- Email: [email protected]
Rocket observations of propagating waves in the night time equatorial F-region over Brazil
Available online at www.sciencedirect.com
Advances in Space Research 42 (2008) 1194–1201 www.elsevier.com/locate/asr
Rocket observations of propagating waves in the night time equatorial F-region over Brazil P. Muralikrishna
*
Instituto Nacional de Pesquisas Espaciais – INPE/MCT, Division of Aeronomy, Av. dos Astronautas, 1758, C.P. 515, 12201-970, Sa˜o Jose´ dos Campos-SP, Brazil Received 1 November 2006; received in revised form 25 July 2007; accepted 27 July 2007
Abstract A Brazilian SONDA III rocket carrying plasma diagnostic experiments was launched from the Brazilian rocket launching stations in Alcaˆntara (2.31°S, 44.4°W Geog. Lat.) to measure the height profiles of electron density, electron temperature and the ambient electric field. High frequency capacitance probe was used to measure the height profile of the electron density and the Langmuir probe was used to measure the electron density and the spatial structures of plasma irregularities. An electric field double probe was used to measure the electric field fluctuations associated with the F-region plasma irregularities. Spectral analysis of the fluctuations in electron density and electric field indicated the presence of propagating waves in the night time F-region over a large height range. The electron temperatures estimated from the LP data showed abnormally high values in the base of the F-region during the upleg of the rocket and practically normal values in the same height region during the downleg. A brief study of the characteristic features of the spectra of electron density and electric field fluctuations and the associated electron temperature variations are presented and discussed here. Ó 2007 COSPAR. Published by Elsevier Ltd. All rights reserved. Keywords: F-region; Plasma waves; Electron temperature; Rayleigh–Taylor instability; Plasma irregularities
1. Introduction Spread-F is considered the most important of the plasma instability phenomena occurring in the equatorial F-region. Several linear and non-linear theories have been developed to explain the wide spectrum of electron density irregularities observed in the night time F-region associated with spread-F activity (Reid, 1968; Hudson et al., 1973; Sudan et al., 1973). Haerendal (1974) suggested a multistep process to explain the large range of wavelengths observed, from several kilometres down to few centimetres. The post-sunset equatorial F-layer can become unstable under the influence of any disturbance produced by gravity waves, neutral winds or electric field fields, and can generate plasma irregularities through the Rayleigh–Taylor instability (RTI) mechanism (Hysell et al., 1990; Singh *
Tel.: +55 12 39457148; fax: +55 12 39456990. E-mail address: [email protected]
et al., 1997) resulting in the generation of large scale plasma depletions or plasma bubbles. Bubbles are mostly aligned with the geomagnetic field flux tubes with plasma density decreases of up to three orders of magnitude. These depletions are generally produced over geomagnetic equator and they connect upwards through the F-layer peak to the topside ionosphere, reaching altitudes as high as 1200 km or more (Woodman and La Hoz, 1976). Though the RTI mechanism can explain the large scale irregularities observed in the bottom side F-region, it is known to be stable in the short wavelength range. But coherent radars installed at several locations all over the world have observed echoes from irregularities in the short wavelength range close to 3 m not only from the bottom side of the F-region but also from the top side. Phase velocities of these irregularities may range from a few m s1 to several hundreds of m s1 close to the ion acoustic speed. If no other wave generation process exists, these short wavelength waves must receive energy from the longer scales
0273-1177/$34.00 Ó 2007 COSPAR. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.asr.2007.07.039
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
through a cascade process of some sort. Considerable effort has gone into studying the possible generation of short wavelength waves by the larger primary waves. Chaturvedi and Kaw (1976) put forward a two-step theory of longer wavelength RTI modes directly coupling with kinetic collisional drift waves to explain the measured k2 spectra of the electron density irregularities. Both theory and experiments show that steep electron density gradients can develop in the medium as a result of the RTI process and most studies have involved gradient driven instabilities, under the general term of drift waves, as responsible for the shorter wavelength irregularities (Costa and Kelley, 1978a; Huba and Ossakow, 1981; LaBelle et al., 1986). The question of wave generation at wavelengths of a few meters, close to the ion gyro radius, is discussed by Costa and Kelley (1978b). Theoretical studies indicate that the relationship between the electron density perturbation and the fluctuations in the electric field associated with the short wavelength waves in the F-region is a linear one, though the exact relationship between them may depend on the plasma instability mechanism responsible for their generation (see Kelley and Heelis, 1989). Hence the power spectrum of the fluctuations in electric field and electron density should be similar except for the power contained in the spectra. Their spectral signatures may be used as a test for sorting out the electrodynamic processes that operate at any given altitude as a function of wave number of the waves. From simultaneous in situ measurements of electron density and electric field fluctuations Hysell et al. (1994) showed that irregularities in the scale size range of 100 m–2 km display a power law behaviour with spectral index n 2 that increased to 4.5 for wavelengths less than around 100 m when the F-layer is high. Measurement of the spectral parameters of the electron density and electric field fluctuations can, thereby, give us valuable information on the plasma instability mechanism responsible for the generation of these irregularities. In situ measurements of the height variation of the ionospheric electron density and electric field variations were made with two different types of rocket-borne electron density probes and an electric field double probe from the equatorial region in Brazil to study the relationship between the spectral features in the electron density and electric field variations in the F-region dominated by large scale plasma bubbles. The results obtained from this comparative study are presented and discussed here. 2. Experiment and flight details A Brazilian SONDA III rocket carrying three plasma experiments in addition to other airglow experiments was launched on 18 December 1995 at 2117 hrs (LT) from the equatorial rocket launching station, Alcaˆntara (2,31°S; 44,4°W) in Brazil. The principal objective of the plasma experiments was to measure the electric field, the electron density, and the spectral distribution of plasma irregulari-
1195
ties associated with plasma bubbles. The payload consisted of the following experiments: electric field double probe (EFP), Langmuir probe (LP) and high frequency capacitance probe (HFC). The experiment and data analysis details are described in Muralikrishna et al. (2003) and Muralikrishna (2006). The basic principle of operation and the details of the electronic subsystems of the LP and HFC experiments are given in Muralikrishna and Abdu (1991). Several ground equipments were operated during the launch campaign with the specific objective of knowing the ionospheric conditions at the time of launch and thereby to launch the rocket into an F-region prone to the presence of plasma bubbles. Ionograms were obtained at Sa˜o Luis, a station close to the launch site and at Cachoeira Paulista, a station outside the low latitude belt, but more or less at the same geomagnetic longitude as Sa˜o Luis. Appearance of spread-F traces in the ionograms recorded at Sa˜o Luis and the consequent appearance of spread-F traces at Cachoeira Paulista was considered to be a strong indication for the presence of plasma bubbles in the ionosphere over that launch station. The rocket was launched under ionospheric conditions favourable for the presence of bubbles. The rocket reached an apogee altitude of 557 km and covered a horizontal range of 589 km. The mean azimuth angle of the plane of the trajectory was about 61.2°. The rocket in fact passed through several medium scale plasma bubbles mainly during the downleg. The upleg electron density profile showed the presence of a very clearly defined base for the F-region around 300 km, while the downleg profile showed the presence of a wide spectrum of electric field and electron density irregularities in this height region as well as in the upper F-region. An FFT algorithm was then used to estimate the spectral distribution of the electric field and electron density fluctuations. It should be noted here that the LP and EF sensors were mounted away from the rocket body in a plane perpendicular to the rocket spin axis. The electric field measured thus represents only the component of the field perpendicular to the spin axis of the rocket. The rocket had a spin rate of less than 3 per second, which in fact, can be clearly seen in the total electric field data, but the amplitude of this modulation is reduced very much in the electric field fluctuation data that is separated through a high pass filter. The rocket spin and precession were monitored through onboard magnetometers and their effects on the electron density and electric field data were seen to be rather low and easily identifiable in the spectra. The LP sensor was operated in a swept voltage mode with a sweep cycle duration of about 2.5 s. During each operation cycle the sweep voltage applied to the sensor increased linearly from 1 V to +2.5 V in about 1.5 s and remained steady at +2.5 V for about a second. The fluctuating components of the LP and EF data corresponding to the time intervals when the LP sensor is in saturation current mode (at fixed +2.5 V bias) are chosen for the spectral analysis, while the LP data
1196
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
collected during the voltage sweep was used to estimate the electron temperature.
