Nonlinear effects associated with the finite frequency pump kinetic Alfvén wave and turbulent spectrum in solar wind

The paper contains a numerical simulation of the nonlinear coupling between the kinetic Alfvén wave and the ion acoustic wave for an intermediate β-plasma (m _e/m _i?β?1). For this study, we have introduced the nonlinear ponderomotive force (due to the finite frequency (ω ₀<ω _ci) kinetic Alfvén wave) in the derivation of the ion acoustic wave. The main aim of the present paper is to study the nonlinear effects associated with the different driving finite frequencies (ω ₀<ω _ci) of the pump kinetic Alfvén wave on the formation of localized structures and a turbulent spectrum applicable to the solar wind around 1 AU. As a result, we found that the different driving frequencies of the pump kinetic Alfvén wave affect the formation of the localized structures. We have also studied the turbulent scaling which follows (～k ^?3.6) for ω ₀/ω _ci≈0.2, (～k ^?3.4) for ω ₀/ω _ci≈0.3 and (～k ^?3.2) for ω ₀/ω _ci≈0.4, at small scales. Further, we have also found that different finite driving frequencies of the pump kinetic Alfvén wave affect the turbulence scaling at small scales, which may affect the heating of the plasma particles in solar wind. The present study is correlated with the observation made by the Cluster spacecraft for the solar wind around 1 AU.     

Variability in the power spectrum of solar five-minute oscillations

Two-dimensional power spectra of solar five-minute oscillations display prominent ridge structures in (k, ω) space, where k is the horizontal wavenumber and ω is the temporal frequency. The positions of these ridges in k and ω can be used to probe temperature and velocity structures in the subphotosphere. We have been carrying out a continuing program of observations of five-minute oscillations with the diode array instrument on the vacuum tower telescope at Sacramento Peak Observatory (SPO). We have sought to establish whether power spectra taken on separate days show shifts in ridge locations; these may arise from different velocity and temperature patterns having been brought into our sampling region by solar rotation. Power spectra have been obtained for six days of observations of Doppler velocities using the Mgi λ5173 and Fei λ5434 spectral lines. Each data set covers 8 to 11 hr in time and samples a region 256″ × 1024″ in spatial extent, with a spatial resolution of 2″ and temporal sampling of 65 s. We have detected shifts in ridge locations between certain data sets which are statistically significant. The character of these displacements when analyzed in terms of eastward and westward propagating waves implies that changes have occurred in both temperature and horizontal velocity fields underlying our observing window. We estimate the magnitude of the velocity changes to be on the order of 100 m s^-1; we may be detecting the effects of large-scale convection akin to giant cells.     

Parametric excitation of alfvén wave by magnetosonic wave with oblique propagation

A parametric excitation of the Alfvén wave (k^A, ω_a) by the magnetosonic wave (K₁^fs, ω_1fs), which propagates obliquely to the static magnetic field, has been analyzed. The theoretical model for a one-fluid with uniform, unbounded, ideally conducting and compressible plasma is employed. The resonance conditions are chosen such as, $\text{k}{}_1{}^{\mathrm{fs}}\text{=}\text{k}{}_1{}^{\mathrm{fs}}\text{+}\text{k}{}^{\mathrm{A}}$ and ω_1fs ? ω_A = δ ～ ω_2fs. The p wave is assumed to be strong enough, so that the pump wave is given as a constant. In both the case of the standing and the propagating pump the growth rates of the excited waves depend on not only the pump power but also the β-ratio. In the standing pump the threshold pump intensity of the oscillating instability is zero at the perfect matching. It is found that we can obtain a larger growth rate of the parametric excitation of Alfvén wave by the fast magnetosonic pump wave for θ_1f ～ 70–80° and the occurrence regions of parametric excitations are localized at the resonance point in the magnetosphere (β ～ m_e/m_i). It is concluded that the parametric instability of Pc3 range HM-waves is the more possible theory than the linear resonance theory.     

