VARIATION OF A METEOR SPECTRUM DURING THE TIME OF FLIGHT

NASA TECHNICAL TRANSLATION

NASA TT F-il+l+90

VARIATION OF A METEOR SPECTRUM DURING THE TIME

OF FLIGHT

by
v. A. Smirnov and Ya. A. Sm? rnova

Translation of "Ob izmenenii spektra meteora za vrera-
ya poleta", Astronomicheskl|. Vestnik, Vol. 3> No,
k, Oct. -Dec. 1969, pp. 230-235.

i(NASA-TT-F-1il490) VARIATION Of" A "me7eor"^ ' "
SPECTRUM DURING THE TIME OF FLIGHT V A
|Smirnov, et al (NASA) Mar. 1972 19 p CSCL
^ 03A

N72-32819

Unclas
G3/30 43310

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
WASHINGTON, D.C. 205^6 MARCH 1972

VARIATION OF A METEOR SPECTRUM
DURING THE TIME OF FLIGHT

B Y
V.A. SMIRNOV AND YA.A. SMIRNOVA

Translated June 1970 by F.D. Jones from
the Russian Ob izmenenii spektra meteora
za vremya poleta. Astronbmicheskiy vest-
nik, vol. 3, No. A (Oct-Dec 1969), pages
230-235. 21 references. English abstract
A70-20946, International Aerospace Abstracts.

AMES RESEARCH CENTER

National. AeAonaaticU) and Spacz Admlnl6tAation

MoUeXt Flold, CaliioKnla 94035

Tfum^JUvtion 6uppZlQ.d by

addis TRANSLATIONS International
Post Office Box i+097
Woodside, California 9U062
{kl5) 851-'10i+0

Ob izmenenii spektra meteora za vremya poleta (Variation of a Meteor
Spectrum during the Time of Flight) by V.A. Smirnov and Ya. A. Smirnova.
AstronomicheskiV Vestnik, vol. 3» no. 4 (I969) 230-255. Translated from
the Russian (June 1970) by F.D. Jones.

tumtm*****************!********************** *****************************
***********9*************************************************************
*t^i**********************************************************************

VARIATION OF A METEOR SPECTRUM DURING THE TIME OF FLIGHT

by
V.A. Smirnov and Ya. A. Smirnova

(Odessa Astronomical Observatory, Odessa, Ukrainian SSR)

The character of the dependence of a meteor's color on its
brightness is determined on the basis of a spectrophoto-
metric investigation of 4O meteor spectrograms. The mode
of the change of a meteor spectrum along its trajectory
in the atmosphere is compared with experimental data on
collisions between nitrogen atoms and meteoric gases.

Study of meteor spectrograms has shown that the character of the

1-5
spectra undergoes changes while the meteorite traverses the atmosphere

Investigations of the color indices of various meteors have revealed that

these are dependent on the meteor brightness. Since the brightness of any

meteor may change sharply during the time of flight, it is natural to expect
an analogous dependence for individual meteors. L.Yakkiya explained this
phenomenon by the Purkinje effect'^. In 1959, Z, Ceplecha used a panchromatic
emulsion to photograph meteors. The dependence of the color indices on the

brightness obtained by the latter author proved to be different from the
dependence obtained by Yakkiya and by J. Davis, who used an emulsion sensi-
tive to blue light. This showed that the meteors' color does in reality
change with the brightness^. It was found that the fainter meteors are
relatively redder than bright ones. The same dependence is observed during

the time of flight of one and the same meteor.

The use of photoelectric photometers has made it possible to
exclude the subjectivity of visual estimates of meteor magnitudes while
retaining an observation efficency close to the visual one . As is well
known, in visual observations, the brightness of a moving meteor is
compared with the bri. ghtness of fixed stars. Owing to a physiological
effect the estimate of the brightness, of the moving source, is low, and
this introduces a systematic error"^. A low value &t the brightness of visual
meteors is also obtained because the radiation of bright meteors (for
example the H and K lines of Call) lies beyond the short-wave limit

of twilight vision.

The fainter sections of the meteor trajectories thus appear more
red. As has been shown by the observations reported by Z. Ceplecha and

o
his associates , the reddening of meteors with decreasing brightness is

also found for telescopic meteors.

Study of meteor spectrograms with identified lines confirms the
above data. The present authors examined approximately 4O spectrograms
obtained at Odessa, Ashkhabad (A. P. Savrukhin, K.A. Lyubarskii), and

Simferopol (V.V. Martynenko, L.A. Pushnoi). All the spectra were photo-
graphed on pginchromatic aerial photography emulsions with the aid of short-
focus prism spectrographs (using mainly NAFA 3S/25 cameras).. The angles
between the flight direction and the dispersion varied from 30 to 90 .
The dispersion of the spectrograms is 200-700 X/mm. Transverse photo-
metric sections of the spectrograms were carried out on an MF-2 micro-
photometer, moving the slit along the dispersion direction. Measurements
were also conducted on individual lines silong the' meteor's trajectory.
•Readings were taken every O.Ol - O.O5 mm.