HFC Electron Density (m-3) 9
10 600
3. Results and discussion 500
3.1. Electron density and electric field fluctuations
HFC Electron Density (m-3) 9
10 600
500
1010
1011
1013
Alcantara, Brazil 18 December 1995 2117 LT Upleg
400
Height (km)
1012
D
HFC
LP
C B 300
A
200
100
0 8 10
10
9
10
10
10
11
10
12
-3
LP Electron Density (m ) Fig. 1. Upleg electron density profiles estimated from the Langmuir probe (LP) and the high frequency capacitance (HFC) probe data. Note that the x-axis (top) for the HFC profile is shifted by one order to the right.
11
12
10
13
10
10
Alcantara, Brazil 18 December 1995 2117 LT Downleg HFC C
400
Height (km)
The electron density profiles estimated from the LP and HFC experiments for the rocket upleg and downleg are shown in Figs. 1 and 2, respectively. The general features of the electron density profiles measured by the HFC experiment are similar to those measured by the LP, but the absolute values of the electron densities at different height regions showed some differences. Muralikrishna and Abdu (1991) tried to explain similar differences observed in one of their earlier in situ measurements as due to certain inherent problems associated with the two techniques of measurement. However, what is more important in the present studies is the relative variation of the electron density with altitude. As can be seen from Figs. 1 and 2 the upleg electron density profile shows the presence of a rather steep F-region base, free of any large scale electron density depletions or bubbles, while the downleg profile shows the presence of a large number of plasma bubbles. The downleg profile does not show a sharp F-region base. Power spectra of electron density and electric field fluctuations observed during the upleg and downleg of the rocket at a few selected height regions are compared in Figs. 3–6. Though normally the spectra are plotted on a
10
10
LP
D
B 300
A
200
100
0 8 10
109
1010
1011
1012
-3
LP Electron Density (m ) Fig. 2. Downleg electron density profiles estimated from the LP and HFC probe data. Note the shift in the x-axis for the HFC profile to the right.
log–log scale, in the present case they are plotted with linear scale along the x-axis (wave number) and log scale along the y-axis (spectral power) with the main purpose of highlighting the structure of the spectral peaks. The two panels on the left represent the upleg spectra and the two panels on the right the downleg spectra corresponding to more or less the same altitude region. The data-sampling rate is 1250 s1 that corresponds to an upper frequency limit of 625 Hz. Assuming that at the rocket speed of 2– 3 km s1, the ambient plasma irregularities can be considered practically stationary, the temporal variations observed by the experiments can be converted into spatial variations. The frequency of the LP current fluctuation or the E-probe potential fluctuation can be converted to wavelength k or wave number k (k = 2p/k) of the waves knowing the rocket velocity. The spectra of electron density and electric field fluctuations are estimated from data blocks each containing 1024 points collected in about 0.8 s. Shown along the x-axis is the wave number k that varies from 0 to more than 2200 km1, the upper limit being a function of the rocket velocity. Along the y-axis the spectral power is shown in arbitrary (relative) units. Spatial fluctuations in both electron density and electric field can be seen in both Figs. 3 and 4. The electron density and E-field spectra in Fig. 3 correspond to a mean altitude of about 295 km. This height region is identified by the letter A in the electron density height profiles shown in Figs. 1 and 2. The plasma regions represented by the two panels correspond to the base of the F-region and have a horizontal separation of more than 400 km. Remembering that the y-axis representing the spectral power have logarithmic scales, one can see that the spectral peaks have higher
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
(δE)2Hz-1
10
1
Electric Field
Electric Field
10
1
10
-1
10
-1
10
-3
10
-3
10
-5
10
-5
10
1
10
-1
10
-3
10
-5
10
(δne)2Hz-1
1197
Electron Density
1
10
-1
10
-3
10
-5
Electron Density
0
400
800
1200
1600 0
Wave Number k (km-1)
400
800
1200
1600
Wave Number k (km-1)
Fig. 3. Power spectra of electric field (top) and electron density (bottom) observed at the mean height of about 295 km during upleg (left) and downleg (right) of the rocket.
(δE)2Hz-1
10 1
10
1
10
-1
10
-3
10
-5
10
5
10 3
10
3
1
10
1
10
-1
10
-1
10
-3
Electric Field
Electric Field
10-5
(δne)2Hz-1
10
10
Electron Density
5
10-1
0
500
1000
1500
2000 0
Wave Number k (km-1)
Electron Density
400
800
1200
1600
Wave Number k (km-1)
Fig. 4. Power spectra of electric field (top) and electron density (bottom) observed at the mean height of about 333 km during upleg (left) and downleg (right) of the rocket.
powers in the right panels than in the left panels. While the E-spectrum observed during the upleg is practically devoid of any spectral peak, one can see the existence of two rather very low power peaks in the upleg ne spectrum corresponding to the mean wave numbers of 1470 km1 and 1655 km1. This probably is due to the fact that the RTI mechanism has not yet started or is only at its very initial state of development. During the downleg two common spectral peaks are seen in the E and ne spectra more or less at the same wave numbers and with the same relative power as seen during upleg. But the power in the spectral peaks during downleg is higher by orders of magnitude.
The downleg ne spectrum shows at least three additional spectral peaks corresponding to the wave numbers of about 575 km1, 660 km1 and 740 km1 that are not seen in the E-spectrum. From the fact that while the E-spectrum is completely devoid of any spectral peaks the ne spectrum exhibits a few spectral peaks and the observation of difference of orders of magnitude in spectral powers during upleg and downleg one can rule out the possibility of the sharp spectral peaks being caused by any mutual interference between the EF and LP sensors or by interference with other experiments on board. The existence of wave structures in both ne and E at the same wave numbers is
1198
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
10
1
1
Electric Field
Electric Field
10
-1
10
-1
10
-3
10
-3
10
-5
10
-5
10
5
Electron Density
Electron Density
10
5
10
3
10
3
10
1
10
1
10
-1
δ
δ
10
10
-1
0
500
1000
1500
2000 0
Wave Number k (km-1)
500
1000
1500
2000
Wave Number k (km-1)
Fig. 5. Power spectra of electric field (top) and electron density (bottom) observed at the mean height of about 376 km during upleg (left) and downleg (right) of the rocket.
