CC BY
27
Lower thermosphere and ionosphere behaviour during a strong magnetic storm of March 31, 2001: Modelling and comparison with the Millstone Hill incoherent scatter radar measurements
12 2 1 A.A. Namgaladze ' , A.N. Namgaladze , Yu.V. Fadeeva ,
3 3
L.P. Goncharenko , J.E. Salah
1 Murmansk State Technical University, Murmansk, Russia
2
Polar Geophysical Institute, Murmansk, Russia
3
Haystack Observatory, Massachusetts Institute of Technology, Westford, USA
Abstract. The numerical global Upper Atmosphere Model (UAM) has been used for studying the lower thermosphere and ionosphere behaviour during a strong magnetic storm of March 31, 2001. A comparison of the calculation results with the Millstone Hill IS radar data obtained during this magnetic storm has been carried out. A satisfactory agreement between the measured and calculated ion temperatures, electron and ion drift velocities has been obtained. The calculated electric field variation is similar to the observed one but is twice less in magnitude. As to the neutral horizontal wind velocities, there is a large difference between the calculated values and data obtained by the Millstone Hill IS radar, down to the different signs of the wind components. This difference can be bound up to distinctions in the frequencies of the ion-neutral collisions dependent on the neutral atmosphere parameters and the ion composition, and in the electric fields measured at Millstone Hill and theoretically calculated by the UAM. The reasons of the discrepancy of the thermospheric winds simulated by the UAM and calculated at Millstone Hill have been examined. The results of this examination show that the Millstone Hill wind results are very sensitive to the measured electric field values especially when the last are so large as they were in the case, and therefore even a moderate difference between the calculated and observed electric fields may cause a significant discrepancy of the thermospheric winds simulated by the UAM and calculated at Millstone Hill.
1. Introduction
The magnetic storm of March 31 - April 1, 2001 was chosen for a numerical examination of behaviour of the lower ionosphere and thermosphere parameters for two reasons. First, it was enough isolated and distinguished by an extremely major degree of perturbation: the Dst-variation almost reached value -400 nT, and the peak value of Kp-index was 8.7 (see Fig. 1). Secondly, there was an opportunity to compare the results of our calculations to the data of the Millstone Hill incoherent scatter radar measurements (geographical coordinates: 43°N, 289°E, geomagnetic coordinates: 54.4°, 358.5°), concerning to the height region of 100-150 km (Goncharenko et al., 2001).
2. Model calculations
The calculations have been performed by using the numerical global Upper Atmosphere Model (UAM) (Namgaladze et al, 1988, 1998a,b). This model calculates numerically the time-dependent global distributions of the densities and temperatures of the neutral and charged components of the upper atmosphere in the height range from 80 km up to the geocentric distance of 15 RE from the corresponding continuity and heat balance equations as well as the electric fields from the electric potential equation and the thermospheric wind and ion drift velocities from the corresponding momentum equations. The model takes into account the offset of the geographic and geomagnetic axes of the Earth. The computational spatial grid has been taken with a variable size of the mesh elements along the geomagnetic latitude (from 1° at about 60°-latitude to 3° near the geomagnetic poles, 5° and 10° at the equator for obtaining ionospheric and thermospheric parameters, correspondingly), a size of 15° in the geomagnetic longitude, and a variable height size (from 3 km at the 80 km-altitude to about 20 km at the altitude of the F2 layer maximum).
0 24 48 72 9® UT
30.03 31.03 1-.D4 ' 2.04 3.04
Fig.1. Variations of Kp and Dst indices during the magnetic storm of March 31 - April 3, 2001
Fig. 2. The time variations of the model input electric field potential drop across the polar cap and the polar cap boundary and the zone 2 field-aligned currents magnetic latitude locations
UT, hour
Fig. 3. The electric field modulus variations, measured by the Millstone Hill ISR (curve 1) and calculated by the UAM (curve 2)
The locations of the polar cap boundary, precipitating electron zones, field-aligned zone 2 currents and their intensities have been taken as the model input parameters dependent on Kp-index as well as the electric field potential drop across the polar cap. These dependencies have been chosen firstly to be consistent with the DMSP data (when such data were existing) and secondly to get the model electric field values to be similar to those measured by the Millstone Hill radar. Fig. 2 shows the time variations of the electric field potential drop across the polar cap and the locations of the polar cap boundary and the field-aligned zone 2 currents, as they have been set in the model.
3. Comparison with the Millstone Hill ISR data
Fig. 3 presents both the UAM calculated and Millstone Hill ISR measured electric field variations. As it can be seen in this figure, these variations have a similar form but there are some differences in magnitude. The maximal measured electric field values were 40 and 50 mV/m (very large values for the subauroral geomagnetic latitude) at 06 and 21UT whereas the corresponding UAM values at these UT moments were of about 25 and 35 mV/m.
