Jo~~rnal of Atmospheric andTerrestrial Physics, 1071, Vol.33,pp.197-203. Pergamon Press.Printed inNorthern Ireland
Auroral oval planetary energetics YA. I. FELDSTEIN P/o Akadamgorodok, Izmiran, Moscow, U.S.S.R.
and G. V. STARKOV The Polar Geophysical Institute,
Academy of Sciences, Apatiti, U.S.S.R.
(Received 17 June 1970) Ab&&--The aurora.
energy associated with a polar substorm is deduced from photographs
A MAGNETOSPHERIC substorm is accompanied by magnetospheric generation of intensive fluxes of electrons and protons entering the upper Earth’s atmosphere. Entrance of these particles causes a set of geophysical events called the polar substorm (AKASOFU, 1968). The most important phenomenon of themagnetospheric substorm is activation of auroras occurring along the oval due to entrance of electrons with E > 1 keV into the atmosphere. The data on the intensity of the luminescence and its extent make it possible to estimate the energy flux of the entering particles and hence to obtain an idea about the energy context of the phenomena. In O’BRIEN et al. (1964) and O’BRIEN (1964) the energy flux carried in by the electrons with E > 1 keV in to the aurora1 zones and the planetary energy release have been estimated as 4 erg/cm2 set and 4 x 1017-1018 erg/set respectively. In some cases the energy of entering flux may be 2000 erg/cm2 set (O’BRIEN et al., 1962).In AXFORD (1964) the total energy release in aurora has been estimated as 1017erg/set. KRASOVSKY (1967,1968) has shown that in the case of a special type of aurora where the radiation is predominantly in the red oxygen line the energy release may reach 300 erg/cm2 set as it did during the geomagnetic storm on February 11, 1958 and the maximum all-planetary flux may be 1.5 x 10zo erg/ sec. In this case the energy release in this emission reached 2 x 1O23erg for a period of ~10 hr. SHARP and JOHNSON (1968) have obtained the planetary energy release in both hemispheres for electron fluxes with E > 80 eV as 3 x 1016 erg/set at K, N 1 and 6 x 1017erg/set at 4 < K, Q 5. On the basis of the data on the brightness of auroras in individual emissions one may determine the energy carried in by the primary electron flux (DALQARNO, 1965) and the total energy release in the aurora1 zone at various magnetic activity levels. To obtain such estimates one should have the data on both the size of the aurora1 region and the aurora1 brightness. The size of the aurora region in zenith has been obtained in FELDSTEIN and STARKOV (1967) as a function of the intensity of the geomagnetic activity index Q. The variations in the aurora1 brightness along the oval for the appropriate Q index may be obtained by processing of ascafilms of the U.S.S.R. networkstations in which the photometric standard is imprinted. For this purpose the mean blackening of a section of the ascafilm was measured using a densitometer (NADUBOVICH, 1962). To eliminat,ethe influence of distortion and outside exposure at the horizon, only zenith 197
YA. I. FELDSTEXNmd G. V. STAREOV
anglesless than 60’ were used. The mean intensity of the luminescence from the surface of #IO” km*, was measured in this way. The calibration was made using a sensitometric standard imprinted on each film. The calibration curve was plotted on the basis of comparison of the readings of the densitometer for the frames with relative uniform blackening with the blackening of the corresponding steps of the calibrat#ion standard. A special study has shown that the estimate of the amoral brightness obtained in measuring the negative and positive copies differs not more than by 20 per cent. Such a small difference is due to the fact that the appearance of the discrete bright aurora1 form is usually accompanied by an increase in the background of all-sky luminescence and, hence, t,henegative copies may be used for our purposes. To obtain the aurora1 brightness for various sections of the oval zone the ascafilms from stations Pyramida (the day se&ion of the aurora1 oval) and Dixon The aseafilm from Murmansk station were (the night section) were measured. also used for the near midnight hours on disturbed days. The photographic density of the film was recalculated into the primary electron flux energy with the aid of the standard sensitometer using the following techniques. The sensitometer has a calibrated lamp with spectral distribution which after a special blue levelling filter, corresponds to colour temperature R - 6000°K. The blackening of the film A, at a distance I from the lamp can be determined from relation