1
10
δ
-3
10
-5
10
Electron Density
5
Electron Density
3
δ
10
1
10
-1
10
0
500
1000
1500
2000
Wave Number k (km-1)
1
10
-1
10
-3
10
-5
10
5
10
3
10
1
10
-1
Electric Field
Electric Field -1
10
10
10
0
500
1000
1500
2000
Wave Number k (km-1)
Fig. 6. Power spectra of electric field (top) and electron density (bottom) observed at the mean height of about 404 km during upleg (left) and downleg (right) of the rocket.
indicative of propagating waves. The right panel also exhibits spectral peaks seen only in the ne spectrum, not associated with corresponding peaks in the E-spectrum. These peaks are not exhibited by the left panel spectrum and therefore do not seem to represent propagating waves, but only electron density irregularities. Fig. 4 corresponds to the mean height region of about 335 km, indicated by letter B in the ne profiles shown in Figs. 1 and 2. The upleg spectra on the left panels, still exhibit low power spectral peaks in both ne and E, while the downleg spectra exhibit higher power, especially in the ne spectrum. The wave numbers corresponding to these spectral peaks in E and ne are almost same in the upleg and
downleg spectra, while the higher peak in the upleg spectrum seems to appear at a wave number slightly smaller than that in the downleg spectrum. This again indicates that they are associated with propagating waves. It should be noted here that the horizontal separation of the height regions represented by the left and right panels in this case is about 380 km. The slight difference in the wave numbers of the spectral peaks during upleg and downleg may probably be caused by the upward velocity of these drift waves in association with the upward rocket velocity and/or by the changes occurring in the rocket trajectory as discussed later. The estimated wavelength corresponding to the spectral peak at wave number 1470 km1 is about 4.3 m.
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
3.2. Electron temperature profile Electron temperature was estimated from the slope of the current–voltage characteristics of the Langmuir probe. In estimating Te from the slope of the I–V characteristic curve one generally assumes that the positive ion current is much less than the electron current. The exact shape of the current–voltage characteristic of a Langmuir probe is decided by the plasma temperature and the shape and size of the plasma sheath that forms surrounding the sensor surface. By solving Poisson equation for the distribution of potential in a plasma sheath, one can obtain the following approximate relation for the current of particles moving in a retarding field as ene v eV Ip ¼ a: exp ; ð1Þ 4 kT e
Thus measuring the probe current collected at two different probe potentials one can, in principle, estimate both the plasma number density and the temperature using the above relations. The LP data corresponding to the time intervals when the LP sensor is in the sweep potential mode (1 V to +2.5 V bias) only are chosen for estimating the electron temperature reported here. Muralikrishna (2006) presented the LP current–voltage characteristics for selected height regions and estimated the electron temperature profiles from the LP data obtained during the flight reported here and found that the electron temperature showed an abnormal increase, probably just before the onset of the RTI mechanism responsible for the bubbles observed during the rocket downleg. The upleg and downleg Te profiles are shown in Figs. 7 and 8 for comparison with the spectral data presented above. Comparison of Figs. 3–6 with Figs. 7 and 8 brings out some interesting observations. While Fig. 3 corresponding to the height region of about 295 km does not show large power spectral peaks in neither ne nor E upleg spectra, Fig. 7 shows abnormally large electron temperature in the same height region. In the same height region more than 400 km away horizontally, the downleg spectra show the presence of spectral peaks with significant power, but, as can be seen from Fig. 8, the abnormal electron temperature seen in the base of the F-layer has disappeared. This probably can be explained as due to the conversion of the electron thermal energy into wave energy during the operation of the RTI mechanism. Another observation that can be made by comparing the corresponding spectral peaks in ne and E given in Figs. 3–6 600 ALCANTARA, BRAZIL 18 December 1995 2117 LT LP Upleg
500
D
400
C
Height (km)
The ne and E spectra corresponding to the height regions of about 375 km and 404 km (indicated by letters C and D in the ne profiles of Figs. 1 and 2), respectively, are shown in Figs. 5 and 6. It can be seen from these figures that the propagating waves continue to exist in these height regions. The horizontal separation of the height regions shown in Fig. 5 is about 330 km and that for the height regions shown in Fig. 6 is about 304 km. The power of the spectral peaks observed during the upleg and downleg in these height regions is almost same except that the corresponding wave numbers are slightly smaller in the upleg spectra as seen in Fig. 4. But the electron density spectra corresponding to these height regions show several additional spectral peaks at lower wave numbers (at higher wavelengths), mainly in the downleg spectra, few of which are accompanied by spectral peaks in the E spectra. These probably correspond to plasma irregularities generated by other instability mechanisms. One should remember here that the electron density profile observed during the upleg of the rocket does not indicate the presence of neither plasma bubbles nor large or medium scale plasma irregularities. However, the downleg profile indicates the presence of a large number of plasma bubbles and associated large, medium and small-scale irregularities. The spatial and temporal separation of the upleg and downleg trajectories of the rocket being small, the rather complete absence of plasma bubbles in the upleg profile and the presence of a large number of bubbles in the downleg profile indicate that the bubbles could be recently generated and be in their developing phase.
1199
B A
300
200
Ne
Te
100
where a is the surface area of the probe, e is the electronic charge, ne is the electron density. This relation is valid regardless of the shape of the probe and consequently ln Ip has a linear dependence on the potential V applied to the probe (in a given potential range). From this linear relation one can get the electron temperature as, 5040 Te ¼ dðln I p Þ=dV
ð2Þ
IRI Model
0 9 10
10
10
10
11
10 -3
Electron Density(m )
12
0
1000
2000
3000
Electron Temperature (οK)
Fig. 7. Night time electron temperature height variation observed on 18 December 1995 during the rocket upleg. Also shown in the figure are the upleg electron density profile of Fig. 1 and the IRI model temperature profile.
1200
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201 600 ALCANTARA, BRAZIL 18 December 1995 2117 LT LP Downleg
500
D
Height (km)
400 C B
300
A
Ne
200
Te
100
IRI Model
0 9 10
10
10
10
11
Electron Density(m-3)
0
400
800
1200
Electron Temperature (οK)
Fig. 8. Night time electron temperature height variation observed on 18 December 1995 during the rocket downleg. Also shown in the figure are the downleg electron density profile of Fig. 2 and the IRI model temperature profile.
is that the relative amplitudes of the spectral peaks are more or less same. Also one can see that the spectral power decreases with increasing wave number or with decreasing wavelength thereby showing negative spectral indices as expected from the existing theories. The spectral peaks observed in electron density and electric field spectra could represent waves generated during the initial phase of the development of plasma bubbles or during the operation of other instability mechanisms responsible for the drift waves. The appearance of these spectral peaks in both the upleg and the downleg of the rocket trajectory indicates that these are propagating waves. Since the E-field measurements are made on board a moving space vehicle, it is difficult to estimate the actual wave numbers of these waves without knowing their phase velocities (Fredericks and Coroniti, 1976; Temerin, 1979). However, one can infer that the real wave numbers are close to the ones observed, since the ne spectra also exhibit spectral peaks at the same wave numbers and these are not affected by the velocity of the moving frame of reference. Electrostatic waves in the Earth’s ionosphere exist over a large range of wavelengths, from centimetres to hundreds of kilometres, and over a large range of phase velocities, from quasi-static to speeds of hundreds of kilometres per second. Because of the motion of the spacecraft relative to ambient plasma, even quasi-static waves appear as temporal fluctuations in the spacecraft frame. For electrostatic waves with frequencies below the ion plasma frequency or the lower hybrid frequency, whichever is lower, both ions and electrons participate in the response of the medium to electrostatic waves. This implies that there are density and velocity fluctuations associated with the wave.