On the influence of nonlinearities on the eigenfrequencies of five-minute oscillations of the Sun

Fitting the results of linear normal-mode analysis of the solar five-minute oscillations to the observed k - ω diagram selects a class of models of the Sun's envelope. It is a property of all the models in this class that their convection zones are too deep to permit substantial transmission of internal g modes of degree 20 or more. This is in apparent conflict with Hill and Caudell's (1979) claim to have detected such modes in the photosphere. A proposal to resolve the conflict was made by Rosenwald and Hill (1980). They pointed out that despite the impressive agreement between linearized theory and observation, nonlinear phenomena in the solar atmosphere might influence the eigenfrequencies considerably. In particular, they suggested that a correct nonlinear analysis could predict a shallow convection zone. This paper is an enquiry into whether their hypothesis is plausible. We construct k - ω diagrams assuming that the modes suffer local nonlinear distortions in the atmosphere that are insensitive to the amplitude of oscillation over the range of amplitudes that are observed. The effect of the nonlinearities on the eigenfrequencies is parameterized in a simple way. Taking a class of simple analytical models of the Sun's envelope, we compute the linear eigenfrequencies of one model and show that no other model can be found whose nonlinear eigenfrequencies agree with them. We show also that the nonlinear eigenfrequencies of a particular solar model with a shallow convective zone, computed with more realistic physics, cannot be made to agree with observation. We conclude, therefore, that the hypothesis of Rosenwald and Hill is unlikely to be correct.     

Analysis of the Helioseismic Power-Spectrum Diagram of a Sunspot

The continuous high spatial resolution Doppler observation of the Sun by the Solar Dynamics Observatory/Helioseismic and Magnetic Imager allows us to compute a helioseismic k–ω power-spectrum diagram using only oscillations inside a sunspot. Individual modal ridges can be clearly seen with reduced power in the k–ω diagram that is constructed from a 40-hour observation of a stable and round sunspot. Comparing this with the k–ω diagram obtained from a quiet-Sun region, one sees that inside the sunspot the f-mode ridge is more reduced in power than the p-mode ridges, especially at high wavenumbers. The p-mode ridges all shift toward lower wavenumber (or higher frequency) for a given frequency (or wavenumber), implying an increase of phase velocity beneath the sunspot. This is probably because the acoustic waves travel across the inclined magnetic field of the sunspot penumbra. Line-profile asymmetries exhibited in the p-mode ridges are more significant in the sunspot than in the quiet Sun. Convection inside the sunspot is also highly suppressed, and its characteristic spatial scale is substantially larger than the typical convection scale of the quiet Sun. These observational facts demand a better understanding of magnetoconvection and interactions of helioseismic waves with magnetic field.     

The stability of differential rotation with non-axisymmetric perturbations

A plane-wave analysis on a simplified scheme based on the Boussinesq approximation and shallow convection is used to establish the necessary conditions for stability of a differentiallyrotating, compressible flow between two coaxial cylinders subject to non-axisymmetric perturbations. To test the adequateness of this simplification, the sufficient conditions for stability are again established which agree with those obtained by a normal-mode analysis on an exact scheme in an earlier paper by the author. This model is applicable to stellar models with rotation Ω=Ω(ω), where ω is the radial distance from the axis of rotation (thez-axis). A necessary condition for stability, in the non-dissipative case, is found to be that $$\frac{1}{\varrho }G_\varpi S_\varpi + \frac{{k_z^2 }}{M}\Phi - \frac{1}{4}\frac{{m^2 }}{M}\left( {D\Omega } \right)^2 \geqslant 0$$ everywhere. Here,m andk _z are the wave numbers in the ø- andz-direction, \(M \equiv k_z^2 + m^2 /\varpi ^2 ,D \equiv d/d\varpi ,G_\varpi \equiv - \varrho ^{ - 1} Dp,\varrho \) the density,p the pressure,S _ω and Φ the Schwarzschild and the Rayleigh discriminants defined as \(S_\varpi \equiv \left( {\gamma p/\varrho } \right)^{ - 2} Dp - D\varrho and \Phi \equiv ^{ - 3} d\left( {\varpi ^4 \Omega ^2 } \right)/d\varpi \) respectively, γ the ratio of specific heats. This condition is also a sufficient one. Some conjectures regarding the stabilizing influence of uniform rotation and the destabilizing influence of differential rotation are also verified. The most striking instability mechanism introduced by shear forces and by radiative dissipation is the excitation of the stable motion of small oscillations into that of oscillations with growing amplitude, i. e., overstability. In the case of radiative dissipation and axisymmetric perturbations, the Goldreich-Schubert criterion is only necessary but not sufficient for stability. Instability sets in as soon as the Schwarzschild criterion is violated. When the perturbations are non-axisymmetric, instability always sets in as overstability as long as rotation is differential. This may explain the convective turbulence in the upper atmosphere where the radiation is active.     

