S. I. Kozlov and N. V. Smirnova
Institute of Applied Geophysics, Moscow, Russia
When a rocket moves through the stratosphere, the high-temperature flame of the rocket engine and the shock wave, as well as the products of fuel combustion, which spread away to long distances from the trajectory, lead to destruction of ozone O 3 . There have been numerous studies of such impacts on O 3 for the liquid- and solid-fueled rockets with varying degrees of completeness (see, e.g., Aleksandrov and Upenek , Burdakov et al. , Deminov et al. , Karol et al. , Pollack et al. , Potter , Prather et al. , and Whitten et al. ). The experimental data recently published [ Vetchinkin et al., 1993] demonstrate that apparently there may exist another mechanism of impact on O 3 of rocket launches. This mechanism is discussed in this paper.
On February 3, 1984, more than 2 hours after the launch of the Space Shuttle system, optical observations by the UV telescope at the "Astron" astrophysical station began. The region of the flight including the active part of the trajectory were observed [ Vetchinkin et al., 1993]. The measurements were performed in three spectral ranges, when the station was at altitudes of about 40,000 km. An important result of the observations is that there was an increase by about 40-60% relative to the background of the UV emission in the l 241-350 nm range from the spatial region with dimensions L 1000 km. Taking into account considerable spatial irregularities and fluctuations of the UV emission, the effect was registered during about 1-2 hours after the beginning of the measurements ( 3.5-4.5 hours after the launch of the Space Shuttle system).
It is worth noting that Vetchinkin et al.  took as a background the results of UV measurements (scans 1-3) in the spatial region on the opposite side relatively to the trajectory of the system active flight.
The above indicated spectral range contains the Hartley band and the Huggins bands of the ozone dissociation and plays a very important role in the ozone photochemistry. It should be expected that any variations of the solar radiation flux in this range may influence in a complicated way the [O 3](h) vertical distribution in the stratosphere. However, the principal factor in this case is the altitude of the region, which scatters the UV radiation. If the region is situated at stratospheric heights ( h < 40-50 km) and the scatter occurs at disperse (aerosol) particles, then, as the calculations of Pollack et al.  show, variations of the solar radiation fluxes, which influence O 3 , should be insignificant. The situation is quite different if the scattering formation is situated at h > 50 km. The fact of an increase of the UV emission intensity with l 241-350 nm, registered from space, means that there was an attenuation of this emission at lower h , that is, at stratospheric heights.
The experimental data on the emission unfortunately do not allow for a direct determination of the altitude of the scattering formation. However, this altitude may be fairly correctly evaluated by two indirect ways.
First, the total duration of the effect registration was several
hours. Therefore, independently on physical characteristics of the
formation, which was able to screen the UV emission, and its
altitude, there is a diffuse stage of development of the given
formation. It is known that the diffusion characteristic time is
determined by the following formula: td L2/4D ,
where D is the
diffusion coefficient. According to
Vetchinkin et al. , td 3.5-4.5 hours, L 1000 km, which gives D 10
The second indirect method is based on a determination of the
minimum height, from which the emission in the 241-350 nm range
is able to reach the altitude of about 40,000 km, where the
astrophysical station "Astron" was located. The experimental data
on an increase of the emission intensity were presented by
Vetchinkin et al. 
for l = 275 nm. This emission is
mainly absorbed by ozone, the absorption cross section being s O3 6 10
Thus both indirect methods indicate the fact that the scatter of the UV radiation occurs at altitudes that exceed the stratospheric level. Therefore the solar radiation in the Hartley band and Huggins bands should be weakened when it comes to the stratosphere.
The universal aeronomical model of Kozlov et al. [1988, 1990] is used to evaluate the influence of this effect on variations of [O 3 ]. Description and verification of the model were published by Kozlov [1984, 1987], Kozlov and Smirnova , Kozlov et al. [1982, 1983, 1984], and Smirnova et al. . Here we mention only that the model includes 15 minor neutral and excited constituents (O( 3 P), O( 1 D), O 3 , O 2 ( 1 Dg ), H, OH, H 2 , H 2 O, HO 2 , H 2 O 2 , N( 4 S), N( 2 D), NO, N 2 O, NO 2 ) and is short-circuited to the natural undisturbed distribution of these constituents.