In the case of 6O-70% of the total number of meteor spectrograms
considered in the present work the brightness can be seen to undergo
sharp changes adong the trajectory. In flare-ups the brightness increases

by several stellar magnitudes, and this is in most instances accompanied

by a sharp rise in the intensity of radiation in the blue-violet region
of the spectrum (in about 50^ of the spectrograms). About 209o of the

meteors end in a single flare producing radiation in the range of 4400-3900 A.
In the cases indicated the type of the meteor spectrum changes from X to Y

(refs. 9 and 10).

About 20% is accounted for by faint meteors (0 , - 1 ), emitting
mainly in the red part of the spectrum, and Jfo by bright "pulsating"
meteorites, most often of Millraan's type Z. The spectrum of the several

flare-ups in these meteors changes along the trajectory in exactly the same
way at all wavelengths, but the red part of the spectrum flares up before
the other lines. It seems that the different modes of the change in the
type of the meteor spectrum with brightness are due to differences in the
meteor structure.

Photometry of meteor spectrograms makes it possible to detennine
the color indices using the formula:

CI--2.51g^i 'I ^^'

I E,QTP(n'] E^PVdX

1). .0

in which E| is the energy distribution in the meteor spectrum, E, is

the energy distribution in the spectrum of an AO star, V is the spectral
sensitivity of twilight vision, Q is the spectral sensitivity of the

photographic emulsion used, T is the spectral transmission of the aerial

11
camera optics, and P is the spectral transmission of the atmosphere .

The formula given by (1) is valid if the E| values have been corrected
for the spectral properties of the photographic system (atmosphere, optics,
photosensitive emulsion). If the E| values have not been corrected in
this way, the formula msiy be written as follows:

;y)v/^5'.x^'''I E^d\ : I E,PVd\

rci=::-.2.5fe

' ■■!■'■'■ ' "■; . ■'' ■ ' ti .
The values entering into equations (l) and (2) may be expressed

in relative units. This is particularly important in the determinations of
E. values*, since special photometric standards modeling meteor exposiures axe

Translator's note ; Sic. This should presumably be "E| "

12
required to obtain absolute E| values. In relative photometry of

meteor images it is necessary to determine as precisely as possible the

contrast coefficient of the characteristic curve. For this purpose we used

experimentally determined contrast coefficients in dependence on exposure

m the meteor exposure range .

V.I. Ivainikov ^ used formula (l) to determine the panchromatic
color index for one meteor; this proved to be equal to +0.2.

The described method of determining color indices was used by ~

the present authors to determine the meteor brightness dependence of the
color indices on the basis of spectrophotometric data. The color indices
obtained by equation (1) or (2) are special and express the difference
between the stellar magnitudes of meteors in spectral systems k and 1,
where k is the photographic system used agxi 1 is the visual system.

The quantities appearing in formula (2) were determined as
follows: The energy distribution in the spectrum of ocLyr was found

from the data reported in ref. 14» The sensitivity curve of the eye for

twilight vision was obtained from ref. 11. The spectral sensitivity of

panchromatic emulsion and the transmission of the aerial camera optics

(in relative units) were determined from the results of laboratory investi-
gations''^. In ref. 15 the authors also gave curves characterizing in
relative units the spectral properties of the entire photographic system
used, as well as the spectral transmission of the atmosphere.

The method described in ref. 12 was used in the processing of 12
spectra obtained by V.A. Smirnov at the Odessa Observatory. When process-
ing the spectra obtained at Simferopol and Ashkhabad use was made of
stellar images in view of the absence of laboratory standards, for the

construction of characteristic curves. . ,

As is well known, a parallel shift of the characteristic curve

with respect to the intensity coordinate does not affect the relative
intensities. During the determination in relative units of the intensity

distribution in meteor spectra it is not necessary to find the systematic
error on account of the difference between the exposures of the meteor and
the star. In the indicated method of measurement an error may only arise

owing to a difference in the contrast coefficients of the characteristic

12
curves obtained during meteor and stellar exposures .

The brightness at selected points along the meteor's trajectories
was determined as follows: Intensity distribution curves were constructed
in the same scale along the dispersion of selected sections of meteor spectro-
grams. These curves were then corrected on the basis of data on the spectral
properties of the photographic system in question. Numerical integration
was next used to find the integral intensities of certain spectral
intervals of the selected sections on the meteor spectrograms, and their
summation gave the meteor's brightness.

