Rl. Dowden et Cdd. Adams, SIZE AND LOCATION OF LIGHTNING-INDUCED IONIZATION ENHANCEMENTS FROM MEASUREMENT OF VLF PHASE AND AMPLITUDE PERTURBATIONS ON MULTIPLE ANTENNAS, Journal of atmospheric and terrestrial physics, 55(10), 1993, pp. 1335-1359
A 600-km array of five Trimpi receivers (''elements'') has been set up
in New Zealand broadside to the VLF (22.3 kHz) transmitter, NWC, some
6000 km west, with element separations varying from 8 km to 550 km. A
lthough such a five-element array is inadequate for imaging of lightni
ng-induced ionisation enhancements (LIEs) by VLF holography, or invers
e scattering, estimates of LIE size and location can be made if the sh
ape and form of the LIE can be guessed or assumed, with even fewer ele
ments. With five elements, tests of the assumed model can be made as w
ell. Owing to its transform properties, the simplest model to use for
scattering inversion is the Gaussian LIE distribution. For this model,
and for single mode propagation, an inversion process is derived here
for the full range of LIE and path dimensions, ranging from those for
which the receiver is in the diffraction far field to those in which
''geometric optics'' dominate. This inversion process has some validit
y for small LIEs of other shapes of simple form. For more extreme mode
ls, the dominance of geometry or diffraction can usually be establishe
d in individual cases which then allows simple scaling procedures to b
e used in scattering inversion. Some 70 Trimpi events were observed on
all five elements during a single night in July, 1991 (late winter).
These were used to determine LIE location and size, and to test the ap
plicability of various LIE models. It was found that most LIEs that ni
ght occurred over the Tasman Sea near the great circle from the VLF tr
ansmitter, NWC, to Wellington, generally some 500 to 2000 km from Well
ington, and with north-south dimensions of 100-250 km. Much longer eas
t-west dimensions (oriented towards NWC) are suggested to account for
the very strong Trimpis observed. While about half of these LIEs that
night could have had a smooth lateral spread (e.g., Gaussian), the rem
ainder required varying degrees of fine structure, from ''flat'' or Bu
tterworth LIEs to multiple LIEs as might be expected from multiduct wh
istlers, to explain the observed diffraction pattern exhibiting maxima
and minima as well as the wide angular range over which simultaneous
Trimpis were observed.