Synoptic radio observations as proxies for upper atmospheric modelling

The solar radio flux at 10.7 cm is one of the most widely used solar proxies for upper atmospheric modelling. In comparison, very little attention has been given to the flux at other centimetric wavelengths, which provides important complimentary information on the solar radiative output. Synoptic radio observations have been made almost continuously since the late 1950's by several observatories, and in particular at Toyokawa and Nobeyama. We have merged different observations into a single continuous composite dataset of daily radio fluxes at 30, 15, 10.7 and 8 cm (without flares), ranging from 1957 up to today. A blind source separation analysis shows that the variability on time scales from days to several months can be decomposed into two contributions, which can be readily identified as thermal Bremsstrahlung and gyro-resonance emission. Most solar indices, such as the MgII index and the H Lyman-alpha line intensity, can be remarkably well reconstructed by linear combination of these two contributions. We advocate the flux at 30 cm for upper atmospheric modelling since it performs better than the 10.7 cm flux when it comes to describing the solar forcing of the thermospheric density; this is revealed by comparing neutral density observations with reconstructions from the DTM (Drag Temperature Model) model. These results show that synoptic radio observations greatly add to the spectral variation that is available in the 10.7 cm flux. Because of their long-term stability and their reliability, they are also excellent candidates for operational space weather use.