Investigation of the source Investigation of the source region region

of ionospheric oxygen outflow of ionospheric oxygen outflow

in the cuspin the cuspusing multi-spacecraft using multi-spacecraft

observations by CIS onboard observations by CIS onboard ClusterCluster

COSPAR, 2002, Houston, Texas, Paper D3.1-0048-02

Authors

Y.V. Bogdanova (1), B. Klecker (1), G. Paschmann (1), E. Georgescu (1),

M.I. Kubyshkina (2), L. M. Kistler (3), C. Mouikis (3), E. Moebius (3),

H. Reme (4), J.M. Bosqued (4), I. Dandouras (4), J.A. Sauvaud (4),

N. Cornilleau-Wehrlin (5), H. Laakso (6), A. Korth (7),

M.B. Bavassano-Cattaneo (8), T. Phan (9),

C. Carlson (9), G. Parks (9), J.P. McFadden (9),

M. McCarthy (10), R. Lundin (11)

(1) Max-Planck Institute for Extraterrestrial Physics, Garching, Germany

(2) St.Petersburg State University, St.Petersburg, Russia

(3) Space Science Center, University of New Hampshire, Durham, NH,

USA

(4) C.E.S.R.., Toulouse, France

(5) C.E.T.P., Velizy, France

(6) ESA, ESTEC, Noordwijk, Netherlands

(7) Max-Planck Institute for Aeronomie, Kaltenburg-Lindau, Germany

(8) I.F.S.I., Rome, Italy

(9) University of California, Berkeley, CA, USA

(10) University of Washington, Seattle, WA, USA

(11) Swedish Institute of Space Physics, Kiruna, Sweden2

Abstract

Beams of singly ionized oxygen with very narrow energy distributions originating in

the dayside cusp region are observed frequently in the cusp and polar cap regions by

the CODIF sensor of the CIS instrument onboard Cluster. During summer nad fall of

2001 the high separation distances of ~1 Re between the spacecraft provided a good

opportunity to estimate the size and location of the source of the outgoing O+ ions.

Statistical study based on 10 events shows that the source region is located near the

equatorward edge of the cusp region. Using Tsyganenko T96 magnetospheric

magnetic field model we have estimated the size of the acceleration region at the

ionosphere level (100 km): latitudinal size is around 1,5 deg and longitudinal size is

around 10 deg. Inside the source region Cluster observations at 5-6 Re show high

transverse heating of the O+ population. This process is colocated with the sharp

enhancement of the low-frequency wave fields measured by STAFF and EFW

instruments. We conclude that oxygen ions outflow is caused by resonant heating by

broadband extra low frequency (BBELF) wave fields at altitudes of 5-6 Re and below.

3

Introduction• Oxygen outflow from the cusp/cleft region, one of the main sources of the magnetospheric oxygen population,

was discovered two decades ago by Lockwood (Lockwood et al., 1985) and Moore (Moore et al, 1986), who used data from Dynamic Explorer-1 (DE-1) satellite and named this outflow as “cleft ion fountain” or “upwelling ions”.

• This type of the O+ outflow was extensively studied (see reviews Andre and Yau, 1997; Yau and Andre, 1997; Moore et al., 1999) and some common features of this outflow are well-known but still need additional investigation :

1. The origin of the upward moving ions is inside the cusp/cleft region ( Dubouloz et al., 1998; Bouhram et al., 2002 and references therein) or near the equatorward edge of the cusp/cleft (Mats et al., 1990). 2. Oxygen outflowing population originates from a narrow (<2 deg) latitudinal interval and wide MLT interval (Lockwood et al., 1985; Andre et al., 1990; Bouhram et al, 2002) , or from wide latitudinal interval (Dubouloz et al., 2001). 3. The O+ outflows are associated with the transverse energization processes . The main dominant processes are resonant heating by the broadband extremely low frequency electric field turbulence, by electrostatic waves near and just above low hybrid frequency and by electromagnetic ion cyclotron waves (Moore et al., 1999; Norqvist et al., 1998). Contributions of the different acceleration processes to the ion heating is still under study. 4. Oxygen heating was observed at the different altitudes, from 2000 km up to 15 000 km. Dubouloz et al, 2001 and Bouhram et al., 2002 showed that heating processes take place inside a broad range of altitudes.• The main goal of this study is an improvement of our knowledge about oxygen upward moving ions in the cusp

