Radio Images of the Globular Cluster 47 Tucanae
D. McConnell ,
R. Deacon ,
J.G. Ables
, PASA, 18 (2), in press.

Next Section: Discussion
Title/Abstract Page: Radio Images of the
Previous Section: Observations and Data Reduction
Contents Page: Volume 18, Number 2

Image Analysis and Results

We describe the analysis and results from the 1.4 GHz image. (The same reduction procedures were used for the 1.7 GHz image and its relevance in this paper is restricted to the measurement of spectral indices of sources in the cluster centre.) Over the 53 $^\prime$ field (2208 arcmin²) the image noise fluctuations range from

$\sigma_{1.4} \simeq\$ 18 $\mu\mbox{Jy}$ over the cluster centre to $\sim$ 60 $\mu\mbox{Jy}$ at the field edge. Within this area 305 sources were detected above the 5 $\sigma$ threshhold. After correction for the varying sensitivity across the field we found a mean density of sources with S_1.4 > 90 $\mu\mbox{Jy}$ of 0.25 $\pm$ 0.01 arcmin^-2. The sensitivity to radio sources is limited by image fluctuations originating from thermal noise in the receiving equipment and radiation from the sky and ground. The detectability of weak sources is also limited by the extent to which corrections can be made for the sidelobe response to strong sources in the field. While image fluctuations are dominated by sources of random noise the sensitivity $\sigma$ is expected to improve with integration time T as

$\sigma \propto T^{-1/2}$ . The image formed from the 1992 observations had an effective integration time on the cluster core of 33 hours and at 1.4 GHz the image rms was 42 $\mu\mbox{Jy}$ . The new image has T = 170 hours and

$\sigma_{1.4} \simeq\$ 18 $\mu\mbox{Jy}$ close to the value expected from the

$\sigma \propto T^{-1/2}$ relation. Note that this is well above the rms value of $\sim$ 10 $\mu\mbox{Jy}$ expected from considering only the thermal contributions to noise.

**Figure:** The inner part of the imaged field. The radio emission is shown as the 1.4 GHz brightness contours at 54, 90, 144, 216, 306 $\mu\mbox{Jy}$ /beam (that is
$\sigma \times 3, \times 5, \times 8, \times 12, \times 17$ ). The size of the synthesized beam at 1.4 GHz is indicated at lower left. The central circle marks the position (De Marchi et al., 1996) and size ( $r_{c} \simeq$ 23

$^{\prime \prime }$ , Howell, Guhathakurta and Gilliland, 2000) of the cluster core. The positions of pulsars with known timing positions are indicated by crosses and labelled with their alphabetic names. Other 1.4 GHz sources are numbered as described in the text.
$\begin{figure} \begin{center} \psfig{file=fig1.ps,height=14cm} \end{center} \end{figure}$

Figure 1 shows the centre of the 1.4 GHz image - within 2 $^\prime$ of the cluster centre. The positions of pulsars with known timing positions are indicated and labelled with their alphabetic names. Sources not coincident with known pulsar locations are labelled with their respective numbers from our list of 305 sources.

**Table 2:** Flux measurements at 1.4 GHz and 1.7 GHz and spectral indices of the sources in Figure 1. The fluxes of pulsars L and O were estimated by modelling the associated extended source with two point sources constrained to lie at their timing positions. Although pulsars O and U are not detected at 1.7 GHz the table gives meaningful (but broad) limits to their spectral indices. Column (2) gives the source numbers from Table 2 of MA2000. Sources 9 and 11 from MA2000 are not detected (see text). Other sources reported in MA2000 were outside the area analysed here.
(1)	(2)	(3)	(4)	(5)
Source		S_1.4	S_1.7	$\alpha$
	( $\mu\mbox{Jy}$ )	( $\mu\mbox{Jy}$ )
C	4	243	122	-3.6 $\pm$ 0.8
D	8	369	200	-3.2 $\pm$ 0.5
E		84	61	-1.7 $\pm$ 2.1
F		176	50	-6.5 $\pm$ 2.2
G/I		102	65	-2.3 $\pm$ 1.9
J	5	330	202	-2.5 $\pm$ 0.5
L		95	79	-1.0 $\pm$ 1.7
O		77	(22)	(-6.4 $\pm$ 7)
Q		116	59	-3.5 $\pm$ 2.3
U		114	(28)	(-7.3 $\pm$ 5.4)
142	3	727	628	-0.8 $\pm$ 0.2
176	7	202	201	0.0 $\pm$ 0.7
178		92	116	1.2 $\pm$ 1.4
183		108	36	-5.7 $\pm$ 5.5
186		101	59	-2.8 $\pm$ 2.2

Table 2 gives, for each source in Figure 1, measured values of flux at 1.4 GHz and 1.7 GHz and their spectral indices. The relatively large uncertainties in spectral index result from the small frequency separation of the flux measurements and the low signal-to-noise ratio for most sources.

Next Section: Discussion
Title/Abstract Page: Radio Images of the
Previous Section: Observations and Data Reduction
Contents Page: Volume 18, Number 2

ASKAP

Public

Australia Telescope National Facility

Image Analysis and Results