Ionospheric absorption measurements during a sunspot cycle

Abstract

The results of a comprehensive series of observations on the absorption of radio waves during the process of ionospheric reflection are described. Such observations have included experimental studies of the dependence of atmospheric absorption ¦ log ρ ¦ (ρ being the effective ionospheric reflection coefficient), on the radio wave frequency ƒ, on solar zenith distance χ and on solar activity as expressed by the sunspot number R. For the last-mentioned purpose the time-series of comparable observations has extended over a period of nearly two sunspot cycles. It is found that the variation of ¦ log ρ ¦ with ƒ is adequately explained by the magneto-ionic theory of ionospheric dispersion and absorption, provided that ν 2 ⪡ (ƒ + ƒL) 2 , where ν is the electron collision frequency and ƒ L is the gyro-frequency corresponding to the vertical component of the geomagnetic field. On the other hand, the dependence of ¦ log ρ ¦ on cos χ, examined either diurnally or seasonally, is found to be irreconcilable with simple theory if ν is so restricted and the absorbing layer is formed in an isothermal atmosphere of uniform composition in which the electron recombination coefficient is independent of pressure. Such a simple theory predicts that ¦ log ρ ¦ should be proportional to (cos χ) n where n is 1·5, whereas some hundred diurnal examinations of this relationship yield an average value for n of the order of 0·75. Possible explanations of this discrepancy are that the electron recombination coefficient falls off to some extent with increasing height and that the temperature of the absorbing stratum is not uniform, but increases upwards. It seems, however, clear that the removal of electrons from the absorbing layer cannot take place by way of a process described by an attachment law. Seasonal studies of the dependence of ¦ log ρ ¦ on cos χ have confirmed the existence of a winter anomaly, which has been evident during the months of November to February ever since the series of observations was started in 1935. This winter anomaly is characterized by the occurrence of excessive absorption on certain groups of days during the months mentioned, a phenomenon which is not paralleled during the rest of the year. A statistical study shows that the occurrence of these days of high absorption is directly correlated with the occurrence of sporadic reflecting strata below the level of the E Layer. A series of noon measurements of absorption, of eighteen years' duration, has demonstrated a general direct correlation of ¦ log ρ ¦ with sunspot number R. Sunspot-cycle control is, however, more evident in summer than in winter. Various lines of evidence emerging from the experimental results show that the absorbing stratum must be located below the E Layer, and the general conclusion reached is that it lies within the D Region. Arguments are also advanced for concluding that the D Layer cannot be the seat of the currents responsible for the quiet-day solar magnetic variations S q ; while other considerations indicate that the layer of temporarily enhanced ionization caused by solar flares can, likewise, not be identified with the S q current layer, as has previously been widely supposed. It does, however, seem most probable that sudden ionospheric disturbances associated with solar flares are due to augmentations of ionization in the normal absorbing layer, the ionization density being increased there by a factor of from 5 to 10.