low velocity gas in K6 (K6: 64 km s−1). Towards K6, it follows that signiﬁcant differ- ences in gas phase abundances and/or excitation exist in the two gas clouds. As the low and high velocity clouds are obviously distinct bodies, this result is not suprising.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	More interestingly, the line proﬁles towards L and K4 look very distinct compared to those in K6. We ﬁrst compare only the low velocity gas in all three regions. In the CH3CHO line, the NLRs with respect to K6: 64 km s−1are 50% for the low velocity gas in both L and K4 (L: 56 km s−1 and K4: 62 km s−1, respectively). In the line of CH3CH2CN, however, absorption by L: 56 km s−1 and K4: 62 km s−1 is considerably weaker, at ∼25 and 30%, respec- tively. Whereas the low velocity component absorption by CH2NH is weak in L: 56 km s−1 (at ∼22% the strength observed in K6), it is very strong in K4. On the other hand, the line of H2CS is strong in the low velocity gas towards L: 56 km s−1 (at ∼ 75% the strength observed in K6), and weak in K4: 62 km s−1 (at ∼25% the strength observed in K6). From these four lines alone, it is very clear that differences in excitation and/or gas phase molecular abundances exist on small angular scales of <10 arcsec.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Comparing the high velocity gas in the three regions produces a similar result. With respect to K6: 82 km s−1, the high velocity components in L and K4 have 40 and 80% deeper line absorp- tion by CH3CHO. In the CH3CH2CN line, K6: 82 km s−1 and L: 76 km s−1 have similar line strengths, while the high veloc- ity gas in K4 (K4: 82 km s−1) shows signiﬁcant enhancement. L: 76 km s−1 and K4: 82 km s−1 have lower line strengths of CH2NH 111-202, and while H2CS is undetected in K6: 82 km s−1, it is observed towards L: 76 km s−1 and K4: 82 km s−1 . This re- sult does not support the hypothesis that the high velocity gas is an extended cloud located in the envelope of Sgr B2.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	components if they have enhanced line optical depths compared to the median. This does not imply that these points are unreliable, but implies that the lower left corners of the plot cannot be ﬁlled in.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	We consider the line radiation with respect to the median ra- tio, as differences in the median NLR (such that median (cid:54)= 1) do not necessarily indicate differences in the relative chemical abundances or excitation. Instead the median NLR may correspond to the rel- ative optical depth of molecular gas, and it also holds the factor offset introduced by [f (TEx) − f (TCM B)] in Equation 3 towards L and K4. However, statistically signiﬁcant differences with respect to the median NLR indicate either (1) excitation differences, with distinct excitation of some molecules, or (2) chemical differences, with different relative gas phase molecular abundances present in distinct gas components. If statistically signiﬁcant scatter is ob- served in lines of a single molecule (e.g. differences in the NLR of different lines of CH3CHO), excitation differences are impli- cated. Although low excitation conditions are common in the gas components discussed here, differences in the excitation conditions can impact the excitation of different K-states, and weak masing ef- fects may selectively enhance or deplete the populations of certain states (Menten 2004; Faure, Remijan & Szalewicz 2014). Where line excitation is not implicated, the differences are attributable to chemical abundance differences.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Trends in Figure 14: Transitions that appear to vary the most between different panels of Figure 14 include CH2NH 111 – 202, CCS 32 – 21, HCS+ J = (1-0), H2CS 101 – 000, and weaker transi- tions of large aldehyde and O-bearing species. Of these species, CH2NH is known to be weakly masing at 1 GHz (Godfrey et al. 1973) and is masing in this survey as well; H2CS has been re- ported to exhibit non-thermal behavior at cm wavelengths (Sin- clair 1973); and transitions of the O-bearing species t-CH3CH2OH and H2COH+ exhibit weak masing in this work. While the lines included in Figure 14 appear in absorption and are therefore not themselves masing, the energy state populations may be very sen- sitive to the physical conditions. Signiﬁcant effort will be required to determine the excitation conditions responsible for the observed line strengths, providing powerful constraints on the physical con- ditions in the gas.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	While aware of the presence of non-LTE excitation in these lines, we expect that a few of the trends indicate true gas phase relative abundance differences. In this discussion, relative abun- dance differences are relative to the median abundance ratio of molecules sampled in two components of gas. In most panels, tran- sitions of CH3CH2CN appear consistently below the median value, indicating that K6: 64 km s−1 likely has a high gas phase abun- dance of the species compared to other gas components. Similarly, K6: 64 km s−1 likely has enhanced relative abundances of most S- bearing species, with only L: 56 km s−1 having NLR values that are similar to the median. As many of the O-bearing species have multiple lines with high NLRs, we expect that these do reﬂect true abundance differences.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	than they are to the ﬁrst set of components. Evidence for this follows from the signiﬁcant amount of scatter observed in lines of CH3CHO (K6: 82 km s−1), NH2CHO (K6: 82 km s−1 and K4: 62 km s−1), CH2CHCN (K4: 82 km s−1), CH3CH2CN (K4: 62 km s−1), HC5N (K4: 62 km s−1 and K4: 82 km s−1), and cis-CH2OHCHO (K4: 62 km s−1).
