1. Introduction

[2]  A rocket flight with working engine causes some powerful disturbances in the ionosphere. A disturbance in the ionosphere at a large distance from the launch location and flight trajectory was for the first time detected in 1959 during the launching of the Vanguard 2 satellite [Karlov et al., 1980]. During more than 40 years since that time, the observations of the upper atmosphere state using various radiophysical methods were intensively conducted in the world during carrier rocket (CR) launchings. The study of the ionosphere response to rocket launchings is of a great importance for the atmosphere science, because it would make it possible to study in detail various physical processes occurring in the upper atmosphere. The Earth atmosphere is an unique laboratory for studies of many physical processes, such as variation in photochemical reactions, propagation of shock waves (SW), formation of unstable structures in the ionosphere plasma, generation and propagation of large-scale atmosphere waves, etc. The important feature of rockets, as a source of a disturbance, is that they are located directly in the ionosphere F region.

[3]  Analyzing the results of numerous experimental works performed in different time using various observation methods [Adushkin et al., 2000; Karlov et al., 1980], one can state that disturbances in the upper atmosphere are observed in all cases of CR launching. These disturbances are mainly of two types: the first and second types are the generation of long-living large-scale irregularities in the ionosphere and generation of wave-like attenuating oscillations in the upper atmosphere propagating to large distances from the source, respectively. The cause of the generation of the first type is the development of the Rayleigh-Taylor and Perkins instabilities in the plasma [Adushkin et al., 2000] and distortion of the ionosphere photochemistry caused by the release of the combustion products of the rocket propulsion system [Karlov et al., 1980], etc. The second response arises due to the propagation of SW in the atmosphere with the following generation of acoustic gravity waves (AGW). A typical feature of this type of disturbance is that AGW are observed at large distances (about 1000 km) from the rocket trajectory.

[4]  According to the observation data [Adushkin et al., 2000], during rocket launchings a few wave packets are detected in the ionosphere. SW are registered the first out of them. The type of rocket, the orientation of the orbit plane relative to the rocket motion plane, and the length of the oblique short-wave radio path influence weakly the studied parameters of SW. The main input is provided by the diurnal seasonal behavior. No responses to signal of oblique short-wave sounding similar to the response to SW were detected in the natural conditions. It is due to the fact that in the ionosphere plasma there are absent disturbances with timescales of 1-4 min with the form similar to the profile typical for SW. Weak wave disturbances belonging to the low-frequency acoustic range (LFA₁ ) are observed after SW propagation. The second group of waves (LFA₂ ) in the signal response appears in a few tens of minutes after the start. Also the arrival of one more group of waves (LFA₃ ) is possible on the background of LFA₂. The periods of these acoustic disturbances vary in the range 2.9-5.3 min. Internal gravity waves (IGW) are also inherent part of wave disturbances accompanying rocket launching. According to the data available the period range of IGW observed in the disturbances spectrum splits into three bands: up to 10 min, 15-30 min, and 75-100 min. The first packet (IGW1) appears in the signal spectrum after the intersection of the CR plane by the reference radio path and SW passage. This wave packet was detected not only in the direct vicinity of the active part of rocket motion trajectory but also at distances of about 1000 km and more. The second wave packet (IGW2) appears in a few tens of minutes after the first one and can be registered during a few hours. At remote distances of about 1000 km and more, the front of the disturbances is almost vertical with a small advance in time relative to the disturbance appearance at higher altitudes.

Figure 1

[5]  IGW with periods of 15-75 min were registered over the Arecibo observatory (at a distance of > 1000 km from the source) by the incoherent scatter radar after the launching of the space shuttle CR from the Kennedy Space Center (KSC) site on 27 June 1982 [Noble, 1990]. Similar results were obtained at the launching of the Soyuz CR from Baikonur site on 12 December 1990. Figure 1 shows the variations in the Doppler frequency registered during this launching at the Tashkent-Tomsk circuit crossing the active part of the rocket motion trajectory [Adushkin et al., 2000].

[6]  The above mentioned disturbance properties are also confirmed by other researchers. Acoustic waves (AW) with period about 1.5-4 min were revealed during launchings time of spacecrafts Apollo 12 and Apollo 13 using the Doppler sounding of the ionosphere at frequencies of 4824 and 6030 kHz and microbarographs net [Karlov et al., 1980]. The experimental data obtained give values of the phase and group velocities of the wave equal to 700-800 and 220-450 m s^-1, respectively.

