ISSN 00167932, Geomagnetism and Aeronomy, 2013, Vol. 53, No. 6, pp. 711–715. © Pleiades Publishing, Ltd., 2013. Original Russian Text © V.G. Vorobjev, A.S. Kirillov, Ju.V. Katkalov, O.I. Yagodkina, 2013, published in Geomagnetizm i Aeronomiya, 2013, Vol. 53, No. 6, pp. 757–761.
Planetary Distribution of the Intensity of Auroral Luminosity Obtained Using a Model of Aurora Precipitation V. G. Vorobjev, A. S. Kirillov, Ju. V. Katkalov, and O. I. Yagodkina Polar Geophysical Institute, Kola Scientific Center, Russian Academy of Sciences, ul. Fersmana 14, Apatity, Murmansk oblast, 184200 Russia email:
[email protected] Received April 3, 2013
Abstract—A model of auroral precipitation (AP) developed on the basis of statistical processing of DMSP F6 and F7 satellite data (Vorobjev and Yagodkina, 2005, 2007) was used for the calculation of the global distri bution of the auroral luminosity in different spectral ranges. The algorithm for the calculation of the integral intensity in bands N2 LBH (170.0 nm), ING N2+ (391.4 nm), 1PG N2 (669.0 nm), and (OI) 557.7nm emis sion is shown in detail. The processes of formation of electronically excited atoms O(1S) as a result of the transport of excitation energy from metastable state N2(A3Σ +u ), excitation of O(3P) by primary and secondary electrons, and dissociative recombination were taken into account to calculate the intensity of emission at 557.7 nm. A high correlation between the model distribution of the auroral luminosity in the UV spectral range and the observations of the Polar satellite is demonstrated. DOI: 10.1134/S0016793213060169
1. INTRODUCTION Empirical models of the planetary distribution of precipitating particle characteristics can be conve nient for studying (monitoring and forecasting) the peculiarities of the spatial distribution and brightness characteristics of the auroral luminosity during peri ods of magnetospheric perturbations. The first attempt to develop a global model of the auroral luminosity was made in (Ivanov et al., 1993). In their calculations, the authors used the model of precipitations suggested in (Spiro et al., 1982), in which the spatial distribution of the mean energies and energy fluxes of precipitating electrons was presented for four levels of magnetic activity based on the data of satellites AEC and AED. In the model suggested in (Spiro et al., 1982), as well as in the latest precipitation models, hourly value of the AE index and the Kp index were used to arrange satellite data relative to the level of magnetic activity. However, a satellite crosses the zone of auroral precip itation only during 2–5 minutes; therefore, the real level of magnetic activity during the passage of a satel lite does not correspond to the values of the used indi ces. This is the cause that such models can give only a rough estimate of the characteristics of precipitating particles at different latitudes despite a sufficiently good spatial resolution. In this work, we used the Auroral Precipitation Model (APM) that we developed earlier (Vorobjev and Yagodkina, 2005, 2007) to obtain the spatial distribu tion of the auroral luminosity; the model is available at http://apm.pgia.ru. The model makes it possible to obtain the global characteristics of precipitating elec
trons in coordinates of corrected geomagnetic lati tude–magnetic local time (CGL–MLT) for different levels of magnetic activity determined by the Dst value and 5min values of the ALindex. These data are the basis for calculating the brightness of different auroral emissions and bands. 2. ALGORITHM FOR THE CALCULATION OF THE INTENSITY OF THE AURORAL LUMINOSITY We apply the model of the kinetics of electronically excited molecular nitrogen for heights of the polar ionosphere presented in (Kirillov, 2008, 2010; Kirillov, 2011a) as the basis for calculating the intensity of the auroral glow. This model takes into account five triplet ( A 3Σ +u , B3Πg, W3Δu, B' 3Σ −u, and C3Πu) and three singlet (a '1Σ −u, a1Πg, and w1Δu) electronically excited states of the N2 molecule. The principal difference of these cal culations from the cited papers is in the fact that here we did not take into account the contribution of molecular collisions for the following states: B3Πg, W3Δu, B ' 3Σ −u and a '1Σ −u, a1Πg, w1Δu. Thus, in the calcu lations, we considered the following processes of for mation and deactivation of electronically excited states of N2 molecules. (1) Electronic excitation of N2(X 1Σ +g ) during colli sions with primary and secondary electrons. The exci tation rates were calculated using the method of “energetic casts” (Gordiets and Konovalov, 1991; Ser gienko and Ivanov, 1993); the energy losses of elec
711
712
VOROBJEV et al.