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	lines of NH2CHO transitions of small molecules are and CH3CHO, and most consistent with the median NLRs, although NH3 transi- tions show some variability. Observed lines of the S-bearing molecules SO, HCS+ and H2CS are comparatively depleted in L: 76 km s−1. Lines of CH3CH2CN are depleted in both L: 56 km s−1 and L: 76 km s−1 relative to K6: 64 km s−1, and lines of CH2CHCN may be slightly depleted in L: 76. km s−1. On the other hand, multiple lines of O-bearing species, in- cluding c-H2C3O (L: 56 km s−1 and L: 76 km s−1), H2COH+ (L: 76 km s−1), CH3OCHO (L: 56 km s−1), and t-CH3CH2OH (L: 56 km s−1), are enhanced in both components in L. While fur- ther analysis is required, this likely indicates higher gas phase abun- dances of the speciﬁed O-bearing species, particularly where mul- tiple lines consistently demonstrate enhancement (i.e. for c-H2C3O and t-CH3CH2OH in L: 56 km s−1). Particularly because no maser emission is detected at either component towards L, we do not an- ticipate that the energy state populations are signiﬁcantly enhanced or depleted by these processes, further supporting the interpretation of the results as indicating true differences in relative abundances. In the second set of components, inclduing K6: 82 km s−1, K4: 62 km s−1, and K4: 82 km s−1, we note low line ratios for some S-bearing species including H2CS, CCS (K6: 82 km s−1 and and HCS+ (K6: 82 km s−1). Additionally, K4: 62 km s−1), CH3CH2CN lines are lower than the median NLR towards K6: 82 km s−1 and SiO isotopologues have somewhat low values. three regions, multiple O-bearing species exhibit Towards all In K6: 82 km s−1, H2COH+, cis-CH2OHCHO, high NLRs. t-CH3CH2OH, and the nitrile species C3N, t-CH2CHCHO, exhibit enhanced line radiation relative to the median. In K4: 62 km s−1, cis-CH2OHCHO and H2COH+ have high NLRs, and in K4: 82 km s−1, lines of H2CCO and c-H2C3O have to one another, K6: 82 km s−1 and high NLRs. With respect K4: 82 km s−1 exhibit similar excitation (evidenced by NLR values for molecules with multiple line transitions), but very different line strengths for some S-bearing species. With respect to one another, they also show differences in the relative abundances of some aldehyde and nitrile species, with NLR values that are consistently offset from the median. They additionally show differ- ences in which weak O-bearing transitions are detected, possibly indicating relative abundance differences. K4: 64 km s−1 has a very similar molecular inventory to that of K6: 82 km s−1, but the excitation is somewhat distinct. Besides the relative depletion of SiO isotopologues in K6: 82 km s−1, the relative abundances are quite consistent with those in K6: 64 km s−1, indicating that these two regions have similar chemical conditions.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	cient variability is observed so that signals from different positions cannot be combined, particularly for transitions that exhibit greater degrees of variability.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Implications for the Structure of Sgr B2: If the structure of Sgr B2 included two clouds in the envelope that are extended over the continuum structure, we would observe that the molecular abundance ratios and excitation should be self consistent in the low velocity material and in the high velocity material. The low velocity material could show a very distinct pattern from the high velocity gas however. If this were the case, the panels corresponding to low velocity material in Figure 14 (panels b and d) should have very few points outside of the shaded 3σ errors. While the high velocity ma- terial (panels a, c, and e) could have points outside of this, the same points should appear outside of the shaded region in all three pan- els. As is evident from Figures 14 and from the discussion above, this is not observed. The differences in line radiation thus indicate that we cannot adopt the most simple treatment of the molecular material, namely as two extended sheets of gas that are not inter- acting directly with the structure of the core. Instead, we observe signiﬁcant variation on size scales of <30 arcsec, indicating that both the low velocity component and the high velocity component are likely located proximate to the star forming core and are not envelope components.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	While K6: 82 km s−1 and K4: 82 km s−1 are more similar to one another, differences nontheless persist. The chemistry and ex- citation may be localised to a lesser degree than the low velocity gas. While this gas does not appear to be located in the extended envelope, it may be located further from the cores than the low ve- locity gas in Sgr B2(N). As L: 76 km s−1 shows a distinct pattern of excitation and gas phase abundance, it is not clear from our obser- vations that it is related to the high velocity gas observed towards K6 and K4. However, the work of Hasegawa et al. (1994) and Sato et al. (2000) provides evidence connecting the gas clouds, indicat- ing that the high velocity material may also be part of an extended structure (> 40 arcsec). It is plausible however that a very clumpy structure in the envelope, as proposed by Goicoechea, Rodr´ıguez- Fern´andez & Cernicharo (2003), could account for the differences between the high velocity gas in K6 and K4 while allowing the gas to be located in the envelope.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	If we adopt this structure, in which the low velocity material is located closest to the HII regions, then the expanding K6 ion- ization shell would interact mechanically and radiatively with the 64 km s−1 molecular gas given the HII region kinematics compiles in Tables 5 and 6, driving shocks into the system and contribut- ing near-UV radiation to generate a PDR. These conditions could produce the more highly varying line ratios for molecules towards K6. Towards L, the low velocity molecular material would not be expected to mechanically interact with the HII regions given the recombination line kinematics measured in §4.1, although it very likely sets the radiative environment.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Furthermore, if the high velocity components are indeed fur- ther from the HII regions than the low velocity components, we may be directly observing the colliding molecular clouds that are triggering star formation in Sgr B2, consistent with the scenario ﬁrst proposed by Hasegawa et al. (1994). In this scenario, an ex- tended high velocity cloud (at 75 < vLSR < 90 km s−1) collides with the body of Sgr B2 (at 40 < vLSR < 50 km s−1), generat- ing star formation in processed material at an intermediate velocity of 55 < vLSR < 70 km s−1. The observed 2-component molecu- lar gas thus may be a direct observation of remaining material in the high velocity cloud continuing to collide with the processed cloud, a process that may trigger yet more star formation. This scenario is consistent with the observation of molecular line mate- rial at ∼65 km s−1 towards Sgr B2(M) (H¨uttemeister et al. 1995). However, signiﬁcant pieces of evidence for this scenario remain missing. For example, there should be observational signatures of a shock at the interface of the colliding gas clouds at a velocity intermediate to the high and low velocity gas components. While the line proﬁle of 28SiO is wing-broadened, the optically thin iso- topologues of SiO do not show such a signature. Additionally, to elucidate the interaction between the HII region and the low veloc- ity component molecular gas, high spatial and spectral resolution observations of CI, CII, NI, NII, and OI are required to provide in- formation on the PDR at the surface of the HII region. Notably, however, carbon recombination lines are not observed in this mate- rial.