6.2 Further Research

Below we conclude by outlining some promising future investigations which follow from work detailed in this dissertation.

Three-dimensional modeling of VLF propagation past diffuse sprite regions: Early/fast events

As discussed in Section 2.5.2, the nature of the ionospheric disturbances causing early/fast VLF phase and amplitude changes following cloud-to-ground lightning is still under debate. Our suggestion that modeled ionization changes in observed sprite halos may produce such VLF perturbations with characteristics similar to recorded events can easily be addressed more quantitatively using available models of VLF scattering [Poulsen et al., 1993].

Telescopic high-speed video

Recent high-speed video observation of sprite filamentary development [Stanley et al., 1999] suggests that the apparent continuous filaments often visible in 17 ms video fields may in fact be due to temporal integration of short propagating streamer heads.

Photometers bore-sighted on a video system will accurately record the duration of optical emission of a new streamer channel, but not necessarily resolve whether it glows only during development or persists after the initial breakdown. Recent telescopic video observations [Gerken et al., 1998; Gerken et al., 2000] reveal a number of poorly understood features such as ``beads'', rebrightening regions, and sharply truncated columns in high spatial resolution. An obvious strategy for simultaneously resolving time and spatial scales comparable to those of the development of some of these phenomena [Stanley et al., 1999] is to combine the proven telescopic and high speed techniques. For instance, a high speed camera operated at 4000 frames per second with a 0.7$^\circ$ vertical field-of-view can resolve streamer propagation with velocities between ${\ensuremath{2\!\times\!10^{5}}}$ m/s and ${\ensuremath{2\!\times\!10^{7}}}$ m/s at a range of 500 km. Cameras with variable exposure times shorter than the frame rate are also available now, allowing for ``strobe'' imagery of streamer propagation and sprite development. High resolution imaging studies may also be applied to negative sprites in order to compare their fine structure to that of positive sprites.

Multi-anode photometric arrays

Since the Fly's Eye's design in 1995, new technology has become available which is perfectly suited to the construction of a photometric array without multiple optics and separately sighted photometers. These multi-anode photomultipliers are available as linear arrays and square arrays of up to 16 or 64 elements. With a single set of optics and a multi-anode photomultiplier, crude imagery could be obtained continuously at rates up to 10$^6$ s$^{-1}$. Vertical arrays of this design have been operated since 1996 [Fukunishi et al., 1996a; Fukunishi et al., 1998].

A pair of vertical arrays with fields-of-view of $\sim$4$^\circ$$\times$0.5$^\circ$ and two optical filters similar to the red and blue filters used on the Fly's Eye would be well suited to studies of bright sprites exhibiting an exponential optical relaxation. A narrow vertical field-of-view is crucial in constraining the elevation angle used to account for the effects of atmospheric scattering, and in constraining the source altitude for determination of the electric field associated with a measured relaxation time constant. The combination of these techniques could prove complimentary for remotely sensing sprite electric fields and could serve as a powerful probe of the energetics of sprites at high time resolution. Sprite halos could also be resolved and their relationship to any initial pulse in blue emissions could be investigated.

In addition it may be interesting to deploy such a system, possibly oriented as a horizontal rather than vertical array, in a study of storms which have not necessarily been selected as sprite-producing candidates. This effort could help to assess the prevalence of (negative) sprite halos (as well as elves) on a global scale.