the mass and speed dependence of meteor air plasma temperatures

ASTROBIOLOGYVolume 4 Number 1 2004copy Mary Ann Liebert Inc

Research Paper

The Mass and Speed Dependence of Meteor Air Plasma Temperatures

PETER JENNISKENS1 CHRISTOPHE O LAUX2 MICHAEL A WILSON3

and EMILY L SCHALLER4dagger

ABSTRACT

The speed and mass dependence of meteor air plasma temperatures is perhaps the most im-portant data needed to understand how small meteoroids chemically change the ambient at-mosphere in their path and enrich the ablated meteoric organic matter with oxygen Suchchemistry can play an important role in creating prebiotic compounds The excitation condi-tions in various air plasma emissions were measured from high-resolution optical spectra ofLeonid storm meteors during NASArsquos Leonid Multi-Instrument Aircraft Campaign This wasthe first time a sufficient number and range of temperature measurements were obtained tosearch for meteoroid mass and speed dependencies We found slight increases in tempera-ture with decreasing altitude but otherwise nearly constant values for meteoroids with speedsbetween 35 and 72 kms and masses between 1025 g and 1 g We conclude that faster and moremassive meteoroids produce a larger emission volume but not a higher air plasma tempera-ture We speculate that the meteoric plasma may be in multiphase equilibrium with the am-bient atmosphere which could mean lower plasma temperatures in a CO2-rich early Earth atmosphere Key Words Prebiotic moleculesmdashOrigin of lifemdashMeteorsmdashPlasma tempera-turesmdashMultiphase medium Astrobiology 4 81ndash94

81

INTRODUCTION

INFALLING EXOGENOUS MATTER responsible formeteors is capable of altering the atmospheric

composition in its wake creating compoundssuch as NO that are important for nitrogen fixa-tion In the process the mixture of air plasma and

ablation products can chemically change the or-ganic molecules ablated from the meteoroid (Jen-niskens et al 2004) Meteoroids 150 mm in sizerepresented a significant source of mass influx atthe time of the origin of life rivaling the amountof material delivered by less frequent comet andasteroid impacts (Chyba and Sagan 1992 Love

1Center for the Study of Life in the Universe SETI Institute Mountain View California2High Temperature Gasdynamics Laboratory Mechanical Engineering Department Stanford University Stanford

California3NASA Ames Research Center Moffett Field California4NASA Ames Astrobiology Academy Dartmouth College Hanover New HampshirePresent address Ecole Centrale Paris Laboratoire EM2C Chatenay-Malabry FrancedaggerPresent address Astronomy Department California Institute of Technology Pasadena California

and Brownlee 1993 Jenniskens et al 2000) To-day only a small mass fraction (8) of mete-oroids in the 10ndash50 mm range survive entry inEarthrsquos atmosphere and arrive unaltered as mi-crometeorites (Anders 1989) An even smallerfraction of large meter-size asteroidal debris sur-vives in the form of meteorites The bulk of in-falling meteoroids is ablated in the Earthrsquos at-mosphere and in the process creates brief flashesof light called meteors

Some of the relevant chemistry occurs in awarm (T 4400 K) air plasma thought to resultfrom expansion into the ambient atmosphere ofimpinging air molecules after their interactionwith the meteoroid and its ablation vapor cloud(Boyd 2000 Jenniskens et al 2000 Popova et al2000) The physical conditions in this plasma setthe tone for the subsequent chemical evolution ofablated compounds and the atmosphere in themeteorrsquos path The initial impact excitation is ahighly non-local thermodynamic equilibrium(LTE) process but the product is a warm plasmawith air plasma and metal atom emissions thatare nearly in equilibrium (Jenniskens et al 2000)From this plasma the site of the bulk of hot-tem-perature chemistry the bulk of optical emissionsis observed The plasma only gradually coolswith distance away from the meteoroid (Jen-niskens and Stenbaek-Nielsen 2004)

Earlier excitation temperature measurementssummarized in Table 1 were based on analysis ofthe curve of growth of radiation from metal abla-tion lines in photographic spectra of bright fireballs(Ceplecha 1964 1968 Harvey 1973) The brightermetal atom lines of meteors as faint as 10 magni-tude are affected by self-absorption and dampingdue to significant absorption of emitted photons inthe plasma itself (Ceplecha 1973) Values of themetal atom excitation temperature of such fireballswere calculated to range from 3000 to 5500 K withno obvious dependency on meteor velocity ormass The only significant dependence was noticedby Ceplecha (1965) who found a decrease in theexcitation temperature as a function of meteor lu-minosity during a flare However these and othermeasurements of temperature are model depen-dent with early measurements later corrected up-wards to match the luminous efficiency by as-suming that the effective radiating volume increaseswith decreasing lower energy level of the excita-tion The adjusted values were around 5500 K andpertain to bright fireballs at a typical altitude of 75km and speed 32 kms (Ceplecha 1973)

More recent excitation temperatures were de-rived by fitting synthetic spectra of optically thinradiation (ie no self-absorption) to the metalatom ablation lines (BoroviIumlcka 1993) Borovi Iumlcka(1994) concluded that several lines including Ca1

and Mg1 originate from a separate hot gas com-ponent with T 10000 K This high-temperatureplasma makes up only 002 (in number not vol-ume) of the meteor vapor envelope in slow me-teors but accounts for more than 5 in fast me-teors BorovicIuml ka (1994) considered the O and Natomic lines in meteor spectra to be part of thishigh-temperature gas because of their high exci-tation energies (11ndash12 eV) Moreover results fromcurve of growth analysis of atomic oxygen linesby Ceplecha (1973) also pointed to an air plasmatemperature of about 14000 K

The recent introduction of cooled charge cou-pled device (CCD) detectors in meteor spectro-scopy (Jenniskens et al 2000) provides the near-infrared sensitivity and detector linearity necessaryto derive air plasma temperatures from the shapeof the first positive bands of molecular N2 (firstidentified by Cook and Millman 1954) and therelative intensity of the O and N atomic linesFrom the first such spectra recorded during the1998 Leonid Multi-Instrument Aircraft Campaign(MAC) (Jenniskens and Butow 1999) we foundthat the chemical equilibrium of N and N2 waswell described by an LTE temperature of about T4300 K This happens to coincide with thechemically most interesting temperature domainin air plasma chemistry for a CO2-rich early Earthatmosphere making meteors a relatively efficientsource of reduced molecules in an oxidizing at-mosphere (Jenniskens et al 2000)

Here we use the same measures of tempera-ture to consider the mass and speed dependenceat the peak of optical luminosity Extrapolationto smaller sizes provides the relevant conditionsfor meteoroids that supply most of the mass in-flux which are particles 150 mm in size with amedian speed of about 25 kms rather than themuch brighter or faster meteors that can be mea-sured by optical techniques

METHODS

A large sample of high-quality Leonid spectrawere measured during the 2001 Leonid MAC(Jenniskens and Russell 2003) all of which en-tered the Earthrsquos atmosphere at 716 6 04 kms

JENNISKENS ET AL82

METEOR AIR PLASMA TEMPERATURES 83

TABLE1

SUM

MARY

OFCURVE-O

F-G

ROW

THTEM

PERATUREM

EASU

REM

ENTS

FROM

METALA

TOM

LIN

EEMISSIONSFO

UND

INTHELIT

ERATURE

Mv

Vinf

Meteor

magnitude

(kms)

H(km)

Warm

Hot

Methoda

Reference

Com

etary

S6212

220

76

297

06

160

c-Fe

ICep

lech

a (196

4)760

0bCa

Ca1

Saha

Cep

lech

a (196

4)EN39

421

23

326

75

399

06

800

c-Fe

ICep

lech

a (197

1)438

06

100

Fe I no wak

eCep

lech

a (197

1)528

06

600

c-Ca I

Cep

lech

a (197

1)412

06

200

c-Na I

Cep

lech

a (197

1)14

000

6 3000

c-O I

Cep

lech

a (197

1)549

06

130

c-Fe

ICep

lech

a (197

3)Sa

mple

mdashmdash

mdash350

0ndash6000

Harve

y (197

1)EN 151

068

209

190

40

350

0ndash4700

s-Fe

Ca M

gBorov

icIumlka

(19

93)

