COSPAR COLLOQUIA SERIES VOLUME 9
MAGNETOSPHERIC RESEARCH WITH ADVANCED TECHNIQUES
PERGAMON
Sponsored by THE CHINESE N A T I O N A L C O M M I T T E E FOR COSPAR THE C O M M I T T E E ON SPACE R E S E A R C H (COSPAR) and
THE N A T I O N A L N A T U R A L SCIENCE F O U N D A T I O N OF CHINA Chairman Yixun Yah (China) G. Haerendel (Germany) Program Committee Ronglan Xu (China, Main Scientific Organizer) G. Rostoker (Canada, Deputy Organizer) A. T. Y. Lui (USA, Deputy Organizer) G.K. Parks (USA) B. Wilken (Germany) Wenyao Xu (China) Shui Wang (China) D. N. Baker (USA) R. Lundin (Sweden) Heng Du (China) U. Inan (USA) E. T. Sarris (Greece) V. N. Oraevsky (Russia) H. Matsumoto (Japan) Zhenxin Liu (China) L. Zelenyi (Russia) Zuyin Pu (China) C. Z. Cheng (USA) Guocheng Zhou (China) Associate Editor Lei Li (China)
MAGNETOSPHERIC RESEARCH WITH ADVANCED TECHNIQUES Proceedings of the 9th COSPAR Colloquium held in Beijing, China, 15-19 April, 1996 Edited by
R. L. Xu Centerfor Space Science and Applied Research, Chinese Academy of Sciences, P. O. Box 8701, Beijing 100080, China and
A. T. Y. Lui Applied Physics Laboratory, The Johns Hopkins University, Johns Hopkins Road, Laurel, Maryland 20723-6099, USA
PERGAMON
U.K
Elsevier Science Ltd. The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, U.K
U.S.A
Elsevier Science Inc., 655 Avenue of the Americas, New York, NY 10010, USA
JAPAN
Elsevier Science Japan, 9-15 Higashi-Azabu 1-chome, Minato-ku, Tokyo, 106 Japan
Copyright 9 1998 COSPAR All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means; electronic, electrostatic, magnetic tape, mechanical, photocopying, recording or otherwise, without permission in writing from the publishers.
First edition 1998 Library of Congress Cataloguing in Publication Data A catalogue record for this book is available from the Library of
Congress British Library Cataloguing in Publication Data A Catalogue record for this book is available from the British Library. ISBN 0 08 0433 308 In order to make this volume available as economically and as rapidly as possible the authors' typescripts have been reproduced in their original form. This method unfortunately has its typographical limitations but it is hoped that they in no way distract the reader. Whilst every effort is made by the publishers and editorial board to see that no inaccurate or misleading data, opinion or statement appears in this publication, they wish to make it clear that the data and opinions appearing in the articles and advertisements herein are the sole responsibility of the contributor or advertiser concerned Accordingly, the publishers, the editorial board and editors and their respective employees, officers and agents accept no responsibility or liability whatsoever for the consequences of any such inaccurate or misleading data, opinion or statement.
CONTENTS Preface R. L. Xu, and A. T. Y. Lui Opening Address of the COSPAR President to the COSPAR Colloquium on Magnetospheric Research with Advanced Techniques G. Haerendel Section I. Magnetospheric Observation and Measurement Techniques
Investigation of a Substorm Following an Extended Interval of Northward Interplanetary Magnetic Field A. T. Y. Lui, D. J. Williams, R. W. McEntire, S. Ohtani, L. J. Zanetti, W. A. Bristow, R. A. Greenwald, P. T. Newell, S. P. Christon, T. Mukai, K. Tsuruda, T. Yamamoto , S. Kokubun, H. Matsumoto, H. Kojima, T. Murata, D. H. Fairfield, R. P. Lepping, J. C. Samson, G. Rostoker and G. D. Reeves Structure and Evolution of the Current Sheet by Multi-Spacecraft Observations X. Y. Zhou, C. T. Russell and J. Gosling
17
Energetic Oxygen Ion Bursts in the Distant Magnetotail Q. G. Zong and B. Wilken
23
Present and Future Research Program in Solar-Terrestrial Physics at Zhongshan, Antarctica R. Y. Liu Evidence for Resonant Absorption of VLF Waves Obtained at Zhongshan Station, Antarctica K. Y. Tang, F. L. Peng, Z. L. Ning, W. Z. Cao, Q. F. Meng, Y. H. Yang and C. M. Jiao
33
37
Global-Scale Imaging: New Approaches in Magnetospheric Research J. L. Green, S. F. Fung, D. L. Gallagher, M.-C. Fok, G. R. Wilson, G. R. Gladstone, J. D. Perez, P. H. Reiff, J. L. Burch and T. E. Moore
41
UV Auroral Imaging J. S. Murphree
51
vi
Contents
High Altitude Electrostatic Fields Driving Subauroral Ion Drifts J. F. Lemaire, M. Roth and J. De Keyser
61
Search for Lunar Pickup Ions E. Kirsch, B. Wilken, G. Gloeckler, A. B. Galvin, J. Geiss and D. Hovestadt
65
The Tether System Used in the Magnetospheric Studies S. I. Klimov, F. L. Doudkin, V. E. Korepanov, A. A. Petrukovich, M. L. Pivovarov and A. V. Prudkoglyad
71
Simulations of a Spherical Section Electrostatic Analyzer J. H. Vilppola, P. J. Tanskanen and B. L. Barraclough
75
A Numerical, Double-Layer Solution for Plasma Contactors Collecting Electrons in Electrodynamic Tether Applications A. J. Li and D. W. You
85
Section H. Active Experiments
Modification Natural And Manmade EM Environment due to Ionospheric Plasma Barrier Transparency for Groundbased Transmitter Emission V. N. Oraevsky, Yu. Ya. Ruzhin, V. S. Dokukin, Kh. D. Kanonidy, B. P. Singh and G. S. Lakhina Intmins Project. The Three Levels Active Experiment in the Magnetosphere S. I. Klimov, O. V. Lapshinova, Yu. Lissakov, S. A. Romanov, W. W. L. Taylor, N. N. Antropov, F. L. Doudkin, V. E. Korepanov, M. N. Nozdrachev, W. E. Pine and M. P. Gough
91
97
Asymmetric Structure of the Luminous Neutral Barium Cloud in Chemical Release Experiments R. L. Xu and L. Li
107
Acoustic Gravity Wave (AGW) Generation during the Barium Injection Experiments Yu. Ya. Ruzhin, A. Kh Depueva and L. Palasio
111
Minimizing the Adverse Effect of the Photosheath around a Spacecraft H. Zhao, K. Torkar, R. Schmidt, C. P. Escoubet and W. Riedler Ti-C Reaction in Laboratory as a Heating Technique in Space Chemical Release Experiments F. Wu, R. L. Xu, L. Li, Z. G. Zhang and Y. B. Liu
115
119
Contents
vii
Section III. Numerical Simulation and Theoretical Modeling
Ion Beam Velocity Distributions in Plasma Sheet Boundary Layer L. M. Zelenyi, A. L. Taktakishvili, E. M. Dubinin, E. Yu. Budnik and L Sandahl Generation Mechanism of the Field-Aligned Current System Deduced from a 3-D MHD Simulation of the Solar Wind-Magnetosphere-Ionosphere Coupling T. Tanaka Configuration Instability of the Near-Earth Magnetotail in the Presence of an Earthward Plasma Flow and Substorm Onset M. H. Hong, Z. Y. Pu, X. M. Wang, Z. X. Chen, Z. X. Liu, A. Korth and R. H. W. Friedel
125
133
143
Generation of Electrostatic Waves by Nongyrotropic Protons P. D. Convery, M. Ashour-Abdalla and D. Schriver
153
Model Study of Polar Cap Arcs L. Zhu, R. W. Schunk, J. J. Sojka and D. J. Crain
159
Precipitation of Hot Protons from a Stretched Near-Earth Current Sheet W. W. Liu, G. Rostoker and J. C. Samson
165
Computer Simulation of the Three-Dimensional Decay of Thin Collisionless Current Sheets J. Bfichner and J.-P. Kuska
177
Magnetic Reconnection and Plasmoid Events in the Magnetopause Boundary Layer Z. X. Liu, H. Zhang, T. Chen, Z. Y. Pu and S. Y. Fu
187
A Simulation Study of Magnetic Reconnection in a Multiple Current Sheet System H. Zhang and Z. X.Liu
193
Theoretical Study of Vortex Induced Reconnection Processes C. Shen and Z. X. Liu
197
A Modulated Whistler Wave Model on Envelope Soliton in CRIT II Observation D. Y. Wang, N. Brenning, M. Raadu and O. Bolin
201
Author Index
205
List of Participants
207
List of Unpublished Papers
209
This Page Intentionally Left Blank
Prof. R. L. Xu and A. T. Y. Lui, editors of the COSPAR Colloquia Series #9 on Magnetospheric
Research with Advanced Techniques, in the COSPAR Colloquium on Magenetospheric Research with Advanced Techniques, 15-19 April, 1996, Beijing, China
Prof. G. Haerendel, President of COSPAR, giving opening address in the opening ceremony of the COSPAR Colloquium on Magnetospheric Research with Advanced Techniques
This Page Intentionally Left Blank
PREFACE
The COSPAR Colloquium on Magnetosphere Research with Advanced Techniques was held in one of the biggest garden hotels in Asia, Beijing Friendship Hotel, on April 15-19, 1996. The Colloquium was sponsored by the Chinese National Committee for COSPAR, COSPAR, and the National Natural Science Foundation of China. The theme of the meeting focused on four areas of modern magnetospheric studies, namely: (1) multi-point observation, (2) innovative measurement techniques, (3) active experiments in space, and (4) numerical simulation and theoretical modeling. This meeting was held after the launch of several major magnetospheric satellites which set the stage for most intensive investigations ever of the Earth's plasma environment in the framework of the International SolarTerrestrial Physics Program. These exciting new results and the ongoing discussions of innovative approaches to scientific instrumentation and spacecraft technology also provided the opportunity for the scientists, especially the Chinese Scientists to join the international space community and develop various international collaborative programs. More than seventy scientists from all over the world participated, from countries of Austria, Belgium, Canada, China, Finland, Germany, Japan, Korea, Russia, Sweden and the United States. A total of 93 papers were presented to provide a broad spectrum of exciting new multi-platform observations from the International Solar-Terrestrial Physics Program, state-of-the-art instrumentation design, and simulation and modeling on supercomputers. It was a busy time for meetings around April 1996 when this COSPAR Colloquium was held. Many colleagues who expressed an interest to attend this meeting felt unfortunate that they could not come due to conflicting commitments. Thanks to the hard work of the Scientific Organizers, Session Coordinators, and the Local Committee, the meeting preparation went smoothly without a hitch. We had received many letters and messages highly praising the organizational arrangement and scientific significance of this meeting. Fruition of this meeting includes plans of joint studies and protocols on some joint programs signed by both sides. We organized the manuscripts submitted to this Monograph into: (1) Magnetospheric Observation and Measurement Techniques, (2) Active Experiments, and (3) Numerical Simulation and Theoretical Modeling. Papers presented in the meeting but not submitted to this Monograph are listed by title as unpublished papers at the end of this Monograph. We would like to thank the Chairman and Co-Chairman of the Colloquium and the members of the Scientific Program Committee listed in the first page of this Monograph, the Chairman of the Local Organization Committee Houying Zhang and the committee members: Changchun Ai, Yidong Gu, Jicheng Kang, Kangwen Chen, Hai Lin, Quan Lin, Houren Pan, Faren Qi, Daheng Wang, and Yongren Zhao, the Executive Secretary: Fangyu Liao, Deputy Executive Secretary: Lei Li and Wei Li for an excellent job done in less than ideal circumstances. During the meeting the local Committee arranged a special program to visit the Great Wall and other historical sites around
2
Preface
Beijing, allowingthe participants to gain an impression on the long historical tradition and recent progress of China. We are grateful to W.-H. Ip for his help during the beginning of this Colloquium, and to S. Grzedzielski for his encouragement during preparing this Colloquium Monograph. We deeply appreciate the effort ofD. N. Baker, P. A. Bernhardt, J. F. Carbary, C.-L. Chang, J. Chen, S. P. Christon, R. B. Decker, R. E. Erlandson, W. R. Hu, Y. Q. Hu, E. P. Keath, K. Liou, A. T. Y. Lui, Z. Y. Pu, G. Rostoker, J. Wanliss, D. G. Wei, R. H. Wei, P. H. Yoon, L. J. Zanetti, H. Zhao, G. C. Zhou and anonymous referees for their help in improving the quality of the manuscripts published in this Monograph. In particular, the Associate Editor of this Monograph Lei Li, deserves special mention for her major contribution in bringing together the materials for this Colloquium Monograph for publication by Elsevier Science in early 1998.
