Warm and dense molecular gas in NGC 1068

Coordinator: J.-U. Pott , M. Krips , A. Eckart & L. Lindt


Program is available and data products can be downloaded

Scientific justification: The Seyfert galaxy NGC 1068

NGC 1068, at a redshift of z=0.004, is one of the best studied Seyfert galaxies. It is regarded to be the archetype for unification schemes of Seyferts. Khachikian & Weedman (1974) first classified NGC 1068 as Seyfert 2 due to the detection of unpolarized (direct-path) emission lines with narrow widths associated with the Narrow-Line-Region (NLR). Also, polarized (scattered-path) emission lines with broad line widths were found (Antonucci & Miller et al. 1985) strongly suggesting a hidden Seyfert 1 nucleus in NGC 1068. This nicely fits into the unified theory generally proposed for Seyfert galaxies. The nuclear activity in NGC 1068 is quite pronounced. Its bolometric luminosity is estimated to a few 10 L indicating - in terms of Eddington luminosity - a relatively strong accretion. To understand the accretion processes, the analysis of the molecular gas content is an essential and indispensible piece of information. The molecular gas in NGC 1068 was mapped at high angular resolution with the IRAM Plateau de Bure Interferometer (PdBI) by Schinnerer et al. (2000, and references therein) using CO(1-0) and CO(2-1) . It is distributed in a two arm spiral, along a bar and in a warped nuclear ring with two emission peaks. The distribution of dense gas traced by HCN(1-0) and CO(1-0) has also been observed (e.g. Tacconi et al. 1997, Krips et al in prep.). While HCN appears to be more centralised, the CO(1-0) is more prominent in the spiral arms. Together with the data from Schinnerer et al. (2000), we have the most sensitive maps at the so far highest available angular resolutions for the CO(1-0) , CO(2-1) and CO(1-0) line emission in NGC 1068. With this detailed knowledge about the molecular gas properties, NGC 1068 is an ideal candidate to reveal further information about its nature by focussing on the warm and dense part of its molecular gas content.

Scientific goals
We aim at detecting the HCN(4-3) and CO(3-2) line in NGC 1068 to investigate the warm and dense parts of the molecular gas. We plan to record a 3x3 point map in both lines enabling us to differentiate between the center and the spiral arms and to search for spatial variations in the emission. HCN(4-3)
Higher transitions of HCN tracing warmer parts of the dense gas have never been observed before in NGC 1068. Since HCN is not only known to be an ideal complement to CO to analyse the chemistry of the gas (e.g. Kohno et al. 1999 & 2003), it is also supposed to be more strongly correlated with star formation than CO (e.g. Solomon et al. 1992, Curran et al. 2000). While CO emission is known to trace regions with a relatively low gas density (n(H )$ 200cm ), HCN is associated with denser gas regions (n(H ) 10% cm ). Kohno et al. (1999 & 2003) and Tacconi et al. (1994) find in the cases of NGC 6951, NGC 1097 and NGC 1068 that the HCN emission unveils a different gas structure compared to what is obtained via the CO lines. In NGC 1097 and NGC 1068 (compare Fig. 1 for NGC 1068) for instance, HCN emission is enhanced toward the nucleus with respect to the CO emission while the CO morphology is distributed on larger scales, in- cluding the spiral arms for instance. This implies spatial variations of the HCN to CO integrated intensity ratio as pointed out by Kohno et al. (1999 & 2003). NGC 1068 and NGC 1097 reveal unusually large HCN-to- CO ratios at the center. The enhancement of HCN towards the nucleus can be created in two different ways: either through strong X-ray radiation from an AGN or simply by enhancing the dense gas fraction which is strongly linked to star formation, i.e. Photon-Dominated-Regions (PDRs). In non-Seyfert galaxies including nuclear starburst galaxies, i.e. without AGN, HCN-to-CO ratios with R(HCN/CO)>0.3 have never been observed (Kohno et al. 1999 & 2003). Together with the fact that NGC 1068 and NGC 1097 show HCN-to-CO ratio (XDRs; i.e., R(HCN/CO)>0.3; Kohno et al. 2003 & 1999) and PDRs (i.e., R(HCN/CO)0.3). This provides a new diagnostic tool to seperate "pure" AGNs without composite nuclear starburst (NGC 1068 and NGC 1097) from those with associated starburst (NGC 6951). The already strong correlation of HCN(1�0) with star formation suggests that higher transitions of HCN might be even more tightly connected to star forming regions since their excitation does not only enforce dense but also much warmer environments. This will help to constrain HCN as starburst tracer on a much tighter basis.
12CO(3-2) The 12CO(3-2) line is an indispensible and complementary piece of information in studying the warm parts of the molecular gas in NGC 1068. Similar to HCN(4-3) , 12CO(3-2) can only be excited by sufficiently warm and/or dense gas environments (excitation temperature: 33 K; critical density: 5x10^4 cm ). While the lower transitions give rather insights on the "overall" distribution of the molecular gas, 12CO(3-2) is, due to its excitation conditions, most likely stronger connected to star forming regions in which the gas properties may be different. As the frequency of 12CO(3-2) is similar to the one of HCN(4-3) , the beam sizes will almost be identical so that a more direct comparison will be possible without any artefacts due to different filling factors for instance. This will allow us to investigate the role of star formation in NGC 1068 through two independent gas tracers.
NGC 1068 with its bright molecular lines is the ideal testbed to analyse the warm and dense molecular gas in nearby active galaxies and to understand the role of star formation in AGN.

