This first KAT-7 HI observation of the nearby galaxy, NGC 3109, provides a unique opportunity to simultaneously achieve HI spectral-line science verification and an original scientific result. The 12-hour observation was taken using the newly deployed eight thousand channel wideband mode (WBC8K) of the KAT-7 correlator. This gives us a spectral resolution of 48 kHz, equivalent to a velocity resolution for this spectral line of 10km/s, while still having a bandwidth of 256 MHz. Figure 1 shows the distribution of the HI gas, which is more extended than the optical stellar disk, while Figure 2 illustrates its rotation pattern. By modeling this velocity field, it will be possible to derive a model of how the dark matter, which is the main mass component, is distributed.
Figure 1: The green contours show the first observation of the HI (neutral hydrogen) emission in the nearby galaxy NGC 3109. It is superposed on a three-colour composite image from the Digital Sky Survey . It can be seen that the HI emission extends well beyond the stellar disk.
Figure 2: Velocity field derived from the HI observations showing evidence of rotation. The red side is receding from us and the blue side is moving toward the observer. Such a map will allow studying different models of the distribution of the dark matter in this late-type spiral galaxy.
NGC 3109 is a low surface brightness late-type spiral and is of significant scientific interest for two main reasons. Firstly, Jobin & Carignan (1990) used observations with an hybrid VLA C-D configuration (beam of 36” x 27” & velocity resolution of 10.3 km/s) to perform a dynamical study of this galaxy, comparing the rotation curve derived from a tilted-ring analysis to models composed of a luminous galaxy disk (stars & gas) and of a dark halo. With higher velocity resolution data, it should be possible to compare those results to a MOND (no dark matter) model of the rotation curve.
Secondly, Barnes & de Blok (2001) used 21cm Multibeam data with the Parkes 64m dish (beam ~15.5’ & velocity resolution of 1.1 km/s) to study the environment of NGC 3109. They provide a compelling argument that the warp in the HI disk of NGC 3109 could be due to a dynamical encounter with the Antlia dwarf. With high sensitivity and better spatial resolution observations it should be possible to map much better the traces of that encounter.
In order to sort out whether the conventional (dark matter) mass models (isothermal or NFW) or the MOND models give the best representation of the observed kinematics, we need both the spatial resolution (to avoid beam smearing) and the velocity resolution (since Vmax < 70 km/s) for a late-type spiral like NGC 3109, which has a nearly solid-body type rotation curve. As shown in Figure 3, the HIPASS data have the necessary velocity resolution but poor spatial resolution while it is the contrary for the VLA data.
Figure 3: A "resolution" comparison of previous observations (VLA & HIPASS) with the current 8k mode and the future KAT-7 observations
Additional correlator modes, to be released later this year (see Table 1), will give us a higher spectral resolution (~1 km/s) while the beam size ~3.5 arcmin will provide sensitivity to low surface brightness emission. In this way, KAT-7 will provide the necessary accuracy for the crucial inner parts of the rotation curve, which really constrains the mass models and will also reveal in the outer parts the possible interaction with its dwarf companion galaxy Antlia.
Table 1: Expected correlator modes for KAT-7 to be commissioned before the end of 2012.
Barnes, D. G. and de Blok, W. J. G. 2001, Astronomical Journal, 122, 825
Jobin, M. and Carignan, C. 1990, Astronomical Journal, 100, 648