If one assumes that the spectral peaks are caused by drift waves propagating upwards, the wavelength of fluctuations as measured by a rocket moving upward will appear longer (wave number smaller) and that as measured by a rocket moving downward will appear shorter (wave number larger). This is exactly what is seen in the spectra presented in Figs. 3–6. The spectral peaks appear at slightly lower wave numbers during the upleg than during the downleg. In addition to this, changes in the trajectory of the rocket will also cause changes in the measured wave number of the waves. As the rocket moves to higher altitudes its trajectory becomes more and more horizontal and for a given vertical wavelength of the waves, the measured wavelength becomes longer and longer or the wave number becomes smaller and smaller. The observed shift in the spectral peaks between the upleg and downleg may be partly due to this too. It should be noted here that the spectral peaks seen in the spectra presented here will practically disappear if one presents them in a conventional log–log plot. For example, a close look at the electron density spectra presented in Kelley et al. (1982) clearly indicates the presence of a large number of spectral peaks near the high wave number end. They are not dominantly seen only because of the logarithmic scale used. The strongest spectral peaks observed in the present case are also at the high wave number end of the spectra and are seen always practically at the same wave numbers. The important features of the power spectra of electron density and electric field fluctuations presented in Figs. 3–6 and the electron temperature profiles shown in Figs. 7 and 8 can be summarised as follows. At most of the heights the spectral data show the presence of sharp spectral peaks in both ne and E probably caused by drift waves. The origin of these waves is unknown. One of the possibilities is that these are generated during the development phase of plasma bubbles. The large density gradients associated with the walls of the bubbles (in both east–west and vertical directions) in conjunction with the east–west or vertical E-field may also trigger some plasma instability process responsible for the drift waves. Alternatively the large scale bubbles, in course of time, may break into smaller scale waves (cascade process) losing their energy. This may finally lead to a steady state distribution (without spectral peaks) of energy in a large range of wavelengths. This final state maybe characterised by one or more characteristic spectral indices for the irregularities reported in the literature. The spectral peaks observed seem to represent propagating waves and are seen with more or less same relative amplitudes in the electron density and electric field. At those height regions where no spectral peaks are observed neither in ne nor in E-field, and their fluctuation amplitudes are relatively high, it seems that the
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
situation has already reached a steady state, wherein the spectral features are those that are expected from the existing theories. The abnormally high electron temperature observed below the base of the F-region seems to be a precursor to the development of plasma bubbles. Once the bubbles develop and start rising up the electron temperature falls down. The large kinetic energy of electrons may be transported to other height regions through the rising bubbles or propagating waves. The presence of the abnormally large electron temperatures at the base of the F-region before the development of bubbles, and its disappearance in the same height region (horizontally separated by a few hundred kilometres) during the rocket downleg, also supports this hypothesis.
4. Conclusions
Bubble regions are associated with a wide spectrum of both electron density and electric field fluctuations probable caused by drift waves. The absence of sharp spectral peaks in ne and/or E-field probably indicates that the cascading process of the generation irregularities has already reached the final state or that the physical conditions in the particular height region are not favourable for the operation of any plasma instability mechanism. Abnormally large electron temperatures observed at the base of the F-region seem to be a precursor to the development of plasma bubbles.
Acknowledgements The author is grateful to the Directors of IAE/CTA and CLA, Alcantara for providing the rockets and the launch facilities respectively and to the staff of IAE and CLA for their help during the pre-launch tests of the experiments, and during the launching of the rockets. Sincere thanks to Sinval Domingos, Agnaldo Eras, and Narli Baesso Lisboa for their technical help in the development testing and integration of the experiments. The work reported here was partially supported by FINEP under contract FINEP-537/CT, and by CNPq under process 300253/89-3/
1201
GM/FV. The authors are thankful to the referees for their very useful suggestions and comments. References Chaturvedi, P.K., Kaw, P. An interpretation for the power spectrum of spread-F irregularities. J. Geophys. Res. 81, 3257–3260, 1976. Costa, E., Kelley, M.C. On the role of steepened structures and drift waves in equatorial spread F. J. Geophys. Res. 83, 4359, 1978a. Costa, E., Kelley, M.C. Linear theory for the collisionless drift wave instability with wavelengths near the ion gyro radius. J. Geophys. Res. 83, 4375, 1978b. Fredericks, R.W., Coroniti, F.V. Ambiguities in the deduction of rest frame fluctuation spectrum from spectrums computed in moving frames. J. Geophys. Res. 81, 5591–5595, 1976. Haerendal, G., Theory of equatorial spread-F, Report of Max Planck Institu¨t fur Physik und Astrophysik, Garching, West Germany, 1974. Huba, J.D., Ossakow, S.L. Lower hybrid drift waves in equatorial spread F. J. Geophys. Res. 86, 829, 1981. Hudson, M.K., Kennel, C.F., Kaw, P.K. Two step drift mode theory of equatorial spread-F. Trans. Am. Geophys. Soc. 54, 1147, 1973. Hysell, D., Kelley, M.C., Swartz, W.E., Pfaff, R.F., Swenson, C.M. Seeding and layering of equatorial spread-F by gravity waves. J. Geophys. Res. 95, 17253–17260, 1990. Hysell, D., Kelley, M.C., Swartz, W.E., Woodman, R.F. Steepened structures in equatorial spread-F, 1. New observations. J. Geophys. Res. 99, 8827–8840, 1994. Kelley, M.C., Heelis, R.A. The Earth’s Ionosphere: Plasma Physics and Electrodynamics. Academic Press, 1989. Kelley, M.C., Pfaff, R., Baker, K.D., Ulwick, J.C., Livingston, R., Rino, C., Tsunoda, R. Simultaneous rocket probe and radar measurements of equatorial spread F – transitional and short wavelength results. J. Geophys. Res. 87, 1575–1588, 1982. LaBelle, J., Kelley, M.C., Seyler, C.E. An analysis of the role of drift waves in equatorial spread F. J. Geophys. Res. 91, 5513–5525, 1986. Muralikrishna, P., Abdu, M.A. In-situ measurement of ionospheric electron density by two different techniques – a comparison. J. Atmos. Terr. Phys. 53, 787–793, 1991. Muralikrishna, P., Vieira, L.P., Abdu, M.A. Electron density and electric field fluctuations associated with developing plasma bubbles. J. Atmos. Solar Terr. Phys. 65, 1315–1327, 2003. Muralikrishna, P. Electron temperature variations in developing plasma bubbles – rocket observations from Brazil. Adv. Space Res. 37, 903– 909, 2006. Reid, G.C. Small-scale irregularities in the ionosphere. J. Geophys. Res. 73, 1627–1640, 1968. Singh, S., Bhamgboye, D.K., McClure, J.P., Johnson, F.S. Morphology of equatorial plasma bubbles. J. Geophys. Res. 102, 20019–20029, 1997. Sudan, R.N., Akinrimisi, J., Farley, D.T. Generation of small-scale irregularities in the equatorial electrojet. J. Geophys. Res. 78, 240, 1973. Temerin, M. Doppler shift effects on double-probe measured electric field power spectra. J. Geophys. Res. 84, 5929–5934, 1979. Woodman, R.F., La Hoz, C. Radar observations of equatorial F-region irregularities. J. Geophys. Res. 81, 5447–5466, 1976.