where I is the current intensity in the lamp, F is the voltage, K is the proportionality coefficient ; the ray incidence angle is considered zero. The proportionality coefficient K can be determined from relation K = 0.34 (TV. TV . 7%)/t where 71 is the transmission of the blue filter; 7s is the transmission of a grey filter, 7, is the optical density of the n-th step of the standard; t is the exposure time. It was assumed that the film was sensitive to the radiation in the 3900-6700 A wavelength range (the short-wave radiation is absorbed with a glass filter and the opties S~~bstituting of the camera; the film is not sensitive to the longwave radiation). the known numerical values of I, V, 1, TV,TV,t and proceeding to erg/cm” see we shall obtain for the blackening beyond the n-th step of the standard A = 1.12 T, erg/cm2 sec. Assume that the photographic density in the film caused by aurora is the same as that beyond the n-th step of the standard. In this case t.he blackening can be determined by a set of radiations in spectral lines and hence
where (Pi is the transmission of the atmosphere, S, spectral sensitivity of ascafilm, This Q spectral transmission coefEeient of C-180 camera, IA-intensity of aurorae. relat,ion can be rewritten as 1.12 7, = 1, ; PA * 8, - 7.4* wbJ where I, is the intensity
of the 3914 A emission (INCAr,+).
In the calculations
Aurora1 oval planetary energetics
values of p1 were taken according to FRISHMAN (1959), S, according to STARKOV (1964a), TV according to STARKOV (1964b). The most important contribution to the C-180 camera film are from emissions 3914 8, 4278 d, 5577 A, 6300 8, 6364 A. According to CHAMBERLAIN (1961) the relation I(5577 A)/1(3914 A) = 1 and the intensities of the emission pairs I(3914 A)-I(4278 A) and I(6300 b)-I(6364 A) are in a functional dependence. For the day section of the oval where high auroras of A-type are often observed (EVLASHIN, 1961) the relation I(6300 8)/1(3914 8) is assumed to be 1.5 and for the night section of the oval it is 0.4. If one converts the values of I, into rayleighs (R) per cm2 set, it is possible to calculate the energy flux of ent’ering electrons which cause the observed luminescence. The effectiveness of the corpuscular flux energy conversion to the 3914 A radiation in air has been assumed to be 2 x 1O-3 (DALGARNO, 1965; DALGARNO et al., 1965; DAVIDSON, 1966; MILLER et al., 1968). Figure 1 shows the energy flux densities for the particles causing the aurora in the case of an increase in the intensity of magnetic disturbance evaluated by the magnetic activity index Q at the night side of the aurora1 oval at # N 65”. In the day hours the energy flux density changes from 0.45 to 1.4 erg/cm2 set as the magnetic activity index Q increases from 0 to 7. The root mean square error is ~0.3 erg/cm2 sec. The most pronounced increase in the energy flux density is observed at the mid-night hours where E changes from l-4 to 7-O erg/cm2 sec. The minimum variations occur at 0200-0600 hr LT and in t,his section of the oval the energy flux density is usually a maximum which is in good agreement with Sandford’s data (SANDFORD, 1964). These energy flux densities are several times higher than those measured directly onboard a satellite (JOHNSON et al., 1969; SHARP et al., 1969). The mean values of the particle energy flux were possibly underestimated because of fast satellite movement and small width of the aurora1 forms. On the day side of the oval the energy flux density in auroras is 4-5 times less than on the night side. This coincides with the results of the satellite measurements of particle fluxes (SHARP et al., 1968; EVANS, 1967) where the ratio of the energy fluxes at night and day hours was -5. CHUBB and HICKS (1970) measured the aurora1 radiation intensity in the far ultraviolet in the 1375-1500 A spectral range and determined the maximum luminescent intensity when crossing the aurora1 oval at night hours. At K, = 0 the luminescence intensity was 0.15 kilorayleighs (kR) at K, = 6 it was 2.12 kR. According to MILLER et al. (1968) the main contribution to the radiation in this spectral range is from the bands of the Lyman-Birge-Hopfield system of the N, molecule. If according to MILLER et al. (1968), one assumes the ratio of the 3914 A band intensity to the intensity of the 1375-1500 A spectral interval to be 4.9 and the effectiveness of the corpuscular flux energy transfer to the radiation 3914 A to be 2 x 10m3 we obtain that the energy flux is changed from 2 erg/cm2 set at K, = 0 to 27.8 erg/cm2 set at K, = 6 which exceeds somewhat the values obtained on the basis of the measurements of the all-sky camera frames. According to CHUBB et cd. (1970) the luminescent intensity at night hours is 2-3 times higher than at the day hours. It is known that the brightness of aurora1 forms may change by 3-4 orders of magnitude and the entering electron flux may change by 5-6 orders while according to the data of Fig. 1 the mean energy release is changed by less than one order
YA. I. FELDSTEIN and G. V.