Technical justification
We ask in total for 9 hours of observing time. This splits up into 4.5 hours for the HCN(4-3) and 12CO(3-2) line, respectively. We plan to observe a 3x3 map (see 13CO(1-0) map in Fig. 1) by spending 10-15 minutes on source per pointing plus 30-50% extra time for calibration purposes, focussing, pointing etc. resulting in 25-30 minutes per map if we also acount for the same amount of time for "on-sky" as for "on-source" measurements. CO as well as HCN have been very strong in previous observations with brightness temperatures of the order of 1 Kelvin or more (e.g., Usero et al. 2004; see also Fig. 2). Assuming the worst case, where the higher transitions are less excited, 12CO(3-2) and HCN(4-3) would be still of the order of >>100 K. With an assumed system temperature of approximately 250K at 350 GHz, a (smoothed) bandwidth of 10 MHz per channel (using the 2048 channel configuration with a total bandwidth of 1028 MHz) and an on-source integration time of 10-15 minutes per pointing, we would expect the noise level to be at 4-5 mK. This will result in excellent signal-to-noise ratios. The full linewidths (at zero maximum) of the 12CO and HCN lines are of the order of 500-600 km s^-1 leaving still enough room to fit the baseline within a 1025 MHz bandwidth.
NGC 1068 is visible during 4-4.5 hours at night for APEX so that a whole map (per line) can be finished within one night. The HCN(4-3) line is redshifted to 353.0931 GHz and the 12CO(3-2) to 344.4184 GHz (taking z=0.004) resulting in beams of 15" . The spacings between two pointings are choosen more generously to cover already most of the molecular line emission including the spiral arms with 9 maps (see Fig. 1).

Antonucci, R.R.J. & Miller, J.S., 1985, ApJ, 297, 621
Khachikian, E.Y., Weedman, D.W., 1974, ApJ, 192, 581
Kohno, K., Kawabe, R., Vila-Vilar� , B. 1999, ApJ 511, 157
Kohno et al. 2003 PASJ 55 L1
Schinnerer, E.,Eckart, A. & Tacconi, L.J., 2000, ApJ 533, 826
Solomon, P. M., Downes, D., Radford, S.J.E, 1992, ApJ, 387,55
Tacconi, L.J., Schinnerer, E., Gallimore, et al., 1997, AAS, 191, 7803
Tacconi, L.J., Genzel, R., Blietz, et al., 1994, ApJ, 426, 77