Ronglan XU A. T. Y. LUI
OPENING ADDRESS OF THE COSPAR PRESIDENT TO THE COSPAR COLLOQUIUM ON MAGNETOSPHERIC RESEARCH WITH ADVANCED TECHNIQUES
Vice-President Yan, Chairman Xu, Ladies and Gentlemen, This is the Ninth COSPAR Colloquium and the second one I am attending since I became president of COSPAR. I am most pleased to be here and to speak to you, in particular, since the subject of this conference is in my field of research and since I am here also in a second function, namely as CoChairman of this event together with Vice-President Professor Yan. In this second capacity, I would like to welcome you and wish you an interesting and stimulating scientific, social and also touristic event. COSPAR was conceived at the beginning of the space era in order to promote space research and related applications with the emphasis on the exchange of results, information and opinions. By concentrating on data, methods and scientific insights COSPAR has always managed to keep clear of political interference, on the contrary, to build bridges across political barriers. Lately, after the fall of the iron curtain the bridging function of COSPAR in the area of space research may appear to have lost importance, but some barriers continue to exist and new challenges are emerging which need the platform offered by the Committee in order to bring together colleagues form East and West, from South and North, from poor and rich, from well-established and newly entering space-faring nations. In this context, COSPAR is watching with particular attention the developments in Eastern and SouthEastern Asia. Japan and China have both been highly active and successful in space for several decades, have developed powerful launchers, associated infrastructure and space industry, and have established a still growing space research activity. But other nations are just entering, like Thailand, Indonesia and Korea. China took the initiative to call in 1992 in Beijing an Asia-Pacific Workshop on Multilateral Cooperation in Space Technology and Applications. A follow-up conference was held in Islamabad last year and a third one will be held in Seoul at the end of May. This conference is meant as a forum for concerned space agencies, other institutions and administration with the aim to establish closer cooperation and to learn from each other in science and applications. COSPAR applauds such efforts and tries to raise the awareness that science is one of the best technology drivers. Therefore, even when applications like telecommunications, weather services and Earth's observations are the immediate goals of a young space nation, it should allocate a sensible fraction of the resources to basic research, because besides the technology impact there is nothing better to motivate the young generation and future technology leaders of a nation than an early exposure to the great questions and goals of basic research. But of o~ourse, to be able to pursue one's career and expand one's knowledge in research, largely free of commercial and political interests, is a great privilege. Let me now turn to this COSPAR Colloquium on Magnetospheric Research and Advanced Techniques with a few reflections. First on Magnetospheric Research. This field of research popped up with the very
4
G. Haerendel
first space vehicles. For almost four decades, it has produced an impressive list of discoveries of objects as well as of processes. It has profoundly affected our notion of the nature of the space between stars. At this point in time the field has entered its programmatic culmination with the Inter-Agency SolarTerrestrial Physics Program, with the successful launches of Geotail, Wind, Interball Tail Probe, SOHO, Polar and with CLUSTER, Interball Auroral Probe, FAST and Equator-S to follow soon. It is natural that in this phase, when thinking about he future of our field, we feel some anxiety. What will be the challenges after reconnection, flux-transfer events, boundary layer formation, substorm onset, plasmoids, coronal mass ejections, auroral acceleration and kilometric radiation have been largely understood? Of course, this understanding will not be achieved immediately. But did we not promise, when proposing all these missions, that this was what we were after, and that thesr missions, once successfully executed, would bring us close to the desired goals? What will be our new questions, the new frontiers? When I say "'our", I mean the younger ones among us, our students and successors. It is my belief that the more exciting tasks of the future will lead space plasma physics farther away from the Earth, to the planets, the outer heliosphere and above all to the Sun. We have already now an impressive program of ongoing missions in the solar system and very ambitious ones like Mars 96, Cassini and Rosetta in preparation. Solar Probe, Pluto Express, Mercury Orbiter, Intermarsnet, etc. are prospective candidates for the more distant future. All of them offer, at least in principle, opportunities for studying new aspects of plasmas and fields. But with the continuation of flight opportunities near Earth, also here new goals can be set and pursued, like the study of magnetic field-aligned processes as proposed in the IBIZA/IMPACT mission. A field of research is as good as its tools. In magnetospheric research the development of the key instrumentation has gone through several, if not many iterations. The progressive advancement of techniques, experimental, computational and theoretical, is the subject of this conference and rightly so. In addition, I regard it as highly significant that it is held in Eastern Asia, in the region of new space markets. With the expansion and increasing resolution of the accessible parameter spaces comes, almost necessarily, new knowledge and advancement of understanding. But in view of the growing distances from r arth of our future activities, the frequency of flight opportunities is bound to decrease, even with "'faster, better and cheaper" approaches. So, we are all challenged to work economically, proceed in the direction of miniaturizing, of data compression, of low-cost developments of systems and subsystems. Small satellites, use of flight opportunities mainly designed for other purposes, complementary groundbased work, these are some ways to go. I am confident that space plasma physics will not have to complain about a lack of tasks. For a while, however, the emphasis will be on harvesting and not on seeding. Let us enjoy this phase and exploit it fully! Finally, a word about COSPAR. Unified by the applications in, on and from space, COSPAR covers a wide variety of disciplines. Organized in seven Commissions, various Sub-Commissions, Panels and Task Groups, the authority for the definition of the scientific program lies entirely with the scientific community. The topics that need to be highlighted at the bi-annual Scientific Assemblies or the thematically more confined Colloquia and Workshops, they are chosen by the about 4000 Associates, who elect their Commission chairs and organize themselves in the business meetings during the Assemblies and bi correspondence in between. The Council, COSPAR's highest authority, assisted by the Bureau and Executive have the task to establish the right balance between disciplines. They, of course, also provide the necessary infrastructure and raise and distribute financial support for conference
Opening Address of the COSPAR President to the COSPAR Colloquium
5
attendees. The latter is one of the greatest concerns of the COSPAR Bureau, since the spirit of donation is fading rapidly in the present climate of economic crises. Nevertheless, we try hard to mitigate the hardships introduced by recent economic-political developments. I do not want to miss this chance to remind you all to our forthcoming Scientific Assembly in Birmingham (14 - 21 July) as well as of the preceding 10th COSPAR Colloquium on Asteroids, Comets and Meteorites at Versailles. The following Scientific Assembly, 1998, will be held in Japan, in Nagoya. This is my first visit to Mainland China. Although brief, I look forward with great expectations to this first glimpse of a great country with great people and, of course, to an exciting conference for all of us. G. Haerendel President of COSPAR
This Page Intentionally Left Blank
Session I Magnetospheric Observation and Measurement Techniques
This Page Intentionally Left Blank
INVESTIGATION O F A S U B S T O R M F O L L O W I N G A N E X T E N D E D INTERVAL OF NORTHWARD INTERPLANETARY MAGNETIC RELD A. T. Y. Lui 1, D. J. Williams 1, R. W. McEntirel, S. Ohtani 1, L. J. Zanetti 1, W. A. Bristow 1, R. A. Greenwald 1, P. T. Newell 1, S. P. Christon 2, T. Mukai 3, K. Tsuruda 3, T. Yamamoto 3, S. Kokubun 4, H. Matsumoto 5, H. Kojima 5, T. Murata 5, D. H. Fairfield 6, R. P. Lepping 6, J. C. Samson7, G. Rostoker7, G. D. Reeves 8
1The Johns Hopkins UniversityApplied Physics Laboratory, Laurel, MD 20723, USA 2Department of Physics, University of Maryland, College Park, MD 20742 USA 3The Institute of Space and Astronautical Science Kanagawa 229, Japan 4Solar-Terrestrial Environment Laboratory, Nagoya University, Toyokawa 442, Japan 5RASC, Kyoto University, Uji, Kyoto 611, Japan 6NASA/GSFC, Greenbelt, MD 20770, USA 7Department of Physics, University of Alberta, Edmonton, Canada T6G 297 8Los Alamos National Laboratory, Los Alamos, NM 87545, USA ABSTRACT Strong northward interplanetary magnetic field was observed for an extended period by the Wind spacecraft at an upstream distance of--200 RE from February 8-10, 1995. Within this period was a brief break of southward IMF on February 9 which led to a substorm of moderate inten,::.y " 500 nT) with its expansion onset at -0431 UT. In this paper, this substorm is examined with data flora eleven spacecraft in space and two networks of ground stations coveting both the northern and southern hemispheres. Detailed analysis of this event shows (1) an unusually long duration of the magnetospheric reconfiguration prior to expansion onset for this isolated substorm (2) new e,,:aence far mllltiple particle acceleration sites during substorm expansion, and (3) indications for sunward plasma flow in the plasma sheet during the late expansion phase of a substorm not related to a single acceleration site (e.g., an X-line) moving from the near-Earth tail to the more distant tail. INTRODUCTION One of the main objectives of the ISTP (International Solar Terrestrial Physics) program is to investigate the flow of energy, momentum, and mass from the Sun through the magnetosphere to the ionosphere and the atmosphere. Achieving this objective in the ISTP era has the distinct advantage over previous attempts because of the unprecedented multi-point measurements available and planned for ISTP activities. In this paper, we address this ISTP task by studying an isolated substorm with eleven spacecraft (Wind, IMP-8, Geotail, six geosynchronous satellites, one DMSP satellite, and Freja) and two networks of ground stations (Canopus and SuperDARN) from the ISTP data base. Studying an isolated substorm is particularly appropriate to address this ISTP task because a large amount of energy, momentum, and mass from the solar wind pass through the magnetosphere during a substorm episode. Furthermore, studying an isolated substorm instead of a substorm embedded within a sequence of substorm disturbances allows one to eliminate the possible interference from preceding substorm activity. Consequently, a clearer identification of features genuinely associated with the different phases of a substorm can be made without the ambiguity introduced by the remnants of previous activities. Before the onset of this substorm under study, a magnetic cloud passed over the Earth, engulfing the Earth's magnetosphere with a prolonged period of steady northward interplanetary magnetic field (IMF). This sets up an ideal situation for our study since the magnetosphere was then
10
A.T.Y.
Lui
et al.
brought to a state with very little, if not the minimum, activity present. The extensive coverage of substorm activity during this event provides us with results which range from verifying some of our conventional expectations about substorm phenomena as well as revealing some surprises which are different from the anticipation of the traditional model for the substorm expansion and recovery development. OBSERVATIONS The suite of data available for this event may be divided into four categories according to the region where measurements were taken. These are solar-wind/magnetosheath, ground-based/auroral-region, geosynchronous altitude, and magnetotail. Figure 1 shows the locations of the high-altitude spacecraft Wind, IMP-8, Geotail, and six geosynchronous satellites during this event. Wind was in the far upstream region at -200 RE in front of the magnetosphere. IMP-8 was near the dusk meridian, initially in the solar wind but later entered the magnetosheath before the substorm onset. Geotail was near the midnight meridian at X -- -32 RE. In the geosynchronous orbit were four LANL satellites (1984-129, 1987-097, 1989-046, and 1990-095), GOES-7, and GOES-8 distributed at various local times. Not shown in the sketch are two low-altitude satellites DSMP F11 and Freja. This suite of satellites makes a total of eleven spacecraft in space gathering data for this event. Supplementing these data are observations from two networks of ground stations, Canopus and SuperDARN, which cover both the northem and southem hemispheres. These constitute a rich data set, truly unprecedented in terms of coordinated observations. Spacecraft Locations at Substorm Expansion Onset February 9, 1995 at 0431 UT
1995 Feb 8-9 (39-40) Solar Wind Observations WIND
O
Wind (193,-44,-18)
~,~o
-40
-30
,--"r:
200 190)/[ fo9v~ I -'~0~10 -20 I
Xgsm
I/{
t
't
'/
""e"
'
~ ~ \ ~.~.[_~ / / \ \ 046"I--G7
Low Altitude SIC - Freja -DMSP
I
I
: ....
_. ....
1
-30 I
Geotail
],,,. IMP-8 _~aa'~-13,21,29) 1"" ~
Ground Station Network + 40 - CANOPUS SuperDARN Ygsm
1
~;'~176 r
12:00 2/8/95
18:00
00:00 2/9/95
12.'00
O6:00
IMP-8
,oh
12:00 2/8/95
.
.
.
.
.
1
, ~,.~.-
18.0o
,
9 ~
9 '
i
oo:oo
.
.
.