Quiet auroral arcs: Ionosphere effect of magnetospheric convection stratification

A detailed study of the mechanism of electromagnetic stratification of the large-scale stationary magnetospheric convection due to a friction of the convective flow in the ionosphere layer was performed. Magnetosphere-ionosphere interaction was taken into account by means of the effective boundary conditions on the ionosphere top and bottom boundaries including the actual height profile of charge particles velocity in the ionosphere. It has been shown that the magnetospheric convection is stratified into small-scale current sheets which are respective in the linear approximation to an oblique Alfvén wave. The dispersion equation was deduced for the Alfvén mode and its solution obtained determining the space-time scales and the increment of instability. The maximum increment is realized for the disturbances stretched along the convection velocity that is correspondent to the actual orientation of the auroral arcs. In the conditions of rapid growth of Alfvén velocity above the maximum of the ionosphere F layer, it was shown that small-scale disturbances with the transverse scales l_⊥ ? 1 km are localized at the altitudes up to several thousand kilometers whereas the large-scale stratification penetrate into the equatorial plane of the magnetosphere. A mechanism is proposed to intensify the parallel electric field acting at that stratification stage when the field-aligned currents in the Alfvén wave are sufficient to form abnormal resistance along geomagnetic lines of force.     

Jovian seismology

Standing acoustic waves, with periods between about 4.5 and 9 min, may be trapped in a wave duct beneath Jupiter's tropopause. Detection of these oscillations by observations of Doppler shifting of infrared and ultraviolet absorption lines would offer a new important method for probing the giant planet's deep atmosphere and interior. Information would be revealed on Jupiter's thermal and density structure and the depth to which its zonal winds penetrate. Standing oscillations in the molecular hydrogen envelope are modeled and their theoretical eigenfrequencies are presented as they might appear in actual data analysis. Several forcing functions for wave generation are considered. These include coupling with turbulent and convective motions, thermal overstability due to radiative transfer, effects of wave propagation in a saturated atmosphere, and consequences of ortho- to parahydrogen conversion. Altjough the forcing mechanisms couple well with the acoustic waves, allowing for possible maintenance of the oscillations, the contribution they make to velocity amplitudes is very small, between 1.0 and 0.1 m sec⁻¹. This implies that the Doppler shifting caused by the waves may be unresolvable except, perhaps, by methods of superposing time records of oscillations to enhance acoustic signals and diminish random noise.     

On pulsar-driven isothermal blast-wave models of supernova remanants

We investigate the ‘equilibrium’ and stability of spherically-symmetric self-similar isothermal blast waves with a continuous post-shock flow velocity expanding into medium whose density varies asr ^?ω ahead of the blast wave, and which are powered by a central source (a pulsar) whose power output varies with time ast ^ω?3. We show that:

1. for ω<0, no physically acceptable self-similar solution exists;
2. for ω>3, no solution exists since the mass swept up by the blast wave is infinite;
3. ? must exceed zero in order that the blast wave expand with time, but ?<2 in order that the central source injects a finite total energy into the blast wave;
4. for 3>ω_min(?)>ω>ω_max(?)>0, where $$\begin{gathered} \omega _{\min } (\varphi ){\text{ }} = {\text{ }}2[5{\text{ }} - {\text{ }}\varphi {\text{ }} + {\text{ }}(10{\text{ }} + {\text{ 4}}\varphi {\text{ }} - {\text{ 2}}\varphi ^2 )^{1/2} ]^2 [2{\text{ }} + {\text{ (10 }} + {\text{ 4}}\varphi {\text{ }} - {\text{ 2}}\varphi ^2 {\text{)}}^{{\text{1/2}}} ]^{ - 2} , \hfill \\ \omega _{\max } (\varphi ){\text{ }} = {\text{ }}2[5{\text{ }} - {\text{ }}\varphi {\text{ }} - {\text{ }}(10{\text{ }} + {\text{ 4}}\varphi {\text{ }} - {\text{ 2}}\varphi ^2 )^{1/2} ]^2 [2{\text{ }} - {\text{ (10 }} + {\text{ 4}}\varphi {\text{ }} - {\text{ 2}}\varphi ^2 {\text{)}}^{{\text{1/2}}} ]^{ - 2} , \hfill \\ \end{gathered} $$ two critical points exist in the flow velocity versus position plane. The physically acceptable solution must pass through the origin with zero flow speed and through the blast wave. It must also pass throughboth critical points if \(\varphi > \tfrac{5}{3}\) , while if \(\varphi< \tfrac{5}{3}\) it must by-pass both critical points. It is shown that such a solution exists but a proper connection at the lower critical point (for ?>5/3) (through whichall solutions pass with thesame slope) has not been established;
5. for 3>ω>ω_min(?) it is shown that the two critical points of (iv) disappear. However a new pair of critical points form. The physically acceptable solution passing with zero flow velocity through the origin and also passing through the blast wave mustby-pass both of the new critical points. It is shown that the solution does indeed do so;
6. for 3>ω_min(?)>ω_max(?)>ω it is shown that the dependence of the self-similar solution on either ω or ? is non-analytic and therefore, inferences drawn from any solutions obtained in ω>ω_max(?) (where the dependence of the solutionis analytic on ω and ?) are not valid when carried over into the domain 3>ω_min(?)>ω_max(?)>ω;
7. all of the physically acceptable self-similar solutions obtained in 3>ω>0 are unstable to short wavelength, small amplitude but nonself-similar radial velocity perturbations near the origin, with a growth which is a power law in time;
8. the physical self-similar solutions are globally unstable in a fully nonlinear sense to radial time-dependent flow patterns. In the limit of long times, the nonlinear growth is a power law in time for 5<ω+2?, logarithmic in time for 5>ω+2?, and the square of the logarithm in time for 5=ω+2?.

The results of (vii) and (viii) imply that the memory of the system to initial and boundary values does not decay as time progresses and so the system does not tend to a self-similar form. These results strongly suggest that the evolution of supernova remnants is not according to the self-similar form.     

On the excitation of oscillations of the Sun (numerical models)

Numerical solutions of the general time-dependent gas-dynamical equations in linear adiabatic approximation are given for initial conditions imitating: (a) a central perturbation, (b) a boundary perturbation (in the convective envelope), and (c) a ‘shrinking’ of the Sun as a whole. For a variety of models of the Sun it is found that at the surface the radial component v _r of velocity is much greater than the tangential component v _t , and that the period T of stationary oscillations does not exceed 131^m. The appearance at the surface of a g mode with period 160^m is found to be improbable. With the initial conditions adopted, a propagating wave is produced which is reflected successively from the centre to the periphery and back, producing 5-min oscillations at the surface of the Sun. Expansion of this wave into separate modes leads to a power spectrum qualitatively similar to that observed.     

Wave packet propagation in an amplifying medium and its application to the dispersion characteristics and to the generation mechanisms of Pc 1 events

The formulae which give the propagation characteristics of a wave packet in a dispersive and amplifying medium, are established. Application is made to the propagation of Pc 1 elements through a magnetosphere constituted of a cold plasma and a high energy proton population. It is shown that the spectral shape, in a frequency-time coordinate system, of the Pc 1 elements is related to two terms : v = d²ω/dk², which represents the variation of the group velocity with frequency and which depends only on the cold plasma characteristics, and μ = -d²γ/dk², in which γ is the amplification coefficient depending on the frequency and which is related to the high energy particle distribution function. When v ? μ, only the usual dispersion effects occur, but a new method is found for determining the line of force on which the micropulsations are generated, without making any assumption about the cold plasma density distribution inside the magnetosphere. It is also possible to deduce some characteristics about the high energy proton distribution. Theoretical computations are presented, which give the frequency variation of the amplification coefficient as a function of the e-folding energy and the anisotropy factor of these high energy protons. Applications are made to ~30 pearl events which are analysed in detail according to this theory. When μ ? v, other effects do appear. After a preliminary phase, the pearl elements can become parallel for a while, or even re-erect before lying again; the duration of each element gives an indication about the number of interacting particles. The conditions for the validity of the quasi-linear theory, and some other non-linear effects related with the interpretation of Pc 1 micropulsations are also discussed.     