trons in 1 cm3 were calculated according N to (Kirillov et al., 1984). (2) Spontaneous transitions between states with radiation in the bands of the first and second positive systems (1PG and 2PG), WuBenesch (WB), after glow (AG), and Vegard–Kaplan (VK) for triplet states and with radiation in the McFarlane (MF), Ogava– Tanaka–Wilkinson–Mulliken (OTWM), and Lyman–Birge–Hopfield (LBH) bands for singlet states. The Einstein coefficients for spontaneous transi tions were taken from (Gilmore et al., 1992; Kirillov, 2011a). (3) Deactivation of states A 3Σ +u (ν = 0–6), W3Δu(ν = 0), a '1Σ −u (ν = 0, 1), and a1Πg(ν = 0) during collisions with unexcited molecules N2 and O2 and atoms O. The deactivation of other vibrational levels of states W3Δu, 1 − 3 − a ' Σ u, and a1Πg and all levels of states B3Πg, B' Σ u, C3Πu, and w1Δu were not taken into account owing to high rates of radiation processes compared to the rates of collisions at altitudes of the upper atmosphere. The deactivation coefficients of states by molecules N2 and O2 were taken according to the quantum–chemical calculations in (Kirillov, 2008; Kirillov, 2011b); the deactivation constants by oxygen atoms were taken from (Thomas and Kaufman, 1985) for the A 3Σ +u state and from (Gudipati et al., 2002; Khachatrian et al., 2003) for the a '1Σ −u and a1Πg states. It was assumed in the calculations that deactivation rates of W3Δu, ν = 0 by atoms O are equal to the deactivation rates A 3Σ +u , ν = 0. The solution of balance equations, taking into account the mechanisms of formation and deactiva tion (Kirillov, 2008, 2010), allows us to calculate vibra tional populations of N νY of triplet states. Radiation in the 1PG band at a wavelength of 669.0 nm is related to spontaneous transitions from B3Πg, ν = 5 to A 3Σ +u , ν' = 2. Therefore, the intensity of the 669.0nm band was cal culated according to the following equation: I669.0 = A669.0[N2(B3Πg, ν = 5),
(1)
where A669.0 = 5.93 × is the Einstein coefficient for this transition (Gilmore et al., 1992). A similar solution for the corresponding balance equations allows us to calculate the vibrational N νY populations of singlet states. It is possible to calculate the intensities of the LBH bands using the Einstein coefficients for the radiation conditions. In this work, we took into account 21 bands in the spectral interval 160–180 nm: ν' = ν + 3, ν + 4, and ν + 5 for ν = 0–3 and ν' = ν + 2, ν + 3, and ν + 4 for ν = 4–6. Contribu tion function f for each band was calculated as 104 s–1
f(λ) = exp[–(λ – 170)2/23], (2) where λ is the wavelength (in nm) and 170 nm is the center of the selected spectral interval.
Radiation N 2+ in 1NG bands is related to spontane ous transitions:
N 2+ (B 2Σ +u , ν) → N 2+ ( X 2Σ +g , ν') + hν1NG.
(3)
Let us use the following equation to calculate the intensity of the 391.4nm band of the 1NG system (ν = ν' = 0): I391.4 = Q*q ν*A391.4/Atot,
(4)
where Q* is the excitation rate of the B 2Σ +u state of N 2+ ions by primary and secondary electrons (Gordiets and Konovalov, 1991); q ν* = 0.883 is the Franck–
Condon factor for transition X 1Σ +g , ν = 0 → B 2Σ +u , ν = 0; A391.4 = 1.14 × 107 s–1 is the probability of spontaneous transition B 2Σ +u , ν = 0 → X 2Σ +g , ν' = 0; Atot = 1.60 × 107 s–1 is the sum of the probabilities of transitions B 2Σ +u , ν = 0 → X 2Σ +g , ν' ≥ 0 (Gilmore et al., 1992). The greenline emission at 557.7 nm of atomic oxygen is related to the spontaneous transition O(1S) → O(1D) + hν557.7.