region using data from multi-spacecraft observation from Cluster

4

CIS Cluster observation of O+ outflow

North hemisphere South hemisphere

Event Time Kp Event Time Kp

02 July 2001 03:45-05:20 UT 1+ 16 July 2001 06:30-08:30 UT 2-

30 July 2001 17:00-18:00 UT 2- 28 September 2001 00:30-03:00 UT 3-

04 August 2001 11:35-13:00 UT 1+, 2- 30 September 2001 10:00-11:40 UT 2-

23 August 2001 12:50-14:00 UT 2- 19 October 2001 11:00-12:55 UT 2-, 3-

28 September 2001

05:40-08:00 UT 3-, 4- 14 November 2001 15:15-17:05 UT 0+

Table 1

During summer and fall 2001 the Cluster satellite very frequently observed O+ upward moving ions in the cusp and polar cap regions. We have selected 10 events, five from each hemisphere with the clearest O+ outflow events. These events occured during times with quiet or moderate geomagnetic activity and have common features. At first, O + outflow has a very narrow limited energy range, it means that the whole population was accelerated to the same bulk velocity, which is typical for the upwelling ions. Moreover, during these events Cluster S/C had optimal orbits to study O + outflow: CIS detected the onset of the O+ outflow, outflow development which displays in enhancement of the ions parallel velocity and energy , and then long-live oxygen beam with gradual drop of the energy.

5

North hemisphere, one example, 23 August 2001CIS overview

One typical example of the O+ outflow from the north hemisphere: energy-time spectrograms of the H+ and O+ populations from S/C 1, 3 and 4. Oxygen outflow is observed at S/C 1 and 4 simultaneously, and at S/C 3 about 30 minutes later.

Fig. 1

6

North, 23 August 2001, orbit and 3D moments

Figure 2 shows the Cluster orbit in the North hemisphere: dayside inner magnetosphere, cusp, polar cap. S/C moves mainly in the Z direction. Separation distances between S/C during event 23 August 2001 were: S/C 1-3 – 1,56 Re,

S/C 3-4 – 1,79 Re, S/C 1-4 – 0,39 Re. Figure 3 presents CIS data from S/C 4 for this event: energy-time

spectrograms of H+ and O+ ions, parallel velocity of O+ (black) and H+ (green) and three components of the O+ perpendicular velocity (convection velocity). During the beginning of the O+ beam observation convection velocity measured by CIS was highly variable with a mean value around zero. EDI data for this period are not available.

S/C1

S/C3

S/C4

12:00

13:30

Fig. 3

Fig. 2

7

North, 23 August 2001, O+ outflow study Figure 4 presents CIS data from S/C 4

for this event: energy-time spectrograms for H+ and O+ ions, and parallel velocity of both populations. Protons exhibit typical cusp behaviour, showing energy-time-latitude dispersion. Oxygen ions spectrogram gives us the possibility to track an evolution of the O+ population: we observe the beginning of the oxygen outflow near the equatorward boundary of the cusp region at 12:46 UT, then development of the outflow over 2 minutes, displayed in the enhancement of the energy and parallel velocity to the maximal value at 12:49 UT, and finally we observe in the cusp and polar cap regions a long-live O+ beam with gradual drop of energy and velocity.