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The evidence that the low velocity material is closer to the HII regions remains tenuous however, and additional evidence sug- gests the situation may be yet more complex. Recombination lines towards K6 have 2-component proﬁles centered at velocities of 60 and 84 km s−1, while molecular gas is observed at 64 and 82 km s−1. If the K6 shell is an expanding HII region that only inter- acts mechanically with the low velocity molecular gas, then it is a coincidence that the high velocity molecular gas is at nearly the same velocity as the far side of the expanding shell. Likewise, this scenario requires that the L: 56 km s−1 molecular gas component is more closely associated with the HII region (at 77 km s−1) than the L: 76 km s−1 molecular gas component. The velocity corre- spondence between the HII region and the L: 76 km s−1 molecular gas may be merely coincidental, however, a deeper look at structure is warranted.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	literature (Greaves & Nyman 1996; Wirstr¨om et al. 2010; Gerin et al. 2010; Menten et al. 2011). The clouds are typically translu- cent (with 102 < n < 104 cm−3 and 2 < AV < 5) and are highly interesting for understanding the environments and chemistry in regions that can only be studied through line-of-sight absorption techniques. To date, authors have reported primarily di- and tri- atomic molecules in the foreground of Sgr B2 and other Galactic sources. Furthermore, it is typically presumed that the absorbing gas is extended and homogenous over the ﬁeld of view (Qin et al. 2010). The ATCA survey provides the ﬁrst sub-10” resolution view of line-of-sight molecular absorption towards any region in the Galaxy.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Additionally, the ∼2 km s−1 component is observed to pro- duce maser emission by the Class I methanol maser at 36 GHz towards K6 and K4 (see Appendix B). Maser emission has not previously been associated with a line-of-sight absorption cloud, indicating that this component is likely quite distinct from other line-of-sight clouds. Previous authors have proposed that the ∼0 km s−1 component consists of ejecta of Sgr B2 (Wirstr¨om et al. 2010). Recently, however, a molecular gas component at ∼0 km s−1 has been observed to be widespread across the Galac- tic Center (Jones et al. 2012) and diffuse recombination line emis- sion has been observed at a similar velocity throughout the Galactic Center (Royster & Yusef-Zadeh 2014).
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	the continuum and emission from one or both hot cores. Transi- tions of NH3, HCS+, CH3CN, CH2CHCN, CH3CH2CN, HC3N, NH2CHO, CH3OCHO, and CH3OCH3 are prominent in absorp- tion in the foreground of the HII regions and in emission from the LMH and usually “h” (notably, NH2CHO is not detected in “h”). Additionally, detected transitions of OCS, HNCO, and NH2D pro- duce strong emission towards the LMH and “h” and very weak ab- sorption against the radio continuum. The data also contain multi- ple instances in which molecules whose spatial distributions might be expected to be correlated exhibit very distinct morphologies.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The molecules SiO and HNCO are widely regarded as excel- lent shock tracers and therefore might be expected to exhibit sim- ilar spatial distributions. Enhanced gas phase abundances of SiO are achieved by grain sputtering upon passage of a shock (Schilke et al. 1997; Caselli, Hartquist & Havnes 1997; Pineau des Forets et al. 1997), and HNCO is observed to be enhanced in multiple re- gions that have known shocks (Rodriguez-Fernandez et al. 2010; Minh & Irvine 2006; Mart´ın et al. 2008). The spatial distribution of HNCO has been observed to follow that of SiO in multiple sources via observations with single dish telescopes (Zinchenko, Henkel & Mao 2000). However, Figure 15 shows that the distributions of tran- sitions between low-energy states of SiO and HNCO in the core of Sgr B2(N) are anti-correlated. The HNCO 202-101 transition pro- duces very strong line emission in both hot cores, whereas only very weak absorption is detected towards K6 (the absorption con- tour of HNCO is at 2.5% of the peak emission line strength). On the other hand, for the J = (1-0) transitions of all isotopologues of SiO, the images include absorption against the continuum struc- ture and a negative “bowl” indicating extended emission that could not be recovered with the interferometer. No compact emission by SiO isotopologues is observed towards the “h” hot core. We note the presence of a weak emission feature on the high velocity wing of 28SiO eminating from the northwest edge of the LMH. Due to the line density towards the LMH and the absence of an emission feature on the low velocity wing, which would be more consistent with the velocity of the LMH, we cannot ﬁrmly identify the noted feature as an SiO line.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The distributions indicate chemical differentiation in Sgr B2(N). HNCO is highly abundant in the LMH and “h” hot cores and depleted in the molecular gas observed against the HII re- gions; on the other hand, even if the noted wing feature is from SiO, the species has a signiﬁcantly lower column density in the LMH than in the absorbing gas. Any emission generated by SiO in the LMH is entirely overwhelmed by the absorbing gas com- ponents, including in the optically thin transitions of the isotopo- logues. Although the higher energy states (100 K < EL < 300 K) of SiO should be preferentially populated in the LMH and “h”, Neill et al. (2014) reported the non-detection of SiO in HEXOS data sampling SiO transitions with T >160 K. SiO thus appears to be depleted from the gas phase in the LMH and “h” hot cores, but highly abundant in molecular material in the low velocity (50 - 65 km s−1) and high velocity (70 - 85 km s−1) gas observed in ab- sorption towards the HII regions, and in extended material around the core. A similar distribution of gas phase SiO has been observed in the source G34.26+0.15 (G34.26), which contains a hot core, UC HII regions, collimated outﬂows, and molecular gas interacting with stellar winds (Hatchell, Fuller & Millar 2001; Watt & Mundy 1999). Towards G34.26, SiO was observed to be widespread and highly abundant in the material interacting with stellar winds; it was not detected in the hot core, however (Hatchell, Fuller & Mil- lar 2001).