.
.
i
o6:oo
9 '
'''
4
'
12:00
2/9/95
-
Fig. 1 Spacecraft locations at substorm onset
Fig. 2 The solar wind parameters as monitored by WIND and IMP-8.
Solar Wind/Magnetosheath Observations The IMF and the solar wind plasma parameters from Wind are given in Figure 2 for February 8-9, 1995. It can be seen that the IMF tumed northward at --14 UT on February 8 and remained strongly northward
Investigation of a Substorm Following an Extended Interval of Northward IMF (Bz in the range of 2-10 nT) for an extended period with a brief--2-hr break at --02-04 UT on February 9 when Bz was slightly negative. The IMF B Y component was . . quite strong, in the neighborhood of --412 nT throughout most of this period. The solar wind velocity was quite steady at the nominal speed of --400 km/s. The solar wind dynamic pressure varied somewhat from --1 to 4 nPa, mostly due to the variation in the solar wind number density. This is part of an interplanetary magnetic cloud studied earlier by Lepping et al. (1996). IMP-8 observations showed similar characteristics of the solar wind as that seen by the Wind spacecraft in spite of the fact that IMP-8 was separated from Wind by --65 RE in GSM Y-coordinate. The only major difference is on the duration and magnitude of the southward IMF seen by the two satellites during this time interval. While Wind observed the brief southward excursion of IMF for less than 2 hrs, IMP-8 detected the southward IMF for a duration of--4 hr, as shown in the bottom panel of Figure 2, which is about twice as long as that seen by Wind. This difference may be related to the bow shock crossings of IMP-8 as indicated by the abrupt changes in the Bz component and in the other two magnetic field components as well. Ground-Based/Auroral-Region Observations The brief southward turning of IMF led to an isolated substorm, as indicated in Figure 3. The IMF Bz component from Wind is reproduced in the top panel. The solar wind energy transfer function, Akasofu's e parameter (Akasofu, 1981), and the auroral kilometric radiation (AKR) index based on the observed kilometric radiation from Geotail, are also shown. The epsilon parameter, which measures the solar wind input energy to the magnetosphere, was quite small (below 103 MW) before the southward turning of the IMF. Afterwards, the epsilon parameter reached to large values (near 105 MW) where substorm activity is anticipated. Indeed, the AKR index at 64s resolution indicates an intensification of AKR emission starting at-0432 UT, followed by another one at--0436 UT. The magnetic stations from Canopus, which were in the pre-midnight sector (-23.5 MLT at the substorm onset time), registered this isolated substorm activity. Shown in the middle of the figure are the magnetograms from Eskimos, Fort Churchill, Back, and Gillam, illustrating a negative bay o f - 5 0 0 nT started near the time of the increase in the AKR. There were two Pi2 micropulsation onsets detected at Gillam, the first one commencing at -0431 UT and the next one at -0436 UT, similar to the onset times indicated by the AKR index. 1995 FEB 9 (40) OBSERVATIC..,.-.,
5
Ikli
i
,
i
a
,
,
,
i
. . . . .
I
:
~
:
"~ o '10
I
'
IJ~az='
'
',
I
L_'
'
'
,
,
. . . . . . . . . . . . :
:
'.
:
',
:
'
~
9 '
'
'. . . . .
,
,
-
"
.
L
"~'104 ~
10 e
"~ -140
Jl~
-15o -160
J
00:00
06.00
9 i
FcHuESKI
L..
9 ,
9 i
,
,
,
i
18:00
9 ,
,
i
,
9
oo oo
9 1
,
,
,
. . . .
BACK
~ ,
02 00
i
12:00
03:00
.
I 450 nT . . . .
,
04:00
05:00
06.'00
07"00
08"00
Fig. 3 Shown here are the Bz component in the solar wind, Akasofu's e parameter, the AKR index (in units of V/m/~(kHz) observed by GEOTAIL), and magnetograms from Canopus
12
A.T.Y. Lui et al.
Observations from Meridian Scanning Photometer (MSP) chain from Gillam and Rankine in the Canopus network (not shown) indicate that at 0200 UT the auroral luminosity was very low but still noticeable at-660-75 ~ (Pace latitude: Baker and Wing, 1989). A t - 0 3 0 0 UT, the auroral emission became enhanced at -690-72 ~ and moved equatorward. This equatorward motion is consistent with the expected expansion of the auroral oval in response to southward IMF. Comparison between the 5577 and 6300 A emissions indicate the polar cap boundary to be at least 72 ~ implying that the substorm expansion onset took place well within the closed field line region. The backscattered signals from Goose Bay radar of SuperDarn indicate that the auroral electrojet was at -68.5 ~ Pace latitude at -0200 UT and began equatorward movement a t - 0 3 0 0 UT, like the auroral luminosity seen by the Canopus MSP. By the end of the substorm growth phase, the auroral electrojet had moved -3 ~ equatorward, in agreement with the equatorward displacement of the auroral emissions from the Canopus stations. The radar observations from Halley station (conjugate to Goose Bay) show equatorward movement of the auroral electrojet from about-68 ~ at 0300 UT to a b o u t - 6 5 ~ (also 3 ~ equatorward displacement) at 0430 UT just before the substorm onset. This shows that the auroral electrojet in both hemispheres responded to southward turning of the IMF in a very similar manner. DMSP F11 crossed the auroral oval in the dusk-midnight sector. Using the identification scheme of Newell et al. (1996), particle precipitation from this satellite indicate that the plasma sheet boundary (defined as the latitude where the electron average energy is neither increasing or decreasing with latitude) expanded from --73.8 ~ Pace latitude (22.6 MLT) at 0008 UT to --67.7 ~ (23.4 MLT) at 0334 UT. This lowering in latitude of the plasma sheet boundary is quite consistent with the equatorward movement of the auroral electrojets in both hemispheres. There are two relevant passes of Freja over the auroral region. The first one at 0224-0254 UT crossed the midnight meridian at 71 o Pace latitude and showed no large-scale field-aligned current (FAC). The second pass at 0414-0444 UT crossed the auroral oval at -03 MLT between 0433:55 and 0435:00 UT, just after substorm onset on the ground. An east-west magnetic field perturbation o f - 1 2 0 nT at -67-69 ~ Pace latitude was observed. This low value implies the large-scale FACs to be still rather weak in the early morning local time a few minutes after the substorm onset. Geosynchronous Al:ltude Observations In the geosynchronous altitude at the time of substorm onset, there were six satellites distributed in local times of 3.4 hr (1990-095), 5.0 hr (1984-129), 11.4 hr (1987-097), 17.5 hr (1989-046), 19.5 hr (GOES7), and 23.5 hr (GOEs-8). As shown in Figure 4, dispersive energetic electron injections were seen by three of the four LANL satellites, namely, 1984-129 (LT=UT+0.5), 1987-097 (LT-UT+6.9), 1989-046 (LT=UT-11.0). For the fourth LANL satellite (1990-095), there was a data gap during this time. The onset time of electron injection based on the dispersive feature (i.e., extrapolating injection time to infinite energy particles) is found to be -0444 UT, which is -12-13 min delay with respect to the first Pi2 onset or AKR increase, and -8 rain delay with respect to the second Pi2 onset and AKR increase. The magnetic field measurements from GOES-8 (LT=UT-5.0) show a gradual depression starting a t - 0 3 UT, coinciding with the equatorward expansion of the auroral oval. The magnetic field dipolarized at -0440 UT, which is -9 min delay from the first Pi2 onset and --4 min delay from the second one. GOES-7 (LT=UT-9.0) which was 4 hr in LT to the west of GOES-8 also detected a substantial decrease in the Hp component just like GOES-8 during the growth phase. The H component is positive northward parallel to the Earth's spin axis, He is positive radially inward, and ~ is positive eastward. Magnetotail Observations For this event, Geotail was located slightly in the pre-midnight mid-tail region where magnetic reconnection is anticipated to occur at substorm expansion onset. The plasma, magnetic field, and electric field measurements from Geotail are displayed in Figure 5. Geotail was in the plasma sheet at the start of the interval and crossed the neutral sheet at -0408 UT. A noticeable sudden decrease in plasma density occurred at-0415 UT, accompanied by a temperature increase and small disturbances in
Investigation o f a S u b s t o r m F o l l o w i n g an E x t e n d e d Interval o f N o r t h w a r d I M F
13
the electric and magnetic fields. These are probably the signatures of Geotail's entry into the plasma sheet boundary layer. At -0425 UT, Geotail entered a low density region representative of the tail lobe. 1995 FEB 9 (40) GEOTAIL OBSERVATIONS
1 9 9 5 F e b 9 (40) Geosynchronous Observations ~o ~ ,.;
--.
. . . . . . . .
,
101
105
Z 10-1
104
,"r
lo~
r
10 '4
"6 ~)
10 ~
.-:~176176
'
|
~.
,
P~,A,I
"
7
;: o ~500
I .
._. 400 [-
2
~o~ 104
-
,
1989-046
TI
-
-
g
o -2oo
"r
80
d) UJ 0
60
10 a ~-~ 102 20
100 ~
g
9
~60
o
,,,~
o lO
&
5
~ o (9
4
40
~"
o
.
.
.
& E
0
"r
-8
.
"~-5
8
,;0:1~
~:1~
04:00
06:00
08:00
g tit"
o 4
04:00
Fig. 4 Geosynchronous observations.
04:30
05:00
05:30
06:00
Fig. 5 Geotail observations
?.~ --0435 UT, -4 min after the substorm onset, Geotail re-entered the plasma sheet boundary layer where a brief interval of weak (-100 km/s) sunward flow was observed. Moderate (-300-500 km/s) tailward flow occurred immediately afterwards but it did not last for the entire period of Geotail's residence in the plasma sheet boundary layer (0435-0445 UT). Therefore, the tailward flow burst was extremely transient and/or localized. Plasma flow reversal occurred not just in the x-component but also in the y-component with a smaller magnitude and a shorter duration. The local magnetic field turned southward a t - 0 4 3 5 UT before the start of the weak sunward flow. The southward field was accompanied by significant changes in the By component, indicating that the field change was due to a three-dimensional field structure rather than a two-dimensional field geometry as envisioned for an Xline. Southward field was associated with enhancements of the dawn-dusk electric field to an average of several mV/m (up to 10 mV/m) in the duskward direction. The computed total magnetic and plasma pressure indicates a relatively steady increase from --0.12 nPa at 0400 UT to -0.21 nPa at 0438 UT. This steady increase in total pressure is consistent with the expected increase in the tail lobe field strength during the growth phase. Since expansion onset occurred at 0431 UT, the pressure increase in the mid-tail region as monitored by Geotail continued for -7 min after the first Pi2 (substorm expansion) onset and for --2 min after the second Pi2 onset in this event. Sunward-dawnward flow occurred at the -0435 UT re-entry of the plasma sheet boundary. A careful examination of the ion and electron energy spectrograms indicates that the strong tailward flow seen after re-entry was mainly from the high energy (> --4 keV) portion of the ion population while the lower energy ion population was rather isotropic, much like what was observed prior to the plasma dropout at -0425 UT. This indicates that in spite of the large southward field ( - - 1 0 nT) seen at this time, the plasma in the plasma sheet boundary layer remained rather undisturbed, just like before the dropout at
14
A.T.Y. Lui et
al.
-0425 UT. The tailward flow was mainly due to a high-energy tailward flowing population streaming past the satellite (this is verified by an examination of the ion velocity distribution which shows distinctly the presence of two ion populations; the transient passing of this tailward flowing population is further indicated by its disappearance well before Geotail exited the plasma sheet boundary layer and while the magnetic field was still strongly southward). The measurements of energetic ions and electrons from the Energetic Particles and Ion Composition (EPIC) instrument on Geotail shown in Figure 6 substantiate the above interpretation of transient energetic particle population passing through an ambient plasma. The anisotropy spectrograms indicate that tailward streaming of energetic ions (-67 keV) and electrons (> 38 keV) were both quite intermittent. At-0522 UT, Geotail re-entered the plasma sheet once again and detected rather isotropic energetic electrons. However, tailward streaming of energetic ions was still quite prominent and lasted till -0541 UT. This implies that the process leading to acceleration of energetic particles and their tailward streaming during substorm was still active and lingered on in the region earthward of the Geotail location. In addition, intermittent sunward streaming of energetic particles were simultaneously seen to mingle with the tailward streaming activity. The co-existence of intermittent tailward and sunward streaming of energetic particles indicates that there were more than one particle source responsible for the streaming activity seen. The plasma at this time had characteristics quite similar to the pre-substorm activity at-0400 UT with no significant plasma flow.