Besides the comparison of the electric field variations, the UAM calculated time variations of the E-region ionosphere parameters have been compared with corresponding parameters measured in the 100-150 km altitude region. Fig. 4 presents the time variations of the ion temperature (top panel) and electron density (bottom panel), measured at Millstone Hill (curves with dots) and calculated by the UAM (bold curves) at the fixed altitudes 108, 120 and 135 km. A satisfactory accordance of the measured and calculated time variations of the electron density and the ion temperature is seen from the figure.
Figs 5 a,b show a behaviour of the horizontal velocities (a - the zonal components, b - the meridional components) of the ion drift at the altitude 127 km and Figs 6 a,b show a behaviour of the horizontal velocities (a - the zonal components, b - the meridional components) of the neutral wind at the altitude 127 km, measured at Millstone Hill and obtained in the model calculations for the point with magnetic coordinates ^=55°, ^=0° during the examined magnetic storm.
Fig. 4. Time variations of the ion temperature (top panel) and electron density (bottom panel) during the magnetic storm on March 31, 2001 at three altitudes of 108, 120 and 135 km. The curves with dots represent the variations measured by the Millstone Hill incoherent scatter radar, the bold curves - the variations calculated by the UAM at the point with magnetic coordinates 0=55°, ^=0°.
a) Eastward velocity, m/s
Ion drift
b) Northward velocity, m/s
800 -i
400
24
-600 -1
-•■ Experiment — Model
-600 -1
-•■ Experiment — Model
Fig. 5. Time variations of the horizontal velocities (a - zonal, b - meridional) of the ion drift at the altitude 127 km, obtained at Millstone Hill and calculated by the UAM for the point with magnetic coordinates (P=55°, ^=0° during
March 31, 2001 magnetic storm
a) Eastward velocity, m/s
Neutral wind
b) Northward velocity, m/s
400 -,
200 -
400 -,
200 -
-400 -1
Experiment Model
fe -200 -
-400 -1
Experiment Model
Fig. 6. Time variations of the horizontal velocities (a - zonal, b - meridional) of the neutral wind at the altitude
127 km, obtained at Millstone Hill and calculated by the UAM for the point with magnetic coordinates (£=55°,
/1=0° during March 31, 2001 magnetic storm
12h 12h
T
06h
500ms"' QOh
300ms"
00h
Fig. 7. Comparison between the winds calculated by the UAM at 130 km altitude on March 31, 2001 (a) in geomagnetic latitude and local time coordinates and WINDII winds at same altitude on April 5th (c) and April 8th (d), 1993 (Zhang and Shepherd, 2000).
As it can be seen there is a very good agreement between the measured and calculated ion drift velocities, both the zonal component and the meridional component. It relates to the character of the behaviour and to the velocity values.
As regards the comparison of the horizontal neutral wind velocities (Fig. 6), the strong difference between the calculated model values and the data obtained at Millstone Hill is revealed especially for the meridional wind after 12UT when even signs of meridional wind are different.
At the same time the global wind vector picture obtained in the UAM calculations for the disturbed ionospheric E-region is similar to that obtained on board of the WINDII satellite during the April 4-5, 1993 magnetic storm (Zhang and Shepherd, 2000) (see Fig. 7).
It is important to notice that Millstone Hill winds are not directly measured ones. But the formula connecting the ion drift and wind velocity used at Millstone Hill and in the UAM is the same. The difference by the use of the formula may be related with the ion-neutral collision frequencies dependent on the neutral atmosphere parameters and electric fields measured at Millstone Hill and theoretically calculated by the UAM.
4. Investigation of causes of the difference between the thermospheric wind velocities calculated by the UAM and obtained at Millstone Hill and discussion
During the March 31, 2001 magnetic storm the Millstone Hill incoherent scatter radar directly measured the following ionospheric E-region parameters: electron density, electric field, ion drift velocity and ion temperature. The meridional and zonal thermospheric winds were calculated on the base of the observed electric field and ion drift velocity by the following formula:
U„ = Vi -(©,/v,) • (E + Vi x B) /
(1)
where Un - the thermospheric wind velocity; Vi - the ion drift velocity; co¡ - the ion gyro-frequency; v¡ - the ionneutral collision frequency; E, B - the electric and magnetic field vectors.
When simulating thermospheric winds the UAM uses the theoretical neutral densities in order to calculate the ion-neutral collision frequency, while at the Millstone Hill observatory the neutral densities are calculated with the help of the empirical model MSISE-00 (NRLMSISE-00). To find the sources of the differences of the neutral winds calculated from the measured ion drifts at Millstone Hill and simulated by the UAM we have performed additional calculations of winds from the formula (1) using two last empirical atmospheric models MSISE-90 (Hedin, 1991) and MSISE-00 to obtain the neutral densities. The ion drift velocity and electric field values were taken from the UAM. The thermospheric wind velocities calculated in this way have been compared with the fully theoretically calculated ones. Fig. 8 shows zonal and meridional wind velocities at the altitude 153 km calculated with the theoretical neutral densities (UAM) and with the neutral densities from the empirical models MSISE-90, MSISE-00. In Fig. 9 we can see the same variations but at the altitude 127 km and with the Millstone Hill data. As it can be seen there are no significant differences between three variants of both horizontal wind components excepting the time interval 19-23 UT, but the differences here don't exceed values of about 50 m/s. At other altitudes the large differences are absent, too.
