IMF direction derived from cycloid-like ion distributions observed by Mars Express

M. Yamauchi, Y. Futaana, R. Lundin, S. Barabash, M. Holmström (IRF-Kiruna, Sweden)

A. Fedorov, J.-A. Sauvaud (CESR, Toulouse, France)

R.A. Frahm, J.D. Winningham (SwRI, San Antonio, USA)

E. Dubinin (MPS, Katlenburg-Lindau, Germany),

E. Kallio (FMI, Helsinki, Finland)

and ASPERA-3 team

MotivationTo understand Mars Express (MEX) hot plasma and energetic neutral atom (ASPERA-3) data, it is important to have the magnetic field (B) measurement. However, MEX does not measure B (because of bureaucracy), and B is measured only on Venus Express. Therefore, any method to estimate B on MEX is helpful. Using Mars Global Surveyor (MGS) data is one method, but we here take another approach :

Analyze gyromotion of ions observed by ASPERA-3 ion mass analyser (IMA) because ion motion in velocity space is a ring! (spiral motion in real space)

Actually, the ring distribution is the "clearest signature" among many interesting phenomena seen in IMA Quick Look (mass and azimuth are integrated)

2005-4-27 1331-1344 UT

2005-7-12 1133-1146 UT

2005-6-3 0602-0615 UT

B

ion

2005-4-27 (easy case) single azimuth plots

= 5~15

= 4

= 3

= 2

= 1

SW

ringMEX orbit (light blue) during 1329-1357 UT in the cylindrical MSO coordinates. R2 = Y2+Z2.

average IMB

average BS

Mars

Mars

IMA energy-time spectrograms for proton cannel at 0.7-7 keV during 1334:00-1343:40 UT on 27 April 2005. Five different azimuthal sectors (φ) are presented except sector 0. The nearly 3-min cycle seen in the IMA data is due to the electric scan of the entrance direction from nearly -45° (elevation=0) to nearly +45° (elevation=15). The counts that are seen above the solar wind in the proton channel are due to contamination from alpha particles.

Ion motion with V0=0

velocity spacetrajectory

V*X

VY

VX

V*Y

V*ZVZ

V (SW frame) = V* (Martian frame) + VSW

B

Ions with V0=0 in Martian frame have an

initial velocity -VSW in SW rest frame where E

(electric field) =0. Ions in SW rest frame make simple spiral trajectory (simple gyration + constant parallel velocity), i.e, a simple ring in velocity space as indicated by blue dashed lines.

Ring = within a plane the minimum variance direction (N) is parallel to B!

Viewing from +B

Viewing from +E

3-D view

Initial velocity to B is V0cos

Velocity space plots of the data

Velocity (in km/s) scatter plots of the H+ data (mass range 0.5-1.1: slightly contaminated by SW He++) during 1334-1357 UT, 27 April 2005, in (a) MSO coordinates, and maximum(L)-medium(M)-minimum(L) variance coordinates determined from (b) manually selected ring data, and (c) automatically filtered data during 1337-1357 UT. Azimuth=0 data are not shown (total 450 points).

The circles and triangles denote the points belonging to the ring distribution by a manual inspection, with circles corresponding to 7-40 counts, filled triangles corresponding to more than 40 counts, and empty triangles corresponding to 4-6 counts.

Non-ring data are shown using plus marks (blue: count is 7-40, light blue: count more than 40).

Coordinates given (b) arranges the ring data fitting to a circle best passing through the origin.

However, IMA orientation is not always ideal

Aligned in the azimuthal direction (points 2, 3, 4, 5 in the ion-motion illustration)

Linear alignment (small data range in M direction)

Minimum variance method gives clear L direction (// ±E), but not M & N directions. We may not use ±N as estimate of B direction, but(1) E x VSW gives BY/BZ orientation ! (2) BX/B (or ) can be obtained intuitively

IMA all-mass & all-azimuth data on 2005-6-3 , 0545-0645 UT

IMA light ion (H+ and He++) data (3~8 keV) for each azimuth at ~0608 UT & ~0623 UT

=5

~1

4

=1

5

=1

=2

=

3

=4

cf. beam-origin (V0 ≠ 0) ring distribution case

QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

Alignment direction of the ring data by IMA

≈ Maximum variance direction (LX, LY, LZ)

where LX << 1

BT=(0, BY, BZ) // VSW x E // (0, LZ, -LY)

Sign of B

If L = evolution direction, sign(BZ) = - sign(LY)

Tilt toward X (=)

* ' = angle between max energy (MAX)

direction and solar wind direction

* Ratio k = VMAX/VSW= (MAX/SW)1/2

If V0 = 0, two angles must be the same,

i.e., ' = , and k = cos()

If cos(') = k , most likely = '

MAX

Procedure & result

(1) Manually select the ring distribution. (2) Apply the minimum variance method to determine L, M, and N. If the ring data is well arranged into a partial circle, ±N // B

If not, (3) Examine |LX| < 0.3. If yes, L // -ESW. If not, remove the direction that corresponds to the lowest energy from the selected set of data and re-calculate a new L. (4) Manually obtain MAX and '. (5) Check if MAX /4SW ≈ cos2(') is satisfied. If yes, we have ≈ ', where BX/|B| = sin().