Fig. 4

Proton data show that at 12:46 UT H+ ions started to be accelerated upward together with oxygen ions, and only at 12:48 UT there is first strong injection of protons and second one at 12:50 UT. Note, that O+ outflow starts before protons injection, it means that source of the upward moving ions is located partially on the closed magnetic field lines. The behaviour of the O+ outflow we explain as following: oxygen ions were accelerated in the rather narrow region located near the equatorward edge of the cusp, when the Cluster spacecraft crossed this region we observe beginning and enhancement of the outflow to the maximal value and when Cluster left this region gradual decreasing of energy and velocity is observed. 8

North, 23 August 2001, O+ distribution functionsFig. 512:46:31-12:47:00 12:47:00-12:47:40 12:47:32-12:48:00

12:48:04-12:48:32 12:51:05-12:51:33

Figure 5 shows oxygen distribution functions in a plane (Vpara, Vperp) for the five time intervals for event 23 August 2001, S/C 4. The distributions are averaged over 30 seconds. The first four panels present distribution functions when Cluster crossed the source region and data show high transverse heating of the O+ population. The last panel displays oxygen distribution function in the cusp region when Cluster left acceleration region, we can see beam-like behavior.

9

North, 23 August 2001, S/C 4, wave activity

Figure 6 presents data from Electric Field and Wave (EFW) and Spatio-Temporal Analysis Field Fluctuations (STAFF) experiments. The panels show the electric field power spectral density in the spin plane in the frequency bands 1-10 Hz (top panel) and 10-180 Hz (bottom panel). At 12:46 UT the sharp enhancement of the electric field power spectral density in both frequency bands in 2-3 orders was observed, this means that at this time Cluster S/C crossed region with the high plasma turbulence and with the broadband low-frequency electric wave fields activity. We suppose that O + transverse heating on this region observed by CIS at 3-4 Re is caused by resonant heating by BBELF electric field waves.

Fig. 6

10

South hemisphere, one example, 16 July 2001CIS overview

Figure 7 shows one typical example of the O+ outflow from the south cusp region, 16 July 2001 event. Panels show energy-time spectrograms of the H+ and O+ ions measured by S/C 1, 3, and 4. Similar to the northern hemisphere, oxygen upward moving ions were observed simultaneously at S/C 1 and 4 and with 30 minutes delay at S/C 3.

Fig. 7

11

South, 16 July 2001, orbit and 3D moments

Figure 8 presents Cluster S/C orbit during 16 July 2001 event: polar cap, cusp, inner dayside magnetosphere. Satellites moved mainly in X and Z directions, separation distances between S/C were: S/C 1-3 – 1,76 Re, S/C 1-4 – 0,45 Re, S/C 3-4 – 1,40 Re.

Figure 9 presents CIS data for this event, S/C 3: energy-time spectrograms of the H+ and O+ ions, parallel velocity of the both populations and three components of the O+ perpendicular (convection) velocity. At the beginning of the O+ outflow observation perpendicular velocity was rather high and unsteady.

S/C1

S/C3

S/C4

07:00

08:30

Fig. 9

Fig. 8

12

South, 16 July 2001, O+ outflow study

Figure 10 presents CIS data from S/C 3 for 16 July 2001 event: energy-time spectrograms of the H+ and O+ ions and parallel velocity of the O+ (black) and H+ (green) populations. Oxygen outflow stats at 08:27 UT near the equatorward edge of the cusp. Up to 08:24:40 UT there is sharp enhancement of the energy and parallel outflow velocity and after this time there is a gradual decrease of the outflow parameters. First injection of protons is observed at 08:21 UT. We explain the O+ behavior as following: S/C came from the north to the narrow acceleration or source region at 08:24:40 UT and left this region near the equatorward boundary of the cusp at 08:27 UT.

