In studies of the mechanisms responsible for the formation of the ionized regions of the atmos phere, one of the problems encountered is the solution of the set of differential equations describing the time dependence of the number density of the chemical species present. For simple cases where only one or two species are involved, there is no particular problem. When all the important constituents are included, however, the solution of these equations becomes very complex. A computer technique is pre sented that will solve any size set of these differential equations involving any number of chemical reactions. (Author).

One way to acquire a better understanding of the formation and destruction of ionization in the atmosphere is through the solution of the system of time-dependent reaction rate equations. These ordinary differential equations form a simultaneous set each question of which describes the time rate of change of a particular atmospheric constituent. In the general problem, all the molecules and atoms whether neutral, charged, or excited, as well as the free electrons would be included. A computer program is presented for developing the numerical solution to this problem. The method of solution of the set of equations uses a fourth order Runge Kutta integration with a variable mesh. When a species enters its quasi-equilibrium state, its differential equation is removed from the set and its equilibrium equation is inserted into the simultaneous algebraic set. The algebraic set is solved by the method of successive substitutions. The over-all solution is obtained by iteration between the differential and the algebraic sets. The ability of the computer program to develop extensive solutions is demonstrated by several examples taken under different conditions.

One of the Keneshea computer codes (see AD-424 173) was adapted for use on the Ballistic Research Laboratories Electronic Scientific Computer. Using this modified code, reaction rate equations were solved for the following 15 species: e, NO2( - ), O( - ), O2( - ), O3( - ), N2(+), NO(+), O(+), O2(+), N, NO, N2O, NO2, O, and O3. The calculations were made for a 4:1 mixture of N2 and O2 at 1 torr total pressure and 300K. Rate constants as given by Keneshea and Fowler (see AD-646 975) were used. The solutions are presented as number densities versus time after the start of the irradiating electron beam. A description of the modified code is presented. (Author).

With the availability of numerical techniques for solving an extensive set of nonlinear differential equations and high-speed computers for performing the calculations, interest in solving the unrestricted reaction-rate equations is growing among ionospheric researchers. In view of this a refinement is given of the techniques previously developed in PB-163 071 and AD-424 173. The computer code is written to solve the photochemical behavior of 15 atmospheric species; these species are electrons, O( - ), O2( - ), O3( - ), NO2( - ), O(+), O2(+), N2(+), NO(+), NO, N, NO2, O3, N2O, and O. Built into the code are 168 reactions that can conceivably take place among these constituents. Several examples of the results obtained using the code are presented, including the buildup of ionization from zero concentrations at altitudes in the D and E regions and the deionization of an atmosphere with high initial electron densities. The diurnal variation of the atmospheric constituents is also presented along with profiles for the above-mentioned species from 60 km to 120 km. The computer codes are included in their entirety with complete explanations on their usage. (Author).

This report describes a radio investigation of traveling ionospheric disturbances carried out near Boulder, Colorado, over a 1-year period from June 1967 to June 1968. The three-dimensional motions of F2 layer disturbances were measured by the high frequency Doppler technique with spaced transmitters and at several probing frequencies. Horizontal motions were determined by cross-correlating three signals on frequencies near 5 MHz, whose reflection points were approximately at the corners of a horizontal equilateral triangle with 40-km sides. Vertical motions were determined from cross-correlation of signals on frequencies of 3.3, 4.0, and 5.1 MHz, whose reflection points were aligned vertically.