(5)
The main formation processes of electronically excited atoms O(1S) in the upper polar atmosphere are (a) excitation energy transfer from the metastable state of N2( A 3Σ +u ), (b) excitation of O(3P) by primary and secondary electrons, and (c) dissociative recombina tion of ionospheric electrons and ions O 2+ (Sharp et al., 1979). Thus, the following equation was used in the calcu lation of the I557.7 intensity:
I 557.7
=
⎛ A557.7 ⎜ ⎝
∑ k*(ν)N
⎞ + Q1S + 0.06 α N e[ O 2+]⎟ (6) ⎠ , + kO* [ O 2] + kO*[ O ]
A ν [O]
ν
A557.7 + A297.2
2
s–1
where A557.7 = 1.215 and A297.2 = 0.076 s–1 are the probabilities of the spontaneous 1S → 1D and 1S → 3P transitions, respectively (Bates, 1992); kO*2 , and kO* are the constants of the deactivation rates of electronic excitation of O(1S) atoms during collisions with O2 and O(3P); k*(ν) are the coefficients of the rates of pro cesses (a) for ν = 0–6, Q 1S is the rate of process (b) (Gordiets and Konovalov, 1991; Sergienko and Ivanov, 1993); α is the constant of dissociative recom bination rate (c) of ionospheric electrons with the Ne concentration and molecular oxygen O2+ ions; and 0.06 is the quantum output of O(1S) for process (c) (Petrig nani et al., 2005). Constants kO*2 and kO* were taken according to (Bates, 1992), and coefficients k*(ν)
GEOMAGNETISM AND AERONOMY
Vol. 53
No. 6
2013
PLANETARY DISTRIBUTION OF THE INTENSITY OF AURORAL LUMINOSITY
respect to the pitch angles and Maxwell distribution by energy. The albedo fluxes for the isotropic distribution of particles precipitating in the atmosphere were taken into account according to (Sergienko and Ivanov, 1993). The concentration of the main atmospheric components—N2, O2 and O(3P) —and the tempera ture profile for altitudes from 90 to 200 km were accepted according to the data in (Morrill and Bene sch, 1996).
I, kR 1.0 0.8 391.4 0.6
557.7
0.4 669.0
0.2
LBHL 10
0
20
30 E, keV
Fig. 1. Brightness of the auroral emission at 391.4, 557.7, and 669.0 nm and LBH (170.0 nm) produced by a unit energy flux (1 erg/cm2 s) of precipitating electrons vs. their mean energy (E, keV). The brightness of the glow is shown in kilorayleighs (kR).
were calculated using the Landau–Zener approxima tion from (Kirillov and Aladiev, 1998). The calculated intensities of auroral emission glow for precipitating electrons with energy fluxes of 1 erg/cm2 s, depending on their mean energy, are shown in Fig. 1. The calculations were made for the isotropic distribution of precipitating electrons with
The relations deduced above and the characteris tics of precipitating electrons deduced from the APM model of auroral precipitations were used to calculate the planetary distributions of the auroral luminosity at different levels of magnetic activity. Figure 2 demon strates an example of the global distribution of inten sity for four auroral emissions at low (upper panel, AL = –200 nT, Dst = –5 nT) and high (lower panel, AL = –1000 nT, Dst = –100 nT) levels of magnetic activity. The vertical blackandwhite scale in the fig ure shows the brightness in kilorayleighs (kR). The investigation of largescale characteristics of the auroral luminosity was one of the main goals of the projects involving the Polar and Image satellites. These satellites were launched to a highapogee orbit where they recorded auroras in different intervals of the visual and UV spectrum ranges. Function f in Eq. (2), which shows the percentage of the glow of LBH(L) bands in the range 160.0–180.0 nm, was selected sim ilar to the configuration of the filter used on the Polar satellite (Torr et al., 1995). A satellite image of the
391.4 nm
557.7 nm
669.0 nm
LBH(L)
12
12
12
12
50°
50°
70°
50°
70°
18
kR 4
50°
70°
06 18
713
70°
0618
06 18
06
2 0
00
00
00
00
12
12
12
12
50°
50°
70°
18
50°
70°
06 18
50°
70°
0618
kR 12
70°
06 18
8 06 4 0
00
00
00
00
Fig. 2. Planetary distribution of the brightness of auroral emissions at low (upper panel, AL = –200 nT, Dst = –5 nT) and high (lower panel, AL = –1000 nT, Dst = –100 nT) levels of magnetic activity. The local magnetic noon (1200 MLT) is in the upper part of each figure, and the morning (0600 MLT) one is on the right. The vertical blackandwhite scale shows the brightness in kilorayleighs. GEOMAGNETISM AND AERONOMY
Vol. 53
No. 6
2013
714
VOROBJEV et al. POLAR
APM 12
12
970110 030038 UT
30 70
25
18
70°
20
80 60
15
30
60°
20
80°
18
06
10
10
5 0
0 00
Fig. 3. Planetary distribution of the auroral brightness of LBH (L) bands on January 10, 1997, obtained from the observations from the Polar satellite and calculated using the APM. The brightness of the luminosity is shown with a blackandwhite vertical scale.
