Observations on the Ionospheric F Layer at Night: II—The Law of Electron Loss in the F Layer

Abstract

h'(f) curves taken at night at Cambridge during the years 1954–55 were analysed for true-heights using the modified Kelso method and the results concerning the height of maximum ionization and the ‘semi-thickness’ were given in Part I of this paper. In Part II an attempt is made to determine, if possible, whether the loss process operating in the F layer is one of attachment-type or the recombination-type. It is found that an attachment-like law operates in the F layer and that attachment-like coefficients are height-dependent, being represented by the following exponential law: where K₁ is the attachment-like coefficient at the height h, K_o refers to the datum level h_o, and H_k is defined as the scale-height of the loss coefficients.

It is possible to determine the loss coefficients accurately as there are N(h) curves at every 10 min. throughout the night. The scale-height H_k determined from the loss coefficients is found to depend on the season, being the greatest in summer. Also, it is shown that the scale-height determined for the night 18/19 May 1956 is much bigger than the value for the years 1954–55. As the solar activity was high during the year 1956, as indicated by the provisional sunspot number, it is inferred that the scale-height of the loss coefficients varies with the solar activity, being the greatest for the sunspot maximum years.

Electrons may disappear at the F layer due to the atomic and molecular processes, by which they recombine with positive ions, or by vertical transport. Some arguments are put forward to show that the vertical movements, if present, are not important. It is shown that the observed distribution of loss coefficients is consistent with the loss processes suggested for the F layer by Bates and Massey.² H_k would then be identified with the scale-height of the catalytic molecules O₂, N₂ and NO which are known to exist at the F region heights.

The scale-height for 1956 is in good agreement with the value expected from the recent rocket and satellite observations,⁵ whereas those for the years 1954–55 are found to be too low compared with the values for the corresponding heights. If the scale-height varies with the solar cycle then this is understandable, as the rocket observations were made in the years of high solar activity.