GEOPHYSICAL RESEARCH LETTERS, VOL. 19, NO. 6,PAGES 605-608, MARCH 20,1992 THE CIV PROCESSES IN THE CRIT EXPERIMENTS K. Papadopoulos Science Applications International Corporation and University of Maryland, Department of Physics Abstract. Recent in situobservations during Critical Ion- ization Velocity (CIV) experiments under the CRIT project produced puzzling results, apparently at odds withcurrent ClV theories. It is shown that incorporation in thetheory of lower hybrid wave(LHW) instabilities of strong turbulence and finite size effects results in reconciliation of the theory with the observations. Introduction Thecritical ionization velocity(CIV) effectis thought to occur when neutral gas streams through a magnetized plasma, with cross field drift energy exceeding the ioniza- tion energy of the gas [Alfven, 1954]. Laboratory exper- iments [Fableson, 1961; Angerth et al., 1962; Danielson, 1970; Eselevich and Fainshtein,1986] and theoretical anal- ysis [Galeev, 1981;Formisano et al., 1982; Papadopoulos, 1983, 1984; Raadu, 1978; Abe and Machida, 1985; Go- enz et al., 1985; Machida and Goertz, 1986, 1988; Mobius et al.,1987] indicate thata plasma instability driven by the streaming newly ionized neutrals, heats theplasma electrons to energies in excess of theionization potential on collision- less timescales. The energetic electrons colliding with the neutrals further ionize the gas,inducing a self supporting discharge. The instability responsible for the efficient en- ergy transfer from the streaming ionized gasmolecules to electrons has been identified as thelower hybrid instability. While laboratory experiments andtheory seem to be in relatively good agreement withtheabove physical descrip- tion, this isnot the case for space tests ofthe CIV hypothesis [Haerendel, 1982; Newell and Torbert, 1985; Wescott et al., 1986a, b;Kelley et al., 1986]. A review of the observations md the discrepancies withtheconventional picture can be found inNewell [1985] and Torben [1988]. Outstanding is- sues concern theprimaryinstability resulting in electron en- ergization, the ionization rates, the scaling with the ambient electron density, and theCIV triggering process. The re- cent CRIT I and II sounding rocket experiment experiments [Swenson et al., 1990; Stenbaek-Nielsen et al., 1990; Swen- son et al.,1991; Brerming et al., !991a, b; Kelley et al., 199!] have been unique in that they provided a good complement of in situ diagnostics, including complete vector electric field measurements for frequencies up to 12 kHz [Swenson et al., 1990, 1991]. The purpose of this letter is toreconcile the theoretical foundations of the CIVhypothesis tothe space experiments. This letter is a qualitative preliminary presen- tation of the concepts. Comprehensive quantitative analysis is ongoing and will be presented elsewhere. Copyright 1992 by the American Geophysical Union. 0•9• numb• 92GL1)0405 •: 8534/92/92GL-00405503.00 CRIT II Experimental Evidence andIssues A comprehensive presentation of the CRITexperiments and data can be found in thepapers referenced above. We review below selectiveCRIT II observations taken at a distance approximately 1.7km from theburst [Swenson et al., 1990], which highlight theissues relevant to this letter 1. The electrostatic wave measurements indicate first a weak activityat .095 secs after the release with a fre- quency in the vicinity 6-7 kHz. This wasfollowed by the appearance of an oscillatory electric field growing with time. The frequency of the growing oscillations started at !.6 kHz andgradually dropped to 300 Hz at about .12 secs. The electric field peaked at.12 secs with amplitude of about 400--600mV/m. The wave structure wasvery bursty andthe polarization almost isotropic. Following .13 secs the wavepattern changed with the waves having lower amplitude and lessstructure. 2. The ambient electron density was seen to rise from an ambient value of 5.4 x 10 • cm -3 to a peak value of 2.8 x 106 cm 4. The rise occurred .11secs following the burst, while the maximum occurred at about .18 secs. The peak production rate was estimated at5 x 107 crn 4 S -1' 3. The interaction was seen to behave as an ionization front riding on the neutral barium with a spatialsize of 600-700 m in the stream. Importantdam concerning the DC electric and magnetic structures have been alsopresented in the CRIT I papers mentioned above. Of importance here is that in CRIT ! there was no evidence for increase of the electron density past its ambient value which was only 4 x 104 #/era 3. The wave spectra were similar to CRIT II, but the maximum amplitude was only 150mV/m. Thesemeasurements posea seriesof theoretical ques- tions, such as: 1. Why wasthefrequency of the observed waves variable and low? For low hybridinstabilities, the frequency should beclose to either the barium or theoxygen lower hybridfrequencies, whichfor the CRIT iI conditions were 2.3 and 6.9 kHz correspondingly. 2. Why wasthe wavepolarization almost isotropic? Ac- cording to lineartheory the wave polarization in lower hybrid instabilities is strongly anisotropic, with thefield aligned electric field smaller by morethan a_n order of magnitude than the transverse component. 3. Can the electronenergization be reconciled with the observed wave spectrum? ff the wave spectrum is isotropic the phase velocity of the waves .in the magnetic fielddirection would be toolow to energize electrons to ionizing energies, at least within the context of quasi- 605