LBH(L) emission obtained on January 10, 1997, at 0300:38 UT and the planetary distribution of the glow intensity in this spectral range calculated using the APM for the same level of magnetic activity (AL = –224 nT; Dst = 8 nT) are shown in Fig. 3. The brightness in units of the satellite instrument Polar UVI (photon/cm2 s) is shown in the right part of each figure, using a blackandwhite vertical scale. A comparison of the two figures demonstrates that they are similar both in the spatial distribution of the luminosity and its intensity. However, in the morning and day sectors, the diffuse luminosity to the equator from the oval at latitudes of approximately 70° CGL is distinguished only in the distribution obtained from the APM. This is related to the relatively low sensitiv ity of the satellite instrument. It was shown in (Torr et al., 1995) that only intensities of ~8 photon/cm2 s can be measured by the Polar UVI instrument at a sig nal/noise ratio equal to 5. This is the reason why the diffuse luminosity is weakly pronounced in the satel lite images in the sector ~0600–1500 MLT. This is also observed in numerous examples comparing data at different levels of magnetic activity. 3. CONCLUSION We developed an algorithm for calculating the inte gral intensity for the emission at 557.7 nm and the N2 LBH (170.0 nm), 1NG N 2+ (391.4 nm), and 1PG N2 (669.0 nm) bands. This algorithm was used in the APM model of auroral precipitation to calculate the global distribution of the intensity of the auroral lumi nosity at different levels of magnetic activity. Good agreement between the model distributions of the auroral luminosity in the UV spectrum range and observations from the Polar satellite was shown. A comparison of the data demonstrated that in the day– time sector, the diffuse luminosity equatorward of the
oval is better distinguished in the distribution of the luminosity obtained from the precipitation model, which is likely to be related to the relatively low sensi tivity of the satellite instruments. ACKNOWLEDGMENTS This study was supported by the Russian Founda tion for Basic Research (project no. 1205000273a) and the Presidium of the Russian Academy of Sci ences (programs nos. 4 and 22). REFERENCES Bates, D.R., Nightglow emissions from oxygen in the lower thermosphere, Planet. Space Sci., 1992, vol. 40, no. 23, pp. 211–221. Gilmore, F.R., Laher, R.R., and Espy, P.J., Franck–Con don factors, rcentroids, electronic transition moments, and Einstein coefficients for many nitrogen and oxygen band systems, J. Phys. Chem. Ref. Data, 1992, vol. 21, no. 5, pp. 1005–1107. Gordiets, B.F. and Konovalov, V.P., Ionospheric gas excita tion and ionization by highenergy electrons, Geomagn. Aeron., 1991, vol. 31, no. 4, pp. 649–656. Gudipati, M.S., Copeland, R.A., and Ginter, M.L., Colli sional removal rate constants for N2 (a, ν = 0 and 1) with N2, O2 and O colliders at 300, 240 and 150 K, EOS Trans. AGU, 2002, vol. 83, p. 236. Ivanov, V.E., Kirillov, A.S., Malkov, M.V., Sergienko, T.I., and Starkov, G.V., Auroral oval boundaries and the luminosity intensity planetary model, Geomagn. Aeron., 1993, vol. 33, no. 5, pp. 80–88. Khachatrian, A., Wouters, E.R., Gudipati, M.S., Ginter, M.L., and Copeland, R.A., Temperature dependent colli sional energy transfer of N2 (a1Πg and a'1Σ −u, ν = 0 and 1), EOS Trans. AGU, 2003, vol. 84, pp. F1149–F1150. Kirillov, A.S., Electronically excited molecular nitrogen and molecular oxygen in the highlatitude upper atmo
