Portal de Periódicos da CAPES

Comment on “On the origin of whistler mode radiation in the plasmasphere” by Green et al.

2006; American Geophysical Union; Volume: 111; Issue: A9 Linguagem: Inglês

10.1029/2005ja011477

2156-2202

R. B. Horne, R. B. Horne, Nigel P. Meredith,

Earthquake Detection and Analysis

[1] Green et al. [2005] have recently performed a statistical analysis of the global distribution of whistler-mode radiation in the plasmasphere. They use the results to draw inferences on the origin of different emissions and the potential impact of such waves on scattering loss of resonant electrons from the radiation belts. Specifically, they present evidence that the average intensity of waves at frequencies near 3.0 kHz exhibits a spatial distribution in both MLT and geographical longitude similar to the distribution of lightning. Green et al. [2005] claim that the properties of such waves are representative of plasmaspheric hiss, an important magnetospheric emission, which has previously been shown to be a major scattering agent in the slot region between the inner and outer radiation belts [Lyons and Thorne, 1973; Abel and Thorne, 1998]. We disagree with both the interpretation made by Green et al. [2005] of the nature of the reported waves and their conclusion that lightning is ultimately responsible for radiation belt loss in the slot region between the inner and outer belts. Because their work has been made public via a NASA press release and reported in popular science news articles, we feel it is important to voice our strong disagreement with their conclusions, which give an erroneous impression about the role of lightning in the loss of radiation belt electrons. [2] Plasmaspheric hiss is a broadband whistler-mode emission, primarily confined within the high-density plasmasphere, with peak power spectral intensity near a few hundred Hz [Thorne et al.,1973]. The intensity of plasmaspheric hiss falls rapidly above 1 kHz, and the waves are rarely seen above 3 kHz. Plasmaspheric hiss is easily identified on high time resolution spectrograms and clearly distinguishable from more discrete signals originating from lightning, which can occur over a similar frequency range, but which generally extend to higher frequency. Lightning-associated emissions that enter the magnetosphere are strongest near a few kHz [Edgar, 1976] and become very weak below 1 kHz, where plasmaspheric hiss emissions have their peak spectral intensity. A second distinguishing characteristic is the pronounced dependence of the intensity of plasmaspheric hiss on both the flux of resonant electrons [Cornilleau-Wehrlin et al., 1985], and on the level of geomagnetic activity [Smith et al., 1974; Thorne et al., 1976; Meredith et al., 2004]. Discrete whistlers originating from lightning, and magnetospherically reflected whistlers that merge into a broadband distribution following many internal reflections [Bortnik et al., 2003], show little dependence on geomagnetic activity and their power spectral intensity is usually much weaker than hiss [e.g., Ristic-Djurovic et al., 1998]. [3] Careful examination of the average latitudinal distribution of dayside 1.2 kHz waves in Figure 2 of Green et al. [2005] reveals two distinct populations; one peaked at high latitude (which may best be interpreted as lightning-generated emissions propagating away from the atmosphere) and a second population distributed around the equator, which is remarkably similar to the modeled distribution of obliquely propagating magnetospherically reflected (MR) whistlers [Thorne and Horne, 1994], which also originate from lightning and tend to settle, after several magnetospheric reflections, around a field line where the wave frequency is comparable to the equatorial lower hybrid frequency [Bortnik et al., 2003]. As a consequence, MR whistlers occur in a band, which moves to higher frequency at lower L [Bortnik et al., 2003], while plasmaspheric hiss tends to occur in a broad band which is relatively independent of L. Green et al. [2005] identify both reported populations of 1.2 kHz waves as plasmaspheric hiss, but we assert that they are simply the direct result of lightning emissions. Although the waves have an MLT morphology similar to the distribution of plasmaspheric hiss [Meredith et al., 2004], the average power spectral intensity reported by Green et al. [2005] is more than an order of magnitude smaller than the intensity of lower-frequency plasmaspheric hiss, under moderately active geomagnetic conditions, reported by Meredith et al. [2004]. Because the rate of electron pitch-angle scattering is proportional to the power of resonant waves [Lyons et al., 1972], the waves reported by Green et al. [2005] are relatively insignificant in causing electron loss from the outer plasmasphere, where scattering by the more intense lower frequency (300–700 Hz) plasmaspheric hiss is dominant [Abel and Thorne, 1998]. Nevertheless, these weaker lightning-generated emissions could still contribute to electron loss in the inner portion of the plasmasphere, where hiss becomes weaker [Thorne et al., 1973], and where resonance with energetic electrons becomes less effective [Abel and Thorne, 1998]. [4] Despite considerable theoretical analysis and modeling [Etcheto et al., 1973; Parady, 1974; Thorne et al., 1979; Church and Thorne, 1983; Huang et al., 1983], the origin of plasmaspheric hiss remains controversial and to this date unresolved. Green et al. [2005] claim that their observations strongly support lightning as the dominant source for plasmaspheric hiss, a concept introduced earlier by Sonwalker and Inan [1989] and Draganov et al. [1992]. To further support this assertion, Green et al. [2005] show evidence in Figures 7 and 8 for an enhancement of the high-latitude 3 kHz emissions over geographic longitudes aligned with the continents, where thunderstorm activity is higher, but they omit from their analysis waves measured near the equator; a key population for plasmaspheric hiss. We agree that the 3kHz waves reported by Green et al. [2005] are probably dominated by lightning emissions that propagate directly out from the atmosphere over areas of thunderstorm activity. However, we strongly disagree that lightning is responsible for the most intense hiss band below 1 kHz. [5] As noted above, plasmaspheric hiss has peak power spectral intensity at much lower frequency (<1 kHz), exhibits a pronounced dependence on geomagnetic activity [Meredith et al., 2004], and has typical intensity several orders of magnitude larger than the peak power spectral intensities of 3 kHz waves (∼3 × 10−7 γ2/Hz) reported by Green et al. [2005]. To demonstrate whether lightning is responsible for the origin of plasmaspheric hiss, and by association the ultimate cause of electron loss in the slot region of the radiation belts, a similar analysis would have to be performed on the most intense hiss emissions near 500 Hz. In their discussion section, Green et al. [2005] admit that "the strong geographical mapping" (and hence a correlation to lightning) "only becomes obvious above" the ≈500 Hz range. We contend that this key distinction, together with the pronounced association of (≤1 kHz) hiss intensities with geomagnetic activity [Meredith et al., 2004], points to a natural instability in the magnetosphere for the origin of hiss. The nature of this source still needs to be established, but lightning does not appear to play a significant role, either directly or as an embrionic source. [6] The assertion by Green et al. [2005] that lightning "maintains the slot region in the radiation belt" is not only incorrect but it also ignores the considerable theoretical work that has shown that the slot region between the inner and outer radiation belts is a result of a balance between inward radial diffusion and losses from several different plasma waves [Lyons and Thorne, 1973; Abel and Thorne, 1998]. The slot is most pronounced at energies above an MeV. At such high-energies, electron scattering by low-frequency (∼Hz) electromagnetic ion cyclotron waves, which have no possible connection to lightning, is a dominant loss process [Albert, 2003; Summers and Thorne, 2003]. [7] This research was funded in part by NASA grant NNG04GN44G and NSF GEM grant ATM-0402615. [8] Arthur Richmond thanks Hiroshi Fukunishi for his assistance in evaluating this paper.