Advances in Space Research 42 (2008) 1194–1201 www.elsevier.com/locate/asr
Rocket observations of propagating waves in the night time equatorial F-region over Brazil P. Muralikrishna
*
Instituto Nacional de Pesquisas Espaciais – INPE/MCT, Division of Aeronomy, Av. dos Astronautas, 1758, C.P. 515, 12201-970, Sa˜o Jose´ dos Campos-SP, Brazil Received 1 November 2006; received in revised form 25 July 2007; accepted 27 July 2007
Abstract A Brazilian SONDA III rocket carrying plasma diagnostic experiments was launched from the Brazilian rocket launching stations in Alcaˆntara (2.31°S, 44.4°W Geog. Lat.) to measure the height profiles of electron density, electron temperature and the ambient electric field. High frequency capacitance probe was used to measure the height profile of the electron density and the Langmuir probe was used to measure the electron density and the spatial structures of plasma irregularities. An electric field double probe was used to measure the electric field fluctuations associated with the F-region plasma irregularities. Spectral analysis of the fluctuations in electron density and electric field indicated the presence of propagating waves in the night time F-region over a large height range. The electron temperatures estimated from the LP data showed abnormally high values in the base of the F-region during the upleg of the rocket and practically normal values in the same height region during the downleg. A brief study of the characteristic features of the spectra of electron density and electric field fluctuations and the associated electron temperature variations are presented and discussed here. Ó 2007 COSPAR. Published by Elsevier Ltd. All rights reserved. Keywords: F-region; Plasma waves; Electron temperature; Rayleigh–Taylor instability; Plasma irregularities
1. Introduction Spread-F is considered the most important of the plasma instability phenomena occurring in the equatorial F-region. Several linear and non-linear theories have been developed to explain the wide spectrum of electron density irregularities observed in the night time F-region associated with spread-F activity (Reid, 1968; Hudson et al., 1973; Sudan et al., 1973). Haerendal (1974) suggested a multistep process to explain the large range of wavelengths observed, from several kilometres down to few centimetres. The post-sunset equatorial F-layer can become unstable under the influence of any disturbance produced by gravity waves, neutral winds or electric field fields, and can generate plasma irregularities through the Rayleigh–Taylor instability (RTI) mechanism (Hysell et al., 1990; Singh *
Tel.: +55 12 39457148; fax: +55 12 39456990. E-mail address: [email protected]
et al., 1997) resulting in the generation of large scale plasma depletions or plasma bubbles. Bubbles are mostly aligned with the geomagnetic field flux tubes with plasma density decreases of up to three orders of magnitude. These depletions are generally produced over geomagnetic equator and they connect upwards through the F-layer peak to the topside ionosphere, reaching altitudes as high as 1200 km or more (Woodman and La Hoz, 1976). Though the RTI mechanism can explain the large scale irregularities observed in the bottom side F-region, it is known to be stable in the short wavelength range. But coherent radars installed at several locations all over the world have observed echoes from irregularities in the short wavelength range close to 3 m not only from the bottom side of the F-region but also from the top side. Phase velocities of these irregularities may range from a few m s1 to several hundreds of m s1 close to the ion acoustic speed. If no other wave generation process exists, these short wavelength waves must receive energy from the longer scales
0273-1177/$34.00 Ó 2007 COSPAR. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.asr.2007.07.039
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
through a cascade process of some sort. Considerable effort has gone into studying the possible generation of short wavelength waves by the larger primary waves. Chaturvedi and Kaw (1976) put forward a two-step theory of longer wavelength RTI modes directly coupling with kinetic collisional drift waves to explain the measured k2 spectra of the electron density irregularities. Both theory and experiments show that steep electron density gradients can develop in the medium as a result of the RTI process and most studies have involved gradient driven instabilities, under the general term of drift waves, as responsible for the shorter wavelength irregularities (Costa and Kelley, 1978a; Huba and Ossakow, 1981; LaBelle et al., 1986). The question of wave generation at wavelengths of a few meters, close to the ion gyro radius, is discussed by Costa and Kelley (1978b). Theoretical studies indicate that the relationship between the electron density perturbation and the fluctuations in the electric field associated with the short wavelength waves in the F-region is a linear one, though the exact relationship between them may depend on the plasma instability mechanism responsible for their generation (see Kelley and Heelis, 1989). Hence the power spectrum of the fluctuations in electric field and electron density should be similar except for the power contained in the spectra. Their spectral signatures may be used as a test for sorting out the electrodynamic processes that operate at any given altitude as a function of wave number of the waves. From simultaneous in situ measurements of electron density and electric field fluctuations Hysell et al. (1994) showed that irregularities in the scale size range of 100 m–2 km display a power law behaviour with spectral index n 2 that increased to 4.5 for wavelengths less than around 100 m when the F-layer is high. Measurement of the spectral parameters of the electron density and electric field fluctuations can, thereby, give us valuable information on the plasma instability mechanism responsible for the generation of these irregularities. In situ measurements of the height variation of the ionospheric electron density and electric field variations were made with two different types of rocket-borne electron density probes and an electric field double probe from the equatorial region in Brazil to study the relationship between the spectral features in the electron density and electric field variations in the F-region dominated by large scale plasma bubbles. The results obtained from this comparative study are presented and discussed here. 2. Experiment and flight details A Brazilian SONDA III rocket carrying three plasma experiments in addition to other airglow experiments was launched on 18 December 1995 at 2117 hrs (LT) from the equatorial rocket launching station, Alcaˆntara (2,31°S; 44,4°W) in Brazil. The principal objective of the plasma experiments was to measure the electric field, the electron density, and the spectral distribution of plasma irregulari-
1195
ties associated with plasma bubbles. The payload consisted of the following experiments: electric field double probe (EFP), Langmuir probe (LP) and high frequency capacitance probe (HFC). The experiment and data analysis details are described in Muralikrishna et al. (2003) and Muralikrishna (2006). The basic principle of operation and the details of the electronic subsystems of the LP and HFC experiments are given in Muralikrishna and Abdu (1991). Several ground equipments were operated during the launch campaign with the specific objective of knowing the ionospheric conditions at the time of launch and thereby to launch the rocket into an F-region prone to the presence of plasma bubbles. Ionograms were obtained at Sa˜o Luis, a station close to the launch site and at Cachoeira Paulista, a station outside the low latitude belt, but more or less at the same geomagnetic longitude as Sa˜o Luis. Appearance of spread-F traces in the ionograms recorded at Sa˜o Luis and the consequent appearance of spread-F traces at Cachoeira Paulista was considered to be a strong indication for the presence of plasma bubbles in the ionosphere over that launch station. The rocket was launched under ionospheric conditions favourable for the presence of bubbles. The rocket reached an apogee altitude of 557 km and covered a horizontal range of 589 km. The mean azimuth angle of the plane of the trajectory was about 61.2°. The rocket in fact passed through several medium scale plasma bubbles mainly during the downleg. The upleg electron density profile showed the presence of a very clearly defined base for the F-region around 300 km, while the downleg profile showed the presence of a wide spectrum of electric field and electron density irregularities in this height region as well as in the upper F-region. An FFT algorithm was then used to estimate the spectral distribution of the electric field and electron density fluctuations. It should be noted here that the LP and EF sensors were mounted away from the rocket body in a plane perpendicular to the rocket spin axis. The electric field measured thus represents only the component of the field perpendicular to the spin axis of the rocket. The rocket had a spin rate of less than 3 per second, which in fact, can be clearly seen in the total electric field data, but the amplitude of this modulation is reduced very much in the electric field fluctuation data that is separated through a high pass filter. The rocket spin and precession were monitored through onboard magnetometers and their effects on the electron density and electric field data were seen to be rather low and easily identifiable in the spectra. The LP sensor was operated in a swept voltage mode with a sweep cycle duration of about 2.5 s. During each operation cycle the sweep voltage applied to the sensor increased linearly from 1 V to +2.5 V in about 1.5 s and remained steady at +2.5 V for about a second. The fluctuating components of the LP and EF data corresponding to the time intervals when the LP sensor is in saturation current mode (at fixed +2.5 V bias) are chosen for the spectral analysis, while the LP data
1196
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
collected during the voltage sweep was used to estimate the electron temperature.