95 Feb 9 (040) GEOTAIL/EPIC
Fig. 6 The anisotropy spectrogram of 67 keV ions and energetic electrons > 38 keV measured by EPIC instrument on Geotail. The left labels refer to the particle streaming direction and the line overlays denote the projected direction of the magnetic field. At-0535 UT, strong (>1000 km/s) sunward-dawnward flow began. There was no noticeable tailward flow preceding the start of the strong sunward-dawnward flow even though Geotail was well within the plasma sheet during this time (plasma beta being in the range of 1-100). Furthermore, the strong sunward flow arose at the time when tailward streaming of energetic particles was still detectable. This observation forms a powerful evidence that the process which produces tailward streaming of energetic particles did not move down the tail past the Geotail location to give rise to the observed strong sunward flow. SUMMARY AND DISCUSSION An isolated substorm is investigated with ISTP data from eleven spacecraft and two networks of ground stations covering both the northern and the southern hemispheres. These observations are unprecedented in providing global monitors of the substorm event, extending to all regions in the near-Earth space, namely, the solar wind, the magnetosheath, the magnetotail, the inner magnetosphere, the auroral altitude, and the ground. Analysis of this comprehensive set of measurements leads to confirmation of
Investigation of a Substorm Following an Extended Interval of Northward IMF
15
some expected substorm phenomena, to awareness of some unusual characteristics for this isolated substorm, and revelation of some surprising features which are difficult to reconcile with the traditional substorm model. The growth phase phenomena resulting from southward IMF (McPherron et al., 1973) are well substantiated in this isolated substorm. These phenomena include (a) the equatorward expansion of the polar cap, (b) the equatorward shift of the auroral oval/electrojet in both hemispheres, (c) the enhancement of the large-scale magnetic-field-aligned current system, (d) the increase in the lobe magnetic field strength (and thus the cross-tail current intensity) as monitored in the mid-tail region, and (e) the decrease in the poloidal component of the magnetic field at the geosynchronous orbit. The features (a) and (b) imply indeed energy storage of the magnetosphere during this interaction. Features (c), (d) and (e) indicate that the energy storage phase is associated with intensification of the large-scale magnetospheric current system which encompasses not just the cross-tail current but also regions 1 and 2 field-aligned currents. Feature (e) further indicates the earthward approach of the cross-tail current in this storage phase. The unusual aspect in the growth phase of this substorm is its long duration o f - 9 0 min instead of the typical -40 min (Foster et al., 1971). This long duration is quite comparable with that for another isolated substorm on March 9, 1995 studied by Rodger et al. (1997) in which the growth phase lasted for --80 rain. Therefore, long duration growth phase may be common among isolated substorms. Another unexpected feature found in this study is the difference in duration of southward IMF between Wind and IMP-8 observations. One possibility is that there were spatial structures of the IMF in the solar wind. Solar wind observations taken at locations far from the sun-Earth line should be viewed with caution. Another possibility is that the difference may arise as a result of the IMF being modified after passing through the bow shock. This is an important question because after all it is the magnetic field in the magnetosheath that interacts with the geomagnetic field. The second unusual aspect of this substorm is the time delay between ground indications of substorm onset and substorm injection onset inferred from observations at the geosynchronous orbit. Although some time delay (about a few min) is not unexpected between these onsets because of the intrinsic uncertainties. ~a timing onsets, the delay o f - 8 - 1 3 min is unusually long. It would be interesting to examine ,z- tv.e future whether this long delay may be common for isolated substorms just like the i2-:g duration of growth phase. Relevant to mention are the observations of an isolated substorm after a prolonged quiet period reported earlier (Lui et al., 1975). This earlier analysis suggests that particle injection iv a substorm following a prolonged quiet period may occur further down the tail (as indicat-_,~ by strong earthward flow detected at a further downstream distance o f - 1 1 RE) such that the injected particles do not reach the geosynchronous orbit. The most surprising observed features of this substorm were noted in the mid-tail region. About 4 min after the ground substorm onset, large southward magnetic field and tailward flow occurred in the midtail region. The traditional interpretation would be that these are signatures of an X-line. However, a detailed examination shows that tailward flow arises from a transient energetic ion population streaming down the tail while the ambient plasma appears to be undisturbed by the activity. Moreover, the ambient plasma remained undisturbed after the passage of the tailward streaming population. The lack of response from the ambient plasma is quite contrary to the expectation of magnetic reconnection occurring nearby as implied by the observed magnitude of southward magnetic field. The variation in the dawn-dusk component of the magnetic field simultaneous with the occurrence of southward magnetic field further suggests that the magnetic field structure at that time is complicated and threedimensional, unlike the relatively simple two-dimensional geometry for an X-line. This complex magnetic field structure may represent the diffusion region of the magnetic reconnection process. However, counter to this interpretation is the undisturbed (and unenergized) character of the ambient plasma after the passing of the tailward streaming population when the magnetic field was still strongly southward and complex. One possible reconciliation of the above conflicts is postulating the diffusion region (located at some distance from the Geotail) to be responsible for the intense particle acceleration in the energetic ion beam population while the slow mode shocks associated with magnetic reconnection provide rather insignificant particle acceleration. At the present time, this conjecture has not yet been tested by full particle simulation of magnetic reonnection.
16
A.T.Y. Lui et al.
Another surprising observation in the mid-tail region deals with the occurrence of strong sunward flow in the late expansion phase when Geotail was in the plasma sheet prior to its appearance. This is an extremely fortuitous situation because almost all previous studies of satellite re-entry to the plasma sheet during the late expansion phase showed an expanding plasma sheet with strong sunward flow. The traditional interpretation of these events in previous studies is that an X-line moved past the satellite location to further downstream, allowing the plasma sheet to expand and engulf the satellite and sunward flow arises from magnetic reconnection at the X-line. This interpretation can be verified or refuted if the satellite were within the plasma sheet before the commencement of strong sunward flow in the late expansion phase. The observations reported here provides one such case and they are inconsistent with such a traditional interpretation. Similar findings were reported by Mukai et al. (1996). These observations, however, are quite consistent with the substorm synthesis model in which multiple sites of current disruption occur during the substorm expansion which subsequently lead to magnetic reconnection further downstream at the late-expansion/early-recovery phase of a substorm (Lui, 1991). In this scenario, the process responsible for substorm expansion onset is different from the process which generates strong sunward flow in association with plasma sheet recovery during the late expansion or early recovery phase. Current disruption can generate earthward flow as well as tailward flow depending on the modification of the local Lorentz force. One distinction between current disruption and magnetic reconnection is that no slow mode shock needs to be formed in current disruption for fast conversion of magnetic field energy into particle energy. Finally, another important feature worthy of note from this study is that the observed strong sunward flow appears to be associated with a visible optical feature moving equatorward on the ground as shown by the Canopus MSP observation. This study represents an attempt to use the comprehensive ISTP data set to address several outstanding issues on substorms. It is important to conduct future studies for more isolated substorms to ascertain the general features exhibited by these substorms and to help resolving the various substorm theories currently under consideration. ACKNOWLEDGMENTS This work was supported in part by funding from the Space Physics Division of the National Aeronautics and Space Administration (under the Depa~,nent of Navy Task IAF; Contract N00039-9 lC-001) and in part by the Atmospheric Sciences Division of the National Science Foundation (Grant ATM-9622080) to the Johns Hopkins University Applied Physics Laboratory. REFERENCES Akasofu, S.-I., Energy coupling between the solar wind and the magnetosphere, Space Sci. Rev., 28, 121 (1981). Baker, K. B., and S. Wing, A new magnetic coordinate system for conjugate studies at high latitudes, J. Geophys. Res., 94, 9139 (1989). Foster, J. C., D. H. Fairfield, K. W. Ogilvie, and T. J. Rosenberg, Relationship of interplanetary parameters and occurrences of magnetospheric substorms, J. Geophys. Res., 76, 6971 (1971). Lepping, R. P., A. Szabo, K. W. Ogilvie, R. J. Fitzenreiter, A. J. Lazarus, and J. T. Steinberg, Magnetic cloud-bow shock interaction: WIND and IMP-8 observations, Geophys. Res. Lett., 23, 1195 (1996). Lui, A. T. Y., A synthesis of magnetospheric substorm models, J. Geophys. Res., 96, 1849 (1991). Lui, A. T. Y., S.-I. Akasofu, E. W. Hones, Jr., S. J. Bame and C. E. McIlwain, Observation of the plasma sheet during a contracted oval substorm in a prolonged quiet period, J. Geophys. Res., 81, 1415 (1976). McPherron, R. L., C. T. Russell, and M. P. Aubry, Satellite studies of magnetospheric substorms on August 15, 1968, 9, Phenomenlogical model for substorms, J. Geophys. Res., 78, 3131 (1973). Mukai, T., et al., GEOTAIL and INTERBALL-TAIL correlative observations of magnetotail dynamics and substorms, EOS Supplement, F611 (1996). Newell, P. T., Y. I. Feldstein, Y. I. Galperin, and C.-I. Meng, Morphology of nightside precipitation, J. Geophys. Res., 101, 10737 (1976). Rodger, A. S. et al., 9 March 1995 - a simple isolated substorm? Geophys. Res. Lett., submitted (1997).
STRUCTURE AND EVOLUTION OF THE CURRENT SHEET BY MULTI-SPACECRAFT OBSERVATIONS X.-Y. Zhou l, C. T. Russell 2, and J. Gosling 3
]Institute of Geophysics, The Chinese Academy of Sciences, Beijing 100101, P.R. China 2Institute of Geophysics and Planetary Physics, University of California, Los Angeles, CA 90095, USA 3Los Alamos National Laboratory, Los Alamos, NM 87545, USA
ABSTRACT On April 22, 1979, from 0840 to 1018 UT, ISEE 1, ISEE 2 and IMP 8 were all in or near the magnetotail current sheet at 17 Re, 16 Re and 35 Re respectively while ISEE 3 monitored the solar wind 206 Re upstream of the Earth. A global perspective of the four spacecraft observations and of the ground magnetic records is presented in this paper. The hyperbolic tangent current sheet model of Harris has been used to calculate the current sheet thickness and to analyze the plasma distribution in the vertical direction. It is fbuna that during this event the current sheet thickness varied from 2.5 Re to 1.5 Re for northward IMF but thinned abruptly to 0.5 Re when the IMF turned southward. INTRODUCTION The plasma and current sheets in the geomagnetic tail are the critical regions in both the coupling of the magnetosphere to the solar wind and the storage of energy for later release. The properties and evolution of these layers are intimately involved in the different substorm theories (Russell and McPherron, 1973; Lui, 1988; Kan and Akasofu, 1989). Under normal geomagnetic conditions the current sheet is relatively thick (Fairfield et aI., 1981), but under substorm conditions it can become very thin (McPherron et aI., 1987). This thickness might be controlled by either interplanetary conditions or by the substorm activity itself (Van Hoven et al., 1987; Wiegelmann and Schindler, 1995). But a quantitative determination of the temporal evolution of the current sheet thickness is difficult. In this paper we present a case study based on the observations of ISEE 3 in the solar wind, IMS ground magnetometers at the nightside and ISEE 1, ISEE 2 and IMP 8 in the magnetotail current sheet. We concisely introduce the conditions in the solar wind, in the near and mid-tail, and on the ground. Then we use this rare but ideal condition for calculation of the current sheet thickness and the analysis of the current sheet structure. GLOBAL PERSPECTIVE ..,Spacecraft Positions and,Magnetopause Configurations On April 22, 1979 from 0840 to 1018 UT, ISEE 1 and ISEE 2 stayed in the tail current sheet at about 17 Re, crossing the current sheet center several times while the initially northward I M Bz turned Southward under conditions of constant solar wind velocity and dynamic pressure. In this interval, ISEE-3 17
18
X. Y. Zhou et al.
monitored the solar wind 206 Re upstream and 82 Re to the dawn side of the Earth. IMP 8 was also located in the tail at about 25 Re and 6 Re above the expected position of the current sheet, and 16 Re to dawn side of the Earth. Figure 1 gives the calculated magnetopause and current sheet configurations corresponding to the northward IMF condition and dipole tilt angle of the Earth at 6.2 ~ toward the sun (Petrinec and Russell, 1995; Hammond et al., 1994). Over the event period ISEE 1 and 2 are expected to slowly cross the current sheet in the center of the tail from the north to south lobe and IMP 8 to stay firmly in the northern lobe. But we will see the latter spacecraft does not, implying a significant twist of the tail current sheet.
30
...~. N
' 1 ' ' ' 1 ' ' ' 1 '
30
20
20
10
10i
i
,
,
-
,
i
'
,
,
,
i
T .