Ii "E \" 4.2 0 z 6
Fig. 1. Energy density flux (E) of aurora1 electrons variation at the different parts of aurora1 oval and at the different &-indexes. (a) dayside sector of the oval; (b) nightside sector of the oval. Solid lines give average curves for 1000-1400 hr LT and 2200-0200 hr LT (experimental results denoted by circles), point-lines give those values for 1400-1800 hr LT and 0200-0600 hr LT (triangles), broken-lines give those values for 0600-1000 hr LT and 1800-2200 hr LT (crosses). as the magnetic
activity increases. This difference is due to the fact that this case not the brightness of the discrete form but the mean brightness of a considerable area of the sky was measured. The measurement technique is such that the brightness of the discrete forms was averaged by taking into account the aurora1 background increase in the intervals between the discrete forms over whole surface measured. It should be noted that the averaged amoral brightness intensities according to the data from CHUBB et al. (1970) or the mean electron energy fluxes according to SHARP et al. (1966) are changed by about one order within a broad range of changes in K,.
The area S covered by the discrete aurora1 forms was determined for each of six 4-hr sectors from the relation . 6, - 6, + $2 . sin Si = 8 .rrlP sin 6, ---2-~ 2
where R is the Earth’s radius; 6, and 6, are the angular distance of the equatorial and polar boundaries of the corresponding aurora1 oval section. The values of 6, and 6, were taken from FELDSTEIN and STARKOV (1967). Then the energy flux is Wi = Si . Ei. Figure 2 shows the values of Vi obtained. The mean values of the energy flux in the day hours was ~4 x 10ls erg/set; the minimum values were observed at 0600-1000 hr LT. During variations in the Q-index from 0 to 7 the 16-Z
16 0 i
15.2 t 15-o
Q 17.6 -
Fig. 2. Aurora1electronenerw flux (W) variationsat the differentpartsof auroral oval and at the different &-indexes. See Fig. 1 for descriptionsof the curves.
YA. I. FELDSTEINand G. V. STARKOV
change is somewhat less than one order. At the night side where the mean value is ~5.5 x 1016 erg/set, the most pronounced changes with changing the Q-index are observed near the mid-night hours (from 55 x 10n erg/set at Q = 0 to 3 x 101’ erg/set at Q = 7). The total energy flux entering both aurora1 ovals in the northern and southern hemispheres is changed from -9 x 1016 erg/set at Q = 0 to l-3 x 1Or8 erg/set
1gsH.y Fig. 3. Dependents of the total energy flux which is carried into the aurora1 ovals on the intensity of the polsr magnetic disturbance.