HFC Electron Density (m-3) 9
10 600
3. Results and discussion 500
3.1. Electron density and electric field fluctuations
HFC Electron Density (m-3) 9
10 600
500
1010
1011
1013
Alcantara, Brazil 18 December 1995 2117 LT Upleg
400
Height (km)
1012
D
HFC
LP
C B 300
A
200
100
0 8 10
10
9
10
10
10
11
10
12
-3
LP Electron Density (m ) Fig. 1. Upleg electron density profiles estimated from the Langmuir probe (LP) and the high frequency capacitance (HFC) probe data. Note that the x-axis (top) for the HFC profile is shifted by one order to the right.
11
12
10
13
10
10
Alcantara, Brazil 18 December 1995 2117 LT Downleg HFC C
400
Height (km)
The electron density profiles estimated from the LP and HFC experiments for the rocket upleg and downleg are shown in Figs. 1 and 2, respectively. The general features of the electron density profiles measured by the HFC experiment are similar to those measured by the LP, but the absolute values of the electron densities at different height regions showed some differences. Muralikrishna and Abdu (1991) tried to explain similar differences observed in one of their earlier in situ measurements as due to certain inherent problems associated with the two techniques of measurement. However, what is more important in the present studies is the relative variation of the electron density with altitude. As can be seen from Figs. 1 and 2 the upleg electron density profile shows the presence of a rather steep F-region base, free of any large scale electron density depletions or bubbles, while the downleg profile shows the presence of a large number of plasma bubbles. The downleg profile does not show a sharp F-region base. Power spectra of electron density and electric field fluctuations observed during the upleg and downleg of the rocket at a few selected height regions are compared in Figs. 3–6. Though normally the spectra are plotted on a
10
10
LP
D
B 300
A
200
100
0 8 10
109
1010
1011
1012
-3
LP Electron Density (m ) Fig. 2. Downleg electron density profiles estimated from the LP and HFC probe data. Note the shift in the x-axis for the HFC profile to the right.
log–log scale, in the present case they are plotted with linear scale along the x-axis (wave number) and log scale along the y-axis (spectral power) with the main purpose of highlighting the structure of the spectral peaks. The two panels on the left represent the upleg spectra and the two panels on the right the downleg spectra corresponding to more or less the same altitude region. The data-sampling rate is 1250 s1 that corresponds to an upper frequency limit of 625 Hz. Assuming that at the rocket speed of 2– 3 km s1, the ambient plasma irregularities can be considered practically stationary, the temporal variations observed by the experiments can be converted into spatial variations. The frequency of the LP current fluctuation or the E-probe potential fluctuation can be converted to wavelength k or wave number k (k = 2p/k) of the waves knowing the rocket velocity. The spectra of electron density and electric field fluctuations are estimated from data blocks each containing 1024 points collected in about 0.8 s. Shown along the x-axis is the wave number k that varies from 0 to more than 2200 km1, the upper limit being a function of the rocket velocity. Along the y-axis the spectral power is shown in arbitrary (relative) units. Spatial fluctuations in both electron density and electric field can be seen in both Figs. 3 and 4. The electron density and E-field spectra in Fig. 3 correspond to a mean altitude of about 295 km. This height region is identified by the letter A in the electron density height profiles shown in Figs. 1 and 2. The plasma regions represented by the two panels correspond to the base of the F-region and have a horizontal separation of more than 400 km. Remembering that the y-axis representing the spectral power have logarithmic scales, one can see that the spectral peaks have higher
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
(δE)2Hz-1
10
1
Electric Field
Electric Field
10
1
10
-1
10
-1
10
-3
10
-3
10
-5
10
-5
10
1
10
-1
10
-3
10
-5
10
(δne)2Hz-1
1197
Electron Density
1
10
-1
10
-3
10
-5
Electron Density
0
400
800
1200
1600 0
Wave Number k (km-1)
400
800
1200
1600
Wave Number k (km-1)
Fig. 3. Power spectra of electric field (top) and electron density (bottom) observed at the mean height of about 295 km during upleg (left) and downleg (right) of the rocket.
(δE)2Hz-1
10 1
10
1
10
-1
10
-3
10
-5
10
5
10 3
10
3
1
10
1
10
-1
10
-1
10
-3
Electric Field
Electric Field
10-5
(δne)2Hz-1
10
10
Electron Density
5
10-1
0
500
1000
1500
2000 0
Wave Number k (km-1)
Electron Density
400
800
1200
1600
Wave Number k (km-1)
Fig. 4. Power spectra of electric field (top) and electron density (bottom) observed at the mean height of about 333 km during upleg (left) and downleg (right) of the rocket.
powers in the right panels than in the left panels. While the E-spectrum observed during the upleg is practically devoid of any spectral peak, one can see the existence of two rather very low power peaks in the upleg ne spectrum corresponding to the mean wave numbers of 1470 km1 and 1655 km1. This probably is due to the fact that the RTI mechanism has not yet started or is only at its very initial state of development. During the downleg two common spectral peaks are seen in the E and ne spectra more or less at the same wave numbers and with the same relative power as seen during upleg. But the power in the spectral peaks during downleg is higher by orders of magnitude.
The downleg ne spectrum shows at least three additional spectral peaks corresponding to the wave numbers of about 575 km1, 660 km1 and 740 km1 that are not seen in the E-spectrum. From the fact that while the E-spectrum is completely devoid of any spectral peaks the ne spectrum exhibits a few spectral peaks and the observation of difference of orders of magnitude in spectral powers during upleg and downleg one can rule out the possibility of the sharp spectral peaks being caused by any mutual interference between the EF and LP sensors or by interference with other experiments on board. The existence of wave structures in both ne and E at the same wave numbers is
1198
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
10
1
1
Electric Field
Electric Field
10
-1
10
-1
10
-3
10
-3
10
-5
10
-5
10
5
Electron Density
Electron Density
10
5
10
3
10
3
10
1
10
1
10
-1
δ
δ
10
10
-1
0
500
1000
1500
2000 0
Wave Number k (km-1)
500
1000
1500
2000
Wave Number k (km-1)
Fig. 5. Power spectra of electric field (top) and electron density (bottom) observed at the mean height of about 376 km during upleg (left) and downleg (right) of the rocket.