To estimate the attenuation of the solar UV radiation is the most difficult part of the problem. More or less correct solution of this part requires, first of all, knowledge of characteristics of the artificial formation, that is, definition of the scatter type and the spatial and temporal distribution of the scattering centers. These characteristics depend on the composition and total amount of the combustion products of the rocket engine injected into the atmosphere along the launching trajectory. The Space Shuttle system has both solid- and liquid-fueled boosters [ Ivashkevich, 1993a, 1993b]. The composition of the combustion products of these boosters has common features but significant differences as well [ Smirnova et al., 1995a, 1995b]. Injection of gaseous N 2 , CO 2 , H, H 2 , and H 2 O (naturally, with different relative content in the total mass of the combustion products) is the common feature. The presence of the gaseous chlorine compounds (HCl and Cl) and NO, as well as the disperse particles of the Al 2 O 3 type, provides an example of a difference. It would be quite reasonable to believe that the scatter of the UV radiation in the 241-350 nm range occurs at dispersed aerosol particles. However, the solid-fueled rocket boosters stop their operation at h 60 km [ Ivashkevich, 1993a, 1993b]. With the known values of D at 50 < h < 60 km, one cannot obtain the characteristic dimensions of the artificial cloud of about 1000 km during 3.5-4.5 hours. There is another mechanism of formation of disperse (aerosol) particles at h 80 km: transformation of the gaseous H 2 O into ice crystals (this effect has been registered experimentally during the launch of the Apollo 8 rocket [ Wu, 1975]; some theoretical estimates of the effect may be found in the work of Turco et al.  and Bernhard ). However, in this case it is also hardly possible that such particles at h 80 km would be conserved during t 3.5-4.5 hours under permanent illumination by the Sun of the spatial region related to the Space Shuttle launch. Thus the question on the nature of the "backscatter" (attenuation) of the UV radiation is left open. The question is out of the scope of this study and presents a separate interest. In our opinion, the multiple scatter of the radiation occurs simultaneously at both disperse and gaseous components of the fuel combustion products.
Let us evaluate the ozone concentration at various heights on the basis of the experimental data of Vetchinkin et al.  and elementary energetic relations. Let us assume that at all altitudes the intensity of the emission in the l 241-350 nm range decreases by 50%, because its growth detected from the space was about 40-60% in comparison with the background. Let us also assume that the total time of the effect existence is about 3 hours. The results of calculations of [O 3](t) at two altitudes are shown in Figure 1. It can be seen that owing to a decrease of the O 3 dissociation rate in the Hartley band, the ozone concentrations increase in relation to the background ones by about 25% and 47% at altitudes of 50 km and 30 km, respectively, the characteristic time of the [O 3](t) variations decreasing with an increase of the altitude.
The observed increase of [O 3 ] is due not only to an ordinary decrease if the ozone photodissociation, but to variation in these conditions of some minor neutral compounds. Out of two photodissociation processes (O 3 + h n (l< 320-350 nm) O( 1 D) + O 2(1 Dg) and O 3 + h n (l > 320-350 nm) O( 3 P) + O 2 ), which were taken into account in the calculations, the former one prevails at h = 30 km and the latter prevails at h = 50 km. Therefore a decrease of the O 3 photodissociation rate by the l< 320-350 nm radiation results in a depletion of the total rate of ozone formation by solar radiation only by about 20% at 30 km, but by about 47% at 50 km. That fact accounts for the more rapid initial increase of [O 3 ] at the latter altitude. Moreover, an increase of the O( 3 P) concentration with consequent rapid transformation of these atoms into O 3 is observed at 30 km. The increase in the O( 3 P) formation rate through the second O 3 photodissociation channel because of the [O 3 ] increase is the reason of the above indicated behavior of atomic oxygen. At the same time the O( 3 P) + O3 2 O2 reaction does not compensate this effect, because at the altitude in question the ozone behavior is governed mainly by the nitrogen cycle and photodissociation processes [ Brasseur and Solomon, 1984]. At h = 50 km, vice versa, [O( 3 P)] decreases because of the slowing down of the first O 3 photodissociation process the O( 1 D) oxygen atoms are quickly quenched to the ground state). Thus a simultaneous depletion of the both ozone formation and ozone loss rates take place. The latter changes more significantly, if the decrease in [O 2(1 Dg) ] is taken into account, and finally that leads to further increase in [O 3 ], though in a lesser degree than at h = 30 km.
The obtained estimates undoubtedly are of a qualitative nature and the maximum ones. There are other suggestions besides the assumption on the value of the UV radiation attenuation. First, owing to a lack of experimental data [ Vetchinkin et al., 1993], the dissociation rate of all the constituents in the l< 240 nm range (including the Herzberg continuum) was assumed to be constant. If in this wavelength range the UV radiation, which comes to ozonosphere altitudes, is weakened by 50%, then as a result of a depletion of the atomic oxygen formation rate, which is the principal source of the O 3 formation at h = 30 km, a weakening of the [O 3 ] increase, or even its absence, is possible. Second, the character of the [O 3](t) variations demonstrates (see Figure 1) that in a general case one has to solve a nonlinear problem of UV radiation propagation through the terrestrial atmosphere.
Nevertheless, the above study is physically justified and indicates a presence of the mechanism (not considered before) of the rocket launch impact on the atmospheric ozone. However further experimental work is needed to make more complete theoretical estimates.
Aleksandrov, E. L., and L. B. Upenek, Evaluations of the influence of rocket fuel compounds and products of their combustion on the ozone content in the stratosphere, in Ecological Problems of Creation and Application of the Rocket and Space Technique, p. 64, NPO Energiya, Moscow, 1991.