10

000

bs-O I N

IBorov

icIumlka

(19

93)

a-C

aprico

rnid

205

230

84

360

0s-Fe

Mg

Borov

icIumlka

and

Web

er (19

96)

Sumav

a (197

4)222

mdash59

400

0ndash5000

s-Fe

Ca M

gBorov

icIumlka

and

Spu

rny(199

6)Perseid

211

610

85

440

0ndash480

0bs-Fe

Mg N

aBorov

icIumlka

and

Betlem (19

97)

10000

bs-Ca II O I

Borov

icIumlka

and

Betlem (19

97)

Perseid

205

610

85

450

0bs-Fe

I Mg I

Borov

icIumlka

and

Majden

(19

98)

10000

bs-O I N

2Borov

icIumlka

and

Majden

(19

98)

Asteroidal

EN32

281

210

289

71

290

0ndash3650

c-Fe

ICep

lech

a (196

5)EN36

221 (iron)

210

319

75

319

5535

0c-Fe

ICep

lech

a (196

7)550

06

370

c-Fe

ICep

lech

a (197

3)EN 130

960

211

300

75

554

06

160

c-Fe

ICep

lech

a (197

3)EN 322

81210

289

75

551

06

200

c-Fe

ICep

lech

a (197

3)EN 070

591

220

210

24

400

0ndash6000

s-Fe

I Ca I

Borov

icIumlka

et al (1998

)

a c c

urve

of grow

th s s

ynthetic spectrum fit to data

bAssum

ed

T(K)

JENNISKENS ET AL84

TABLE2

NEW

TEM

PERATURE(IN

K) M

EASU

REM

ENTSD

ERIV

ED

FROM

THEO N

ANDN

2A

IRPLASM

AEM

ISSIONSSH

OW

NIN

FIG 2

TchemN IN

2TchemO IN

2

Source

Time (UT)

pO I (FW

)Vinf

TvibN

2TeN I (747742)

N 747

N 822

N 868

O 777

O 845

O 616

Gem

inid 200

1071528

470

0 (07)

36470

06

100

bmdash

431

06

10

mdashmdash

mdashmdash

~470

0Pe

rseid 1999

103526

542

5 (13)

61467

06

200

bmdash

434

56

10

mdashmdash

mdashmdash

~471

0Pe

rseid 1999

075238

530

0 (12)

61470

06

200

b410

06

300

432

06

10

mdashmdash

439

5mdash

~471

0Pe

rseid 2000

070057

226

0 (07)

61mdash

mdash432

06

20

mdashmdash

434

0mdash

mdashLeo

nid 199

8174054

96 (06)

72mdash

mdashmdash

443

56

20

445

56

30

mdash438

0mdash

Leo

nid 199

8174706

235 (075)

72mdash

mdash434

06

20

mdash440

06

20

438

0431

0mdash

Leo

nid 199

8180846

10300

(04)

72mdash

mdashmdash

mdash442

06

30

mdash424

0mdash

Leo

nid 199

8195613

50 (07)

72mdash

mdash440

06

50

444

06

20

mdash436

0427

0mdash

Leo

nid 200

1a090557

170 (06)

72mdash

mdashmdash

mdashmdash

mdashmdash

454

0Leo

nid 200

1a091905

540

0 (04)

72605

06

200

bmdash

437

66

10

mdashmdash

mdashmdash

465

0Leo

nid 200

1a094225

313

0 (08)

72612

06

100

b435

56

200

mdashmdash

mdashmdash

mdashmdash

Leo

nid 200

1100430

568 (07)

72mdash

mdashmdash

448

06

10

446

06

30

440

0433

6mdash

Leo

nid 200

1101516

325 (08)

72mdash

mdash438

06

10

433

66

10

434

06

10

430

3414

0mdash

Leo

nid 200

1104458

600 (09)

72445

06

50b0

443

06

500

mdash447

56

10

mdash431

0434

7mdash

Leo

nid 200

1104702

386

0 (04)

72mdash

mdashmdash

mdash450

06

40

mdash435

0mdash

Leo

nid 200

1105231

620 (09)

72559

06

50b

bmdash

mdashmdash

mdashmdash

mdash447

0Leo

nid 200

1105643

203 (06)

72mdash

mdashmdash

458

06

20

459

06

20

442

0431

4mdash

Leo

nid 200

1111100

148 (075)

72mdash

mdashmdash

445

5448

6440

0439

0mdash

Leo

nid 200

1113243

48 (06)

72mdash

mdash443

06

50

453

0mdash

444

0439

5mdash

Leo

nid 200

1113349

164 (04)

72mdash

mdashmdash

450

0457

0mdash

438

0mdash

Leo

nid 200

1a123506

168

6 (055)

72mdash

432

06

100

434

76

10

mdashmdash

441

6mdash

mdash

FW full width at ha

lf m

axim

um

a Secon

d ord

er

bCon

tamination from

CaO


In addition a small number of spectra from theannual Perseid (606 6 06 kms) and Geminid(363 6 06 kms) showers were measured in re-cent ground-based campaigns which are in-cluded in order to address the issue of the dependence of excitation temperatures on mete-oroid speed The two-stage thermoelectricallycooled CCD slit-less meteor spectrograph for as-tronomy and astrobiology (ldquoASTROrdquo) and its in-strumental properties are described in detail inJenniskens et al (2004) In summary a Pixelvisioncamera with two-stage thermoelectrically cooled1024 3 1024 pixel back-illuminated SI003ABCCD with 24 3 24 mm pixel size (245 3 245 mmimage region) was used in un-intensified modeto keep as high a spectral resolution as possibleTo limit the readout time (dead time) to less than1 s we recorded images with 43 binning in thedirection perpendicular to the dispersion direc-tion The imaging optics was an AF-S Nikkorf28300D IF-ED 300-mm telephoto lens whichprovided a field of view of 5 3 5deg In front of thelens was mounted an 11 3 11 cm plane trans-mission grating (35-54-20-660 by RichardsonGrating Laboratory Rochester NY with alu-minum coating) on a 12-mm BK7-type glass sub-strate We measured a dispersion of 540 lmmand a blaze wavelength of 698 nm (34deg) This con-figuration provided a full second order spectrumout to about 925 nm

RESULTS

ASTRO was deployed onboard the US AirForce 418th Flight Test Squadron-operated NKC-135 ldquoFISTArdquo research aircraft over the continen-tal United States during the 2001 Leonid MAC

mission at the time of the intense November 181040 universal time (UT) Leonid storm peak (Jen-niskens and Russell 2003) This storm was causedby dust released from the 1767 return of comet55PTempel-Tuttle Twenty-three high-resolutionspectra were obtained 13 of which contain oneof the First Positive bands of N2 This sampledataset (Table 2 which includes four spectrafrom the 1998 Leonid MAC mission and fourspectra from ground-based campaigns duringthe Perseids and Geminids) more than doubledthe number of available spectra from prior mis-sions and ground-based campaigns While Table1 provides data for 25 to 222 magnitude fire-balls the newly acquired data are for fainter me-teors of magnitude 13 to 22 Each spectrum cov-ers a spectral range of about 200 nm with acentral wavelength determined by the positionof the meteor on the sky relative to the viewingdirection of the camera

Figure 1 shows the raw data of one of our bestresults a Leonid meteor spectrum in second or-der that is centered on the Dn 5 12 First Positiveband of N2 meaning the transitions from the B RA electronic state of N2 that also correspond to achange of 12 in the vibrational quantum number(Herzberg 1950) The meteor moved from bot-tom to top The images from background starswere removed by subtracting an image takenshortly after the meteor On top of the broad dif-fuse emission of N2 are lines of atomic nitrogenNearly all of the detail in this spectrum is due tothe various rovibrational transitions in the nitro-gen molecule Atomic lines from ablated metalatoms are few and faint in this part of the spec-trum

Figure 2 shows the reduced spectra derivedfrom data such as that shown in Fig 1 acquired