Sunspot as an isolated magnetic structure: Stability and oscillations

A simple energy model of a sunspot as a compact magnetic feature is described where the main energy contribution is provided by the coolest and most compressed part of the magnetic force tube of the spot at depths ranging from Wilson’s depression level (300–500 km) down to 2–3 thousand km. The equilibrium and stability conditions for such a system are analyzed using the variation principle, and oscillations of the system as a whole about the inferred equilibrium position are studied. The sunspot is shown to be stable in the magnetic field strength interval from 0.8–1 to 4–5 kG. The dependence of the eigenfrequency on magnetic field strength ω(B) is computed for the main oscillatory mode, where only the umbra of the sunspot takes part in oscillations, ω = ω ₁ (B). Lower subharmonics may appear in the case where penumbra too becomes involved in the oscillatory process: ω ₂ = ω ₁/2, ω ₃ = ω ₁/3. Theoretical curves agree well with the observational data obtained in Pulkovo using various independent methods: from temporal variations of sunspot magnetic field and from line-of-sight-velocity measurements. The periods of oscillations found range from 40 to 200 minutes.     

Lunar photoelectron layer dynamics

Possible waves and oscillations in the lunar photoelectron layer (PEL) are investigated. The steady state PEL is reviewed as a basis for discussing PEL motions. Magnetic fields are neglected, so that there are four possible wave modes to consider. The propagation through the PEL of the two electromagnetic modes is discussed. Positive-ion waves, the third mode, are dismissed and plasma waves are considered at length. It is concluded that there are no propagating waves in the PEL other than electromagnetic. However, there is a type of oscillation which appears to be new and which may not be strongly damped. With these oscillations, termed flight-time oscillations, the height of the PEL fluctuates as does the electric field. These oscillations appear to be analogous to the height oscillations of the vertical jet of water in a city park water fountain. If flight-time oscillations are not much damped then it would be simplest to interpret them as plasma oscillations continually driven by the upwelling photoelectron stream. A possible laboratory investigation of these oscillations is discussed. For the surfaces of the Moon and the planet Mercury, the flight-time oscillation frequency,ω _F, is found to be respectively ç 4 × 10⁶ and ç 10⁷ rad s^?1. The PEL's of those surfaces may be in a state of continual vertical ‘quivering’ due to flight-time oscillations, or may be quiescent.     

Magnetic field in a fluctuating <Emphasis Type="Italic">ABC</Emphasis> flow

The standard dynamo models that explain the origin of the large-scale magnetic fields of celestial bodies are related to the view of turbulent or convective flows as a locally statistically homogeneous and isotropic, but not mirror-symmetric, random field. Using an ABC flow, which is a classical example of a flow with deterministic chaos, we ascertain the extent to which the behavior of the magnetic field in such a flow is similar to the behavior of the magnetic field in mirror-asymmetric turbulence. Such a similarity has been found to be achieved if its coefficients A, B, and C are assumed to be random processes.     

Torsional Oscillations in a Global Solar Dynamo

We characterize and analyze rotational torsional oscillations developing in a large-eddy magnetohydrodynamical simulation of solar convection (Ghizaru, Charbonneau, and Smolarkiewicz, Astrophys. J. Lett. 715, L133, 2010; Racine et al., Astrophys. J. 735, 46, 2011) producing an axisymmetric, large-scale, magnetic field undergoing periodic polarity reversals. Motivated by the many solar-like features exhibited by these oscillations, we carry out an analysis of the large-scale zonal dynamics. We demonstrate that simulated torsional oscillations are not driven primarily by the periodically varying large-scale magnetic torque, as one might have expected, but rather via the magnetic modulation of angular-momentum transport by the large-scale meridional flow. This result is confirmed by a straightforward energy analysis. We also detect a fairly sharp transition in rotational dynamics taking place as one moves from the base of the convecting layers to the base of the thin tachocline-like shear layer formed in the stably stratified fluid layers immediately below. We conclude by discussing the implications of our analyses with regard to the mechanism of amplitude saturation in the global dynamo operating in the simulation, and speculate on the possible precursor value of torsional oscillations for the forecast of solar-cycle characteristics.     

