Low-frequency electrostatic waves in the ionospheric E-region: a comparison of rocket observations and numerical simulations

Abstract

Abstract. Low frequency electrostatic waves in the lower parts of the ionosphere are studied by a comparison of observations by instrumented rockets and of results from numerical simulations. Particular attention is given to the spectral properties of the waves. On the basis of a good agreement between the observations and the simulations, it can be argued that the most important nonlinear dynamics can be accounted for in a 2-D numerical model, referring to a plane perpendicular to a locally homogeneous magnetic field. It does not seem necessary to take into account turbulent fluctuations or motions in the neutral gas component. The numerical simulations explain the observed strongly intermittent nature of the fluctuations: secondary instabilities develop on the large scale gradients of the largest amplitude waves, and the small scale dynamics is strongly influenced by these secondary instabilities. We compare potential variations obtained at a single position in the numerical simulations with two point potential-difference signals, where the latter is the adequate representation for the data obtained by instrumented rockets. We can demonstrate a significant reduction in the amount of information concerning the plasma turbulence when the latter signal is used for analysis. In particular we show that the bicoherence estimate is strongly affected. The conclusions have implications for studies of low frequency ionospheric fluctuations in the E and F regions by instrumented rockets, and also for other methods relying on difference measurements, using two probes with large separation. The analysis also resolves a long standing controversy concerning the supersonic phase velocities of these cross-field instabilities being observed in laboratory experiments.

Structure functions and intermittency in ionospheric plasma turbulence

“…4, the increase in T e is expected to be minute, but for the somewhat larger down-leg fields, E 0 ≈60-70 mV/m, nontrivial enhancements of T e are anticipated, but not observed for the present conditions. The wave propagation velocities, for instance, as found by Iranpour et al (1997), Krane et al (2000) and Dyrud et al (2006) are best explained by an electron temperature of approximately 400 K. Also other reports (Pfaff et al, 1992) noted the lack of electron temperature enhancements for conditions similar to ours. A previous study (Dyrud et al, 2006) attempted to explain the low electron temperatures by thermal conduction to the colder regions below the enhanced wave activity, but used too low numerical values for the electron energy loss per collision, by taking this energy loss to be at most an order of magnitude larger than for inelastic collisions.…”

“…Typically, the saturated potential fluctuations have a characteristic wavelength of ∼2 m, and a peak value of ∼0.3 V. A typical root-mean-square value of the potential fluctuations is ∼0.08 V, corresponding to acelectric fields ∼3×10 −2 E 0 , for the given conditions. The fluctuations in density are relatively modest, typically below 20%, even though we can observe larger spikes (Dyrud et al, 2006).…”

“…By far, the dominant cooling rate is due to inelastic collisions in the E-region. It may be that the analysis of Dyrud et al (2006) applies for conditions in laboratory experiments, where the dominant collisional electron energy loss will usually be for elastic collisions with neutral inert gases. The collisional energy losses for elastic collisions are generally much smaller than for the inelastic collisions.…”

“…The numerical simulations were conducted in two spatial dimensions in the plane perpendicular to the imposed magnetic field, using a Particle-in-Cell (PIC) code (Birdsall and Langdon, 1991) for the ion component and a fluid model for the electrons (Oppenheim et al, 1995;Oppenheim and Otani, 1996;Dyrud et al, 2006). In the present analysis, the electron inertia is ignored.…”

“…for cases to be discussed in the following (Rose et al, 1992;Krane et al, 2000;Dyrud et al, 2006), assuming locally homogeneous and time stationary conditions.…”