Figure 1 .

Dependence of the color index CI
on the logarithm of meteor
' brightness S

Figure 1 shows the values of CI in dependence on log S, where S

is the meteor brightness. As follows from the described method of treatment,
the values of S are proportionsil to the absolute exposure. From Figure 1

it may be seen that the points tend to form a linear dependence. This

behavior may be explained as follows: The change in CI vdth variation

of the meteor brightness is given by the logarithm of the ratio

r /■;. dh

r' AM'

(3)

The maximum of the V(\) curve is known to lie at 5IOO A. At

the same time, the maximum of the energy distribution in the spectra of
meteors is in the regions of 3900-4400 % and 580O-63OO 2. Therefore, the

numerator in (3) is as a rule greater than the denominator, and the

brighter the meteor the greater does this difference become. If the sign

in front of (3) is taken into account, it is clear that brighter meteors

have smaller values of CI,

Figure 2 .

Dependence of the emission in
the red (•) and blue (o) parts
of the meteor spectrum on the
logarithm of the brightness S

Figure 2 shows the variation of the emission in the "red" and

"blue" parts of the spectrum on the logarithm oif the brightness. The

10
emission in the "red" section was obtained by numerical integration in the
wavelength interval 65OO-52OO S, and that in the "blue" section between
5200 and 3900 S. It may be seen from this figure that, at the same
emission intensity in the "red" and "blue" parts of the spectrum, the "red"
spectrum appears at a lower meteor brightness than the "blue". These
results thus confirm once more the fact that the fainter sections of meteor
trajectory are usually more red, and the brighter sections are more blue.

An attempt may be made to interpret the changes in the form of
the meteor spectra during the time of flight in the atmosphere and the

changes in the glow intensity on the basis of experimental data . For

20

example, B. Yu. Levin notes that the physical processes occurring while

the meteor is moving over the initial part of its trajectory are saialogous
to the cathodic sputtering of metals bombarded by ions, and the intensive
emission of meteors resembles the emission cf an iron acre.

Let us consider the experiments that make it possible to estimate
indirectly the physical conditions of the excitation of a meteor spectrum,
i.e. the spectrum of elements making up the meteorite. Inelastic collisions

'O-

11

of heavy particles are processes accompanied by radiation that are closest
to the meteor phenomenon. Experiments of this kind make it possible in
principle to model the process of entry of the meteorite into the

atmosphere, accompanied by increasing opposing flow of the particles. As
a result of such experiments one can determine the excitation functions on

the effective atomic cross sections, which have been used in a number of

21
cases to calculate the meteor radiation .

The calculations show that the following conditions are necessary
in experiments on the bombardment with nitrogen atoms of a gas target corres-
ponding to the model of meteor gas:

1) the current density of the bombarding particles should be of

-2
the order of 1 a. cm ;

2) the energy of the particles should be about 200 ev.

In practice it is found that it is very difficult to obtain such conditions.
The difficulty lies in the need to produce a high current density at a

small potential difference.

In ref. 21 the experiments were conducted at a considerably lower
current density and greater particle velocities than in a meteor phenomenon.

12

Of course, the energy of the individual particles differed in these
experiments from the corresponding valuespf meteoric particles. In spite
of this, the values for the particle flux, depending on the product of the

density p and the velocity v of the particles, are close to those in a

meteor phenomenon, and this justifies the application to meteors of the

21
effective excitation sections obtained from Derbeneva's experxments .

The experiments quoted make it possible to assess the conditions

of meteor radiation. In point of fact, the character of the changes of the

effective excitation cross sections determined by experiments corresponds

to the change in the radiation of elements as the meteorite penetrates the
atmosphere. The curves of the change in the effective cross sections for
individusil spectral lines reproduce and explain the course of the changes

in the intensity of the same lines in meteor spectra. For. example, when

the energy of the bombarding particles is increased from 4OO to 1200 ev
(ref. 18), the effective excitation cross section for Nal 589O A increases
from 2 X 10~ to IO" cm . Analogous changes occur in the effective

cross

sections for Mgll 448O S, Mgl 5176 2, Cal 4226 2, and so on. As a

13
rule, the effective cross section increases with increasing energy of the
bombarding particles. The lines listed above are relatively the most
intense both in meteor and in experimental spectra.

The cross section for Call 3934 2 has the relatively high value

—17 2
of 2 X 10 cm at 600-700 ev. Consequently, this line should appear

brightly as soon as the interaction energy reaches the required 600 ev.

As is well known from the practice of spectroscopic analysis, the
emission of the spectral lines of a component in a gaseous mixture may affect
decisively the emission of the remaining components. The flare up of the
lines of one of the elements making up the meteorite may lead to the flare
up of the lines of other elements.