Frequency splitting in stria bursts: Possible roles of low-frequency waves

The kinematics of the process L ± F→ L′ are explored where L represents a parallel Langmuir wave, F represents a low frequency fluctuation and L′ represents a secondary Langmuir wave, and the results are used to discuss (a) a possible interpretation of the frequency splitting in stria bursts in terms of the processes L ± F → L′, L′ ± F′ → t, where t represents a transverse wave, and (b) second harmonic emission due to the processes L ± s→ L′, L + L′ → t, where s represents an ion sound wave. The following results are obtained:

1. The processes L ± s → L′ are allowed only for k _s < 2k _L ± k ₀, respectively, with k ₀ = ω _p /65 V _e .
2. The inclusion of a magnetic field does not alter the result (1) and adds further kinematic restrictions related to angles of propagation; the kinematic restriction T _e > 5 × 10⁵ K for second harmonic emission through process (b) above is also unchanged by inclusion of the magnetic field. The effect of a spread in the wavevectors of the Langmuir waves on this restriction is discussed in the Appendix.
3. For parallel Langmuir waves the process L - F → L′ is forbidden for lower hybrid waves and for nearly perpendicular resonant whistlers, and the process L + F → L′ is allowed only for resonant whistlers at ω _F ? 1/2ω _p (Ω _e /ω _p )².
4. The sequential three wave processes L ± s → L′, L′ ± s → t and L + F → L′, L′ ± F′ → t encounter difficulties when applied to the interpretation of the splitting in split pair and triple bursts.
5. The four-wave process L ± F ± F′ → t is kinematically allowed and provides a favourable qualitative interpretation of the splitting when F denotes a resonant whistler near the frequency mentioned in (3) above. The four wave processes should saturate under conditions which are not extreme and produce fundamental plasma emission with brightness temperature T _t equal to the effective temperature T _L of the Langmuir waves.

     

Study of inertial kinetic Alfven waves around cusp region

The effect of electron inertia on kinetic Alfven wave has been studied. The expressions for the dispersion relation, growth/damping rate and growth/damping length of the inertial kinetic Alfven wave (IKAW) are derived using the kinetic approach in cusp region. The Vlasov-kinetic theory has been adopted to evaluate the dispersion relation, growth/damping rate and growth/damping length with respect to the perpendicular wave number k_⊥ρ_i (ρ_i is the ion gyroradius) at different plasma densities. The growth/damping rate and growth/damping length are evaluated for different m_e/βm_i, where β is the ratio of electron pressure to the magnetic field pressure, m_{i, e} are the mass of ion and electron, respectively, as I=m_e/βm_i represent boundary between the kinetic and inertial regimes. It is observed that frequency of inertial kinetic Alfven wave (IKAW) ω is decreasing with k_⊥ρ_i and plasma density. The polar cusp is an ideal laboratory for studies of nonlinear plasma processes important for understanding the basic plasma physics, as well as the magnetospheric and astrophysical applications of these processes.     

Smooth models of overshooting at the base of the solar convective zone

A set of smoothed temperature gradient profiles around overshooting layers at the solar convective zone bottom is considered. In classical local theories of convection the one point defined according to the Schwarzschild criterion is enough to describe a convective boundary. To get a sophisticated picture of the overshooting we use four points to compute the transition overshooting functions. Analyzing the transition gradient profiles we found that the overshooting convective flux may be either positive or negative. A negative overshooting flux appears in nonlocal convective theories and causes a steep temperature gradient profile. But we propose an evenly smoothed gradient which corresponds to a convective flux positive everywhere. To outline the effect of the temperature gradient on the solar oscillations the squared Brunt–Väisälä frequency N ² is calculated. In local convective theories the N ² profile shows the discontinuity of the first derivative at the convective boundary, while all smoothed profiles eliminate the break.     

Structure of the solar photospheric convection on subgranulation scales

We investigate the structure of convective flows in the solar photosphere on subgranulation scales. The solar granulation pattern is reproduced by solving the inverse problem of nonequilibrium radiation transfer on the basis of the profiles of the neutral iron line λ 523.42 nm. The wave motions are excluded by the k-ω filtration. The line-of-sight velocity has an asymmetric distribution inside the convective flows in large granules (1.5″ and larger) in the lower photosphere and at the bottom of the middle photosphere. This asymmetry is weaker in the upper photosphere. For smaller flows the distribution is more symmetric at all heights. The asymmetry of the temperature distribution is less pronounced. Large convective flows were found to have a fine structure: they are fragmentized into several smaller flows. The fine structure of large flows and spatial smearing are responsible for the observed asymmetry of the convection velocity distribution inside flows.