foF2 vs solar indices for the Rome station: Looking for the best general relation which is able to describe the anomalous minimum between cycles 23 and 24

Analyses of the dependence of the F2layer critical frequency, foF2, on five widely used solar activity indices (F10.7, Lym-α, MgII, R and EUV0.1-50)are carried out considering noon values manually validated at the ionospheric station of Rome (41.8°N, 12.5°E, Italy) between January 1976 and December 2013, a period of time covering the last three solar cycles and including the prolonged and anomalous minimum of solar cycle 23/24 (years 2008-2009). After applying a 1-year running mean to both foF2 and solar activity indices time series, a second order polynomial fitting proves to perform better than a linear one, and this is specifically due to the very low solar activity of the last solar minimum and to the remaining saturation effect characterizing the high solar activity. A comparison between observed and synthetic foF2 values, the latter calculated by using the analytical relations found for every index, and some considerations made on the R parameter introduced by Solomon et al. (2013), suggest that MgII is the best index to describe the dependence of foF2 on the solar activity. Three main reasons justify this result: (1) the good sensibility of MgII to the variations of foF2 for low solar activity; (2) the reduced saturation effect characterizing MgII at high solar activity; (3) the poor influence of the hysteresis effect characterizing MgII at medium solar activity. On the other hand, the F10.7 index, widely used as input parameter for numerous ionospheric models, does not represent properly the last minimum; specifically, it is not able to describe the variations of foF2 under a solar activity level of F10.7=82·10-22 [J Hz-1 s-1 m-2].