-10
-I0 -20-
-20 j
-30 .
I , , , -20
1, 0
,
i
,
,
I ~ 20 I
-3oJ Y (Re) '
, I , , , I , , = 1 ,
-2o I
o
'
i
2o ,
.
.25 -20 7-
-:
-15 -10 -5
#
0
~.
Solar Wind.and Geomagnetic Conditions
,
ol
0
.30
:
ISEE.2 e e l S E E . 1 5 10
Solar wind conditions are shown in Figure 2. The time between the two dashed lines is the period corresponding to the ground event duration. The time shown on the top of Figure 2 is the universal time on the Earth. If we assume a solar wind flow propagates along the GSM X axis with a speed Vsw=450km/s, then it would arrive at the nose of the magnetosphere after about 46 minutes. For the whole duration of Figure 2, the dynamic pressure p v 2 remained about 3.5 nPa. The interval from 0840 to 1018 UT is in the recovery stage of a moderately strong storm, and in the middle of a series of auroral electrojet activation with AE index increases from 300 to 600 nT.
15 20 25 30:
'
0
'
I -10
,
I -20
,
~
I
,
. --4O
X (Re) 0840 UT
April 2 2 , 1 9 7 9
Fig. 1. Location of ISEE 1 and 2 and IMP 8 during the observations reported herein. The top panel shows the projections of the positions in the GSM YZ plane. The tail boundary and the current-sheet location are those expected from statistical models under the observed solar wind conditions and tilt of the dipole.
Fig. 2. The interplanetary magnetic field in GSM coordinates as observed upstream by ISEE 3. The black shading denotes the period of strong southward IMF.
19
Structure and Evolution of the Current Sheet by Multi-Spacecraft Observations
A strong westward electrojet with an onset at about 0953 UT is detected by the ground magnetometers as shown by Figure 3. The vertical line drawn at 0953 UT in this figure is the onset time of the southward turning of the magnetic field seen by ISEE 1 and 2 at 17 Re in the tail. From College to Whiteshell the electrojet is very strong but at Ottawa and Saint John's it is weak, perhaps due to those stations being far from the center of the electrojet. I00 _ ~ / i
COL
"--.~"-~,
0 =
I
'
-
1
o ~........~
4oo _=----
H
- 1 0 0 -=_--
[nT] MEA-200
t~
=~
STJ
= -
,
X
0 -~ - . . ~
-100
'
I
,
E-
__.
,~.f-~
-
W H S 4 0 ~ On"
J
-\
~
~
-200
I
I
-2-
s~
-2_ =
-I
400 ~
z [nT]
_--_-= -
~
_
-
=
0 __ -100 =
,
- ~-.--._~.._.__.~_I , =
,
09:00
I 0:00
..
__
I
09:00
11:00 79 112 Apr 22
-= ,
10:00
-11:00
U T ( T I C = 3 0 Min)
Fig. 3. Ground-based measurements of the H component of the magnetic field (left) and Z component of the magnetic field (right) at selected stations at high latitudes near midnight.
Geomagnetic and Plasma Variations in the Tail The magnetic field and plasma observed by ISEE 1, 2 and IMP 8 are shown in Figures 4a and 4b respectively. An earthward propagating dipolarization of the magnetic field occurred at 0847 UT. Before 0920 UT the ISEE pair remained at the center of the current sheet, so that the total pressure is mainly due to the thermal pressure which is shown by the thin line in the bottom panel of Figure 4b. A rapid southward turning of the tail magnetic field occurred at 0953 UT corresponding to the expected arrival time of the IMF Bz southward turning seen at ISEE 3 at 0906 UT. This apparent reconnection event then appeared to move down tail, encountering first ISEE 2 and then ISEE 1. Another very significant phenomena occurred during the northward IMF conditions. IMP 8 entered the southern lobe at 0846 UT and remained there about half an hour. This implies that the tail current sheet is twisted significantly under these IMF Bz and By conditions. This twist is in the same direction as predicted by Cowley (1981). ISEE-1 - - - - - - - ISEE-2 20L,,
'
'
'
I
'
'
'
i
"'
'
'
i
'
'
'
i
'
'
'"
'".I
/1
,
.
,
,
,
,
,
,
,
,
%.,-,v - v- qn
=== 0~
v'~'v ~
m
,
-20
,
' 1
,
=
,
- .... I
,
: ,"-~--.=,.~. ,
,
I
,
~ _ . . ,
I
4O ~
20
0 08:40 79 Apt 22
09:00
09:20
09:40
UniversalTime
10:00
,. t
,
Fig. 4a. The magnetic field at ISEE 1 and 2 from 0840 to 1018 UT on April 22, 1979. The GSM coordinate system is used.
20
X.Y. Zhou et al. ISEE-2 Plasma
400
Data
Fig. ilb. The plasma conditions at ISEE 2 from 0840 to 1018 UT on April 22, 1979. From top to bottom this figure shows the plasma velocity toward and away from the Earth; the number density; the ratio of the maximum to minimum ion temperature in one spin of the spacecraft, the average ion temperature and the plasma thermal and magnetic pressures, and their sum.
-200 e,a '
1.0
c~ ~
0.5 0.0
~z
1.4
1.0 uJ 9 ~ 6x10 7 2x10 7 1.5
~
1.0
m
=..
0.5
+ ~c
0.0
08:40
09:00
09:20
09:40
10:00
Universal Time
79 Apr 22
THE HARRIS C U R R E N T SHEET S T R U C T U R E The Harris model is a simple analytical description of a one-dimensional current sheet model and is self-consistent using either MHD or kinetic theory. In a Harris current sheet, the magnetic field is given by Bx = Botanh(z~)
(1)
where B o is the lobe field, z is the distance between the observation point and the current sheet center, h is the current sheet half thickness. With observations from two vertically spaced spacecraft both in the current sheet, formula (1) can be rewritten
(2)
h = 2Az/ln[(Bo2+BoABx-Bx]Bx2)/(Bo2-BoABx-BxiBx2)]
where Az=zl-z2, zl and z2 are the positions of the two spacecraft in the Z direction of the GSM coordinates, ABx=Bxl-Bx2, B x l and Bx2 are the Bx components observed by the two spacecraft separately. The lobe field B o can be calculated under conditions of constant total pressure through the lobe. When we use a coefficient of 2.15 to recalibrate the plasma density to get a constant total pressure through the plasma sheet from one lobe to the other, the lobe field Bo=52.6 nT. The calculated current sheet center position z o in GSM and the half thickness are given in Figure 5. Up and down motion of the current sheet is very clear in the top panel before 0938 UT. Over the whole event the current sheet center was mainly above the equatorial plane, at an average height of about 1.1 Re. As shown in the bottom panel of Figure 5, before 0920 UT the average thickness is about 2.5 Re but it varied very much. ~o!~ , c
0 i
'
'
I ~_
oF
i
'
!1
~
I
'
'
'
([
"~, ,,
=
I
3
,
,
I
''
~
't
I
'
i
.
J
'
'
-7
.
'
.
.
''
....'
I
'
,s~-~ ......
.
.
.
.
.
.
,
0 79 &or 22
09:00
09:20 Universal T i m e
'
'
-I
-~ I
I --
N
08:40
A
09:40
10:00
Fig. 5. The magnetic field and inferred parameters from the Harris current sheet model. The top panel shows the magnetic field components in the GSM X direction. The middle panel shows the distance of the spacecraft from the center of the current sheet. The bottom panel shows the computed thickness of the current sheet.
21
Structure and Evolution of the Current Sheet by Multi-Spacecraft Observations
Between 0920 to 1000 UT the thickness was relatively smooth and the average value is about 1.5 Re, but oscillated near the time of the reconnection event. After 1000 UT the current sheet monotonically thinned to an average thickness of 0.5 Re. From the Harris current sheet model we also can analyze the structure of the plasma sheet along the vertical direction. We assume the current sheet magnetic structure is antisymmetric around the current sheet center and that the plasma properties as observed by ISEE 2 are symmetric about z o, so we reflect the plasma data about z o. Figure 6 shows the plasma distribution in the vertical direction. In Figure 6 the plasma density and average temperature are high at the center and low at the edges, which is very close to a Gaussian distribution. In Figure 6c the difference between the observed and the theoretical thermal pressure might be because the error in the observed Tperp and the limited detector range. Thus, while not a perfect match to the structure of the magnetotail current sheet the Harris model gives a good zeroth order approximation to it. 2
--- ,. --, .... , .... , .... 1.... I--'.-'-', .... i -~-
1.5 -
2
1.5
e .'
......
I ....
-
| .....
,,
, .....
2.S .
.-.,-...1~--,---,--.
2
.
1.5
- . - , - . . , - -
,
- - , - -
,
- , - - - , -
'"
-
1.5
0.5 '~
O.
0.5 ~
4).5 ~
-1
~
-
~
~
'
*
t
.II -
O
~
g
c~ns~y(cm 4)
"* *~
0.5 ~
0
-0.5
-I
2
-
.0.5
,O~
"~.
.
.
......".......... .
I:i; ;2
"i
3.0 ,O
6.0 10 T , ~ (K)
g o ,O
1.21( 01
O0
0.2
O,
~IS ( 2
~
,0
12
,
P. P, (nP,)
6
Fig. 6. The plasma parameters plotted as a function of the normalized distance from the current-sheet center using the Harris current-sheet model. The left-hand panel gives the plasma number density, the next panel the spin average ion temperature, the next panel the thermal pressure and the last panel the sum of the plasma thermal and magnetic pressures. REFERENCES Cowley, S. W. H., Magnetospheric asymmetries associated with the Y-component of the IMF, Planet. Space Sci., 29, 79-96, 1981. Fairfield, D. H., R. P. Lepping, E. W. Hones, Jr., S. J. Bame, and J. R. Asbridge, Simultaneous Measurements of Magnetotail Dynamics by IMP Spacecraft, J. Geophys. Res., 86, 1396(1981). Hammond, C. M., M. G. Kivelson, and R. J. Walker, Imaging the Effect of Dipole Tilt on Magnetotail Boundaries, J. Geophys. Res., 99, 6079 (1994). Kan, J., and S.-I. Akasofu, Electrodynamics of Solar Wind-Magnetosphere Ionosphere Interactions, IEEE Trans. Plasma Science, 17, 83(1989) Lui, A. T. Y., R. E. Lopez, S. M. Krimigis, R. W. McEntire, L. J. Zanetti, and T. A. Potemra, A Case Study of Magnetotail Current Sheet Disruption and Diversion, Geophys. Res. Lett., 15, 721(1988). McPherron, R. L., A. Nishida, and C. T. Russell, Is Near-Earth Current Sheet Thinning the Cause of Auroral Substorm Onset?, in Quantitative Modeling of Magnetosphere-Ionosphere Coupling Processes, pp. 252-257, Kyoto Sangyo Univ., Kyoto, Japan(1987). Petrinec, S. M., and C. T. Russell, Near-Earth Magnetotail Shape and Size as Determined from the Magnetopause Flaring Angle, J. Geophys. Res., 101,137(1995). Russell, C. T., and R. L. McPherron, The Magnetotail and Substorms, Space Sci. Rev., 15, 205(1973). Van Hoven, G., L. Sparks, and D. D. Schnack, Nonlinear Radiative Condensation in a Sheared Magnetic Field, Ap. J., 317, L91 (1987). Wiegelmann, T., and K. Schindler, Formation of Thin Current Sheets in a Quasistatic Magnetotail Model, Geophys. Res. Lett., 22,2057 (1995).
