A 30m Large Program: The CO Line Atlas of the Whirlpool Galaxy Survey (CLAWS)

. Robust knowledge of the distribution, amount, and physical / chemical state of the cold molecular (H 2 ) gas is key to understanding galaxy evolution. With the help of multi-CO line observations, it is possible to study the molecular gas distribution and disentangle numerous physical and chemical processes that shape and govern the molecular interstellar medium (ISM). For the ﬁrst time, we obtain full-galaxy mapping data of faint CO isotopologues ( 13 CO, C 18 O, C 17 O) at 1mm and 3mm wavelengths across the disk of the nearby spiral galaxy M51. With the help of these CO isotopologues, it is possible to constrain the bulk physical and chemical conditions in the molecular gas. We study potential explanations for why CO isotopologue emission varies. Likely drivers include CO abundance variations due to selective nucleosynthesis and changes in the optical depth. Our analysis concludes that a combination of variation in opacity and relative abundances is the dominant driver for the observed CO isotopologue ratio trends on large (kpc) scales. In contrast, abundance variation due to selective photodissociation and chemical fractionation seem to only play a minor or negligible role on galaxy-wide scales.


Introduction
The CO Line Atlas of the Whirlpool Galaxy Survey (CLAWS), an IRAM 30-m large program (#055-17), targets the entire molecular disk of the nearby spiral galaxy M51. With the help of the Eight MIxer Receiver (EMIR; [1]), several molecular emission lines, in particular the CO isotopologues, are observed in the 1 mm (⇠220−230 GHz) and 3 mm (⇠90−110 GHz) with a total of 149 h (109.2 h on-source time) between 2017 and 2019. We published the survey paper in early 2022 [2]. The galaxy M51 is an excellent candidate for faint CO isotopologue studies: It is relatively nearby (D⇡8.6 Mpc; [3]), oriented edge-on, and the molecular gas dominates the inner ⇠5−6 kpc region [4,5]. Figure 1 shows the galaxy and indicates the observed IRAM 30m field-of-view. A wealth of ancillary data exists for the galaxy across all wavelength regimes. The key science question of the large program using the 30m observations are:

Science with Multi-CO Line Ratios
While H 2 is the most abundant molecule in the interstellar medium (ISM), CO and its isotopologues have become a workhorse tracer of the overall molecular gas amount, distribution, and conditions (such as temperature, density, and opacity) [6]. Line ratios from the emission of di↵erent species and transitions capture information regarding the ISM environment and conditions. Ratios of the optical thick 12 CO line of di↵erent J ! J−1 transitions are sensitive to changes in temperature and density of the emitting gas [7]. Furthermore, CO isotopologues provide constraints on optical depth and the abundance of the di↵erent species, since 13 CO and C 18 O remain optically thin across most of the galaxy's disk [8]. The di↵erent CO isotopologue species originate from various physical and chemical processes, such as selective nucleosynthesis, chemical fractionation, or selective photodissociation. Consequently, trends in abundance variations open a window into studying the enrichment history of the ISM [9].

The CO Line Ratio R 21
The R 21 ⌘ 12 CO(2−1)/(1−0) line ratio has been studied on kpc-scales across M51 before with CO(1-0) observations from the Nobeyama Radio Observatory (NRO) 45m telescope [10] and CO(2-1) data from the IRAM 30m large program HERA CO Line Extragalactic Survey (HERACLES; [11]). Already [12] found a clear variation of R 21 between the arm and interarm region. Using the CLAWS 12 CO(2-1) observations in combination with 12 CO(1-0) data from PAWS [13], we found larger R 21 values in the interarm region (the volume weighted average and 16th to 84th percentile range is hR int 21 i = 0.95 +0.21 −0.10 ) as opposed to the spiral arm region (hR arm 21 i = 0.86 +0.10 −0.07 ). Based on a Kolmogorov-Smirnov test, the R 21 distributions for arm and interarm are di↵erent with a p-value of 4 ⇥ 10 −10 . Fig. 2 illustrates the result of the binned line ratios by spiral phase. The red and blue bands indicate spiral phases that define the spiral arm. Our finding is in some tension with some of the prior work on this galaxy, e.g.
[12], which found exactly the opposite arm-interarm trend. What separates this study is the comprehensive and detailed checking of the contribution of error beams and obtaining even IRAM 30m Director's Discretionary Time (DDT) observations of selected pointings in the arm and interarm regions of M51 [2]. We note that such a finding is actually not unexpected.