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	family present yet another puzzle; whereas transitions of CS, CCS, CCCS, and H2CS are detected in the absorbing gas with no appar- ent emission from the hot cores, line radiation from the J = (3-2) transition of OCS shows a similar behavior to that of HNCO (Fig- ure 16). Very weak line absorption by OCS (with the lowest con- tour at 4.5% of the peak emission) is detected towards K6 and K4, but nearly all of the OCS radiation is emission produced in the hot cores. OCS, like HNCO, is believed to form on grain mantles and has been linked to shock enhancement (Ren et al. 2011; Garozzo et al. 2010). Distinct from other CS-containing species, HCS+ ap- pears both in absorption against the continuum and in emission from the hot cores.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The detection of SiO, CS, and other CS-bearing species and the depletion of HNCO and OCS in the absorbing molecular material provides insight into the physical environment in the absorbing gas. The strong absorption by SiO likely results from the presence of shocks and supersonic turbulence in the gas, although the enhance- ment of SiO may instead be attributed to the high X-ray ﬂux in this X-ray reﬂection nebula (Mart´ın-Pintado et al. 2000; Amo-Baladr´on et al. 2009). Other CS-bearing species found in the absorbing gas, particularly CCS and CCCS may be formed by chemistry driven by the radical CH (Petrie 1996; Yamada, Osamura & Kaiser 2002; Sakai et al. 2007), and multiple larger radical species are directly detected in the spectra towards K6, L, and K4 (Appendix B). The presence of radicals typically indicates X-ray or UV irradiation typ- ical of XDRs or PDRs (Danks, Federman & Lambert 1984; Boger & Sternberg 2005; Meijerink, Spaans & Israel 2007).
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Further, the ratio of HNCO to CS may be an excellent diagnos- tic of the radiative environment. Using a limited sample of Galac- tic Center clouds, in which shocks are believed to be ubiquitous, Mart´ın et al. (2008) demonstrated that the HNCO:CS abundance ratio is high (∼70) in non-PDR shocked regions and an order of magnitude lower in PDRs. While both species can be photodisso- ciated, CS is less easily destroyed by UV radiation and can be re- plenished by a gas phase reaction of S+, an abundant ion in PDRs (Drdla, Knapp & van Dishoeck 1989; Sternberg & Dalgarno 1995; Mart´ın et al. 2008). In the Horsehead Nebula, CS is observed to be nearly equally abundant in the PDR and in the dense UV-shielded core (Goicoechea et al. 2006). On the other hand, HNCO is be- lieved to form efﬁciently on grain mantles (Hasegawa & Herbst 1993) and after being released into the gas phase, it is easily pho- todissociated by UV radiation (Mart´ın et al. 2008; Roberge et al. 1991). While theoretical work contends that the HNCO:CS ratio is time-dependent and more indeterminate (Tideswell et al. 2010), the very high ratio of CS:HNCO in the absorbing gas provides ev- idence that this material has an enhanced UV radiation ﬁeld. Like HNCO, OCS is believed to form on grains, and the species has been directly observed on ice grain mantles (Gibb et al. 2004; Garozzo et al. 2010; Palumbo et al. 2011; Ren et al. 2011). Gas phase OCS is destroyed by near-UV radiation with an even higher photodissoci- ation rate than HNCO (Roberge et al. 1991; Sternberg & Dalgarno 1995; Sato et al. 1995). The nearly complete depletion OCS from the absorbing gas is further evidence for PDR conditions.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	HNCO and OCS exhibit widespread distributions to the west of the core (Jones et al. 2008). In this material, we expect that the X-ray ﬁeld is more typical of the bulk of material in Sgr B2. This expecta- tion is consistent with the prevalence of the X-ray tracer CN in the material to the west of the core (Jones et al. 2008). The widespread distributions thus suggest that X-ray radiation is not responsible for destroying HNCO and OCS; as such, we interpret the ATCA dis- tributions of CS, OCS, and HNCO as suggesting the presence of a strong UV radiation ﬁeld in the absorbing gas.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Additional support for a strong UV radiation ﬁeld in the ab- sorbing material follows from the results of §5.1. First, the highly subthermal excitation indicates that the material is sub-critically dense, with densities typical of PDRs. Without an excitation analy- sis, it is possible to estimate an upper limit of n (cid:46) a few × 105 cm−3 using the critical densities of energy states that had undetected tran- sitions in the 30 - 50 GHz range with high quantum line strengths. H¨uttemeister et al. (1995) determined an upper limit to the den- sity of the absorbing gas of n <104 cm−3 based on the fact that the HC3N J = (3-2) transition is in absorption instead of emission, and independently based on an analysis of the (1,1) through (6,6) NH3 inversion transitions. However, the