ScienceWeek

ATMOSPHERIC PHYSICS: ELECTRICAL BREAKDOWN AND LIGHTNING

The following points are made by A.V. Gurevich and K.P. Zybin (Physics Today 2005 May):

1) In 1749 Benjamin Franklin made a fundamental discovery -- that lightning is an electrical discharge between a thundercloud and Earth. Such a discharge can only occur if the atmosphere, which is normally an insulator, undergoes electrical breakdown. Therein lies the first mystery. The conventional breakdown taught in textbooks originates with free electrons heated in an electric field. Fast electrons in the tail of the thermal distribution function have enough energy -- about 10-20 eV -- to ionize matter and therefore to generate new free electrons. Electrons with lower energies disappear when they recombine with the ionized molecules in the air.

2) When the electric field E exceeds a threshold, E > Ethr, the generation rate of new electrons from ionization exceeds their recombination rate, and the number of free electrons begins to exponentially increase: Electrical breakdown occurs. Because the electrons responsible for ionization are out in the high-energy tail of the distribution, the mean electron energy e at which breakdown occurs does not normally exceed several electron volts. For instance, in air, e is approximately 2 eV. For conventional breakdown, Ethr is proportional to the number density of molecules. In air at atmospheric pressure, Ethr is approximately 2 MV/m. All electric field measurements in thunderclouds, however, reveal values substantially less than those needed for conventional breakdown.[1] This is the long-standing mystery about lightning's origin.

3) Another mystery appeared with the discovery of strong isolated radio pulses generated during thunderstorms but not connected to lightning discharge.[2] Those roughly 5-microsecond radio events -- called "narrow bipolar pulses" -- can have astonishingly high power emissions, up to 100 GW. A closely related radio effect is the lightning-initiation pulse -- recently discovered by the authors and their colleagues[3] -- which is always seen as the first isolated pulse at the beginning of a lightning discharge. That type of pulse is also bipolar, but its duration is only about 0.5 microseconds and its power is less than that of NBPs. What could generate such radio pulses?

4) Still another mystery arose after the discovery of intense x-ray bursts[4] both inside and beneath thunderclouds. With characteristic x-ray energies around 50 keV, the bursts last about 1 minute and are usually well correlated with lightning events. In addition, the Compton Gamma Ray Observatory and the Reuven Ramaty High Energy Solar Spectroscopic Imager satellites detected very intense millisecond gamma-ray bursts (0.05-10 MeV) that appeared at altitudes of about 500-600 km in the ionosphere.[5] The data analysis definitely indicated that the bursts were generated during thunderstorms. The existence of analogous gamma-ray emission (2-10 MeV) accompanying lightning was established by Charles B. Moore and colleagues in natural conditions and by Joseph Dwyer and coworkers in rocket-triggered lightning experiments.

References (abridged):

1. D. R. MacGorman, D. Rust, The Electrical Nature of Storms, Oxford U. Press, New York (1998)

2. D. M. LeVine, J. Geophys. Res. 85, 4091 (1980); D. A. Smith et al., J. Geophys. Res. 104, 4189 (1999); D. A. Smith et al., J. Geophys. Res. 107, D134183 (2002); D. A. Smith et al., Radio Sci. 39(1), RS1010 (2004)

3. A. V. Gurevich, L. M. Duncan, A. N. Karashtin, K. P. Zybin, Phys. Lett. A 312, 228 (2003)

4. M. D. McCarthy, G. K. Parks, Geophys. Res. Lett. 12, 393 (1985); K. B. Eack et al., J. Geophys. Res. 101, 29637 (1996). A. P. Chubenko et al., Phys. Lett. A 275, 90 (2000) ; M. B. Brunetti et al., Geophys. Res. Lett. 27, 1599 (2000); T. Torii et al., J. Geophys. Res. 107, 4324 (2001); V. V. Alexeenko et al., Phys. Lett. A 301, 299 (2002); A. P. Chubenko et al., Phys. Lett. A 309, 90 (2003)

5. G. J. Fishman et al., Science 264, 1313 (1994); D. M. Smith et al., Science 307, 1085 (2005)

Physics Today http://www.physicstoday.org

ScienceWeek http://scienceweek.com