1
10
δ
-3
10
-5
10
Electron Density
5
Electron Density
3
δ
10
1
10
-1
10
0
500
1000
1500
2000
Wave Number k (km-1)
1
10
-1
10
-3
10
-5
10
5
10
3
10
1
10
-1
Electric Field
Electric Field -1
10
10
10
0
500
1000
1500
2000
Wave Number k (km-1)
Fig. 6. Power spectra of electric field (top) and electron density (bottom) observed at the mean height of about 404 km during upleg (left) and downleg (right) of the rocket.
indicative of propagating waves. The right panel also exhibits spectral peaks seen only in the ne spectrum, not associated with corresponding peaks in the E-spectrum. These peaks are not exhibited by the left panel spectrum and therefore do not seem to represent propagating waves, but only electron density irregularities. Fig. 4 corresponds to the mean height region of about 335 km, indicated by letter B in the ne profiles shown in Figs. 1 and 2. The upleg spectra on the left panels, still exhibit low power spectral peaks in both ne and E, while the downleg spectra exhibit higher power, especially in the ne spectrum. The wave numbers corresponding to these spectral peaks in E and ne are almost same in the upleg and
downleg spectra, while the higher peak in the upleg spectrum seems to appear at a wave number slightly smaller than that in the downleg spectrum. This again indicates that they are associated with propagating waves. It should be noted here that the horizontal separation of the height regions represented by the left and right panels in this case is about 380 km. The slight difference in the wave numbers of the spectral peaks during upleg and downleg may probably be caused by the upward velocity of these drift waves in association with the upward rocket velocity and/or by the changes occurring in the rocket trajectory as discussed later. The estimated wavelength corresponding to the spectral peak at wave number 1470 km1 is about 4.3 m.
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
3.2. Electron temperature profile Electron temperature was estimated from the slope of the current–voltage characteristics of the Langmuir probe. In estimating Te from the slope of the I–V characteristic curve one generally assumes that the positive ion current is much less than the electron current. The exact shape of the current–voltage characteristic of a Langmuir probe is decided by the plasma temperature and the shape and size of the plasma sheath that forms surrounding the sensor surface. By solving Poisson equation for the distribution of potential in a plasma sheath, one can obtain the following approximate relation for the current of particles moving in a retarding field as ene v eV Ip ¼ a: exp ; ð1Þ 4 kT e
Thus measuring the probe current collected at two different probe potentials one can, in principle, estimate both the plasma number density and the temperature using the above relations. The LP data corresponding to the time intervals when the LP sensor is in the sweep potential mode (1 V to +2.5 V bias) only are chosen for estimating the electron temperature reported here. Muralikrishna (2006) presented the LP current–voltage characteristics for selected height regions and estimated the electron temperature profiles from the LP data obtained during the flight reported here and found that the electron temperature showed an abnormal increase, probably just before the onset of the RTI mechanism responsible for the bubbles observed during the rocket downleg. The upleg and downleg Te profiles are shown in Figs. 7 and 8 for comparison with the spectral data presented above. Comparison of Figs. 3–6 with Figs. 7 and 8 brings out some interesting observations. While Fig. 3 corresponding to the height region of about 295 km does not show large power spectral peaks in neither ne nor E upleg spectra, Fig. 7 shows abnormally large electron temperature in the same height region. In the same height region more than 400 km away horizontally, the downleg spectra show the presence of spectral peaks with significant power, but, as can be seen from Fig. 8, the abnormal electron temperature seen in the base of the F-layer has disappeared. This probably can be explained as due to the conversion of the electron thermal energy into wave energy during the operation of the RTI mechanism. Another observation that can be made by comparing the corresponding spectral peaks in ne and E given in Figs. 3–6 600 ALCANTARA, BRAZIL 18 December 1995 2117 LT LP Upleg
500
D
400
C
Height (km)
The ne and E spectra corresponding to the height regions of about 375 km and 404 km (indicated by letters C and D in the ne profiles of Figs. 1 and 2), respectively, are shown in Figs. 5 and 6. It can be seen from these figures that the propagating waves continue to exist in these height regions. The horizontal separation of the height regions shown in Fig. 5 is about 330 km and that for the height regions shown in Fig. 6 is about 304 km. The power of the spectral peaks observed during the upleg and downleg in these height regions is almost same except that the corresponding wave numbers are slightly smaller in the upleg spectra as seen in Fig. 4. But the electron density spectra corresponding to these height regions show several additional spectral peaks at lower wave numbers (at higher wavelengths), mainly in the downleg spectra, few of which are accompanied by spectral peaks in the E spectra. These probably correspond to plasma irregularities generated by other instability mechanisms. One should remember here that the electron density profile observed during the upleg of the rocket does not indicate the presence of neither plasma bubbles nor large or medium scale plasma irregularities. However, the downleg profile indicates the presence of a large number of plasma bubbles and associated large, medium and small-scale irregularities. The spatial and temporal separation of the upleg and downleg trajectories of the rocket being small, the rather complete absence of plasma bubbles in the upleg profile and the presence of a large number of bubbles in the downleg profile indicate that the bubbles could be recently generated and be in their developing phase.
1199
B A
300
200
Ne
Te
100
where a is the surface area of the probe, e is the electronic charge, ne is the electron density. This relation is valid regardless of the shape of the probe and consequently ln Ip has a linear dependence on the potential V applied to the probe (in a given potential range). From this linear relation one can get the electron temperature as, 5040 Te ¼ dðln I p Þ=dV
ð2Þ
IRI Model
0 9 10
10
10
10
11
10 -3
Electron Density(m )
12
0
1000
2000
3000
Electron Temperature (οK)
Fig. 7. Night time electron temperature height variation observed on 18 December 1995 during the rocket upleg. Also shown in the figure are the upleg electron density profile of Fig. 1 and the IRI model temperature profile.
1200
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201 600 ALCANTARA, BRAZIL 18 December 1995 2117 LT LP Downleg
500
D
Height (km)
400 C B
300
A
Ne
200
Te
100
IRI Model
0 9 10
10
10
10
11
Electron Density(m-3)
0
400
800
1200
Electron Temperature (οK)
Fig. 8. Night time electron temperature height variation observed on 18 December 1995 during the rocket downleg. Also shown in the figure are the downleg electron density profile of Fig. 2 and the IRI model temperature profile.
is that the relative amplitudes of the spectral peaks are more or less same. Also one can see that the spectral power decreases with increasing wave number or with decreasing wavelength thereby showing negative spectral indices as expected from the existing theories. The spectral peaks observed in electron density and electric field spectra could represent waves generated during the initial phase of the development of plasma bubbles or during the operation of other instability mechanisms responsible for the drift waves. The appearance of these spectral peaks in both the upleg and the downleg of the rocket trajectory indicates that these are propagating waves. Since the E-field measurements are made on board a moving space vehicle, it is difficult to estimate the actual wave numbers of these waves without knowing their phase velocities (Fredericks and Coroniti, 1976; Temerin, 1979). However, one can infer that the real wave numbers are close to the ones observed, since the ne spectra also exhibit spectral peaks at the same wave numbers and these are not affected by the velocity of the moving frame of reference. Electrostatic waves in the Earth’s ionosphere exist over a large range of wavelengths, from centimetres to hundreds of kilometres, and over a large range of phase velocities, from quasi-static to speeds of hundreds of kilometres per second. Because of the motion of the spacecraft relative to ambient plasma, even quasi-static waves appear as temporal fluctuations in the spacecraft frame. For electrostatic waves with frequencies below the ion plasma frequency or the lower hybrid frequency, whichever is lower, both ions and electrons participate in the response of the medium to electrostatic waves. This implies that there are density and velocity fluctuations associated with the wave.
