3 Absolute calibration of the Effelsberg total intensity maps

From the observation and the method of data reduction it is obvious that neither the total intensity nor the polarization maps are on an absolute temperature scale. The total intensity maps are calibrated to an absolute scale using the 1.4 GHz northern sky survey by Reich (1982) and Reich & Reich (1986) carried out with the Stockert 25-m telescope. This procedure has already been described by Reich et al. (1990) when calibrating the Effelsberg 1.4 GHz Galactic plane survey. The Stockert survey has a scale accuracy of 5% and an uncertainty of the absolute zero-level of 0.5 K.

Briefly, the Effelsberg data and the Stockert data are convolved to an angular resolution slightly exceeding that of the Stockert 1.4 GHz survey (HPBW 36$^\prime$) and the difference between the two data sets is added to the Effelsberg data. This procedure also improves the original Effelsberg maps since large-scale distortions by atmospheric and ground radiation variations are removed.

However, still existing faint baseline effects in the low resolution Stockert data are also added to the higher resolution Effelsberg data. Such distortions were found to exist on small scales but not on the large-scale background component, which is the missing component of the Effelsberg maps. A modified version to calibrate the Effelsberg maps absolutely was introduced:

$\bullet$

Convolve the Effelsberg map to the resolution of the Stockert map (36$^\prime$). Decompose this map into a large-scale component (Eff.back) and a map with small-scale structures using the "background filtering method'' (Sofue & Reich 1979) with a $3\hbox{$^\circ$}$ Gaussian beam for smoothing.

$\bullet$

Subtract the 2.8 K isotropic background component from the Stockert data and convert the temperature scale from full beam into main beam ($T_\mathrm{MB}/T_\mathrm{FB} = 1.55$).

$\bullet$

Decompose the Stockert data to obtain the corresponding large-scale component (Sto.back) by using the same filtering parameters as for the Effelsberg data.

$\bullet$

Adjust Eff.back to Sto.back as described above, namely, add the difference of both large-scale component maps to the Effelsberg large-scale data and add the 2.8 K isotropic background component. The resulting map is ES.back.

$\bullet$

The final Effelsberg map (Eff.final) is given by:

Eff.final = Eff.orig + ES.back - Eff.back.

  

Figure 2: A small section from the survey, towards $\ell \sim 50\hbox{$^\circ$}$, illustrates the calibration of the total intensity data to the absolute temperature scale. The panels, from top to bottom, display the Effelsberg (contours start at 100 mK and are plotted in 150 mK steps) and Stockert 1.4 GHz (contours start at 4500 mK in 100 mK steps, full beam brightness scale) measurements and the combination of the two, respectively. Contours for the combined map (bottom) run from 5500 mK in steps of 150 mK

We use the modified method only when there are non-negligible scanning effects in the source component of the Stockert data. A sample region (towards $\ell \sim 50\hbox{$^\circ$}$) is given in Fig. 2 to illustrate the effect of the absolute calibration.