5. Discussion and Conclusions

[70]  We have described a technique to observe the neutral winds and electric fields with backscatter from field-aligned irregularities in the E region. The method does not require additional expense and may be used at any ISR whose altitude resolution is one half of the least allowable resolution of the energy sources: neutral winds and electric fields.

[71]  There are two basic ideas at the heart of our method. The first is that the expression of the phase velocity is, to leading order, the same as for linear processes, independently of the degree of nonlinearity of the actual mechanism causing the small-scale irregularities. The second is that coefficients at each contributor to this phase velocity are fast changing with altitude and show quite different altitude behavior. The second idea allows us to use these coefficients as filters to find the boundary condition, i.e., the altitude(s) at which only one contributor defines this phase velocity.

[72]  Showing that at almost any altitude there are only two contributors dominating the phase velocity, we assume the driving forces for the 3-m waves to be alternately the same at two neighboring altitudes separated by the altitude resolution of the radar. Finally, based on modeling and observations we have assumed the relative density fluctuation in the primary wave to be 5%.

[73]  The above scheme allows us reconstruction of winds and electric fields from the altitude-time distribution of line-of-sight Doppler velocities over the altitude range covered by backscatter with type 2 spectral signature with altitude and time resolutions which are equal to twice the altitude resolution and equal to the temporal resolution of the radar measurements, respectively.

[74]  Our method allows reconstruction of (1) the meridional neutral wind up to 100 km, (2) the zonal polarization electric field caused by the primary waves (driven by the meridional electric field and the zonal neutral wind) up to 120 km, and (3) the background zonal electric field above 100 km.

[75]  We made an attempt to compare the wind velocities derived by our method with the meteor radar data available for the same time and location. However, despite the fact that the time near 0300 LT was the period of the highest meteor occurrence, the time and height resolution of the meteor radar for the time of interest were 2 hours and 2 km, respectively.

[76]  It should be mentioned that, like other wind-measuring methods, our method smoothes shears in the neutral wind to the scale of the altitude resolution of the method. The best resolution in the wind measurements is provided by in situ measurements with the chemical release technique. It allows an altitude resolution of about 100 m [Larsen, 2002]. For remote observations with meteor radar, the altitude resolution depends on the number of meteors and, as we have already mentioned, for the case study above it was 2 km. Note that a large shear in the wind velocity in the lower E region results in sporadic ionization with a typical vertical scale of the sporadic layer ( E_s ) equal to the shear vertical scale length. Observations yield E_s vertical scales L=1-2 km as a rule, with occasionally L=500 m and rare cases of a very complex structure of the underlying ionization clouds [Bakhmet'eva et al., 2003]. Thus we do not expect to lose much information if any about the neutral winds.

[77]  Note that our method is based on the expression for the phase velocity along the radar line of sight. In our description above and in correspondence to most of the observations, we assume that the radar observes in the meridional plane. If the CSR line of sight is not in the meridional plane, then the most general expression (13) should be used. It is important that, independently of the complexity of the processes generating small-scale irregularities (gradient drift, thermal or Farley-Buneman), the expression for the phase velocity/frequency, to leading order, is the same. However, we remind once more, that one should keep in mind that our method is valid only for irregularities producing type 2 backscatter whose phase speed is less then the ion acoustic speed ( V _Dop< C_s ).

[78]  It should be mentioned that for the correct range-to-altitude conversion the preliminary radar imaging technique [Hysell and Burcham, 2000; Hysell et al., 2002; [Kudeki and Sürücü, 1991; Woodman, 1997] should be used when possible for more correct altitude profiles of the Doppler line-of-sight velocity.