FIG 1 Spectrum of the 094225 UT Leonid meteor in the range 699ndash768 nm in second order Most emission linesare rotational and vibrational structure in the Dn 5 12 First Positive band of molecular nitrogen Superposed areemission lines from atomic nitrogen and some meteoric metal atoms including third order lines of iron (466ndash512 nm)Background star images have been removed The wavelength scale runs left to right the meteor moved from bottomto top increasing in brightness The spectrum is terminated by the end of the exposure

during the 1998 (4) and 2001 (13) missions andduring the 1999 (2) and 2000 (1) Perseid and 2001(1) Geminid ground campaigns Superposed oneach reduced spectral plot shown in Fig 2 is thecalculated emission line spectrum for N2 as dis-cussed in the next section The data reductionprocess is described in Jenniskens et al (2004) Insummary each spectrum was reduced by (1) sub-

tracting a background frame taken a second afterthe meteor to remove stars and dark current (2) correcting the rows for vigneting (flatfielding)(3) removing cosmic ray hits and star residues(4) aligning and averaging the CCD rows to cor-rect for the meteorrsquos motion relative to the direc-tion of dispersion and (5) correcting the result-ing one-dimensional spectrum for the instrument

JENNISKENS ET AL86

FIG 2 Meteor spectra containing oneof the N2 First Positive bands Each graphshows the meteor intensity corrected forinstrument response versus wavelength(nm) in arbitrary units Superposed is amodel spectrum displaced upward forclarity

spectral response function The spectrum ob-tained by reducing the image of Fig 1 is shownin the lower left corner of Fig 2

The B R A electronic state of N2 is shown inthe energy diagram of Fig 3 The different rangesof wavelengths in individual spectra cover thedifferent changes in vibrational quantum num-bers between upper and lower state Dn 5 11 (860nm) 12 (770 nm) 13 (680 nm) and 14 (590 nm)bands from the First Positive band system of N2(Herzberg 1950) Superposed on the N2 bands aregroups of atomic nitrogen lines at 746 nm 822nm and 869 nm and unresolved groups ofatomic oxygen at 6158 nm 7773 nm and 8447nm Only the Leonid spectra which were ob-tained at altitude are free of water bands origi-nating from absorption in the Earthrsquos atmosphere(called telluric) which cover significant parts ofthe near-infrared spectrum In addition all spec-tra near 760 nm contain the telluric oxygen A band while ground-based observations alsoshow the telluric oxygen B band at 680 nm

Temperature measurements

Each reduced spectrum was then comparedwith the spectrum generated using an air plasmaemission model The expected relative abundancefor each relevant compound was calculated by as-

suming LTE for an air plasma in equilibrium withits surroundings at a pressure of 1026 atmos-pheres (95 km altitude) The expected emissionintensities of the First Positive bands of N2 andthe atomic nitrogen and oxygen lines were cal-culated with the NEQAIR2 program (as updatedin Laux 1993) and using the 1966 National Bu-reau of Standards database of Einstein coeffi-cients for the atomic lines The results were com-pared with emission intensities calculated withan updated SPECAIR program which containsthe more modern 1996 NIST database (CO Lauxet al manuscript in preparation) The lattershowed differences of 20ndash30 in the Einstein co-efficients of the relevant N I lines leading to sys-tematically lower temperatures by 20ndash40 K

The next step in the data analysis involvedmatching the calculated N2 molecular band to theobserved spectrum and subsequently matchingthe atomic emission lines relative to the best fitof the N2 band (ie shown as superposed plotsfor each example shown in Fig 2) For each emis-sion line a best-fit temperature was calculatedfrom interpolated intensities in the spectrum at4100 4300 4500 and 4670 K The best-fit in-terpolated temperature Tint was calculated bymatching the sum of intensities expected for eachtemperature component to the observed spec-trum

Tint 5 2(Ek)ln [a(2EkTa) 1

b(2EkTb)](a 1 b) (1)

where E is the upper energy level Ta and Tb arethe lower and higher temperature and a and bare constants used to find the interpolated meansThe results are summarized in Table 2 Resultsdiffer slightly from those of Jenniskens et al(2000) because of a better treatment of the in-strument response functions Small errors in theposition of the N2 band result in systematic dif-ferences in all of the emission line temperaturesfor a given spectrum In general the calculatedtemperatures were determined to vary within asmall range for each diagnostic in most cases lessthan 6100 K with small systematic variations onthe order of 100 K between different diagnosticemission line (and band) ratios For example the845 nm line of O I (relative to the N2 bandstrength) suggests a lower temperature than the822 nm line of N I These temperatures arethought to describe the chemical equilibrium in


FIG 3 Simplified energy level diagram of the nitro-gen molecule The First Positive band originates in theB R A transition

the plasma which determines the abundances ofthe various compounds

Another temperature measurement followsfrom the shape of the band profile itself Eachband is well resolved with band heads from in-dividual vibrational transitions The Dn 5 12band in Fig 2 for example has peaks represent-ing the vibrational (02) (13) (24) etc transi-tions the width of each peak determined by themoleculersquos rotational level populations The slopeof the molecular band at high vibrational levelschanges slightly with increasing vibrational tem-perature of the N2 molecule The intensity slopeat low wavelengths is proportional to the popu-lation distribution of the vibrational levels By fit-ting the low wavelength slope of the band to themodel spectra we were able to estimate the vi-brational temperature of N2 (Fig 4)

Some of our spectra showed enhanced emis-sion in the Dn 5 12 band contour at short wave-length above that expected from LTE At least insome of these cases we suspect that excess emis-sion near 620 nm (Fig 2) may be molecular emis-sion from CaO which peaks at 620 nm FeO emis-sion may be present as well Indeed thesesuspected oxide emission bands are strong inthose meteor spectra that have a strong metalatom line component (indicated by footnote b inTable 2) Such contamination can result in mis-leading results for the vibrational temperature

Finally we measured electronic excitation tem-peratures for the nitrogen lines from the relative

line intensity ratio of the 746 nm N I lines Themain 3p4pndash3p4s transitions (vacuum wavelengths74257 74443 and 74704 nm) have the same1200 eV excitation energy but there are smallcontributions from higher levels at 1367 eVThese lines are so close together in the spectrumthat small errors in the instrument response andflatfielding have no effect on the outcome This isimportant because the observed variations aresmall (Fig 5)

In principle the O I lines at 777 nm (1074 eV)and 845 nm (1099 eV) could be used to measure

JENNISKENS ET AL88

FIG 4 Variation of the shape of the model N2 bandprofile for different assumed vibrational temperaturesModels at 4500 and 4300 K are superposed on the ob-served spectrum of the 104726 UT Leonid meteor of No-vember 18 2001

FIG 6 Gradual increase of temperature with decreas-ing height as observed from the variation of the NN2intensity ratio (expressed as the corresponding temper-ature in an LTE model) as a function of height for twometeors

FIG 5 Response of the N I line ratio to variations inthe electronic excitation temperature The observationsof the 104726 UT Leonid meteor of November 18 2001after subtraction of the N2 band profile are comparedwith LTE models at 4300 and 4500 K

the electronic excitation temperature of oxygenatoms (Park et al 1997) but those lines are farenough apart in wavelength to be sensitive to any potential intensity calibration uncertaintiesMoreover only four of our spectra show bothoxygen lines The 616 nm line has a significantlyhigher excitation energy (1275 eV) but it was notmeasured simultaneously with the other O I lines

Temperature dependencies with height mass and speed

The various temperatures derived here aremean values for a part of the meteor trajectory

in particular that part of the trajectory that wasbright enough to be recorded This tends to co-incide with the peak of the meteorrsquos luminositynear 85ndash115 km

There was a small gradual increase in theplasma temperature with altitude judging fromthe relative increase in the strength of the N Ilines Figure 6 shows results for the 2001 Leonidof 094225 UT (Fig 1) and the 1999 Perseid spec-trum at 075238 UT The observed trend (between115 and 85 km) was T (K) 270 3 H (km)