(6) If possible, identify the evolution direction and determine the sign of L and ESW. BT is parallel to VSW x ESW.

Using (3)~(6) Changing IMF for 2005-6-3 event

0608 UT, northward IMF, X tilt ~ +35°~40°, Y tilt ~ -10°

0613 UT, northward IMF, X tilt ~ +35°, Y tilt ~ -45°

0623 UT, northward IMF, X tilt ~ +20°, Y tilt ~ -30°

0633 UT, X tilt > +40°

Using (1)~(2) Constant IMF for 2005-4-27 event

1336-1357 UT, dawn-dusk oriented IMF, with X tilt ~ 20°

In-coming directions of ions

Azimuth scan Elevation scan

Electric scan

Electric scanIM

A

0 3 6 9 12 15

0 3 6 9 12 1501

2

3

4

5

6

1514

summaryMEX/IMA often observed partial ring distributions of protons in velocity space upstream of the Martian bow shock. Since ions are expected to gyrate around the IMF in velocity phase space, an ensemble of ions with nearly the same initial velocity (beam-like or zero velocity) forms such ring distribution when the IMF is relatively constant over a gyroperiod. Inversely, approximate IMF orientation can be derived from the observed ring distribution because the ring's plane must be perpendicular to the IMF.

However, the actual derivation of the ring plane is very complicated because (a) the ring distribution is just one of minor component that IMA detects, and (b) IMA's viewing directions and angular resolution are very limited. Therefore, we developed method to obtain the IMF orientation from actual IMA data (see "procedure" above). There are three important steps:(1) Manually choose the ring data with the same V0: (automatic selection does not work).

(2) Obtain maximum (L) and minimum (N) direction from the selected ring data. (3) After the quality check, use either N or L (see procedure).

We applied actual IMA data. Two examples (2005-4-27 and 2005-6-3) are shown.It is possible to obtain the IMF orientation about 75% of BS traverals when IMA is in the best operation mode (highest post-acceleration to detect proton).

Selected ring data

*1) L = (0.05, -0.02, 1.00) in MSO *2) M = (0.94, -0.34, -0.06) in MSO *3) N = (0.34, 0.94, -0.003) in MSO

El Az energy (keV)

VX (km/s)

VY (km/s)

VZ (km/s)

count VL *1 (km/s)

VM *2 (km/s)

VN *3 (km/s)

02 03 1.37 -414 148 -260 25 -285 -424 0 03 03 1.86 -508 181 -251 55 -281 -524 0 04 03 2.27 -584 208 -218 7 -253 -607 0 05 03 2.49 -630 225 -164 15 -201 -659 0 08 03 2.74 -680 243 39 4 -1 -723 0 09 03 2.74 -673 241 110 4 69 -719 0 10 03 2.74 -659 236 180 40 140 -709 0 12 03 2.49 -585 210 298 32 262 -637 0 13 03 2.27 -529 190 341 20 309 -580 0 14 03 2.00 -464 167 370 7 346 -520 0 15 03 1.68 -392 141 382 5 358 -438 0

2005-4-27, 1344-1347 UT

2005-6-3, ~0608 UTEl Az energy

(keV) VX

(km/s) VY

(km/s) VZ

(km/s) count VL *1

(km/s) VM *2 (km/s)

VN *3 (km/s)

13 01 4771 -711 -410 485 153 -341 -401 372 13 01 4352 -679 -391 464 296 -324 -362 371 14 01 5712 -728 -422 616 15 -336 -495 466 14 01 5225 -696 -404 590 14 -320 -453 461 15 01 5712 -671 -392 696 7 -291 -497 558 13 02 4771 -810 -106 492 13 -46 -470 272 14 02 5225 -793 -106 596 36 -31 -521 363 14 02 4771 -758 -102 569 176 -28 -476 363 14 03 5225 -769 205 602 14 280 -492 336 14 03 4771 -734 196 575 11 269 -449 338 15 03 4771 -677 179 647 27 265 -448 431 15 03 4352 -647 171 618 8 255 -406 428 14 04 4352 -572 442 554 71 517 -297 383

*1) L = (0.06, 0.99, 0.14) in MSO *2) M = (0.79, 0.04, -0.61) in MSO *3) N = (0.61, -0.15, 0.78) in MSO

H+ only (no He++) !

= 3

= 2

= 4 = 3 = 2SW He++

Ring H+

SW H+

Distribution in velocity space

From ring data at ~0623 UT (2005-6-3)

-750

-500

-250

0

250

-500 -250 0 250 500

-750

-500

-250

0

250

-250 0 250 500 750

VL (km/s) VN (km/s)

VM

(km

/s)

2005-4-27 (easy case)

SWring

However, IMA orientation is not always ideal

2005-6-30545-0645 UT

cf. beam-origin (V0 ≠ 0) ring distribution case