GEOMAGNETISM AND AERONOMY
Vol. 53
No. 6
2013
PLANETARY DISTRIBUTION OF THE INTENSITY OF AURORAL LUMINOSITY sphere, Ann. Geophys., 2008, vol. 26, no. 5, pp. 1159– 1169. Kirillov, A.S., Electronic kinetics of molecular nitrogen and molecular oxygen in highlatitude lower thermosphere and mesosphere, Ann. Geophys., 2010, vol. 28, no. 1, pp. 181–192. Kirillov, A.S., Singlet molecular nitrogen in the auroral ion osphere and under the conditions of laboratory dis charge, Zh. Technical Physics, 2011a, vol. 56, no. 12, pp. 1737–1744. Kirillov, A.S., Calculation of the quenching rate coefficients of electronically excited singlet molecular nitrogen, Zh. Technical Physics, 2011b, vol. 56, no. 12, pp. 1731– 1736. +
Kirillov, A.S. and Aladiev, G.A., The role of the N2(A Σ u , ν) + O reaction in the green line luminosity and molecular nitrogen vibrational kinetics in the highlatitude upper atmosphere, Kosm. Issled., 1998, vol. 36, no. 5, pp. 451–457. Kirillov, A.S., Osipov, N.K., and Ivanov, V.E., Energy absorp tion of electron bunch in the O, O2, and N2 gas mixture, Geomagn. Aeron., 1984, vol. 24, no. 4, pp. 625–630. Morrill, J. and Benesch, W., Auroral N2 emissions and the effect of collisional processes on N2 triplet state vibra tional populations, J. Geophys. Res., 1996, vol. 101A, pp. 261–274. Petrignani, A., van der Zande, W.J., Cosby, P.C., Hellberg, F., Thomas, R.D., and Larsson, M., Vibrationally resolved rate coefficients and branching fractions in the disso ciative recombination of O2+ , J. Chem. Phys., 2005, vol. 122, no. 1, p. 014302. 3
GEOMAGNETISM AND AERONOMY
Vol. 53
No. 6
715
Sergienko, T.I. and Ivanov, V.E., A new approach to calculate the excitation of atmospheric gases by auroral electron impact, Ann. Geophys., 1993, vol. 11, no. 7, pp. 717– 727. Sharp, W.E., Rees, M.H., and Stewart, A.I., Coordinated rocket and satellite measurements of an auroral event. 2. The rocket observations and analysis, J. Geophys. Res., 1979, vol. 84A, pp. 1977–1985. Spiro, R.W., Reiff, P.H., and Maher, L.J., Jr. Precipitating electron energy flux and auroral zone conductances– an empirical model, J. Geophys. Res., 1982, vol. 87, no. 10, pp. 8215–8227. Thomas, J.M. and Kaufman, F., Rate constants of the reac tions of metastable N2( AΣ +u ) in ν = 0, 1, 2, and 3 with ground state O2 and O, J. Chem. Phys., 1985, vol. 83, no. 6, pp. 2900–2903. Torr, M.R., Torr, D.G., Zukis, M., Johnson, R.B., Ajello, J., et al., A far ultraviolet imager for the international solar–terrestrial physics mission, Space Sci. Rev., 1995, vol. 71, pp. 329–383. Vorobjev, V.G. and Yagodkina, O.I., Effect of magnetic activity on the global distribution of auroral precipita tion zones, Geomagn. Aeron., 2005, vol. 45, pp. 438– 444. Vorobjev, V.G. and Yagodkina, O.I., Auroral precipitation dynamics during strong magnetic storms, Geomagn. Aeron., 2007, vol. 47, pp. 185–192.
Translated by E. Morozov
2013