Science with Multi-CO Line Ratios
While H 2 is the most abundant molecule in the interstellar medium (ISM), CO and its isotopologues have become a workhorse tracer of the overall molecular gas amount, distribution, and conditions (such as temperature, density, and opacity) [6]. Line ratios from the emission of di↵erent species and transitions capture information regarding the ISM environment and conditions. Ratios of the optical thick 12 CO line of di↵erent J ! J−1 transitions are sensitive to changes in temperature and density of the emitting gas [7]. Furthermore, CO isotopologues provide constraints on optical depth and the abundance of the di↵erent species, since 13 CO and C 18 O remain optically thin across most of the galaxy's disk [8]. The di↵erent CO isotopologue species originate from various physical and chemical processes, such as selective nucleosynthesis, chemical fractionation, or selective photodissociation. Consequently, trends in abundance variations open a window into studying the enrichment history of the ISM [9].

The CO Line Ratio R 21
The R 21 ⌘ 12 CO(2−1)/(1−0) line ratio has been studied on kpc-scales across M51 before with CO(1-0) observations from the Nobeyama Radio Observatory (NRO) 45m telescope [10] and CO(2-1) data from the IRAM 30m large program HERA CO Line Extragalactic Survey (HERACLES; [11]). Already [12] found a clear variation of R 21 between the arm and interarm region. Using the CLAWS 12 CO(2-1) observations in combination with 12 CO(1-0) data from PAWS [13], we found larger R 21 values in the interarm region (the volume weighted average and 16th to 84th percentile range is hR int 21 i = 0.95 +0.21 −0.10 ) as opposed to the spiral arm region (hR arm 21 i = 0.86 +0.10 −0.07 ). Based on a Kolmogorov-Smirnov test, the R 21 distributions for arm and interarm are di↵erent with a p-value of 4 ⇥ 10 −10 . Fig. 2 illustrates the result of the binned line ratios by spiral phase. The red and blue bands indicate spiral phases that define the spiral arm. Our finding is in some tension with some of the prior work on this galaxy, e.g.
[12], which found exactly the opposite arm-interarm trend. What separates this study is the comprehensive and detailed checking of the contribution of error beams and obtaining even IRAM 30m Director's Discretionary Time (DDT) observations of selected pointings in the arm and interarm regions of M51 [2]. We note that such a finding is actually not unexpected.
Previous studies studying other nearby galaxies also found higher R 21 values in the interarm region [14][15][16] and they are not unphysical: Potential explanations could be the presence of di↵use gas at higher excitation temperature and lower optical depths in the interarms which hence leads to brighter CO(2-1).  Figure 3 shows line ratio trends as a function of the star formation rate (SFR) surface density, ⌃ SFR , for four selected CO isotopologue ratios. ⌃ SFR scales with the average gas density and temperature in the molecular ISM [17]. Hence, ⌃ SFR is a potential proxy of varying environmental conditions. All selected ratios show a clear trend with SFR. Regarding global, galaxy-wide drivers, the observed trends agree with abundance variations from selective nucleosynthesis and changes in the opacity of the CO emitting gas. Since 13 CO, C 18 O, and C 17 O are optically thin, their variation in the line ratio is mainly driven by changes in their relative abundance. Chemical fractionation which converts 12 CO into 13 CO is likely not a driver. This process increases the relative abundance of 13 CO . Chemical fractionation is more efficient in the colder gas regions, hence at low ⌃ SFR . So the trend we see in the 13 CO/C 18 O ratio would actually be in agreement with the process as the main driver. However, we would expect the opposite trend in 12 CO/ 13 CO line ratio. In contrast, nucleosynthesis could explain the, relatively speaking, higher C 18 O and lower 13 CO abundances and hence the observed trends. In addition, the opacity could vary. This would a↵ect mainly the optically thick 12 CO emission line. Similarly to the explanation of the increased R 21 ratio, the fact that the ratios with 12 CO(1-0) in the denominator drop toward higher ⌃ SFR could indicate the presence of di↵use, more optically thin CO gas in regions with lower SFR. A combination of the above mentioned e↵ects likely drives the overall variation in CO isotopologue line ratios.

Physical and Chemical Drivers of CO Isotopologue Ratio Trends
We also want to highlight the detection of C 17 O(1-0) across a range of SFRs. The C 17 O/C 18 O(1-0) ratio can trac the primary and secondary processing of oxygen due to nucleosynthesis (similar to 13 CO/ 12 CO, however, since 12 CO is optically thick, that ratio does not necessarily trace changes in abundance). We find a C 17 O/C 18 O(1-0) ratio similar to the average in the Milky Way solar neighborhood [18]. From nucleosynthesis alone, we would, however, expect an opposite trend. Both isotopologues have very low abundance in general, so selective photodissociation likely plays a more relevant role for this ratio trend. Both species are not well shielded by the other, more abundant isotopologues (due to di↵erences in the wavelengths of their ultraviolet absorption lines). The C 18 O isotopologue is more abundant than C 17 O. Consequently, there can be regions within molecular clouds where C 18 O is still self-shielding, while C 17 O molecules are photodissociating. This could explain why we see an increase of relative C 18 O abundance where star formation is more active (and hence the photodissociating radiation field stronger).  [18], (2) from [19], and (3) from [20].