This Page Intentionally Left Blank
ENERGETIC OXYGEN MAGNETOTAIL
I O N B U R S T S IN T H E D I S T A N T
Q.-G. Zong and B. Wilken Max-Planck-Institut fiir Aeronomie, D-37191 Katlenburg-Lindau,
Germany
ABSTRACT This paper reports on the characteristics of O + beam sequences detected in GEOTAIL/HEP-LD observations in the distant tail (X=-60 and -90 Re). The relationship to concurrent plasma and magnetic field data is also discussed. The energetic O + burst usually last only 20 to 30 minutes, exhibit strong beam-like structures, appear after substorm onset, are located in the post plasmoid plasma sheet (PPPS), and are embedded in plasma flowing tailward at very high speeds (exceeding 1000Km/s). The bulk speed of the oxygen beam indicate that these ions were accelerated to the same velocity as the plasma. The energy dispersion in the oxygen bursts allows us to estimate the position of the acceleration source (presumably the acceleration region position) as lying between X=-15 and-20 Re. The distant tail is a unique location for monitoring substorm dynamic processes; details are smoothed out during the downtail propagation and only principal features are conserved. 1.INTRODUCTION Recent observations of the energetic oxygen abundance in flux ropes in the distant magnetotail showed no evidence for a flux increase above the ambient plasmajsheet abundance during normal geomagnetic activity(Lui et al., 1994). However, during an intense substorm,, energetic pxygen bursts with high speed tailward flow were found in both the CPS (Wilken et al., 1995; Zong et al., 1997b) arid post-plasmoid topologies (Zong et al., 1996, 1997a; Wilken et al., 1997). Cold ion beams (probably O + apparently of ionospheric origin) have indeed been observed in the lobes of the distant magnetotail (Mukai et al., 1994). Baker et al. (1996) showed that cold ions from the polar ionosphere can drift slowly toward the plasma sheet and reach locations relevant for the formation of acceleration regions (e.g. neutral lines) in the distant tail. Energetic atomic (0 +1 and N +1) and molecular (0 +1, N O +1 and N +1) ions of undoubtedly ionospheric origin were also observed in the deep tail (X ~ 146 Re) (Christon et al., 1994). In this paper, based on GEOTAIL HEP/LD measurements, we concentrate on the properties of energetic oxygen bursts in plasmoids and in the plasma sheet. 2. GEOTAIL/HEP-LD OBSERVATIONS 2.0 Instrument In this paper we use energetic ion measurements from the HEP-LD instrument on board the GEOTAIL spacecraft. A short account of the HEP-LD spectrometer was given e.g. by Zong et al. (1997a), whereas detailed information on the HEP-LD instrument was presented by Wilken et al. (1993) and Doke et al. (1994). For the easy reference we briefly describe the principal features of the instrument: HEP-LD is an advanced energetic particle spectrometer with time-of-flight(T) and energy(E) detection systems which determine the mass of incident nuclear particles. The novel design of the sensor head as a 'projection' camera allows the linear imaging of particle distributions in the polar angle (reference axis is the satellite spin vector pointing approximately towards ecliptic north): Twelve contiguous angular intervals over a range of 180 ~ Combined with the sectored spin plane of the spacecraft the unit sphere in phase space is covered with a total of 192 contiguous angular pixels (the satellite spin rate is 20 rpm). The energy ranges for hydrogen, helium and oxygen are approximately 40 - 3100 keV, 70 - 4000 keV, and 140 - 4000 keV, respectively. The instrument is not sensitive to the ionic charge state of nuclear particles; the mass resolution is marginal for separating carbon, nitrogen and oxygen ions. For the purpose of this paper the CNO group is referred to as 'oxygen' because this species is considered the major constituent. In addition to the energetic particle data , we use magnetometer measurements from the Magnetic Field Experiment (called MGF) on board GEOTAIL, which makes high resolution 23
24
Q.G. Zong and B. Wilken
vector field measurements. The MGF instrument is described in detail by Kokubun et aL (1994). 2.1 The Oxygen Burst in a Plasmoid
Fig. 1: Overview of the oxygen event on Jan.15, 1994, from 02:00 to 04:00 UT. From the top the panels show: Integral counting rates 'All Ions' (mass integrated) and 'CNO Ions' (the oxygen ion burst is marked in black); counting rates for hydrogen and helium; color coded azimuthal intensity distributions of the 'all ions' counting rate (look directions); GSE components and magnitude of the magnetic field (in nT).The circles in the third panel mark the magnetic field direction projected on the equatorial plane. Figure 1 gives an overview of the HEP-LD and MGF measurements for the time interval 02:00 to 04:00 UT on Jan. 15, 1994, when GEOTAIL was located approximately at (GSE: -96, 7,-5 Re) and a typical plasmoid event as described below was detected. This plasmoid event lasted from 0210 to 0330 UT. From 0210 to 0220 UT the spacecraft was in the northern lobe. In this location it detected , as a precursor of the plasmoid, a total magnetic field B, compression from 8 nT to 10 nT (the Bx and Bz components show an increase as well) prior to the arrival of the plasmoid at 0220 UT. This signature may have been caused by a compression of the lobe magnetic field in front of the moving plasmoid. The relative increase in the field magnitude is ~ 20%. In the plasmoid, the B u and Bz components show (+/-) bipolar waveforms with different times for the zero crossings (inflection point). The peak-to-peak variations are ABv=15 nT and ABe=9.6 nT. The zero crossing of the B u and Bz bipolar signatures occurred at 0243 UT and 022(3 UT, respectively. There is no indication of an enhanced core field at the time of the inflection points. This kind of plasmoid (bipolar signatures in the B v and Bz component), was first investigated by Moldwin and Hughes (1992). Inspection of Figure 1 shows that, during this event, the energetic particle peak count rates occurred when the instrument was looking sunward, i.e., there was a clear tailward flow in a downtail propagating magnetic structure. In particular, the magnetic field of this event is consistent with a 'closed loop' plasmoid with a symmetry axis roughly in the y-z plane forming an angle of approximately 450 with respect to the +z axis(Moldwin and Hughes, 1992). Because the zero crossing of the Bz and Bit bipolar signatures do not coincides, i.e., the spacecraft did not pass through the exact center of the structure (Slavin et al., 1993, 1994). Figure 1 shows the appearance of a strong energetic proton burst associated with this plasmoid event from 0220 UT to 0320 UT; embedded in this particle burst is a short intense increase in the oxygen flux from 0247 to 0310 UT. The
Energetic Oxygen Ion Bursts in the Distant Magnetotail
25
envelope of the energetic particle burst coincides with the interval of the disturbed magnetic field and as such appears to be part of the plasmoid morphology. The oxygen spike started 27 minutes after the onset of the particle burst and lasted for 23 minutes. The oxygen burst seems to be closely correlated with the strong negative B U component which developed after a strong positive excursion. The third panel of Figure 1 shows angular distributions for the particle bursts ('all ions' rate). The energetic particle burst (including the oxygen burst) display mainly tailward anisotropies. It should be mentioned that the (+/-) Bz bipolar signature passed over the spacecraft essentially during the interval for which the By component was positive. Between 0243 UT and the end of the event at 0320 UT the By component was essentially negative while Bz fluctuated with small amplitudes around zero. This indicates that GEOTAIL was traveling through a post-plasmoid-plasma sheet (PPPs). 2.2 Tailward Streaming Oxygen Event in the Plasma Sheet
Fig. 2: Overview of the oxygen from 17:45 to 19:15 UT on February 13, 1994. The format is the same as in Figure 1. Figure 2 gives an overview of the HEP-LD and MGF measurements on February 13, 1994, for the time interval 17:15 to 19:15 UT when GEOTAIL was located at (-63, 7.0,-3.8) Re in GSE coordinates, respectively. In figure 2, the Bx trace and the 'all ions' rate profile indicate that GEOTAIL was travelling in the PS and left the PS for the southern lobes at 19:15 UT. As shown in Figure 2 the angular distributions for hydrogen and helium (no oxygen) are quasi isotropic from 17:15 to 18:30 UT with a small net earthward flow. After 18:30 UT the distributions developed a strong tailward to duskward directed beam structure which ended abruptly at 19:10 UT when the spacecraft entered the southern lobes. The composition in the initial phase of the beam (18:30 to 18:47 UT) was primarily by hydrogen and helium. However, at 18:47 UT strong oxygen fluxes appeared and peaked at 19:02 UT. The oxygen peak coincided with an intensity reduction for hydrogen and helium as indicated in Figure 2. The decline of the oxygen peak after 19:02 UT began in the CPS shortly before the spacecraft passed through the PSBL towards the southern lobes. In antiphase with the decreasing oxygen abundance, the hydrogen intensity shows a transient returned to the pre-oxygen burst level as the spacecraft moved from the CPS to the lobes in the interval 19:05 to 19:10 UT. In the lobes the energetic particle flux dropped to background. In this event, the Bz component of the magnetic field was essentially positive with some fluctuations throughout the initial phase of the event and turned negative for the time of the tailward beam-like
26
Q. G. Zong and B. Wilken
distribution particular when the oxygen was present. The change of the B, polarity to principally negative values and the tailward streaming energetic particles can be explained by the widely accepted picture of a neutral line which has formed earthward of the spacecraft (Hones, 1984, e.g.). This neutral line is correlated with substorm onset at about 18:05 UT as obtained from ground-based and geosynchronous orbit observations (Zong et al., 1997b). 3. C H A R A C T E R I S T I C S O F THE OXYGEN B U R S T S
Color-coded representations of the angular distributions on the unit sphere are shown in Figure 3 for the time interval of the oxygen burst (02:15 to 03:30 UT, look directions). Four pictures show different views of the sphere: from the sun and from the tail (upper displays), from the north (lower left) and the south (lower right). Obviously, oxygen ions were flowing tailward in a highly collimated beam. Approximate values for the width in polar and azimuthal angle (determined at 10% of the maximum) for protons and oxygen are 37o x 125o and 90 o x 75~ respectively. Similar distributions are found for the event on February 13 as well. This difference in streaming direction may be caused by a mass-dependent process in the magnetic reconnection dynamics.
Fig. 3: Three-dimensional velocity distributions of the energetic oxygen burst from 02:15 to 03:30 UT on Jan 15, 1994. Particle Energy Spectra During the Oxygen Bursts Particle energy spectra for the Jan.15, 1994 plasmoid event are plotted in Figure 4. The integration time covers the entire plasmoid including the oxygen burst. As can be seen, the oxygen distribution has a well developed peak at 250 keV. Figure 4 shows no indication for the development of a peak in the proton and helium distribution functions. This can be explained by assuming that both proton and helium ions are flowing with the same bulk velocity as the oxygen. In this case the corresponding peak energies fall below the HEP/LD instrument energy window. Because of the strong collimation, the phase space density is computed only for the angular sector with the maximum flux. Number density ratios for the two oxygen events are calculated for the energy ranges covered by HEP-LD: 40 - 3100 keV for protons; 70 - 4000 keV for helium ions, and 140 to 4000 keV for the oxygen ions. Table 1 lists the results, value for the quiet plasma sheet and the normal solar wind (SW) (Bame et al., 1975) are added for comparison.