[7]  Observations were also carried out using high-frequency and Doppler sounding during launchings of the space shuttle on 28 February 1990 and 28 April 1991 [Jacobson and Carlos, 1994]. The ionosphere wave-like disturbances with periods of 150-250 s were detected in all cases. Calculations show that the N form pulse observed in the ionosphere due to shuttle flight propagates in the upper atmosphere with the speed of sound. Disturbances with a few cycles with a period of about 200 s are observed 700-800 s after that. Other authors also registered SW caused by launchings of space shuttle which propagate in the Earth atmosphere with a horizontal velocity of 600-700 m s^-1. SW with a phase velocity of 700-800 m s^-1 was also detected during the launching of the Apollo CR.

[8]  A new activity in studies of ionosphere disturbances caused by sources of various types including rocket launchings began with development of the Global Positioning System (GPS) in the 1990s. High density of the GPS receivers net and sufficient regularity of rocket launchings from various rocket sites make it possible to study this phenomenon within wide time and spatial scales. Calais and Minster [1998] presented a detailed analysis of the observation results obtained by GPS receivers during the launching of the space shuttle Columbia CR on 18 October 1993. The authors registered two wave packets of oscillations of the total electron content (TEC). The first wave packet is a N form pulse with a period of about 300 s. It does not show any wave dispersion, whereas the second wave packet demonstrates obviously the dispersion and consists of several cycles. The calculations show that the horizontal phase velocity of the first wave propagation is of about 800 m s^-1 at higher altitudes. In 15 min after the passing of the first packet the second wave packet propagating with a horizontal velocity of about 300 m s^-1 is observed. Wave disturbances of TEC during launchings of Soyuz and Proton CR from the Baikonur site were detected using the GPS system [Afraimovich et al., 2001]. On the basis of these results one can state that the ionosphere response during the considered launchings has a shape of N wave. The disturbances amplitude varies from 0.03 TECU to 0.9 TECU (total electron content units is commonly used unit of TEC, 1 TECU is equal to 10¹⁶ m^-2 ). The period of these waves varies from 132 to 288 s.

[9]  In a few papers attempts have been undertaken to explain from the theoretical point of view the observed disturbances in the ionosphere during a rocket flight. The analytic and numerical calculations predict correctly the arrival time of the first ionosphere waves pulse from such sources. However, no model can still explain the cause of the appearance of the second wave packet following the first pulse [Calais and Minster, 1998].

[10]  Arendt [1971] assumes that the shock wave formed at rocket flight splits at a height of about 160 km in the ionosphere to ion acoustic and normal acoustic modes. The facts that the motion velocity of the first disturbance found during the launchings of Apollo 14 and Apollo 15 is close to the velocity of the ion acoustic mode and that the second disturbance velocity is close to the velocity of the normal acoustic wave are considered by Arendt [1971] to be in favor of the proposed hypothesis.

[11]  Some authors interpret the appearance of the secondary waves as a consequence of the reflection from the solid Earth surface. However, the same disturbances are observed during earthquakes and ground explosions, and in this case the waves can not be explained by a reflection [Calais and Minster, 1998].

[12]  Calais and Minster [1998], Nagorsky [1998] and Tolstoy et al. [1970] assume that the second wave packet is a result of the capture of AGW in the atmospheric mesosphere-thermosphere waveguide (MTW), which is located between the mesopause (about 100 km) and the thermocline (about 120 km). Nevertheless, these assumptions are not proved by conclusions of theoretical works. None of the existing models can correctly predict the generation of long-period IGW from short-lived disturbances pulses. Moreover, on the basis of the existing models, it is impossible to describe completely the total picture of ionosphere disturbances.

[13]  The above mentioned facts clearly show that the great experimental material dedicated to the atmosphere and ionosphere disturbances generated at rocket launchings has been accumulated. However, the interpretation of the space-time characteristics of these disturbances is obviously insufficient. The goal of our paper is to simulate the ionosphere wave-like disturbances caused by rocket launchings using contemporary achievements in the field of numerical simulation of geophysical fluid dynamics problems and also to perform a preliminary comparison of the obtained results to experimental data.