We find that all temperatures derived from theN I N2 ratio were in a small range from 4300 to4700 K with most results between 4300 and4400 K All results are summarized in Fig 7 as afunction of visual peak magnitude of meteorbrightness The observed trend with meteor mag-nitude was very weak Earlier data summarizedin Table 1 are also shown Those early data scat-ter widely but center around the same value

The brightness (and meteoroid mass) depen-dence of the chemical equilibrium temperaturesfor individual nitrogen and oxygen lines in rela-tion to the molecular band of nitrogen is sum-marized in Fig 8 Each set of data showed a grad-ual increase of temperatures to smaller meteorbrightness but the trend is very weak with aslope of only about 115 Kmagnitude (Table 3)The result disagreed significantly with the resultsof Ceplecha (1964) who derived a 10 timessteeper dependence versus peak bolometric mag-nitude from the variation of excitation tempera-ture in the flare of a bright fireball T (in K) 5(4427 6 399) 1 (151 6 47)M (in magnitude)


FIG 7 Temperature as a function of meteor peak vi-sual magnitude Small dots show our results while largesymbols are data reported in the literature Literature dataare derived from curve-of-growth analysis of metal atomablation emissions and fits of optically thin emission tometeor spectra (Table 1)

FIG 8 Chemical equilibrium temperatures as a function of the meteor peak visible magnitude Right panel Re-sults from oxygen lines at 616 777 and 845 nm Left panel Results from nitrogen lines at 747 822 and 868 nm

The temperatures calculated for the vibrationalexcitation of N2 electronic excitation for N andthe chemical equilibrium temperatures werefound to be very similar (Fig 9) This result im-plies that the air plasma is nearly in LTE

Small deviations from LTE were determined tobe present in the data The O I 777 and 865 nmlines were systematically weaker than predictedfrom an LTE plasma while the O I 616 nm lineis stronger (Fig 8) Such deviations are importantbecause they refer to the most common compo-nents in the air plasma Less abundant com-pounds including organic molecules from themeteor ablation exhibited a chemistry that islikely more non-LTE due to less frequent colli-sions Some of those systematic deviations maybe due to uncertainties in the Einstein coefficientsWith the newer values the results for 868 and 822nm N I lines would move 10ndash20 K closer to thosederived from the 747 nm lines

We also checked the dependence of the excita-

tion temperature on meteor entry speed fromfour spectra obtained in ground campaigns out-side the Leonid season The sample consisted ofthree Perseid spectra obtained in ground-basedcampaigns in 1999 and 2000 and one Geminidspectrum obtained in December 2001 The spec-tra were distorted by telluric absorptions of wa-ter vapor and the atmospheric O2 B absorptionband which affected the shape of the nitrogenband at 750 nm However precise measure-ments made on the few high signal-to-noise spec-tra acquired to date indicated that the excitationtemperatures were the same as those of fasterLeonid meteors of similar brightness (Fig 10)

DISCUSSION

Oxygen and nitrogen lines are from warm plasma not hot

The results in this paper are contrary to previ-ous findings on several points BorovicIuml ka (1994)

JENNISKENS ET AL90

TABLE 3 TEMPERATURE (IN K) DEPENDENCE AS A FUNCTION OF MAGNITUDE FOR CHEMICAL

EQUILIBRIUM LINE RATIOS RELATIVE TO MOLECULAR NITROGEN EMISSION

Ej (eV) Intensity (arbitrary units) T (m 5 0) DTmagnitude Correlation coefficient

N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075

FIG 9 Comparison of different temperature indica-tors including vibrational excitation ( s ) electronic ex-citation ( n ) and chemical equilibrium in the airplasma ()

FIG 10 Velocity dependence of meteor plasma tem-perature Symbols as in Fig 7 Speeds for natural mete-oroids in elliptic orbits are in the range 11ndash72 kms

assigned the O I (7774 nm) band ldquowithout doubtrdquoto the high-temperature T 10000 K componentresponsible for Ca1 and Mg1 emissions mainlybecause of atomic oxygenrsquos relatively high exci-tation energy of 1074 eV Indeed Ceplecha (1971)concluded on the basis of the O I curve-of-growthanalysis that the oxygen and nitrogen lines cor-respond to a high-temperature (T 5 14000 K)emission However Leonids and Perseids areknown to display strong Ca1 emission aboveabout 0 magnitude With no significant change inthe O IN2 and N IN2 line ratio with increasingmeteor brightness this implies that the source ofthe hot T 10000 K component thought to be re-sponsible for Ca1 emission lacks significant airplasma emissions of atomic oxygen and nitrogenIndeed the match of model and observed airplasma emissions at low temperature implies thatthe oxygen and nitrogen lines are part of thewarm T 4400 K component that also producesthe molecular N2 emission Although faster me-teoroids have stronger Ca1 lines Leonids do notshow a significantly different N IN2 ratio thanGeminids that enter at half the speed We con-clude that the high-energy atomic lines could beobserved at low temperatures in our spectra be-cause oxygen and nitrogen atoms were so muchmore abundant in the air plasma than the metalatoms not because they were excited to higherenergies

The optical thickness of the air plasma

The constant O IN2 and N IN2 intensity ra-tios also imply that the oxygen and nitrogen linesare not optically thick It was previously thought(Ceplecha 1964 BorovicIuml ka and Majden 1998)that the abundance of emitting atoms was suffi-ciently high for other atoms in the gas to absorbthe emitted radiation These earlier investigationssuggested that the more intense metal atom linesof meteors as faint as 15 magnitude show suchself-absorption If self-absorbed a progressivedecrease of N I and O I line intensity relative tothat of the weaker molecular N2 band would havebeen expected with increasing meteor brightnessThat would have produced a false increase oftemperature with meteoroid mass but the oppo-site (if any) was observed

If the plasma was optically thick in the mostintense lines but not in fainter lines then the ob-served radiation would emerge from differentdepths within the plasma volume deeper so for

the lines that originated from higher excitationenergies Within measurement uncertainty thetemperature was not a function of the energy ofthe upper level as expected for an optically thickplasma (Table 3) Instead the meteor air plasmaemission was well described by an optically thinplasma for all observed lines (for meteor magni-tudes of ndash2 and less)

We conclude that the meteor plasma does notbecome significantly hotter with increasing me-teoroid mass over the range 1025ndash1 g This con-clusion is likely valid for an even wider massrange The excitation temperatures derived fromthe metal atom line emissions of bright fireballsby Ceplecha (1973) give very similar tempera-tures (Fig 7) These data suggest that the varia-tion of meteor plasma emission temperature formeteoroids that range in mass from 1025 to 106 gis at most a few 100 K

Introducing possible different phases of meteoricair plasma

The finding that the meteoroid plasma emis-sion temperature does not vary more than a few100 K for meteoroids that range in mass over 11orders of magnitude is perhaps too large a rangeto be consistent with our previous suggestion thata bigger or faster meteor would simply create alarger vapor cloud and thus create more plasmarather than a hotter plasma (Jenniskens et al2000) It is unclear how the measured tempera-ture can remain unchanged when the physicalconditions change from a rarefied flow regime for0 magnitude meteors to a continuous flow regimeaccompanied by the formation of a bow shock for210 magnitude fireballs

One possible explanation is that the plasma tem-perature reflects a balance between heating andcooling in the gas a process similar to the one thatcauses the cool warm and hot phases of the in-terstellar medium which have widely differenttemperatures but are in pressure equilibrium witheach other (Begelman 1990) Field et al (1969) pro-duced the first model for a two-phase equilibriumof the interstellar medium in which radiative cool-ing is balanced by cosmic ray heating The twophases in the model included cold dense clouds atT 100 K and a second intercloud medium withT 10000 K with a third unstable phase T 5000K only expected in transient phenomena

The characteristic temperatures of the warmand cold thermal phases are insensitive to the de-


tails of the heating processes Instead they reflectthe rate of cooling at various temperatures L(T)called the cooling curve a function of the ener-gies of the resonance and fine-structure lines thatare responsible for the cooling of the gas If thegas is in pressure equilibrium the instability cri-terion is (dLdT)p 0 In a meteor plasma theheating is caused by collisional excitation of mol-ecules with the ambient air During radiationthere is still a significant translational velocity dif-ference (5 kms) between air plasma and am-bient air in the direction along the meteorrsquos path

