5 Discussion

In Table 3 we show the available low-frequency continuum surveys below 100 MHz covering a relatively large area of the northern hemisphere. Milogradov-Turin & Smith (1973) used the Jodrell Bank 76 m telescope for their survey at 38 MHz, which gives less than one-half of our resolution. Baldwin (1955) using a cylindrical paraboloid at 81 MHz obtained a survey with even lower resolution. The synthesis surveys are intended to delineate small-scale features, and they are generally not reliable to study large scale structures. The surveys by Williams et al. (1966), by Caswell (1976) and by Blythe (1957) do not cover 24 hours of right ascension, while the maps by Dwarakanath & Udaya Shankar (1990) are seriously contaminated by side lobe response to strong sources.

We believe that the calibration of the data presented here by using the 45-MHz data from the Maipu survey was successful and that the estimated error is less than 15%. This survey shows well the global features so far known, for instance, the north polar spur, the temperature minimum region around $\alpha=9^{\rm h} 30^{\rm m}$, $\delta=35^{\circ}$. Some discrete emission and absorption features are seen and their positions are in good agreement with those in the 22 MHz map of Roger et al. (1999).

A frequency around 45 MHz is not too high to be contaminated with thermal emission, nor too low to be affected by thermal free-free absorption (Alvarez et al. 1987). We therefore believe that the 45 MHz survey represents almost pure synchrotron emission. Comparing this survey with others in different ranges of the electromagnetic spectrum will yield fruitful results. The Galaxy is a mixture of matter, radiation and magnetic fields that interact in complex ways. It is therefore not unexpected to find a correlation between some of the phases of the interestellar medium. For example, a tight linear correlation has been well established in spiral galaxies, between the logarithm of the far infrared (FIR) and the logarithm of the nonthermal radio continuum (NTRC) (e.g. Fitt et al. 1988; Helou et al. 1985; de Jong et al. 1985). Also a correlation between the NTRC and the CO line emission has been found in galaxies though it has not been as profusely investigated as the FIR case (Israel & Rowan-Robinson 1984; Adler et al. 1991).

It is of interest to compare the low-frequency radio map with sensitive H$\alpha$ surveys (e.g. Reynolds 1998) which indicate the distribution of the ionized hydrogen. At high galactic latitude, where optical extinction is small, the intensity of the H$\alpha$ line is directly proportional to the emission measure, therefore also to the free-free opacity along the line of sight. One of the future uses of the present survey, on which we are already working, is that it can be combined with the Maipu survey in order to get a map of the whole sky (except 4.7% of it around the north celestial pole). This will be a meaningful map since the northern and southern surveys have been made with practically the same angular resolution and frequencies. The only published all-sky map that gets close to meeting these conditions is that by Haslam et al. (1982) at 408 MHz. In the near future a 1420 MHz all-sky map will be obtained with the northern data from Reich (1982) and Reich & Reich (1986), and with the southern data from Testori et al. (in preparation). The combination of these unique all-sky surveys will allow us to obtain the spectral index of the radio background and its variation across the sky reflecting the energy distribution of the radiating relativistic electrons.

Acknowledgements

The MU radar belongs to and is operated by the Radio Atmospheric Science Center, Kyoto University. We thank Dr. T. Sato, Dr. M. Yamamoto, and Dr. H. Hashiguchi for many suggestions about our observations. We also thank Dr. S. Maeda and Dr. Y. Muraoka for their contribution in the data processing. The assistance of F. Olmos in the data reduction is highly appreciated. We are specially grateful to Prof. R. Wielebinski for his continuous interest in and support of this project. We thank Prof. T.D. Carr for pointing out that we were doing similar work in both hemispheres. We appreciate helpful comments from Dr. W. Reich and Dr. R. Duncan. J.M, H.A., J.A., and K.M. acknowledge support from FONDECYT-Chile grant 8970017.