In conclusion, we shall mention one other point. Experimental

work has not as yet provided sufficient information on the effective
excitation cross sections of atoms, whose knowledge is necessary for the

determination of the luminous efficiency of the meteor and also of the
individual elements making up the meteorite. To a large extent this is
caused by the difficulty of the experiments themselves. Attention should

14
be turned to the possibility of obtaining the effective excitation cross

sections directly from meteor spectrograms, using the method of absolute

12
spectrophotometry of meteors . However, preliminary experiments should

determine- the model of an "artificial meteor" for which calculations

19
could be carried out

The discussed experiments will thus help to clarify the processes
that accompany meteor raddction. Interpretation of the meteor spectra and of
the radiation intensity is already possible on the basis of existing experi-
mental data.

Odessa Astronomical Observatory . Received ; April 10, I968.

After further improvement,
■ . : ■ resubmitted on March 12, I969.

15

REFERENCES

1. . S.N. Blazhko, On the spectrum of the meteor of August 12, 1917,
Astron. Zh. , 9, no. 3-4, (1932) 146-162.

2I A.N. Vyssotsky, A meteor spectrum of high excitation. Astrophys. J.,
91, no. 2 (1940) 264-266.

3. ' P.M. Millman, Meteor nev/s, J.R. Astron. Soc. Canada, ^1 0940) 3.

4. L. Yakkiya, Fragmentation as the cause of anomalies of faint
meteors. In "Meteory" [Meteors], Moscov;, 1959.

5. Z. Ceplecha, On the color index of meteors, BAC, 10 (1959) 39.

6. J. Davis, Photoelectric meteor observations and the color indices
and visual magnitudes of meteors, M.N., 126 (I963) 5.

7. V.I. Ivanikov, Errors in visual brightness estimates and the
color index of meteors, Byull. Komissii po kometam i meteoram,
no. 10 (1965) 13.

8. Z. Ceplecha, J. Grygar, and L. Kohoutek, Distribution of telescopic
meteors, BAC, 16 (1965) 2, 123.

9. V.A. Smirnov, Identification of the spectral lines of meteors with
the aid of laboratory dispersion standards, Astron. zh. , ^, no. 6
(1967) 1316-1324.

10. P.M. Millman, A general survey of meteor spectra. Smithsonian
Contributions to Astrophysics, 7 (1963) 119-127.

11. D, Ya. Martynov, Kurs prakticheskoi astrofiziki [Course of
Practical Astrophysics], Nauka, Moscow, I967.

12. V.A. Smirnov, On the photometric standardization and calibration
• of meteor spectrograms, Astron. vestnik, 5, no. 1 (1969).

13. V.I. Ivanikov, Spectral energy distribution and the color index
of one meteor, Byull. In-ta astrofiziki, no. 41-42 (1966) 26.

14. A.V. Kharitonov, Energy distribution in the spectrum of aLyr, Izv.
Astrofizicheskogo in-ta, 21 (1962) 52.

15. V.A. Smirnov and Ya. A. Smirnova, Spectrophotometric properties
of a system used for the photography of meteors. Inform, byull.
"Geofizika i astronomiya" , no. I4 (I969).

16. J.M. Sluthers, E. De Haas, and J. Kistemaker, Charge exchange
ionization and electron loss cross sections in the energy range
5 to 24 keV, , Physica 25 (1959) 1376.

16

17. J.M. Sluters, and J. Kistemaker, Excitation mechanism of Ar
ions in He, Ne, Ar, Kr, and Xe, Physica, 25 (1959) 1389.

18. S.H. Neff, Excitation in atomic collisions related to meteor
radiation, Astrophys. J., 140 (I964) 348-360.

19. J.F. Friichtenicht, J.C. Slattery, -and E. Tagliaferri, Meteor
luminous efficiency, Astrophys. J., 15I (1968) 2, 743-758.

20. B. Yu. Levin, Fizicheskaya teoriya meteorbv i meteornoe
veshchestvo v Solnechnoi sisteme [Physical Theory of Meteors

and Meteorites in the Solar System], Izd-vo AN SSSR, Moscow, 1956.

21. A.D. Derbeneva, Meteor emission coefficient, Astron. zh., ^ (I966) 2.

S u m m a r y

Intensity distributions at various points of a meteor trajectory are
constructed from 40 meteor spectra. The dependence of the color indices
CI, determined by formulas (I) and (2), on the meteor brightness S is
obtained. For equal emission intensity in the "red" and "blue" parts of the
spectrum, the "red" spectrum appears at lower meteor magnitudes than the
"blue". The variations of the meteor spectra during the motion are inter-
preted on the basis of experiments involving the bombardment, of a gaseous
target (meteor gases) with nitrogen atoms.