latter estimate may not hold if the H3O+ formation pumping excitation mechanism pro- posed in Lis et al. (2012) proves effective for NH3. Densities of 103 < n < 105 cm−3 are typical of PDRs (Teilens et al. 1993; Hol- lenbach & Teilens 1997), and densities of n ≈ a few × 105 cm−3 are reasonable in the hot, supremely dense environment at the core of Sgr B2(N). In section §5.1, we determined that both absorbing gas components appear to be local to the star forming cores rather than in the extended envelope material, with the low velocity gas possibly closer to the interface of the HII regions. If the absorbing gas clouds are proximate to the HII regions, it is not only reason- able, but structurally required, that they are or contain PDRs.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Second, the species l-C3H+ is detected in the absorbing ma- terial, but not in the hot core. To date, l-C3H+ has only been ﬁrmly detected in the Horsehead Nebula, the Orion Bar, and Sgr B2(N), despite a deep search towards multiple hot cores, hot corinos, and dark clouds (Pety et al. 2012; McGuire et al. 2013, 2014). Addi- tionally, the species was not detected in molecule-rich UV-shielded core in the Horsehead Nebula (Pety et al. 2012). The species is clearly associated with UV-irradiated gas, and its detection in the low and high velocity gas towards K6 supports the hypothesis that both gas components are PDR sources.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Finally, a few of the organic species detected in the absorbing clouds were recently reported to be highly abundant in a PDR envi- ronment (Gratier et al. 2013; Guzm´an et al. 2014). CH3CN, HC3N, H2CCO, and CH3CHO were observed in the Horsehead PDR with abundances enhanced by a factor of 1 - 30 as compared to abun- dances in the UV-shielded core in the Horsehead. While the species are known to be abundant in multiple distinct environments, their detection in the absorbing material in Sgr B2(N) is consistent with the absorbing material being a PDR.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The data and discussions in this work thus support a structure in which molecular gas at 50 < vLSR < 70 km s−1 is located on the interface of the HII regions at the core of Sgr B2. This gas compo- nent is UV-irradiated by the O-stars that power the HII regions, and may mechanically interact with the HII region, particularly towards K6 where the shell-shaped HII region is thought to be expanding. The high velocity cloud (70 < vLSR < 85 km s−1) is also close to the core, although it may not be as deeply embedded as the low velocity material. It may be actively colliding with the low velocity cloud consistent with a cloud-cloud collision scenario (Hasegawa et al. 1994), although additional evidence is required to verify this.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Based on the detection of l-C3H+ and the nearly complete deple- tion of OCS and HNCO, it appears that both absorbing clouds are PDRs. This hypothesis should be further tested via comparison to high spatial and, ideally, high spectral resolution observations of neutral and ionized carbon, sulfur and oxygen.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	However, merely calling the gas clouds PDRs is an over- simpliﬁcation. The clouds contain supersonic turbulence requiring shocks, have a high X-ray ﬂux, and likely have an enhanced cos- mic ray ﬂux as compared to most Galactic sources. It is quite pos- sible that a different factor drives the chemistry of different groups of species; for instance, the strong X-ray ﬂux may produce radi- cals that drive the chemistry of imine molecules (with an X=N– H group), whereas UV radiation may drive the sulfur and carbon- chain chemistry and selectively destroy certain molecules. Mean- while, shocks may liberate silicon-bearing species and dust mantle species to be processed in the soup of ionic, atomic, and molecular gas.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Regardless of the physical parameters driving the formation of each species, the extracted spectra suggest that a multitude of organic molecules, including nitriles, aldehydes, alcohols, and imines, either form or persist in the presence of UV irradiation, X-ray radiation, and shocks. While we expect that some of the ob- served species will not prove ubiquitous in more typical Galactic PDRs, all of the detected species can at least survive moderate UV radiation ﬁelds and are promising targets for exploring the chemical complexity in PDRs.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Continuum images line data cubes are available at at http://cdsarc.u-strasbg.fr/viz- bin/qcat?J/MNRAS/. Spectra extracted from ﬁve positions, re- including the LMH and “h” hot cores and three HII and at gions available https://github.com/jfc2113/MicrowaveLineFitter. ex- traction from the data cubes, the spectra have had a primary beam correction and a baseline correction applied (Figure 2).
