next up previous
Next: References Up: SUPERDARN MEASUREMENTS OF DOUBLE-PEAKED Previous: Discussion

Conclusion and Summary


The SuperDARN radar has been used to analyze Doppler spectra of backscatter from the F-layer. A distinct spectral signature, that of a double-peaked spectrum, has been observed in spatially localized patches or bands encompassing a number of neighbouring radar range cells seen by a single radar. Often, both radars in a SuperDARN pair see the double-peaked spectra in the same patch or band.

Several spectral estimators have been used and it was found that the parametric models are appropriate for this type of analysis. In particular the Burg-algorithm outperformed FFT based methods and produced the best results. The evaluation of the spectral estimators was done using high-resolution autocovariance functions which were generated from special pulse patterns. Several optimized pulse patterns and a general algorithm to generate these patterns have been described.

The influence of particle precipitation on the occurrence of the double-peaked spectra was checked using the DMSP satellite particle data. A set of model calculations gave information about the ionization rates and conductivities of the E- and F-layer as derived from SSJ/4 electron precipitation spectra. The discussion included a description of general ionospheric conditions and F-layer plasma flow patterns during the events.

The results suggest that a vortical flow pattern or the flow along an arc-like structure is a likely candidate for the generation of double peaked Doppler velocity spectra. This is not unexpected because vortices of scale sizes from 1000 km down to 1 km as well as arcs of scale sizes from 10 km to 100 m are commonly observed ionospheric features. The precise physical mechanisms that produce double peaked spectra cannot be determined with certainty, but there is evidence that the production mechanism is linked to regions of inhomogeneous, increased F-layer ionization produced by low-energy electron precipitation as measured by the DMSP satellite.

To correlate the observations of the double peaked spectral features with optical signatures in the aurora, a study that makes use of all-sky camera images is planned. A low-cost CCD imager will be used to image red-line emissions (630 nm) from the lower F-region. These emissions should be associated with the low energy, spatially confined electron precipitation regions that have been shown to correlate with the spectral features. They might also show specific spatial features than are indicative of vortices.

Modeling of power spectra of simulated vortices in the F-region that take into account the three dimensional flow pattern and scattering cross-section will be an important tool for future investigations. Corresponding studies have been performed for the detection of tornados with meterological radars and double peaked spectral signatures were found. A similar analysis will be performed to determine the possible shape of SuperDARN power spectra measurements.

Also, a statistical analysis into these spectral features is currently under way for several reasons. Satellite measurements of vortical features have been made but provide at best a snapshot of a very dynamic event. The SuperDARN radar system has the potential to continously map such features in its field of view. It has also been demonstrated that double-peaked spectra disturb normal ACF processing of SuperDARN data. The inclusion of a double-peaked spectral model into the current data analysis procedure of SuperDARN data would further improve the accuracy of SuperDARN measurements and could provide improved quality to the SuperDARN measurements. As has been pointed out in several recent publications, vortical features seem to play an important role in the dynamics of the F-layer and therefore SuperDARN could be used as an important additional tool for monitoring and analyzing these effects.

next up previous
Next: References Up: SUPERDARN MEASUREMENTS OF DOUBLE-PEAKED Previous: Discussion

Andreas Schiffler
Wed Oct 9 10:05:17 CST 1996