Energetic Oxygen Ion Bursts in the Distant Magnetotail
27
Fig. 4: Particle energy spectra for supra-thermal H +, He, and oxygen ions versus energy obtained for the plasmoid event on Jan. 15, 199~1. The integration time indicated on the upper right includes the oxygen burst. Even though the above number densities obtained from HEP-LD strictly speaking reflect only partial densities the following points with respect to the O / H ratios can be drawn: (1) ratios observed in the plasmoid and in the plasma sheet in the absence of enhanced oxygen abundance (burst) are less than 2.0%. (2) the solar wind ratio is approximately two orders of magnitude smaller than in the plasma sheet. (3) dynamic processes related with the generation of the observed oxygen beams can dramatically increase the oxygen/hydrogen density ratio. Lui et al. (1994) reported density ratios Nhr+ N~ obtained in flux ropes and/or plasmoids. In the energy range from ,-~ 10 keV to 230 keV, the ratio was between 0.07% to 0.75% and in the energy range from ~ 60 keV to 3 MeV (almost the same energy range covered by HEP-LD) the ratio was 1%. This is similar to the HEP/LD values for the quiet plasma sheet reference intervals. Oxygen ion bulk speeds The Geotail HEP-LD instrument measures the ion flux in 12 polar angles and 16 azimuthal sectors (total of 192 pixels); time-of-flight (T) and energy (E) are each analysed in 256 channels. The design allows the imaging of flux distributions over the complete unit sphere in phase space. This excellent angular coverage permits very good determination of
CASE 15 Jan. 1994
13 Feb. 1994
Solar Wind
Table 1: Ne s~~ Nn+ N2/+ 02:20-02:46 1.9% 13.3% 02:47-03:10 15.0% 4.8% Time [UT]
18:05-18:45 18:57-19:07
Comment Plasmoid Post-Plasmoid
0.96% 34.8%
15.5% 5.7%
Plasma sheet oxygen burst
052%
3.3%
Bame et al. (1975)
Q. G. Zong and B. Wilken
28
0247-0310 UT 15 Jan.94
Table 2: . First Order Anisotropies 0.7898
Oxygen Bulk Speed 1060 km/s
Source Location S _ - 1 6 . 7 Re
1857-1907 UT 13 Feb.94
0.890
1197 Km/s
S ~_ -20.9 Re
parameters relevant for particle distributions. The general form of a particle distribution function on the unit sphere for a given energy channel is I(O, r Applying the technique described by Sanderson and Page (1974); Sanderson and Hynds (1977); Sanderson et al. (1985) the expansion into a set of spherical surface harmonics (192 angular pixels) is represented by
I(O, r - E n c~--1AnOgr~O E :+= l e
(AnmYnem + BnmyOm)
(1)
Here, I(O, r is the particle intensity in the direction 0, r measured with respect to an arbitrary spherical coordinate system, Y~,~ are partially normalised surface harmonics, and An,~ and Bnm are the appropriate coefficients. Following (Daly et al., 1984, 1985), the oxygen anisotropy can be defined to first order in terms of the first harmonic A1 = (A10, All, Bll) which results from a spherical harmonic fit to the 192 directional ion flux measurements. The directional intensity
InI = lnAo + A1 9nl + h i g h e r - orderterms
(2)
is observed in the direction of the unit vector nl. With this representation the direction of the pole of the selected spherical coordinate system coincides with the direction of maximum intensity. Assuming that the ion distribution is isotropic in a frame moving with the (bulk) velocity V b relative to the spacecraft, then A1, the first-order anisotropy, can be related to the spectral index 7 and the bulk velocity Vb by writing
A1
-
--2(')' at-
1) -E, vb
where v is the mean ion speed in the given energy channel; the spectral index can be determined from the energy dependence of the anisotropy:
"[ -- lnAlcl-lnAlc2lnEcl"Z,~Ec2, here, subscripts cl and c2 denote different energy channels The oxygen bulk speed VD is calculated from A1 which is obtained from the spherical harmonic fit. Specifically, the bulk speed listed in Table 2 are derived from relatively broad energy channels for statistical reasons: Eel= 138.5397.5 keV, Ec2= 407.7-993.2 keV. As can be seen in Table 2 the bulk speeds of the oxygen beams exceed 1000 Km/s for the both two events (Jan. 15, 1994 and Feb. 13, 1994). Those are reasonable agreement with concurrent GEOTAIL CPI or LEP plasma bulk speed measurements. Geotail CPI measurements show the bulk speeds for those two events are -965 km/s and -1001km/s respectively. And in the same time interval, tailward proton beams at energies approximately 5 to 50 keV are also seen by GEOTAIL/CPI instrument (B. Paterson, Private communication). GEOTAIL/LEP Plasma measurements do not existing for the event on January 15, 1994; however, values for the plasma speed on February 13, 1994, vary between -1000 and-1600 km/s (Zong et al., 1997b). Those results indicate that the oxygen ions were accelerated to the nearly the same velocity as the plasma ions (proton and helium) in the course of these two events. Energy Dispersion of The Oxygen Ions The GEOTAIL satellite experienced numerous encounters with energetic oxygen bursts in the distant magnetotail. These bursts are usually related to substorm activity and exhibit strong beam-like angular distributions. Energy
29
Energetic Oxygen Ion Bursts in the Distant Magnetotail
dispersion effects are rarely seen in the time (T) or energy (E) spectrograms. However, the present two events display some dispersion effects as can be seen in the E versus T scatterplots in Figure 5 and 6.
15.01.1X=-96. 99402:55-03:11 DE,Ed_B,Ms/ode: E+T, Def=lkV Y=7. 2, Z=-4. 2, UT RE 6 1031i"~: 102~~""02:5:!i5i"~s~'~'2!02:56i":2:~..,.~.~:02:57 jj."[s~\.2~I02:58 Geotail HEP-LD
Ill
l
I
tllltl|
I
I
tllll,l
. . . . . .
Ill
I
I
IIIlli[
I
I
IIlllll
I
t
Ill|Ill
I
l
lllllll
,
l
tJdlLI
I021031i~ 50921:] "~ ~"."03: " 0"0!~ .~ 03:01! . ~ ~ 03:02,1 •
.................... ~................... ,................... ~......._.-:_. . . . . . . . . i"" _.. i"" . .... . "" -
~10311 ~~.,,I ~30:30 i ~ i : 03:0I4 ~ ~ 03:0ti5 ~ ~ 03:"0!6 ~,~102 7.:
102 101
2:
":2:
,
,,i
10 0
~ ,
,
,,,,,,|
-.,.....: .
,
,
"
"2: .
.
.
,,,;,,i
101
2::
" ":2: .
.
,
,
.
,,
10 0
.
....
.
|
.,
-
,'"1
101
'
" ":2:
---e
. . . .
?2:
....
'
'
''
10 0
....
|
101
Time of flight [nsec]
'
'
.
- : ......
'''"'|
'
10 ~
'
'''"'!
m ~
101 '
......
_..
Fig. 5" A series of 16 scatter plots showing the mass and energy distributions during the oxygen burst in the post plasmoid on Jan. 15, 1994. Mass line in each frame indicate the loci of hydrogen, helium, CNO and SiFe (from the left). The frames show the oxygen ion with higher energy was detected previously than which with lower energy. With the assumption that all particles left the source region simultaneously the distance to the source region can be estimated (the method is also used by Williams (1981)). Consider two oxygen particles, A and B, with different energies and which leave the source at the same time (the spacecraft position is assumed to be stationary). The observations of the temporal evolution of the oxygen distribution via the time-of-flight effect can be wrote:
S ~r
=
~
S -
vlCos01
(3)
v2Cos02
where the 5T is the time difference between the different ions; S is the distance to the hypothetical source and 0 is the particle pitch angle observed by GEOTAIL. The subscripts 1 and 2 denote the different particles. From this we obtain for the distance S
S = vl v 2 C o s 0 1 C o s O 2 5 T
(4)
v2Co802 - I)1C0801
The results are listed in Table 2. The time intervals from 0259 to 0302 UT for Jan.15,1994 and from 1902 to 1906 UT for Feb.13,1994 and the energy channels (120 KeV and 686 KeV) are used in this back-trace calculation. It should be stressed, however, that the assumption of simultaneous emission of the particles is an essential pre-requisite for this estimate. Furthermore, the particle motion in the central plasma sheet may be nonadibatic or chaotic (Biichner and Zelenyi, 1989), and If the motion is chaotic, the particles will lose all "memory" of previous trajectory(Lyons and Speiser, 1982; Speiser, 1991). 4. DISCUSSION The discussed energetic oxygen bursts consisted of singly-charged O + ions as established by the EPIC instrument (private communication with S.Christon). This is strong evidence for an ultimate origin in the Earth's polar ionosphere. In the two oxygen bursts, the number density ratio No is much higher than that found in the normal plasma sheet. In addition, it is established that the oxygen bursts are closely related to substorms and can be understood as a direct
NH
30
Q. G. Zong and B. Wilken Geotail HEP-LD
13.02.1994 18:54-19:10UT
DE,Ed_B,Mode: s/E+T,Def=lkV :2:
X=-63.Y=7. 2, Z=-3. 0, RE 8
:
~--~102103 k ~5:.:
-4
9
~~_5_:_ ~~.~_
-
~_ 19:01
102 - 2:
-
!-=:
102103.~~6 ~..~::. 100
,,i
,
,
,,,,,,I
101 100 ,
,
,,,,,,I
,
--
!~:
: ~~.:: " ~.~.~~_119:09 ,
,
,,
101 100 101 100 Time of flight [nsec]
....
!
,
,
,,,,,,i
,
,
,,,,,,!
,
,
,,,,,,!
,
,
,,1,,,i
101 ,
",,
,,e_ ,,,,1
Fig. 6" A series of 16 scatter plots showing the mass and energy distributions during the oxygen burst in the central plasma sheet on Feb. 13, 1994. Mass line in each frame indicate the loci of hydrogen, helium, CNO and SiFe (from the left). The frames show the oxygen ion with higher energy was detected previously than which with lower energy.
product of the substorm process. By invoking the neutral line model we can explain the observed energetic oxygen bursts and the relation to substorm activity in this framework" First, the fast earthward flow with positive Bz was observed when GEOTAIL was located in the distant plasma sheet as seen in Figure 2. This is consistent with the existence of a 'constant' quiet time neutral line in the distant tail ( presumably at ~ 100 Re downtail) (Hones and McPherron, 1994; McPherron, 1991; Richardson et al., 1987; Richardson and Cowley, 1987). -
Second, groundbased magnetometer and geostationary particle observations show substorm expansion onsets at 02:10UT on Jan. 15, 1994(Zong et al., 1997a) and 18:05 UT on Feb. 13, 1994 (Zong et al., 1997b). At these times a second neutral line formed earthward of GEOTAIL as evidenced by the signatures of the magnetic field and the onset of tailward particle flow or the distant neutral line have moved Earthward of Geotail at 60 Re. The mean oxygen tailward flow speed is greater than 1000 km/s (Table 2).
-
-Third, magnetic field line reconnection at the new neutral line (closer to Earth) eventually results in the formation and rapid tailward motion of a plasmoid. Magnetic tension causes the plasmoid to move downtail at high speed (-400 to-600 km/s)(Richardson et al., 1987; Richardson and Cowley, 1987). Oxygen ions, extracted by the substorm process from the polar ionosphere, drift slowly (about 150 km/s) tailward in the lobe fields towards the active reconnection region which is located earthward of GEOTAIL and leads to appear after substorm onset about 20 to 30 minutes. In post-plasmoid-plasma- sheet (PPPs), plasma and oxygen ions are accelerated to velocities greater than 1000 km/s, which is far greater than the speed of plasmoids(Moldwin and Hughes, 1992). This process leads eventually to the observed bursts of energetic O + ions in the post-plasmoid plasma sheet (PPPS) characterised by a consistently negative Bz component (Zong et al., 1997a,b). 5.SUMMARY (1) Tailward flowing bursts of energetic oxygen (O +) were observed in the distant magnetotail (-X great than 65 Re) after the onset of a substorm. These particle bursts, embedded in the PPPS, showed bulk speeds of about V x = - l l 0 0 Km/s. The atomic charge state Q = I suggests a polar ionospheric origin. The oxygen bursts observed in the distant tail appear to be the products of global dynamical processes in substorms. The present particle observations are consistent with the widely accepted neutral line model of substorms. (2) The bulk speed of the oxygen burst, at least on February 13, 1994, is similar to the plasma speed; The energy dispersion of the oxygen ions allows us to estimate the position of the acceleration source (probably the neutral line position). Within the constraints of the assumptions, source locations between X=-15 to-20 Re were obtained.
Energetic Oxygen Ion Bursts in the Distant Magnetotail
31
(3) The number density ratio No observed during the bursts was much higher than values usually found in the distant NN+ plasma sheet and in the solar wind. This is interpreted as evidence for substorm associated extraction of oxygen ions from the polar ionosphere and subsequent acceleration in the magnetotail; by this process the oxygen abundance is temporarily drastically increased. (4) The angular distribution of the oxygen bursts is restricted to a rather small part of the unit sphere (about 7% to 10%). We suggest that a large amount of heavy ions from the ionosphere can be transferred to the distant tail and accelerated to high energies during substorm activity and these oxygen (O +) ions from the polar ionosphere can be considered as "Tracer Ions" in the substorm dynamical and magnetic reconnection process. REFERENCES Baker, D. N., T. I. Pulkkinen, P. Toivanen, M. Hesse, and R. L. McPherron, A possible interpretation of cold ion beams in the earth's tail lobe, J. Geomagn. Geoelectr., 48,699-710 (1996). Bame, S. J., J. R. Asbridge, W. C. Feldman, and M. D. Montgomery, Solar wind heavy ion abundances, Solar Physics, 43,463-473 (1975). Biichner, J., and L. M. Zelenyi, Regular and chaotic charged particle motion in magnetotaillike field reversals - 1. basic theory of trapped motion, J. Geophys. Res., 94, 11821-11842 (1989). Christon, S. P., G. Gloeckler, D. J. Williams, T. Mukai, R. W. McEntire, et al., Energetic atomic and molecular ions of ionospheric origin observed in the distant magnetotail flow-reversal events, 9rl, 21, 3023-3026 (1994). Daly, P., T. R. Sanderson, and K. P. Wenzel, Survey of energetic (e 35 kev) ion anisotropies in the deep geomagnetic tail, J. Geophys. Res., 89, 10733-10739 (1984). Daly, P., T. R. Sanderson, and K. P. Wenzel, A method to measure the bulk velocity of an energetic ion distribution in the presence of ion composition mixing, J. Geophys. Res., 90, 1499-1505 (1985). Doke, T., M. Fujii, M. Fujimoto, K. Fujiki, T. Fukui, et al., The energetic particle spectrometer HEP onboard the GEOTAIL spacecraft, J. Geomagn. Geoelectr., 46,713-733 (1994). Hones, E. W., Jr., Plasma sheet behavior during substorms, in Magnetic Reconnection in Space and Laboratory Plasmas, Geophys. Monogr. Set., vol. 30, edited by Hones, Jr., E. W., pp. 178-184, AGU, Washington, D. C. (1984). Hones, E. W., Jr., and R. L. McPherron, Evidence supporting the near-earth neutral line model of substorms: A reminder and update, in Proceedings of International Conference on Substorms 2, Fairbanks, U.S.A., March 7-11, edited by J. R. Kan, J. D. Craven, and S.-I. Akasofu, pp. 167-173, Univ. of Alaska, Fairbanks (1994). Kokubun, S., T. Yamamoto, M. Acuna, K.Hayashi, K. Shiokawa, et al., The GEOTAIL Magnetic Field Experiment, J. Geomagn. Geoelectr., 46, 7-21 (1994). Lui, A. T. Y., D. J. Williams, S. P. Christon, R. W. McEntire, V. Angelopoulos, et al., A preliminary assessment of energetic ion species in flux ropes/plasmoids in the distant tail, Geophys. Res. Lett., 21, 3019-3022 (1994). Lyons, L. R., and T. W. Speiser, Evidence of current sheet acceleration in the geomagnetic tail, J. Geophys. Res., 87, 2276 (1982). McPherron, R. L., Physical processes producing magnetospheric substorms and magnetic storms, Geomagnetism, 4, 593-739 (1991). Moldwin, M. B., and W. Hughes, On the formation and evolution of plasmoids: A survey of ISEE 3 geotail data, J. Geophys. Res., 97, 19259-19282 (1992). Mukai, T., M. Hirahara, S. Machida, Y. Saito, T. Terasawa, et al., Geotail observation of cold ion streams in the medium distance magnetotail lobe in the course of a substorm, Geophgs. Res. Lett., 21, 1023-1026 (1994). Richardson, I. G., and S. W. H. Cowley, Plasmoid-associated energetic ion bursts in the deep geomagnetic tail: Properties of the boundary layer, J. Geophgs. Res., 92, 9997-10013 (1987). Richardson, I. G., S. W. H. Cowley, J. E. W. Hones, and S. J. Bame, Plasmoid-associated energetic ion bursts in the deep geomagnetic tail: Properties of plasmoid and the postplasmoid plasma sheet, J. Geophys. Res., 92, 9997-10013 (1987).