Bernhard, P. A., Modeling of modification experiments involving neutral gas release, Symposium on Active Experiments in Space, Alpbach, May 24-28, 1983, Eur. Space Agency Spec. Publ., ESA SP-195, 271, 1983.
Brasseur, G., and S. Solomon, Aeronomy of the Middle Atmosphere, 452 pp., D. Reidel, Dordrecht, Holland, 1984.
Burdakov, V. P., N. F. Elanskiy, and V. M. Filin, Impact of launches of "Shuttle" and "Energiya" rockets on the ozone layer of the Earth, Vestnik Akad. Nauk SSSR, 12, 72, 1990.
Deminov, I. G., N. F. Elanskiy, Yu. E. Ozolin, and V. K. Petukhov, Evaluations of the impact of regular launches of "Energiya" and "Shuttle" rockets on the ozone layer and the climate of the Earth, preprint, 110 pp., IFA Ros. Akad. Nauk, Moscow, 1992.
Ivashkevich, A. K., Rocket carriers of the USA, Zemlya Vselennaya, no. 3, 85, 1993a.
Ivashkevich, A. K., Rocket carriers of the USA, Zemlya Vselennaya, no. 4, 94, 1993b.
Karol, I. L., Yu. E. Ozolin, and E. U. Rozanov, The space rocket launching effects on ozone and other atmospheric gases, Ann. Geophys., 9, 83, 1992.
Kozlov, S. I., Stratospheric and mesospheric disturbances from the UV source, Kosm. Issled., 22 (5), 712, 1984.
Kozlov, S. I., Photochemistry of the ionospheric D region in conditions of artificial injection of nitric oxides, Geomagn. Aeron., 27 (3), 383, 1987.
Kozlov, S. I., and N. V. Smirnova, A possibility to deplete the electron concentration in the highly ionized D region by an artificial injection of water vapor, Geomagn. Aeron., 30 (3), 522, 1990.
Kozlov, S. I., V. A. Vlaskov, and N. V. Smirnova, Ion kinetics, minor neutral and excited constituents in the D region with increased level of ionization, part I, Kosm. Issled., 20 (6), 881, 1982.
Kozlov, S. I., V. A. Vlaskov, and N. V. Smirnova, Ion kinetics, minor neutral and excited constituents in the D region with increased level of ionization, part II, Kosm. Issled., 21 (6), 892, 1983.
Kozlov, S. I., V. A. Vlaskov, and N. V. Smirnova, Ion kinetics, minor neutral and excited constituents in the D region with increased level of ionization, part III, Kosm. Issled., 22 (4), 565, 1984.
Kozlov, S. I., V. A. Vlaskov, and N. V. Smirnova, Specialized model to study artificial modification of the middle atmosphere and lower ionosphere, part I, Kosm. Issled., 26 (5), 738, 1988.
Kozlov, S. I., V. A. Vlaskov, and N. V. Smirnova, Specialized model to study artificial modification of the middle atmosphere and lower ionosphere, part II, Kosm. Issled., 28 (1), 77, 1990.
Pollack, J. B., et al., Estimates of the climatic impact of aerosols produced by Space Shuttles, SST's and other high flying aircraft, J. Appl. Meteorol., 15 (3), 247, 1976.
Potter, A. E., Environmental effects of the Space Shuttle program, AIAA pap., 81-0241, 1, 1981.
Prather, M. M., et al., The Space Shuttle's impact on the
Smirnova, N. V., V. A. Vlaskov, S. I. Kozlov, and N. A. Ovchinnikov, Variations of the oxygen compounds in the high-latitude mesosphere during disturbances, in Numerical Models of the Dynamical Processes, p. 22, Akad. Nauk SSSR, Apatity, 1984.
Smirnova, N. V., S. I. Kozlov, and E. A. Kozik, An impact of launches of the solid-fueled rockets on the terrestrial ionosphere, part I, Kosm. Issled., 33 (1), 98, 1995a.
Smirnova, N. V., S. I. Kozlov, and E. A. Kozik, An impact of launches of the solid-fueled rockets on the terrestrial ionosphere, part II, Kosm. Issled., 33 (2), 115, 1995b.
Turco, R. P., et al., A study of mesospheric rocket contrails and clouds produced by liquid-fueled rockets, Space Sol. Power Rev., 3, 223, 1982.
Vetchinkin, N. V., L. V. Granitskiy, Yu. V. Platov, and A. I. Sheykhet, Optical events in the terrestrial environment during operation of the boosters of rockets and satellites, 1, Ground-based and satellite observations of artificial formations during rocket launches, Kosm. Issled., 31 (1), 93, 1993.
Whitten, R. C., W. J. Borucki, I. G. Poppoff, and R. P. Turco, Preliminary assessment of the potential impact of solid-fueled rocket engines in the stratosphere, J. Atmos. Sci., 32 (3), 613, 1975.
Wu, B. J. C., Possible water vapor condensation in rocket exhaust plumes, AIAA J., 13 (6), 797, 1975.