ICE FORMATION DURING LAGOPEDO UNO Charles F. Lebeda, EG&G, Los Alamos, NM; Morris B. Pongratz, Los Alamos Scientific Laboratory, Los Alamos, NM Hydrocarbon combustion products such as CO2 and H2O, as emitted from rocket engines, are known to deplete the electron and ion content of the ionosphere. Operation Lagopedo was executed to determine the behavior of these reactive species in the ionosphere under controlled conditions. Operation Lagopedo consisted of two rocket-borne injections of CO? and H2O into the F-2 region of the ionosphere near Hawaii during September of 1977. The first of these injections occurred in sunlight, the second did not. Photographic images of the first injection yield unique experimental information concerning the effectiveness of HoO in depleting the ionosphere. The expanding water vapor froze and was not able to react with the local electrons and ions as ice. Subsequent sublimation of the ice particles to vapor made the HnO capable of reacting with the ambient electrons and ions. Determination of the amount of the water that froze and the duration that it remained frozen defines temporally how much of the water can react. We present the results of the analysis of the images of the scattered sunlight that we used to determine the amount of water that froze and the decay time for the ice. The data used for the early-time analysis consisted of four images of light scattered from the ice particles. These images were recorded on Kodak EKIR film (500 to 900 nm), and covered a time span from two to fifteen seconds after event time. We assumed that all the particles in a column corresponding to a particular line of sight (or a point in the image) were spherical Mie scatterers of identical radius. From the exposures in each of the three independent color bands at each such point in the image, and the properties of Mie scatterers, we were able to determine the radius and number of the scatterers. The particle radius and number in each column were converted to masses of ice. Two-dimensional integration gave total ice mass in the cloud. The maximum amount of ice determined from this analysis was 4.24 kg, which is approximately ten percent of the estimated mass of the water released (45 kg). A fit of total ice mass vs time to a decaying exponential gave a time constant of 10 seconds. The particle radius averaged over the cloud was about o.42 pm, independent of time. There was unexpected structure in the cloud appearing as a fan expanding upward from the center of the cloud. We next assumed that the ice density at a point in space was proportional to the water vapor density theije. 2 We then fit spatial profiles of ice mass density to forms like exp(-|x|/R >. We found that this form, based on a Maxwellian velocity distribution, best fit the early data as opposed to forms that had a peak in the velocity distribution at nonzero velocities. The fit of R vs time for the first three frames was linear with a slope of 2.6 km/s. This fit was good through 6.2 seconds. The next frame (15 s) was inconsistent with this form. Thus, we conclude that the vapor cloud expanded freely prior to 15 seconds and thereafter expanded by another mechanism. To determine the behavior of the ice cloud at later times we analyzed images recorded with intensified cameras filtered to record light in narrow bands at 455.4 and 772.0 nm. A particle size analysis could not be performed on these data because there was not enough information. However, the light output in these two bands could be predicted from the computed properties of the scatterers obtained from the color film. The decay constant predicted for the total light from the cloud in the 455.4-nm band was 8.6 seconds while the intensified camera gave a decay constant of 13.7 seconds. The predicted and measured peak amplitudes agreed to approximately 50%. However, the intensified camera total light peaked approximately 23 seconds after the color pre-
RkJQdWJsaXNoZXIy MTU5NjU0Mg==