If one assumes that the spectral peaks are caused by drift waves propagating upwards, the wavelength of fluctuations as measured by a rocket moving upward will appear longer (wave number smaller) and that as measured by a rocket moving downward will appear shorter (wave number larger). This is exactly what is seen in the spectra presented in Figs. 3–6. The spectral peaks appear at slightly lower wave numbers during the upleg than during the downleg. In addition to this, changes in the trajectory of the rocket will also cause changes in the measured wave number of the waves. As the rocket moves to higher altitudes its trajectory becomes more and more horizontal and for a given vertical wavelength of the waves, the measured wavelength becomes longer and longer or the wave number becomes smaller and smaller. The observed shift in the spectral peaks between the upleg and downleg may be partly due to this too. It should be noted here that the spectral peaks seen in the spectra presented here will practically disappear if one presents them in a conventional log–log plot. For example, a close look at the electron density spectra presented in Kelley et al. (1982) clearly indicates the presence of a large number of spectral peaks near the high wave number end. They are not dominantly seen only because of the logarithmic scale used. The strongest spectral peaks observed in the present case are also at the high wave number end of the spectra and are seen always practically at the same wave numbers. The important features of the power spectra of electron density and electric field fluctuations presented in Figs. 3–6 and the electron temperature profiles shown in Figs. 7 and 8 can be summarised as follows. At most of the heights the spectral data show the presence of sharp spectral peaks in both ne and E probably caused by drift waves. The origin of these waves is unknown. One of the possibilities is that these are generated during the development phase of plasma bubbles. The large density gradients associated with the walls of the bubbles (in both east–west and vertical directions) in conjunction with the east–west or vertical E-field may also trigger some plasma instability process responsible for the drift waves. Alternatively the large scale bubbles, in course of time, may break into smaller scale waves (cascade process) losing their energy. This may finally lead to a steady state distribution (without spectral peaks) of energy in a large range of wavelengths. This final state maybe characterised by one or more characteristic spectral indices for the irregularities reported in the literature. The spectral peaks observed seem to represent propagating waves and are seen with more or less same relative amplitudes in the electron density and electric field. At those height regions where no spectral peaks are observed neither in ne nor in E-field, and their fluctuation amplitudes are relatively high, it seems that the
P. Muralikrishna / Advances in Space Research 42 (2008) 1194–1201
situation has already reached a steady state, wherein the spectral features are those that are expected from the existing theories. The abnormally high electron temperature observed below the base of the F-region seems to be a precursor to the development of plasma bubbles. Once the bubbles develop and start rising up the electron temperature falls down. The large kinetic energy of electrons may be transported to other height regions through the rising bubbles or propagating waves. The presence of the abnormally large electron temperatures at the base of the F-region before the development of bubbles, and its disappearance in the same height region (horizontally separated by a few hundred kilometres) during the rocket downleg, also supports this hypothesis.
4. Conclusions
Bubble regions are associated with a wide spectrum of both electron density and electric field fluctuations probable caused by drift waves. The absence of sharp spectral peaks in ne and/or E-field probably indicates that the cascading process of the generation irregularities has already reached the final state or that the physical conditions in the particular height region are not favourable for the operation of any plasma instability mechanism. Abnormally large electron temperatures observed at the base of the F-region seem to be a precursor to the development of plasma bubbles.
Acknowledgements The author is grateful to the Directors of IAE/CTA and CLA, Alcantara for providing the rockets and the launch facilities respectively and to the staff of IAE and CLA for their help during the pre-launch tests of the experiments, and during the launching of the rockets. Sincere thanks to Sinval Domingos, Agnaldo Eras, and Narli Baesso Lisboa for their technical help in the development testing and integration of the experiments. The work reported here was partially supported by FINEP under contract FINEP-537/CT, and by CNPq under process 300253/89-3/
1201
GM/FV. The authors are thankful to the referees for their very useful suggestions and comments. References Chaturvedi, P.K., Kaw, P. An interpretation for the power spectrum of spread-F irregularities. J. Geophys. Res. 81, 3257–3260, 1976. Costa, E., Kelley, M.C. On the role of steepened structures and drift waves in equatorial spread F. J. Geophys. Res. 83, 4359, 1978a. Costa, E., Kelley, M.C. Linear theory for the collisionless drift wave instability with wavelengths near the ion gyro radius. J. Geophys. Res. 83, 4375, 1978b. Fredericks, R.W., Coroniti, F.V. Ambiguities in the deduction of rest frame fluctuation spectrum from spectrums computed in moving frames. J. Geophys. Res. 81, 5591–5595, 1976. Haerendal, G., Theory of equatorial spread-F, Report of Max Planck Institu¨t fur Physik und Astrophysik, Garching, West Germany, 1974. Huba, J.D., Ossakow, S.L. Lower hybrid drift waves in equatorial spread F. J. Geophys. Res. 86, 829, 1981. Hudson, M.K., Kennel, C.F., Kaw, P.K. Two step drift mode theory of equatorial spread-F. Trans. Am. Geophys. Soc. 54, 1147, 1973. Hysell, D., Kelley, M.C., Swartz, W.E., Pfaff, R.F., Swenson, C.M. Seeding and layering of equatorial spread-F by gravity waves. J. Geophys. Res. 95, 17253–17260, 1990. Hysell, D., Kelley, M.C., Swartz, W.E., Woodman, R.F. Steepened structures in equatorial spread-F, 1. New observations. J. Geophys. Res. 99, 8827–8840, 1994. Kelley, M.C., Heelis, R.A. The Earth’s Ionosphere: Plasma Physics and Electrodynamics. Academic Press, 1989. Kelley, M.C., Pfaff, R., Baker, K.D., Ulwick, J.C., Livingston, R., Rino, C., Tsunoda, R. Simultaneous rocket probe and radar measurements of equatorial spread F – transitional and short wavelength results. J. Geophys. Res. 87, 1575–1588, 1982. LaBelle, J., Kelley, M.C., Seyler, C.E. An analysis of the role of drift waves in equatorial spread F. J. Geophys. Res. 91, 5513–5525, 1986. Muralikrishna, P., Abdu, M.A. In-situ measurement of ionospheric electron density by two different techniques – a comparison. J. Atmos. Terr. Phys. 53, 787–793, 1991. Muralikrishna, P., Vieira, L.P., Abdu, M.A. Electron density and electric field fluctuations associated with developing plasma bubbles. J. Atmos. Solar Terr. Phys. 65, 1315–1327, 2003. Muralikrishna, P. Electron temperature variations in developing plasma bubbles – rocket observations from Brazil. Adv. Space Res. 37, 903– 909, 2006. Reid, G.C. Small-scale irregularities in the ionosphere. J. Geophys. Res. 73, 1627–1640, 1968. Singh, S., Bhamgboye, D.K., McClure, J.P., Johnson, F.S. Morphology of equatorial plasma bubbles. J. Geophys. Res. 102, 20019–20029, 1997. Sudan, R.N., Akinrimisi, J., Farley, D.T. Generation of small-scale irregularities in the equatorial electrojet. J. Geophys. Res. 78, 240, 1973. Temerin, M. Doppler shift effects on double-probe measured electric field power spectra. J. Geophys. Res. 84, 5929–5934, 1979. Woodman, R.F., La Hoz, C. Radar observations of equatorial F-region irregularities. J. Geophys. Res. 81, 5447–5466, 1976.