Fig. 10

13

South, 16 July 2001, O+ distribution functionsFig. 1108:25:02-08:25:35 08:20:08-08:20:30

08:15:01-08:15:33 08:10:05-08:10:37

Figure 11 shows O+ distribution functions in a plane (Vpara,Vperp) for four time intervals during cusp crossing 16 July 2001, data from S/C 3. The data are averaged over 30 seconds. First panel displays distribution function when S/C crossed acceleration region, other panels – during cusp crossing. One can see high perpendicular energization of the O + population during this event. 14

South, 16 July 2001, S/C 3, wave activity

Figure 12 shows EFW and STAFF data for the same time interval. The panels present the electric field power spectral density in the spin plane in the frequency bands 1-10 Hz (top panel) and 10-180 Hz (bottom panel). At 08:27 UT one can observe strong enhancement of the electrostatic wave activity. At the same time CIS data measured the beginning of O+ outflow, we suppose that O+ transverse heating is connected with the broadband low-frequency turbulence observed by Cluster at 3-5 Re.

Fig. 12

15

Statistical study of the location and size of the acceleration region using Tsyganenko models

• For statistical study we have selected 10 events from both hemispheres (see Table 1). During these events the satellites passed the acceleration region completely, from the beginning, which we defined at time of the start of the O+ outflow, to the end, which we marked as a point with maximal outflow energy and parallel velocity. Using these two points we will estimate latitudinal and longitudinal size of the acceleration region based on the data from one S/C. Moreover, as we have observations from three satellites for each event, we will have satisfactory statistics.

• To estimate the size and location of the source we will use Tsyganenko T89 and T96 magnetospheric magnetic field models for the tracing of the two observational points (“start” point corresponds to equatorward boundary of the source region and “end” point corresponds to polarward boundary of this region) to the ionosphere level (100 km).

• Both T89 and T96 models are semi-empirical and include the contributions from external sources: ring current, magnetotail current system, magnetopause currents and large-scale system of field-aligned currents. T89 model depends on the Kp-index and T96 model depends on IMF Bz and By components, solar wind dynamic pressure and Dst index. Solar wind and IMF condition for each event we have specified using data from WIND spacecraft with correct time difference.

• Table 2 contains results of the mapping based on T96 model for selected 10 events: geomagnetic latitude and longitude of the source region and size if this region, small letters “c” or ”o” correspond to “open” or “closed” model magnetic field lines in which O+ outflow is observed.

16

Tracing results from Tsyganenko model T96

Date Kp S/C Time of start/end

GM latitude of start/end

GM longitude of start/end

closed/open field lines

Size in latitude

Size in longitude

1+ 1 03:48-03:56 71.19 / 72.19 240.43 / 237.06 c/o 1.0 3.37 1+ 4 03:45-03:50 71.56 / 72.33 240.48 / 238.06 C/o 0.77 2.42

02 July 2001

1+ 3 04:27-04:34 70.13 / 71.14 230.59 / 227.96 C/o 1.01 2.63

2- 1 17:00- 17:05 76.30 / 77.23 34.06 / 32.34 C/c 0.93 1.72 2- 4 16:59- 17:01 76.90 / 77.30 35.44 / 34.37 C/c 0.4 1.07

30 July 2001

2- 3 17:36-17:38 75.52 / 75.76 19.01 / 19.06 C/c 0.24 0.05

1+ 1 11:40-11:46 72.77 / 73.89 135.31 / 136.24 C/c 1.12 0.93 1+ 4 11:35-11:39 72.63 / 73.47 137.01 / 137.46 C/c 0.84 0.45

04 August 2001

2- 3 12:13-12:18 73.17 / 74.15 127.19 / 127.99 C/c 0.98 0.8

2- 1 12:50 -12:57 79.03 / 80.31 102.45 / 103.52 O/o 1.28 1.07 2- 4 12:46-12:49 79.17 / 79.72 105.58 / 106.14 O/o 0.55 0.56

23 August 2001

2- 3 13:21-13:29 79.09 / 80.60 89.36 / 90.30 O/o 1.51 0.94

3- 1 05:46-05:52 69.47 / 70.58 158.64 / 158.60 C/c 1.11 0.04 3- 4 05:47-05:51 70.66 / 71.47 159.68 / 159.82 C/o 0.81 0.14