2DÜ
E
m
-200 -
-400
— UAM
MSISE90 S MSISE00
-300
Fig. 8. Horizontal neutral wind velocities at the altitude 153 km calculated from the model ion drift velocity and electric field values with neutral densities from the empirical models and UAM
UT, hours
1 - Experiment — MSISE9Û JLMSISEÜÜ
1 - Experiment ; : — MSISE9Ü—MSISE0Û"
Fig. 9. The same as in Fig. 8 but with the Millstone Hill data and at the altitude 127 km
Fig. 10. Horizontal neutral winds at the altitude 153 km calculated with 90, 100 and 110 %
of the electric field values
Fig.11. The same as in Fig. 10 but with the Millstone Hill data and at the altitude 127 km
To estimate the sensitivity of the formula (1) to the electric field values, the thermospheric winds velocities were computed from (1) with 90 % and 110 % of the initial model values of electric field. The results showed a very strong dependence of the wind velocities calculated from the ion drifts on the electric field values. The Millstone Hill wind velocities are large due to very (unusually for the mid-latitude station) large observed values of the electric field. The time variations of the horizontal neutral wind velocities calculated at the altitude 153 km for three variants of the electric field are presented in Fig. 10. It is seen from the figure that both zonal wind velocities and meridional velocities reveal a very large sensitivity to the electric field values. Fig. 11 shows the horizontal neutral wind velocities, calculated by the model for three variants of the electric field and calculated at Millstone Hill at the altitude 127 km.
Figs 10, 11 demonstrate much higher sensitivity of the calculation formula (1) to the electric field values than its sensitivity to the ion-neutral collision frequency, when the ion drift velocities are given. This high sensitivity increases with height. Such sensitivity can explain the disagreement of the thermospheric winds extracted from the global mathematical model during the magnetic storm modeling and the winds calculated at Millstone Hill by the formula with the same values of neutral densities and measured ion drift velocities and electric fields.
Thus, the maximum disagreement of meridional components of wind velocity at 22UT, which amounted more than 400 m/s at the altitude of 127 km, may be provoked by the 10 mV/m disagreement of the corresponding electric field values, measured at the Millstone Hill and simulated by the model.
5. Conclusions
Choosing as model input the dependencies of the locations of the polar cap boundary, precipitating electron zones, field-aligned zone 2 currents and the electric field potential drop across the polar cap on Kp-index we have got a more or less satisfactory agreement between the model electric fields and those measured by the Millstone Hill radar. A rather good agreement has been achieved for the time variations of such ionospheric E-region parameters at the altitudes of 100-150 km as electron density, ion temperature, ion drift velocities. Both the behaviour character and the magnitudes show it. But the large disagreements have been
revealed between the neutral wind velocities simulated by the UAM and obtained at Millstone Hill. After carrying out the additional calculations and their analysis the conclusion have been done that the main reason of this disagreements is the differences of the electric field values, observed by the Millstone Hill radar and theoretically calculated during the March 31, 2001 magnetic storm simulating. The 10 mV/m change of the electric field values can cause a significant change (up to 400 m/s and even to the sign change) of the values of the thermospheric wind velocity calculated by the formula (1).
This work was supported by Grant No.02-05-64141 of Russian Foundation for Basic Research.
References
Goncharenko L.P., Namgaladze A.A., Namgaladze A.N. and Salah J.E. Effects of the March 31, 2001 geomagnetic storm on the lower thermosphere and ionosphere at mid-latitude. Eos Trans. AGU, 82(47), Fall Meet. Suppl., Abstract, SA12B-0688, 2001. Hedin A.E. Extension of the MSIS thermosphere model into the middle and lower atmosphere. J. Geophys.
Res, v.96, p.1159-1172, 1991. Namgaladze A.A., Korenkov Yu.N., Klimenko V.V., Karpov I.V., Bessarab F.S., Surotkin V.A., Glushchenko T.A. and Naumova N.M. Global model of the thermosphere-ionosphere-protonosphere system. Pure andAppl. Geophys., v.127, N 2/3, p.219-254, 1988. Namgaladze A.A., Martynenko O.V. and Namgaladze A.N. Global model of the upper atmosphere with variable latitudinal integration step. Geomagnetism and Aeronomy International, v.1, N 1, p.53-58, 1998a.
Namgaladze A.A., Martynenko O.V., Volkov M.A., Namgaladze A.N. and Yurik R.Yu. High-latitude version of the global numerical model of the Earth upper atmosphere. Proc. of MSTU (http://mstu.edu.ru/publish/vestnik/), v.1, N 2, p.23-84, 1998b. NRLMSISE-00, http://uap-www.nrl.navy.mil/models_web/msis/msis_home.htm
Zhang S.P. and Shepherd G.P. Neutral winds in the lower thermosphere observed by WINDII during the April 4-5th, 1993 storm. Geophys. Res. Lett, v. 27, N 13, p.1855-1858, 2000.
