The observations were performed with one of the 30-m telescopes of
the Instituto Argentino de Radioastronomía (IAR) in May 1992. The
two-channel receiver was operated at a central frequency of 1435 MHz with
a bandwidth of 20 MHz. The temperature of the system was 90 K. The HPBW of
the antenna is 30 arcmin at this frequency, and the 
ratio
of the main beam brightness temperature to the flux density was 0.16
during the observations.
The observing technique consisted in fast (
)
scans in declination, regularly spaced in right ascension. Two ``up'' and
two ``down'' scans were performed at each selected value of right ascension.
The standard sources
Hydra A and PKS 1308-22 were observed for calibration purposes. The data
reduction was carried out following the techniques described by
Combi et al. (1995) and Combi & Romero
(1995), which are based on those developed
by Haslam et al. (1974). The data from the survey by
Haslam et al. (1981)
were processed in the same way as the new 1435 MHz data. In order to
eliminate local contributions a smooth background was subtracted out by
applying the `background filtering' method developed by Sofue & Reich
(1979). The diffuse emission subtracted consists of two components: a
galactic disk contribution and a component resulting from the merging of
unresolved small-scale background sources, mainly extragalactic ones. This
subtraction technique, however, can not suppress the local contribution of
the spur-like feature which appears to point out towards the southern part
of the galaxy (Haslam et al. 1981). Owing to the strength of Cen A at both
observing frequencies we estimate that the total contribution of the spur
is negligible. A complete study of this galactic structure is in progress
in order to determine its nature and spectral characteristics (Combi &
Romero, in preparation).
Figure 1: a) Radio continuum emission at 408 MHz
from Cen A (with the original beam) after subtraction of the smooth
`backgound' emission. Contour
lines are shown at 1, 5,...., 20; 25, 31,....., 67; 75, 83, ...., 107;
120, 170, ....., 500 K, in brightness temperature. b) Radio continuum
emission at 1435 MHz from Cen A (with the original beam) after subtraction
of the smooth `background'
emission. Contour lines are shown at 0.1, 0.2,....., 1; 1.35, 1.65,....,4;
6, 8,.....,18; 21, 24,...., 32 K. Coordinates are referred to 1950.0
The resulting radio continuum maps (see Fig. 1 (click here), where the brightest part of the spur is visible on the left) were used for computing the spectral index distribution in the source, after convolution and re-tabulation of the 1435 MHz map to the same beam and tabular interval of the 408 MHz map. Two extragalactic compact radio sources, 1332-33 and 1334-127 were used for checking the positional accuracy of the maps by comparing the measured positions with the actual positions taken from the 1 Jy Catalogue by Kühr et al. (1981). The second source, the flat-spectrum QSO 1334-127, was also used for matching the different beam shapes.
The spectral index 
(we consider 
) between
408 MHz and 1435 MHz was computed as:
where 
is the brightness temperature at the frequency 
obtained
after the removal
of the diffuse galactic contribution (see Combi et al. 1995 for details
about the procedure).
The reliability of the spectral index computation rests on the
determination of the absolute zero level of each set of data under
consideration. Systematic errors can be introduced by variations of the
zero level error to intensity ratio with frequency. In order to get an
accurate estimate of these errors we have followed the procedure
described by Reich & Reich (1988). The errors were determined as:
where the contributing errors to 
are the result of scale
errors, local scanning errors, local noise, and the error of the absolute
zero level. Values for the 408 MHz data are quoted by Reich & Reich
(1988) in their Table VIII. Errors for our own data are of the same order
as those described in that Table for the 1420 MHz observations, except
for 
, which is in our case 
K. We
estimate the maximum error in our final map as
. The mean error in the map is
. The errors in the Southern Lobe
could be slightly underestimated due to the mentioned weak contribution of
the local spur structure.