A necessary and sufficient condition for the ex-istence of a multiphase equilibrium is that the sys-tem be in pressure equilibrium and thermally un-stable over a finite range of temperature Coolingof the plasma occurs primarily through energytransfer in collisions of atoms with molecules andin radiation from those molecules Hence thepresence of the molecules has a strong cooling ef-fect that will tend to lower the temperature andcreate more molecules thereby creating a ther-mally unstable condition Thus once there is pres-sure equilibrium with the surrounding mediumit is possible that different phases will form in themeteor plasma at temperatures just above the dis-sociation temperature of O2 and N2 While thosephases may not be thermally stable they occurbecause the cooling time scales are long in com-parison with the occurrence of the meteor phe-nomenon

If a mole fraction of 1022 N2 molecules is suf-ficient to create a warm phase at T 4300ndash4450K then a cold phase may exist at the OO2 equi-librium around 2240ndash2300 K Similarly a hotphase may be associated with the N1N andO1O ionization equilibrium at around 8000 Kwith the atoms being more effective radiators

The very small temperature increase with de-creasing altitude (Fig 6) is not a strong argumentagainst such a multiphase model of the airplasma Expansion of the plasma at higher alti-tudes would be expected to cause adiabatic cool-ing thus affecting the equilibrium temperature

IMPLICATIONS

The consistency of excitation temperaturesmakes it possible to extrapolate the observedtrends to fainter meteors of sizes relevant for themass influx at the time of the origin of life If weallow for the weak positive trend with increasing

mass the temperature would be T 5 4500 K fora typical 150 mm Leonid meteoroid (at 112 mag-nitude) Sporadic meteors with V 5 25 kms ofsimilar size would have a magnitude of about116 (Jenniskens et al 1998) Hence temperaturesmeasured for visual meteors are representativefor conditions in the meteors associated with150 mm grains responsible for the peak of themass influx

This 4500 K temperature is favorable for at-mospheric chemistry and will be a factor in thechemistry of ablated organic molecules In par-ticular under LTE conditions atmospheric CO2will dissociate into CO and O and subsequentlyCO will dissociate into its atoms Between tem-peratures of about 3600 and 5000 K the carbonatoms will react with N2 and other molecules toform relatively high abundances of C2 C3 andCN (Jenniskens et al 2000) Yields are highest forT 4100 K The atmospheric carbon atoms canalso react with meteoric organic matter to formlonger carbon chains At the same time nitrogenradicals peak in abundance above 4000 K and canreact with other meteoric organic molecules toform stable CN and C O bonds in reactionswith oxygen atoms and CO2 molecules Thus or-ganic compounds can be created that are enrichedin nitrogen and oxygen compared with the or-ganic matter typically found in the (micro-) me-teorites that are studied in many laboratory in-vestigations How much this process counteractsthe extraction of carbon by reaction with O andOH radicals needs to be investigated in chemicalmodeling Such work is outside the scope of thispaper

The picture could significantly change if ourhypothesis of a multiphase medium is correct Inthat case the meteor plasma temperature may bedifferent in a different atmosphere If the plasmatemperature is lower this would favor the sur-vival of organic molecules that are ablated fromthe meteoroid On the other hand the NO fixa-tion rate which is mostly determined by oxygenatom attack with nitrogen molecules would besignificantly lower perhaps to the point of beingnegligible

In a Mars-like CO2-rich atmosphere on theearly Earth an equilibrium around 2170ndash2230 Kmight have been established just above theCOCO2 dissociation threshold or above theCOC threshold The latter is more likely how-ever because both CO and CO2 are good thermalemitters in which case the temperature would

JENNISKENS ET AL92

have been similar to that observed in todayrsquosEarth atmosphere

We conclude that the excitation and chemicalequilibrium temperatures in meteor plasma re-main in a narrow range for a wide range of en-try speed and initial mass Whether the reason forthis is the presence of a multiphase equilibriumcan be tested for example in future meteor ob-servations on Mars or in laboratory experimentsof artificial meteors in CO2- and O2-rich atmos-pheres

ACKNOWLEDGMENTS

The ASTRO instrument was developed withsupport of NASA Ames Research Centerrsquos Di-rectorrsquos Discretionary Fund in preparation of theLeonid MAC missions We thank Mike Koop andGary Palmer for technical support The 2001Leonid MAC was sponsored by NASArsquos Astro-biology ASTID (Astrobiology Instrument and De-velopment) and Planetary Astronomy programsThe mission was executed by the USAF 418th

Flight Test Squadron This work meets with ob-jective 31 of NASArsquos Astrobiology Roadmap(Des Marais et al 2003)

ABBREVIATIONS

CCD charge coupled device LTE local ther-modynamic equilibrium MAC Multi-InstrumentAircraft Campaign UT universal time

REFERENCES

Anders E (1989) Pre-biotic organic matter from cometsand asteroids Nature 342 255ndash257

Begelman MC (1990) Thermal phases of the interstellarmedium in galaxies In The Interstellar Medium in Galax-ies edited by HA Thronson Jr and JM Shull KluwerAcademic Publishers Dordrecht The Netherlands pp287ndash304

BorovicIuml ka J (1993) A fireball spectrum analysis AstronAstrophys 279 627ndash645

BorovicIuml ka J (1994) Two components in meteor spectraPlanet Space Sci 42 145ndash150

BorovicIuml ka J and Betlem H (1997) Spectral analysis oftwo Perseid meteors Planet Space Sci 45 563ndash575

BorovicIuml ka J and Majden EP (1998) A Perseid meteorspectrum J R Astron Soc Canada 92 153ndash156

BorovicIuml ka J and Spurny P (1996) Radiation study of two

very bright terrestrial bolides and an application to thecomet S-L 9 collision with Jupiter Icarus 121 484ndash510

BorovicIuml ka J and Weber M (1996) An alpha-Capricornidmeteor spectrum WGN J Int Meteor Org 24 30ndash32

BorovicIuml ka J Popova OP Golub AP Kosarev IB andNemtchinov IV (1998) Bolides produced by impactsof large meteoroids into the Earthrsquos atmosphere Com-parison of theory with observations II Beneov bolidespectra Astron Astrophys 337 591ndash602

Boyd ID (2000) Computation of atmospheric entry flowabout a Leonid meteoroid Earth Moon Planets 82-8393ndash108

Ceplecha Z (1964) Study of a bright meteor flare bymeans of emission curve of growth Bull Astron InstCzech 15 102ndash112

Ceplecha Z (1965) Complete data on bright meteor32281 Bull Astron Inst Czech 16 88ndash100

Ceplecha Z (1967) Spectroscopic analysis of iron mete-oroid radiation Bull Astron Inst Czech 18 303ndash310

Ceplecha Z (1968) Meteor spectra (survey paper) In In-ternational Astronomical Union Symposium No 33 Physicsand Dynamics of Meteors edited by L Kresdaggerk and PMMillman D Reidel Publishing Co Dordrecht TheNetherlands pp 73ndash83

Ceplecha Z (1971) Spectral data on terminal flare andwake of double-station meteor No 38421 (OndrejovApril 21 1963) Bull Astron Inst Czech 22 219ndash304

Ceplecha Z (1973) A model of spectral radiation of brightfireballs Bull Astron Inst Czech 24 232ndash242

Chyba CF and Sagan C (1992) Endogenous productionexogenous delivery and impact shock synthesis of or-ganic molecules An inventory for the origins of lifeNature 355 125ndash132

Cook AF and Millman PM (1954) Photometric analysisof a spectrogram of a Perseid meteor Contrib Domin-ion Observatory 2 250ndash270

Des Marais DJ Allamandola LJ Benner SA DeamerD Falkowski PG Farmer JD Hedges SB JakoskyBM Knoll AH Liskowsky DR Meadows VSMeyer MA Pilcher CB Nealson KH SpormannAM Trent JD Turner WW Woolf NJ and YorkeHW (2003) The NASA Astrobiology Roadmap Astro-biology 3 219ndash235