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	In this work, we developed and applied PYTHON scripts to ﬁt the line emision and absorption in 1-d spectra extracted from the data cubes at the positions of three HII regions in Sgr B2. The code provides a purely empirical output rather than comparing with a model, as this was deemed most appropriate at centimeter wave- lengths where line radiation is often nonthermal. The code then compares with output generated by Splatalogue or another line cat- alog after minor formatting. The code is described in §3, released in Appendix A, and available at the same locations as the extracted spectra.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	velocity components. The performance of the code was evaluated based on the handling of 65 recombination line proﬁles and 90 molecular absorption lines, observed against the free-free radio continuum emission. Typical errors on the ﬁts were signiﬁcantly lower than the channel width in this survey. Approximate errors are summarized in Table 7.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Using the output of the line-ﬁtting script, the molecular in- ventories of two primary velocity components of absorbing gas were determined and the approximate excitation conditions were characterised, showing that the gas clouds that give rise to the ab- sorption are highly subthermally excited (TEx ∼ 10 K (cid:28) Tkin). Line radiation was then inspected using the relative line strengths in different gas components to determine whether the molecular line radiation is self-consistent at distinct spatial positions (Figure 14). This methodology can determine the chemical similarity be- tween clouds at different spatial positions or velocities under con- ditions of similar excitation conditions, and the method could be quantiﬁed to further establish the degree of similarity. In the era of broadband spectral line observing, the method could be highly useful for quickly determining if observed clouds are chemically distinct and demonstrating relative enhancement of different fami- lies of molecules. The application of this method toward a variety of sources is necessary to determine its utility.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	We applied the method to the ATCA data in order to test whether the molecular gas radiation is consistent with an extended distribution that is not interacting with the local structure of the core, as would be expected if the absorbing gas is in the envelope. We determined that in both primary velocity components, the three spatial regions show statistically signiﬁcant differences in their line radiation, indicating differences in either line excitation or gas- phase molecular abundances. Many of the differences appear to be due to relative gas-phase abundance differences. The molecu- lar gas in the low velocity component proved more highly variable, indicating that it may be subjected to more localized conditions and therefore located deeper within the star-forming core. It may be on the interface of the HII regions, potentially interacting me- chanically and radiatively with the structure of the HII regions. In agreement with cloud-cloud collision scenarios, the 82 km s−1 gas is likely located further from the core. However, required pieces of observational evidence for the proposed structure remain unavail- able, particularly high spatial (few arcsec resolution) and spectral (10 - 25 km s−1) resolution images of atomic and ionic lines of carbon, oxygen, and sulfur.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Additionally, the chemistry observed in translucent clouds in the line-of-sight to Sgr B2 was brieﬂy discussed. The data contain line-of-sight absorption by transitions of ∼15 different molecules, including transitions from NH2CHO, which was previ- ously not observed in translucent gas. An absorbing component at ∼0 km s−1 proved to vary signiﬁcantly in its optical depth across the spatial ﬁeld, revealing that it contains signiﬁcant density struc- ture.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	ple classes of oxygenated species including aldhydes and alcohols, (5) nitrogen species of pre-biotic importance including nitriles and imines are abundant in the exotic PDR environments.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	This work highlights only a small fraction of the scien- tiﬁc capacity of the dataset presented and provides a tool to en- able researchers to accomplish additional projects. Towards the HII regions, follow-up work including multi-transition analyses of molecules are facilitated by the full-spectrum line identiﬁcations provided herein. Furthermore, the data complement existing single dish data covering the same frequency range, namely the PRIMOS survey (Neill et al. 2012). Finally, the spectra extracted from the LMH and “h” are very line-dense and they remain uncharacterised. These spectra will provide unprecedented detail on hot core chem- istry in Sgr B2(N).
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	We thank the anonymous referee for the helpful comments that im- proved the quality of this work. The Australia Telescope Compact Array is funded by the Commonwealth of Australia for operation as a National Facility managed by CSIRO. J.F.C. gratefully ac- knowledges funding by the Grote Reber Doctoral Research Pro- gram of the National Radio Astronomy Observatory (NRAO) and by the East Asian and Paciﬁc Summer Institude of the National Science Foundation (NSF). The work was supported by the NSF grant OISE-1310963. The NRAO is a facility of the National Sci- ence Foundation operated under cooperative agreement by Asso- ciated Universities, Inc. NL’s postdoctoral fellowship is supported by CONICYT/FONDECYT postdoctorado, under project number 3130540. LB gratefully acknowledges support by CONICYT Grant PFB-06.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Minh Y. C., Irvine W. M., 2006, NewAst, 11, 594 M¨uller H. S. P., Schlder F., Stutzki J., Winnewisser G., 2005, J.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Table 4. Sample page of table providing line identiﬁcations and Gaussian ﬁt parameters towards K6. See Appendix C for full tables towards K6, L, and K4, and for a full description of labeling conventions.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Table 5. Best ﬁts to the composite average of recombination line transitions. The errors quoted are from the diagonal terms of the covariance matrix. However, the matrices contain signiﬁcant covariance terms, causing the error to be highly overestimated. Fits to the composite average of Hα and Hβ transitions agree to a precision that is signiﬁcantly smaller than the covariance matrix output.