at Q = 7. Figure 3 shows the dependence of lg W on lg 6H where 611 is the equivalent amplitude of the Q-index. This dependence one can present as w -N 5.5 x 1oWgH where 68 is in gammas, and W is in erg/see. The data presented make it possible to estimate the integral energy flux entering the upper atmosphere along the oval in the quiet periods and during substorms. According to KHOROSHEVA (1962), FELDSTEIN (1966), FELDSTEIN and STARKOV (1967) in the case of quiet magnetic field the aurora1 zone may be approximated by a single arc along the day side and with two arcs along the night side of the oval which is a ring around the geomagnetic pole with a radius of 16”. Assume the width of the arcs to be 10 km, the aurora intensity to be class 2 (30 erg/cm2 set at the night side and 6 erg/cm2 set at the day side). Then the energy flux along the aurora1 oval is 3.6 x lOla erg/set which is close to the estimate obtained above of 4.5 x lOis erg/see to a single hemisphere at Q = 0. An estimate may also be made of the total energy carried into the upper atmosFor a bay-like disturbance of 1 hr phere by electron fluxes during a substorm. duration and a maximum amplitude of ~200 y which can be described by a set
Aurora1 oval planetary energetic8
of Q-indices 2,5,4,2 the total energy carried in will be -2 x 1021 erg. For a bay-like disturbance of 1.5 hr duration and a maximum amplitude of -600 y which can be described by a set of Q-indices 2, 7, 6, 4, 3, 2 the total energy flux will he ~4 X 1021 ergs. Thus the contribution of the energy of the primary electrons that serves to develop some aspects of the polar substorm is a significant part of t,he total energy of a magnetic storm. REFERENCES S-1.
AXFORD W. J.
Polar and &taynetospheric Substorm. Reidel, New York. Planet. Space Sci. 12, 45.
of the ,4urora
CHUBB T. A. and HICKS G. T. DALGARNO A.
KH~R~SQEVA 0. V. KRA~OVSKY V. I. KRA~~VSKY V. I. MILLER R. E., FASTIE TV. E. and ISLER R.C. NADUBO~I~R Y. A.
75,1290. Phenomena, p. 39.
O’BRIEN B. J. O’BRIEN B. J. and TAYLOR H. O’BRIEN B. J. and LAUGHLIN C. D. SANDFORD B. P. SHARP R. D., EVANS J. E. and JOHNSON R. G. SHARP R. D. and JOI~NSON R. G. SI~ARP R. D., CARR D. L. and JOHNSON
Pla9zet. Space Sci.
Planet. Space Sci. 14, 651. Space Sci. Rev. 7, 263.
1961 1966 1967 1959 1969
Geomag. C&Aeronomy 1,531. Planet. Space Sci. 14, 121. Planet. Space Sci. 15, 209. Optics Spectrosc. 6, 323. Atmosphwic Emissions (Edited
1962 1967 1968
J. geophys. Res. University
DALGARNO A., LATIMER I. D. and MCCONKEY J.W. DAVIDSON G. EVANS J. E., JOHNSON R. G., SHARP R. D. and RERAN J. B. EVLASHIN L. S. FELDSTEIN YA. I. FELDSTEIN YA. I. and STAREOV G. V. FRISIQZAN I. G. JOHNSON R.. G. and SR~RP R. D.
and Press, New York.
MCCORMAC), p. 219. Reinhold, York. Cfeomag. & Aeronomy 2, 839.
1964 1964 1962 1964 1966
Beomag. & Aeronomy 7, 945. Planet. Space Sci. 16, 47. J. geophys. Res. 73,3353. Trans. Yakutskogo FilialaSibirskogo Otdelenia Akademii Nauk, ser Jizicheskaja, p. 26. Nauka, Moscow. J. geophys. Res. 69,13. J. geophys. Res. 69,45. J. geophys. Res. 67,2667. J. Atmosph. Terr. Ph.ys. 26, 749. Pla,net. Space Sci. 14, 85.
J. geophys. Res. J. geophys. Res.
Cteo-Heliophys Effect Cosmic Rays Aurorae, p. 137. Nauka, Moscow. aeo-Heliophys Effect Cosmic Rays Aurorae, p. 130. Nauka, Moscow. Qeomag.& Aeronomy 7, 62.
R. G. STARKOV G. V. STAREOV
STARKOV G. V. and FELDSTEIN YA. I.