Field GB Goldsmith DW and Habing HJ (1969) Cos-mic-ray heating of the interstellar gas Ap J Lett 155L149ndashL154

Harvey GA (1971) The calcium H- and K-line anomalyin meteor spectra Astrophys J 165 669ndash671

Harvey GA (1973) Spectral analysis of four meteors InNational Aeronautics and Space Administration SP 319Evolutionary and Physical Properties of Meteoroids Pro-ceedings of IAU Colloquium 13 edited by CL Hemen-way PM Millman and AF Cook NASA Washing-ton DC pp 103ndash129

Herzberg G (1950) Spectra of Diatomic Molecules D VanNostrand Co New York

Jenniskens P and Butow SJ (1999) The 1998 Leonidmulti-instrument aircraft campaignmdashan early reviewMeteoritics Planet Sci 34 933ndash943

Jenniskens P and Russell RW (2003) The 2001 Leonid


Multi-Instrument Aircraft Campaignmdashan early reviewISAS Rep SP 15 3ndash15

Jenniskens P and Stenbaek-Nielsen HC (2004) Meteorwake in high frame-rate imagesmdashimplications for thechemistry of ablated organic compounds Astrobiology4 95ndash108

Jenniskens P de Lignie M Betlem H BorovicIuml ka J LauxCO Packan D and Kruumlger CH (1998) Preparing forthe 199899 Leonid storms Earth Moon Planets 80 311ndash341

Jenniskens P Wilson MA Packan D Laux COBoyd ID Popova OP and Fonda M (2000) Mete-ors A delivery mechanism of organic matter to theearly Earth Earth Moon Planets 82-83 57ndash70

Jenniskens P Schaller EL Laux CO Schmidt G andRairden RL (2004) Meteors do not break exogenousorganic molecules into high yields of diatomics Astro-biology 4 67ndash79

Laux CO (1993) Optical diagnostics and radiative emis-sion of air plasmas [PhD Dissertation] HTGL ReportT288 Stanford University August 1993 Stanford Uni-versity CA

Love SG and Brownlee DE (1993) A direct measure-ment of the terestrial mass accretion rate of cosmic dustScience 262 550ndash553

Park C S Newfield ME Fletcher DG Goumlkccedilen T andSharma SP (1997) Spectroscopic emission measure-ments within the blunt-body shock layer in an arc-jetflow In AIAA 97-0990American Institute of Aeronau-tics and Astronautics Reston VA p 18

Popova OP Sidneva SN Shuvalov VV and StrelkovAS (2000) Screening of meteoroids by ablation vaporin high-velocity meteors Earth Moon Planets 82-83109ndash128

Address reprint requests toDr Peter Jenniskens

Center for the Study of Life in the UniverseSETI Institute

2035 Landings DriveMountain View CA 94043

E-mail pjenniskensmailarcnasagov

JENNISKENS ET AL94

and Brownlee 1993 Jenniskens et al 2000) To-day only a small mass fraction (8) of mete-oroids in the 10ndash50 mm range survive entry inEarthrsquos atmosphere and arrive unaltered as mi-crometeorites (Anders 1989) An even smallerfraction of large meter-size asteroidal debris sur-vives in the form of meteorites The bulk of in-falling meteoroids is ablated in the Earthrsquos at-mosphere and in the process creates brief flashesof light called meteors

Some of the relevant chemistry occurs in awarm (T 4400 K) air plasma thought to resultfrom expansion into the ambient atmosphere ofimpinging air molecules after their interactionwith the meteoroid and its ablation vapor cloud(Boyd 2000 Jenniskens et al 2000 Popova et al2000) The physical conditions in this plasma setthe tone for the subsequent chemical evolution ofablated compounds and the atmosphere in themeteorrsquos path The initial impact excitation is ahighly non-local thermodynamic equilibrium(LTE) process but the product is a warm plasmawith air plasma and metal atom emissions thatare nearly in equilibrium (Jenniskens et al 2000)From this plasma the site of the bulk of hot-tem-perature chemistry the bulk of optical emissionsis observed The plasma only gradually coolswith distance away from the meteoroid (Jen-niskens and Stenbaek-Nielsen 2004)

Earlier excitation temperature measurementssummarized in Table 1 were based on analysis ofthe curve of growth of radiation from metal abla-tion lines in photographic spectra of bright fireballs(Ceplecha 1964 1968 Harvey 1973) The brightermetal atom lines of meteors as faint as 10 magni-tude are affected by self-absorption and dampingdue to significant absorption of emitted photons inthe plasma itself (Ceplecha 1973) Values of themetal atom excitation temperature of such fireballswere calculated to range from 3000 to 5500 K withno obvious dependency on meteor velocity ormass The only significant dependence was noticedby Ceplecha (1965) who found a decrease in theexcitation temperature as a function of meteor lu-minosity during a flare However these and othermeasurements of temperature are model depen-dent with early measurements later corrected up-wards to match the luminous efficiency by as-suming that the effective radiating volume increaseswith decreasing lower energy level of the excita-tion The adjusted values were around 5500 K andpertain to bright fireballs at a typical altitude of 75km and speed 32 kms (Ceplecha 1973)

More recent excitation temperatures were de-rived by fitting synthetic spectra of optically thinradiation (ie no self-absorption) to the metalatom ablation lines (BoroviIumlcka 1993) Borovi Iumlcka(1994) concluded that several lines including Ca1

and Mg1 originate from a separate hot gas com-ponent with T 10000 K This high-temperatureplasma makes up only 002 (in number not vol-ume) of the meteor vapor envelope in slow me-teors but accounts for more than 5 in fast me-teors BorovicIuml ka (1994) considered the O and Natomic lines in meteor spectra to be part of thishigh-temperature gas because of their high exci-tation energies (11ndash12 eV) Moreover results fromcurve of growth analysis of atomic oxygen linesby Ceplecha (1973) also pointed to an air plasmatemperature of about 14000 K

The recent introduction of cooled charge cou-pled device (CCD) detectors in meteor spectro-scopy (Jenniskens et al 2000) provides the near-infrared sensitivity and detector linearity necessaryto derive air plasma temperatures from the shapeof the first positive bands of molecular N2 (firstidentified by Cook and Millman 1954) and therelative intensity of the O and N atomic linesFrom the first such spectra recorded during the1998 Leonid Multi-Instrument Aircraft Campaign(MAC) (Jenniskens and Butow 1999) we foundthat the chemical equilibrium of N and N2 waswell described by an LTE temperature of about T4300 K This happens to coincide with thechemically most interesting temperature domainin air plasma chemistry for a CO2-rich early Earthatmosphere making meteors a relatively efficientsource of reduced molecules in an oxidizing at-mosphere (Jenniskens et al 2000)

Here we use the same measures of tempera-ture to consider the mass and speed dependenceat the peak of optical luminosity Extrapolationto smaller sizes provides the relevant conditionsfor meteoroids that supply most of the mass in-flux which are particles 150 mm in size with amedian speed of about 25 kms rather than themuch brighter or faster meteors that can be mea-sured by optical techniques

METHODS

A large sample of high-quality Leonid spectrawere measured during the 2001 Leonid MAC(Jenniskens and Russell 2003) all of which en-tered the Earthrsquos atmosphere at 716 6 04 kms

JENNISKENS ET AL82


TABLE1

SUM

MARY

OFCURVE-O

F-G

ROW

THTEM

PERATUREM

EASU

REM

ENTS

FROM

METALA

TOM

LIN

EEMISSIONSFO

UND

INTHELIT

ERATURE

Mv

Vinf

Meteor

magnitude

(kms)

H(km)

Warm

Hot

Methoda

Reference

Com

etary

S6212

220

76

297

06

160

c-Fe

ICep

lech

a (196

4)760

0bCa

Ca1

Saha

Cep

lech

a (196

4)EN39

421

23

326

75

399

06

800

c-Fe

ICep

lech

a (197

1)438

06

100

Fe I no wak

eCep

lech

a (197

1)528

06

600

c-Ca I

Cep

lech

a (197

1)412

06

200

c-Na I

Cep

lech

a (197

1)14

000

6 3000

c-O I

Cep

lech

a (197

1)549

06

130

c-Fe

ICep

lech

a (197

3)Sa

mple

mdashmdash

mdash350

0ndash6000

Harve

y (197

1)EN 151

068

209

190

40

350

0ndash4700

s-Fe

Ca M

gBorov

icIumlka

(19

93)