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Table 6. Mean parameters and standard errors of ﬁts to Hα - γ transitions. Parameters are calculated from 1-component ﬁts towards L, 2-component ﬁts towards K6, and 1-component ﬁts towards K4.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Table 7. Recommended errors to ﬁts to the ATCA data output by the automated line ﬁtter.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Table 8. Kinematic measurements of molecular gas towards K6, L, and K4 from lines in the 25th - 90th percentile signal-to-noise ratio.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 1. Continuum image of Sgr B2(N) and L at 40 GHz showing the elliptical regions from which spectra were extracted. The dashed line arcs point out the K5 and K6 shell-shaped HII regions. The synthesized beam shape is shown in the lower left corner.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 2. A segment of the full spectral coverage towards K6. a. shows the spectrum extracted from the data cubes and b. is the spectrum after baseline removal using a Hodrick-Prescott ﬁlter.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 3. 40% and 80% integrated line emission contours over 40 GHz continuum illustrate the two most common distributions of molecular line radiation. a. CH2CN (2 - 1), CH3CHO (202 - 101), and Z-CH3CHNH (202 - 101) are observed in absorption (dashed contours) preferentially towards the K5 and K6 shells. b. NH2D (313 - 303), CH3CH2CN (431 - 330), and SO2 (193,18 - 183,15) are detected in emission towards the LMH hot core, and NH2D and CH3CH2CN are detected in emission towards “h”. The central positions of the elliptical regions from which spectra were extracted are indicated by the crosses. The synthesized beam shapes for the continuum (thin black ellipse) and spectral lines (colored ellipses) are shown in the lower left corner.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 4. A representative segment of spectra extracted from all ﬁve regions targeted in this study. Whereas the LMH and “h” have line-dense spectra dominated by molecular line emission, K4, L, and K6 have lower line densities with molecular lines observed in absorption and recombination lines in emission.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 5. A sample page of the full data spectrum extracted towards K6 with line identiﬁcations labeled and the output of the automated line-ﬁtter overlaid. See Appendix B for full ﬁgures for K6, L, and K4.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 6. Hydrogen and Helium α recombination lines towards K6. Proﬁles of the H(59) - H(53)α transitions, extracted from high spatial resolution data, show a double peaked proﬁle, whereas the H(52) and H(51)α transitions are well ﬁt by a single component. The averaged proﬁle is best ﬁt by a Gaussian centered at 59 km s−1 and a weaker component at 84 km s−1. The solid and dashed vertical lines represent the H and He α transitions at 75 km s−1 respectively.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 7. Hydrogen and Helium β recombination lines towards K6. Proﬁles of the H(75) - H(66) β transitions, extracted from high spatial resolution data, show a double peaked proﬁle, whereas the H(65) and H(64)β transitions are well ﬁt by a single component. The averaged proﬁle is best ﬁt by a Gaussian centered at 60 km s−1 and a weaker component at 84 km s−1. The solid and dashed vertical lines represent the H and He β transitions at 75 km s−1 respectively.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 8. Hydrogen and Helium α recombination lines towards L are best ﬁt by a Gaussian at 76.5 km s−1. The solid and dashed vertical lines represent the H and He β transitions at 75 km s−1 respectively.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 9. Hydrogen and Helium α recombination lines towards K4 show unusual line shapes that are highly sensitive to the spatial resolution of the image from which they were extracted. The lines are typically best ﬁt by two primary components. In addition, a weaker wing component appears at ∼100 km s−1. The solid and dashed vertical lines represent the H and He β transitions at 75 km s−1 respectively.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 10. Velocity center of H α - γ transitions ﬁt towards L, K6, and K4. a. includes transitions towards K6 that were ﬁt with 2 Gaussian components; b. includes unblended transitions towards L; c. includes all transitions towards K4. Components with v > 80 km s−1 towards K4 are ﬁtting a high velocity wing that is not a primary components of the line proﬁle.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 11. Channel maps of the H(53)α transition at 42951 MHz shown in contours over the 40 GHz continuum in greyscale. The synthesized beams for the continuum and line images are shown in the lower left corner. The ﬁgure shows a strong velocity gradient from the NW to the SE of K4. The crosses mark the center positions of the regions selected for K4 and K6 (at the base of the K6 shell) and indicate the peak position of the 90 km s−1 gas. Notice the offset positions of high and low velocity gas on the K6 shell and the velocity gradient observed at K4.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 12. Kinematic parameters of molecular line ﬁts output by the automated line ﬁtter. 1-component ﬁts shown include ﬁts with a signal-to-noise ratio in the 25th to 90th percentile of all unblended 1-component ﬁts. 2-component ﬁts shown include unblended transitions in which the low velocity component has a signal-to-noise ratio in the 25th to 90th percentile of the low velocity components of all 2-component lines.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 13. Line proﬁles of CH3CHO 211-110E (39362 MHz), CH3CH2CN 404-303 (35722 MHz), CH2NH 111-202 (33704 MHz), and H2CS 101-000 (34351 MHz) reveal differences in the line optical depth ratios between all spatial positions and all velocity components. The data (cyan) are overlaid with the automated line ﬁts (black), and vertical dotted lines indicate the velocities associated with the low and high velocity gas towards each spatial region as reported in Table 8. a. In K6, while CH3CHO and CH3CH2CN have similar line proﬁles, the selected transition of CH2NH has a comparatively enhanced line strength in K6: 82 km s−1. Further, the line radiation by H2CS in the K6: 82 km s−1 is signiﬁcantly depleted to below the detection threshold. Towards L (b.) and K4 (c.) the line proﬁles vary signiﬁcantly, and line strengths do not scale linearly with the strengths observed towards K6. The scale of the ordinate is the same for each line towards all three regions, and the scale is set relative to the normalised line height in the K6: 62 km s−1 component. The root-mean-squared values of the extracted spectra are shown in the lower left corner of each panel.