28 September 2001

4- 3 06:18-06:21 69.26 / 69.87 152.95 / 153.03 C/c 0.61 0.08

2- 1 07:44 -07:37 -68.23 / -69.49 193.11 / 195.76 C/o 1.26 2.65 2- 4 07:51-07:44 -68.47 / -69.72 191.28 / 194.03 C/o 1.25 2.75

16 July 2001

2- 3 08:27-08:24 -68.81 / -69.42 182.98 / 184.00 C/o 0.61 1.02

3- 1 02:22 -02:11 -70.50 / -72.11 201.85 / 205.78 C/c 1.61 3.93 3- 4 02:28-02:21 -71.34 / -72.53 200.98 / 203.72 C/c 1.19 2.74

28 September 2001

3- 3 02:54-02:45 -72.92 / -74.25 196.29 / 200.33 C/c 1.33 4.04

2- 1 10:50-10:40 -81.98 / -83.43 51.80 / 59.72 O/o 1.45 7.92 2- 4 11:05-10:56 -81.61 / -83.14 42.82 / 49.25 O/o 1.53 6.43

30 September 2001

2- 3 11:34-11:15 -81.65 / -84.52 32.24 / 42.64 O/o 2.87 10.4

3- 1 12:12 -12:02 -76.62 / -78.11 1.17 / 0.72 C/c 1.49 0.45 3- 4 12:19-12:11 -76.84 / -78.24 357.31 / 357.10 C/c 1.4 0.21

19 October 2001

3- 3 12:52-12:46 -76.03 / -77.06 348.81 / 349.01 C/c 1.03 0.2

0+ 1 16:14-15:45 -73.55 / -77.68 287.69 / 291.75 C/o 4.13 4.06 0+ 4 16:23-15:57 -74.20 / -78.09 286.38 / 289.49 C/o 3.89 3.11

14 November 2001

0+ 3 16:57-16:45 -73.92 / -75.67 280.34 / 282.72 O/o 1.75 2.38

17

Table 2

North, model results as plots

Figure 13 is a schematic representation of the location and size of the source region in the north hemisphere derived from T89 and T96 models. Location of the acceleration region in this case is model-dependant result, but the difference is very small. Latitude of the source depends on the day and it is in a limit 69-80 deg geomagnetic latitude. During one event latitudinal locations of the source region derived from different S/C are very similar. Longitude of the source also varies depending on date and O+ outflow can occur at various MLT. Longitude of the source measured on S/C 3 is very different from longitudes derived from S/C 1 and 4, difference can achieve 10 degrees.

Fig. 13

18

South, model results as plots

Figure 14 presents results of the mapping based on T89 and T96 models in the schematic form.

Using tracing results from both hemispheres we estimated the mean latitudinal size of the acceleration region in 1,5 degrees and longitudinal size in around 10 degrees.

Fig. 14

19

Conclusion

• This paper presents results of our study of upward moving O+ ions observed by the Cluster

CIS instrument in the mid-altitude cusp and polar cap regions during summer/fall 2001.

• Our investigation shows that the source of these ions is placed near the equatorward boundary

of the cusp region, partially even on closed magnetic field lines before proton injections.

• Statistical study based on Tsyganenko models T89 and T96 shows that the source is very

localized in latitude, 1.5 degrees and extends in longitude up to 10 degrees, O+ outflow can

occur at all MLT and there is no difference between oxygen outflows from the south and

north cusp regions.

• Inside the source region CIS data show high perpendicular heating of O+ ions at altitudes of

3-6 Re.

• In the same region EFW and STAFF instruments observed sharp enhancements of the

electrostatic low-frequency waves activity, the electric field power spectral density increases

2-3 times in magnitude.

• We suggest that transverse heating of the O+ population is caused by resonant energization by

the broadband low-frequency waves at altitudes 3-6 Re and below.

20