10

000

bs-O I N

IBorov

icIumlka

(19

93)

a-C

aprico

rnid

205

230

84

360

0s-Fe

Mg

Borov

icIumlka

and

Web

er (19

96)

Sumav

a (197

4)222

mdash59

400

0ndash5000

s-Fe

Ca M

gBorov

icIumlka

and

Spu

rny(199

6)Perseid

211

610

85

440

0ndash480

0bs-Fe

Mg N

aBorov

icIumlka

and

Betlem (19

97)

10000

bs-Ca II O I

Borov

icIumlka

and

Betlem (19

97)

Perseid

205

610

85

450

0bs-Fe

I Mg I

Borov

icIumlka

and

Majden

(19

98)

10000

bs-O I N

2Borov

icIumlka

and

Majden

(19

98)

Asteroidal

EN32

281

210

289

71

290

0ndash3650

c-Fe

ICep

lech

a (196

5)EN36

221 (iron)

210

319

75

319

5535

0c-Fe

ICep

lech

a (196

7)550

06

370

c-Fe

ICep

lech

a (197

3)EN 130

960

211

300

75

554

06

160

c-Fe

ICep

lech

a (197

3)EN 322

81210

289

75

551

06

200

c-Fe

ICep

lech

a (197

3)EN 070

591

220

210

24

400

0ndash6000

s-Fe

I Ca I

Borov

icIumlka

et al (1998

)

a c c

urve

of grow

th s s


bAssum

ed

T(K)

JENNISKENS ET AL84

TABLE2

NEW

TEM

PERATURE(IN

K) M

EASU

REM

ENTSD

ERIV

ED

FROM

THEO N

ANDN

2A

IRPLASM

AEM

ISSIONSSH

OW

NIN

FIG 2

TchemN IN

2TchemO IN

2

Source

Time (UT)

pO I (FW

)Vinf

TvibN

2TeN I (747742)

N 747

N 822

N 868

O 777

O 845

O 616

Gem

inid 200

1071528

470

0 (07)

36470

06

100

bmdash

431

06

10

mdashmdash

mdashmdash

~470

0Pe

rseid 1999

103526

542

5 (13)

61467

06

200

bmdash

434

56

10

mdashmdash

mdashmdash

~471

0Pe

rseid 1999

075238

530

0 (12)

61470

06

200

b410

06

300

432

06

10

mdashmdash

439

5mdash

~471

0Pe

rseid 2000

070057

226

0 (07)

61mdash

mdash432

06

20

mdashmdash

434

0mdash

mdashLeo

nid 199

8174054

96 (06)

72mdash

mdashmdash

443

56

20

445

56

30

mdash438

0mdash

Leo

nid 199

8174706

235 (075)

72mdash

mdash434

06

20

mdash440

06

20

438

0431

0mdash

Leo

nid 199

8180846

10300

(04)

72mdash

mdashmdash

mdash442

06

30

mdash424

0mdash

Leo

nid 199

8195613

50 (07)

72mdash

mdash440

06

50

444

06

20

mdash436

0427

0mdash

Leo

nid 200

1a090557

170 (06)

72mdash

mdashmdash

mdashmdash

mdashmdash

454

0Leo

nid 200

1a091905

540

0 (04)

72605

06

200

bmdash

437

66

10

mdashmdash

mdashmdash

465

0Leo

nid 200

1a094225

313

0 (08)

72612

06

100

b435

56

200

mdashmdash

mdashmdash

mdashmdash

Leo

nid 200

1100430

568 (07)

72mdash

mdashmdash

448

06

10

446

06

30

440

0433

6mdash

Leo

nid 200

1101516

325 (08)

72mdash

mdash438

06

10

433

66

10

434

06

10

430

3414

0mdash

Leo

nid 200

1104458

600 (09)

72445

06

50b0

443

06

500

mdash447

56

10

mdash431

0434

7mdash

Leo

nid 200

1104702

386

0 (04)

72mdash

mdashmdash

mdash450

06

40

mdash435

0mdash

Leo

nid 200

1105231

620 (09)

72559

06

50b

bmdash

mdashmdash

mdashmdash

mdash447

0Leo

nid 200

1105643

203 (06)

72mdash

mdashmdash

458

06

20

459

06

20

442

0431

4mdash

Leo

nid 200

1111100

148 (075)

72mdash

mdashmdash

445

5448

6440

0439

0mdash

Leo

nid 200

1113243

48 (06)

72mdash

mdash443

06

50

453

0mdash

444

0439

5mdash

Leo

nid 200

1113349

164 (04)

72mdash

mdashmdash

450

0457

0mdash

438

0mdash

Leo

nid 200

1a123506

168

6 (055)

72mdash

432

06

100

434

76

10

mdashmdash

441

6mdash

mdash

FW full width at ha

lf m

axim

um

a Secon

d ord

er

bCon

tamination from

CaO



RESULTS








JENNISKENS ET AL86









b(2EkTb)](a 1 b) (1)










JENNISKENS ET AL88


















DISCUSSION



JENNISKENS ET AL90




N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075


















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94


TABLE1

SUM

MARY

OFCURVE-O

F-G

ROW

THTEM

PERATUREM

EASU

REM

ENTS

FROM

METALA

TOM

LIN

EEMISSIONSFO

UND

INTHELIT

ERATURE

Mv

Vinf

Meteor

magnitude

(kms)

H(km)

Warm

Hot

Methoda

Reference

Com

etary

S6212

220

76

297

06

160

c-Fe

ICep

lech

a (196

4)760

0bCa

Ca1

Saha

Cep

lech

a (196

4)EN39

421

23

326

75

399

06

800

c-Fe

ICep

lech

a (197

1)438

06

100

Fe I no wak

eCep

lech

a (197

1)528

06

600

c-Ca I

Cep

lech

a (197

1)412

06

200

c-Na I

Cep

lech

a (197

1)14

000

6 3000

c-O I

Cep

lech

a (197

1)549

06

130

c-Fe

ICep

lech

a (197

3)Sa

mple

mdashmdash

mdash350

0ndash6000

Harve

y (197

1)EN 151

068

209

190

40

350

0ndash4700

s-Fe

Ca M

gBorov

icIumlka

(19

93)

10

000

bs-O I N

IBorov

icIumlka

(19

93)

a-C

aprico

rnid

205

230

84

360

0s-Fe

Mg

Borov

icIumlka

and

Web

er (19

96)

Sumav

a (197

4)222

mdash59

400

0ndash5000

s-Fe

Ca M

gBorov

icIumlka

and

Spu

rny(199

6)Perseid

211

610

85

440

0ndash480

0bs-Fe

Mg N

aBorov

icIumlka

and

Betlem (19

97)

10000

bs-Ca II O I

Borov

icIumlka

and

Betlem (19

97)

Perseid

205

610

85

450

0bs-Fe

I Mg I

Borov

icIumlka

and

Majden

(19

98)

10000

bs-O I N

2Borov

icIumlka

and

Majden

(19

98)

Asteroidal

EN32

281

210

289

71

290

0ndash3650

c-Fe

ICep

lech

a (196

5)EN36

221 (iron)

210

319

75

319

5535

0c-Fe

ICep

lech

a (196

7)550

06

370

c-Fe

ICep

lech

a (197

3)EN 130

960

211

300

75

554

06

160

c-Fe

ICep

lech

a (197

3)EN 322

81210

289

75

551

06

200

c-Fe

ICep

lech

a (197

3)EN 070

591

220

210

24

400

0ndash6000

s-Fe

I Ca I

Borov

icIumlka

et al (1998

)

a c c

urve

of grow

th s s


bAssum

ed

T(K)