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 14. Line height ratios with respect to the K6: 64 km s−1 component indicate statistically signiﬁcant differences in gas phase molecular abundances or excitation between different spatial positions. In the plot, the abscissa shows the line height-to-continuum ratios of lines detected in K6: 64 km s−1. Given the low excitation temperatures observed in this gas and the high background continuum temperature, this provides a proxy of the line optical depth towards K6: 64 km s−1. The ordinate provides the normalized line ratios (NLRs) of the lines in the relevant gas component with respect to K6: 64 km s−1. For all but the strongest lines, the y-axis values are proportional to the line optical depth ratio with respect to K6: 64 km s−1. The median ratio is represented with the dashed line, and shading indicates the ±3σ range. Represented data only include lines within the high spatial resolution portion of the spectrum (from 30 - 44.6 GHz) for selected families of molecules. Species are numbered as follows: 1 - 4 are SiO, 29SiO, 30SiO, and SiS. 5 - 9 are sulfur-bearing species SO, CCS, HCS+, H2CS, and OCS. 10 - 16 are nitrile species HC3N, CH3CN, CH2CN, CH2CHCN, CH3CH2CN, HC5N, and CCCN. 17 - 21 are aldehyde species NH2CHO, CH3CHO, CH3OCHO, cis-CH2OHCHO, and t-CH2CHCHO. 22 is NH3. 23 - 27 are non-aldehyde oxygenated species H2CCO, c-H2C3O, H2COH+, CH3OCH3, and t-CH3CH2OH. 28 is the imine species CH2NH. Note that the apparent slope towards the low optical depth segment of the panels is related to the detection limit; because K6: 64 km s−1 has the highest optical depth, lines that are weakly detected towards K6: 64 km s−1 are detected towards other components only if they have higher than the median line optical depth. This does not imply that these points are unreliable, but implies that the lower left corners of each panel are below the detection threshold.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 15. Integrated line contours of the (1-0) transitions of SiO isotopologues are anticorrelated with the 202-101 transition of HNCO. a. 29SiO, 30SiO, and Si18O are observed in absorption (dashed contours) against the continuum and exhibit no emission associated with the hot cores. b. HNCO is present in emission towards the LMH and “h” hot cores. A very weak absorption feature (with a contour at 2.5% of the peak line strength) is detected slightly north of K6. The crosses indicate the center positions of the elliptical regions from which spectra were extracted. The synthesized beams for the continuum and line images are shown in the lower left corner.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Figure 16. Integrated line contours of H2CS (101-000), CCS (32-31), and OCS (3-2). a. H2CS and CCS are observed in absorption (dashed contours) against K4, K5, K6, and L, with no detected emission from hot core regions. Contour levels are at 20, 40, 60, and 80 % of the peak line strength. b. OCS appears in emission from the hot core regions, the LMH and “h”, with very weak absorption observed towards K6 and near K4. The contour levels are at 4.5, 7, 20, 40, 60, and 80 % of the peak line strength. The crosses indicate the center positions of the elliptical regions from which spectra were extracted. The synthesized beams for the continuum and line images are shown in the lower left corner.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	All scripts used for automated line ﬁtting and semi-automated line identiﬁcation are provided in the online journal and in GitHub at https://github.com/jfc2113/MicrowaveLineFitter. We provide in- structions for use including a description of the input and output ﬁles and system requirements.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	The scripts have been tested in IPYTHON version 0.12.1 with PYTHON 2.7.3, NUMPY 1.6.1, and MATPLOTLIB 1.1.1rc. PYTHON 2.7.3 includes the MATH and CSV modules that are used in the scripts.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	FullCode requires a spectrum input in standard ascii for- mat with frequency and baseline subtracted intensity. The ﬁle ‘K6 fullSpec.txt’, provided with the scripts, is in the correct format. If the user chooses to perform line identiﬁcation, the script requires ﬁles containing recombination line or molecular line data output in csv format from any spectral line catalog. The line data should be formatted similarly to the ﬁles ‘recombLines.csv’ and ‘molecu- larLines.csv’, provided. If your source has recombination lines, we recommend preserving the ﬁle name of ‘recombLines.csv’, but in- cluding lines that are relevant for your frequency range. All csv ﬁles read into the code should be text csv ﬁles with UNICODE UTF-8 characters and a colon (:) as the ﬁeld delimiter. The code outputs csv ﬁle data with this format as well.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	two measures of the quality of the ﬁt. First, the rms residual between the data and the ﬁt; second, for 1-component ﬁts, the ﬁle provides the offset between the frequency at which the data has its maximum absolute value and the Gaussian center frequency. Fits with abnormally large values may be suspect.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	spectrum. These should be inspected and possibly re-ﬁt manually. Bin = 5 lines are entirely unreasonable, with very broad line widths or unreasonable line heights. They are used as placeholders to en- able the code to proceed, and they will need to be re-ﬁt. Of 617 components used to ﬁt the spectrum towards K6, 360 have bin = 0, 242 have bin = 1, 1 has bin = 2, 12 have bin = 3, and 2 have bin = 5.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	possible atomic or molecular carriers of the detected Gaus- sian features. The output ﬁle lists any transitions from the input re- combLines.csv and molecularLines.csv ﬁles that are within a spec- iﬁed velocity range of the detected features. It does not choose the most probable line at this step, but does so when the second code is run. These can be prioritized before running VelocityFileMaker.py to obtain the ﬁnal line results.
Processed_An_ATCA_Survey_of_Sagittarius~B2_at_7~mm:_Chemical.txt	Upon inspection of the results and manual updates to any bad ﬁts, run VelocityFileMaker to generate a ﬁnal output csv ﬁle con- taining line identiﬁcations and Gaussian ﬁt parameters, with pa- rameters listed in velocity space. VelocityFileMaker generates an output csv ﬁle entitled ‘velocity ALL (r).csv’. The ﬁle contains line rest frequencies, species, transitions, Gaussian ﬁt velocities, heights, widths (in velocity units), center frequencies, and line types.

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.