JENNISKENS ET AL84

TABLE2

NEW

TEM

PERATURE(IN

K) M

EASU

REM

ENTSD

ERIV

ED

FROM

THEO N

ANDN

2A

IRPLASM

AEM

ISSIONSSH

OW

NIN

FIG 2

TchemN IN

2TchemO IN

2

Source

Time (UT)

pO I (FW

)Vinf

TvibN

2TeN I (747742)

N 747

N 822

N 868

O 777

O 845

O 616

Gem

inid 200

1071528

470

0 (07)

36470

06

100

bmdash

431

06

10

mdashmdash

mdashmdash

~470

0Pe

rseid 1999

103526

542

5 (13)

61467

06

200

bmdash

434

56

10

mdashmdash

mdashmdash

~471

0Pe

rseid 1999

075238

530

0 (12)

61470

06

200

b410

06

300

432

06

10

mdashmdash

439

5mdash

~471

0Pe

rseid 2000

070057

226

0 (07)

61mdash

mdash432

06

20

mdashmdash

434

0mdash

mdashLeo

nid 199

8174054

96 (06)

72mdash

mdashmdash

443

56

20

445

56

30

mdash438

0mdash

Leo

nid 199

8174706

235 (075)

72mdash

mdash434

06

20

mdash440

06

20

438

0431

0mdash

Leo

nid 199

8180846

10300

(04)

72mdash

mdashmdash

mdash442

06

30

mdash424

0mdash

Leo

nid 199

8195613

50 (07)

72mdash

mdash440

06

50

444

06

20

mdash436

0427

0mdash

Leo

nid 200

1a090557

170 (06)

72mdash

mdashmdash

mdashmdash

mdashmdash

454

0Leo

nid 200

1a091905

540

0 (04)

72605

06

200

bmdash

437

66

10

mdashmdash

mdashmdash

465

0Leo

nid 200

1a094225

313

0 (08)

72612

06

100

b435

56

200

mdashmdash

mdashmdash

mdashmdash

Leo

nid 200

1100430

568 (07)

72mdash

mdashmdash

448

06

10

446

06

30

440

0433

6mdash

Leo

nid 200

1101516

325 (08)

72mdash

mdash438

06

10

433

66

10

434

06

10

430

3414

0mdash

Leo

nid 200

1104458

600 (09)

72445

06

50b0

443

06

500

mdash447

56

10

mdash431

0434

7mdash

Leo

nid 200

1104702

386

0 (04)

72mdash

mdashmdash

mdash450

06

40

mdash435

0mdash

Leo

nid 200

1105231

620 (09)

72559

06

50b

bmdash

mdashmdash

mdashmdash

mdash447

0Leo

nid 200

1105643

203 (06)

72mdash

mdashmdash

458

06

20

459

06

20

442

0431

4mdash

Leo

nid 200

1111100

148 (075)

72mdash

mdashmdash

445

5448

6440

0439

0mdash

Leo

nid 200

1113243

48 (06)

72mdash

mdash443

06

50

453

0mdash

444

0439

5mdash

Leo

nid 200

1113349

164 (04)

72mdash

mdashmdash

450

0457

0mdash

438

0mdash

Leo

nid 200

1a123506

168

6 (055)

72mdash

432

06

100

434

76

10

mdashmdash

441

6mdash

mdash

FW full width at ha

lf m

axim

um

a Secon

d ord

er

bCon

tamination from

CaO



RESULTS








JENNISKENS ET AL86









b(2EkTb)](a 1 b) (1)










JENNISKENS ET AL88


















DISCUSSION



JENNISKENS ET AL90




N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075


















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94

JENNISKENS ET AL84

TABLE2

NEW

TEM

PERATURE(IN

K) M

EASU

REM

ENTSD

ERIV

ED

FROM

THEO N

ANDN

2A

IRPLASM

AEM

ISSIONSSH

OW

NIN

FIG 2

TchemN IN

2TchemO IN

2

Source

Time (UT)

pO I (FW

)Vinf

TvibN

2TeN I (747742)

N 747

N 822

N 868

O 777

O 845

O 616

Gem

inid 200

1071528

470

0 (07)

36470

06

100

bmdash

431

06

10

mdashmdash

mdashmdash

~470

0Pe

rseid 1999

103526

542

5 (13)

61467

06

200

bmdash

434

56

10

mdashmdash

mdashmdash

~471

0Pe

rseid 1999

075238

530

0 (12)

61470

06

200

b410

06

300

432

06

10

mdashmdash

439

5mdash

~471

0Pe

rseid 2000

070057

226

0 (07)

61mdash

mdash432

06

20

mdashmdash

434

0mdash

mdashLeo

nid 199

8174054

96 (06)

72mdash

mdashmdash

443

56

20

445

56

30

mdash438

0mdash

Leo

nid 199

8174706

235 (075)

72mdash

mdash434

06

20

mdash440

06

20

438

0431

0mdash

Leo

nid 199

8180846

10300

(04)

72mdash

mdashmdash

mdash442

06

30

mdash424

0mdash

Leo

nid 199

8195613

50 (07)

72mdash

mdash440

06

50

444

06

20

mdash436

0427

0mdash

Leo

nid 200

1a090557

170 (06)

72mdash

mdashmdash

mdashmdash

mdashmdash

454

0Leo

nid 200

1a091905

540

0 (04)

72605

06

200

bmdash

437

66

10

mdashmdash

mdashmdash

465

0Leo

nid 200

1a094225

313

0 (08)

72612

06

100

b435

56

200

mdashmdash

mdashmdash

mdashmdash

Leo

nid 200

1100430

568 (07)

72mdash

mdashmdash

448

06

10

446

06

30

440

0433

6mdash

Leo

nid 200

1101516

325 (08)

72mdash

mdash438

06

10

433

66

10

434

06

10

430

3414

0mdash

Leo

nid 200

1104458

600 (09)

72445

06

50b0

443

06

500

mdash447

56

10

mdash431

0434

7mdash

Leo

nid 200

1104702

386

0 (04)

72mdash

mdashmdash

mdash450

06

40

mdash435

0mdash

Leo

nid 200

1105231

620 (09)

72559

06

50b

bmdash

mdashmdash

mdashmdash

mdash447

0Leo

nid 200

1105643

203 (06)

72mdash

mdashmdash

458

06

20

459

06

20

442

0431

4mdash

Leo

nid 200

1111100

148 (075)

72mdash

mdashmdash

445

5448

6440

0439

0mdash

Leo

nid 200

1113243

48 (06)

72mdash

mdash443

06

50

453

0mdash

444

0439

5mdash

Leo

nid 200

1113349

164 (04)

72mdash

mdashmdash

450

0457

0mdash

438

0mdash

Leo

nid 200

1a123506

168

6 (055)

72mdash

432

06

100

434

76

10

mdashmdash

441

6mdash

mdash

FW full width at ha

lf m

axim

um

a Secon

d ord

er

bCon

tamination from

CaO



RESULTS








JENNISKENS ET AL86









b(2EkTb)](a 1 b) (1)










JENNISKENS ET AL88


















DISCUSSION



JENNISKENS ET AL90




N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075


















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94



RESULTS








JENNISKENS ET AL86









b(2EkTb)](a 1 b) (1)










JENNISKENS ET AL88


















DISCUSSION



JENNISKENS ET AL90




N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075


















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94



JENNISKENS ET AL86









b(2EkTb)](a 1 b) (1)










JENNISKENS ET AL88


















DISCUSSION



JENNISKENS ET AL90




N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075


















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94








b(2EkTb)](a 1 b) (1)










JENNISKENS ET AL88


















DISCUSSION



JENNISKENS ET AL90




N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075


















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94







JENNISKENS ET AL88


















DISCUSSION



JENNISKENS ET AL90




N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075


















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94















DISCUSSION



JENNISKENS ET AL90




N I 747 1200 85 4352 137 077N I 822 1184 320 4440 130 022N I 868 1176 1480 4446 156 031

O I 616 1275 50 4633 2380 084O I 777 1074 1700 4371 56 022O I 845 1099 880 4293 157 075


















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94
















IMPLICATIONS






JENNISKENS ET AL92



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94



ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES









































JENNISKENS ET AL94














JENNISKENS ET AL94

the mass and speed dependence of meteor air plasma temperatures

Documents