32
Q.G. Zong and B. Wilken
Sanderson, T. R., and R. J. Hynds, Multiple telescope measurements of particle anisotropies in space, Planet. Space Sci., 25,799-807 (1977). Sanderson, T. R., and D. E. Page, Spherical harmonic analysis of satellite anisotropy measurements, Nuclear Instruments and Methods, 110, 177-182 (1974). Sanderson, T. R., R. Reinhard, P. van Nes, and K. P. Wenzel, Observations of three-dimensional anisotropies of 35 to 1000-kev protons associated with interplanetary shocks, J. Geophys. Res., 90, 19-27 (1985). Slavin, J. A., M. F. Smith, E. L. Mazur, D. N. Baker, J. E. W. Hones, et al., Isee 3 observations of traveling compression regions in the earth's magnetotail, J. Geophys. Res., 98, 15425-15446 (1993). Slavin, J. A., C. J. Owen, and M. Hesse, Evolution of the plasmoid-lobe interaction with downtail distance, Geophys. Res. Lett., 21, 2765-2768 (1994). Speiser, T. W., Particle motion in the tail current sheet, Adv. Space Res., 9, 151 (1991). Wilken, B., W. Giittler, A. Korth, S. Livi, W. Weiss, et al., RAPID: The imaging energetic particle spectrometer on Cluster, in Cluster: Mission, Payload and Supporting Activities, pp. 185-217, Eur. Space Agency Spec. Publ., ESA SP-1159 (1993). Wilken, B., Q.-G. Zong, I. A. Daglis, T. Doke, S. Livi, et al., Tailward flowing energetic oxygen ion bursts associated with multiple flux ropes in the distant magnetotail: Geotail observations, Geophys. Res. Lett., 22, 3267-3270 (1995). Wilken, B., Q.-G. Zong, T. Doke, K. Maezawa, G. D. Reeves, et al., A isolating substorm occurred at first phase of cir: Geotail observations, J. Geophys. Res., P r e p a r i n g (1997). Williams, D. J., Ring current composition and sources: An update, Planet. Space Sci., 29, 1195-1203 (1981). Zong, Q.-G., B. Wilken, I. Daglis, S. Livi, J. Woch, et al., Geotail observation of energetic ion specics and magnetic field in plasmoid-like structures in the course of a substorm, in Proceedings of the third international conference on substorms (ICS-3), pp. 619-624, ESA SP - 389, ESA (1996). Zong, Q.-G., B. Wilken, G. Reeves, I. Daglis, T. Doke, et al., Geotail observation of energetic ion specics and magnetic field in plasmoid-like structures in the course of a substorm interval, J. Geophys. Res., 102, 11,409-11,428 (1997a). Zong, Q.-G., B. Wilken, T.Mukai, G. Reeves, T. Doke, et al., Energetic oxygen ion bursts in distant magnetotail as a product of intense substorms: Three case studies, J. Geophys. Res., 102, In Printing (1997b).
PRESENT AND FUTURE RESEARCH PROGRAM IN SOLARTERRESTRIAL PHYSICS AT ZHONGSHAN, ANTARCTICA R. Y. Liu
Polar Research Institute of China, 451 Jinqiao Road, Shanghai 200129, China
ABSTRACT Zhongshan Station has an ideal location for ground-based measurements in studying important problems related to solarterrestrial physics. The observations at Zhongshan Station include solar radiation, magnetic field variations, aurora data, ionospheric monitoring, middle and low atmospheric measurements and radio propagation. The present and future research include the program of global character research of solar-terrestrial system in the Antarctic, polar ionospheremagnetosphere coupling and auroral dynamics as well as two international collaborative studies. INTRODUCTION The Antarctic Zhongshan Station was established in February, 1989. It is located in Princess Elizabeth Land Larsemaan Hills of East Antarctica, having geographic coordinates of 69~ and 76~ In Fig.1 is shown the location of Zhongshan Station with respect to the Antarctic geospace network (AGONET) stations and the HF radar field of view (Dudeney, 1994). The annual mean temperature at Zhongshan Station is -9.5 ~ with maximum of 9.5 ~ in summer and minimum of-33.6~ in winter. The midnight sun lasts for 54 days, while the number of days with "Polar night" is approximately 58. The corrected geomagnetic latitude of Zhongshan is about 75 ~ and the equivalent L value is 14. Zhongshan Station is situated under the ionospheric projection of the magnetospheric cusp region at noon, and the polar cap region at midnight, twice passing through the auroral oval during a day. Zhongshan Station is also located in the area which is affected by ozone depletion. The weather is clear in most of the time. It is suitable for both optical and electromagnetic observations. Therefore Zhongshan Station is an ideal ground base for studying important problems related to solar-terrestrial physics. Another advantage of Zhongshan Station is that the distance between Zhongshan and the Australian Antarctic station, Davis, is about 100km. Both Zhongshan and Davis are located under the cusp region, having similar measurement instruments such as magnetometer, scanning photometer, all sky TV camera, riometer, ionosonde and so on. The Zhongshan-Davis pair will make a significant contribution to high latitude ionospheric and magnetospheric studies. It is also shown in Fig. 1 that Zhongshan Station has the same L value as South Pole Station with CGMT difference of about 5 hours. Coordinate observations at Zhongshan and South Pole would enlarge the observation time of high latitude phenomena. In Fig.2 the Antarctic continent is projected along the earth magnetic field lines to the Arctic (Dudeney, 1990). It can be seen that the conjugate point of Zhongshan is near Svalbard, a well equipped international ground base for studying solarterrestrial physics. This figure clearly illustrates the possibilities for carrying out magnetic conjugate observations between Zhongshan and the stations at Svalbard. THE PRESENT OBSERVATIONS AT ZHONGSHAN STATION A program named the global character research of solar-terrestrial system in the Antarctic was conducted over the past five years. The purpose of this program is to study the basic behavior of the coupling and interaction between various regions of solar-terrestrial system, and inquire into the response processes and the global effects of the system to the solar electromagnetic and particle radiation, using both ground based measurements and satellite in situ measurements. The main 33
R. Y. Liu
34
Fig. 1. Location of Zhongshan Station with respect to the AGONET stations and the HF radar field of view. (after Dudeney, 1994)
Fig. 2. Projection of Antarctica along the magnetic field lines on to the northern hemisphere. (after Dudeney, 1990)
research topics include: solar radiation measurement, cusp dynamic observations and modeling, polar ionospheric behavior and its effects on radio propagation, mechanisms of Antarctic ozone depletion hole and the effects of solar activity on the middle and low atmosphere, coordinate observations and analysis of solar bursts and responses of various regions of the system. This program was supported by the National Committee of Science and Technology and the National Committee on Antarctica, and 7 research institutes and universities took part. Now a composite ground based measurement system has been built which contains following observations: 1) Solar Radiation -- a telescope working at 10cm wavelength -- solar radiometer -- UV visible spectroscope -- solar UV spectroradiometry 2) Magnetic Field Variations -- standard magnetograms -- magnetic pulsations -- ELF/VLF emissions
3) Auroral Data -- all sky TV camera -- meridian scanning photometers (6300/?I,5577/~ ,4278/~ ) 4) Ionospheric Monitoring -- ionogram, using analog ionosonde TD-4 in early time, now using a digisonde DPS-4 -- drift measurement, digisonde DPS-4 -- absorption data, riometers (-30MHz and-~50MHz)
5) Radio Propagation -- HF field strength and time delay -- VLF field strength and time delay, navigation receiver.
Present and Future Research Program in Solar-Terrestrial Physics
35
6) Middle and Low Atmosphere -- column amounts of ozone, Brewer ozone spectrophotometer -- ozone profile, ozone-sonde -- stratospheric Lidar -- atmospheric electric field The measurement instruments at Zhongshan Station are listed in Table 1 with the responsible organizations and operating durations. Table 1.
Instrumentation on solar-terrestrial physics at Zhongshan Station
Item
Start Time
10 centimeter solar telescope Quartz photoelectric variometer Induction magnetometer Plasma wave measurement VLF receiver Ionosonde TD-4 Digisonde DPS-4 Imaging Riometer Riometer (-30MHz) Riometer (-50MHz) VLF navigation signal receiver HF field strength meter All sky TV camera Scanning photometer Surface ozone detector Ozone-sonde Brewer ozone spectrophotometer Solar UV spectroradiometry Solar radiometer UV visible spectroscope Stratospheric Lidar Atmospheric electric field mill
1993.4 1991 1992 1996.1 1991.4 1990.3 1995.1 1997.1 1995 1990 1990 1990 1995 1995 1995.1 1993.3 1993 1990.2 1990.2 1991.3 1993.3 1990.2
Responsible Organization BO IG IG PRIC IG CRIRP PRIC PRIC PRIC CSSAR CRIRP CRIRP PRIC PRIC PRIC lAP CAMS IAP IAP CSSAR lAP IAP
Remark
in coop. with UON
in coop. with NIPR
in coop. with NIPR in coop. with NIPR in coop. with NIPR
Abbreviation of Organizations: BO CAMS CRIRP CSSAR IAP IG NIPR PRIC UON
Beijing Observatory, Academia Sinica, Beijing 100080 Chinese Academy of Meteorological Sciences, Beijing 100081 China Research Institute of Radiowave Propagation, Xinxiang 453003 Center for Space Science and Applied Research, Academia Sinica, Beijing 100080 Institute of Atmospheric Physics, Academia Sinica, Beijing 100029 Institute of Geophysics, Academia Sinica, Beijing 100101 National Institute of Polar Research, Japan Polar Research Institute of China, Shanghai 200129 University of Newcastle, Australia
Most of the measurement instruments in Table 1 are used in regular observation and long-term monitoring. Some instruments, for example the solar telescope and solar UV spectroradiometry, will be .stopped during polar night. Based on the exiting measurement data from Zhongshan Station combined with other data from other places, we have done data analysis and theoretical and model studies and got some preliminary results which have shown various distinctive features of geospace in Antarctica (Liu, 1996).
36
R.Y. Liu
FUTURE PLANS FOR SOLAR-TERRESTRIAL PHYSICS AT ZHONGSHAN STATION The program of global character research of solar-terrestrial system in the Antarctic will be continued. The research emphases in the future plans are on polar ionosphere-magnetosphere coupling and auroral dynamic processes, conjugate observations, monitoring and modeling of Antarctic Ozone, UV radiation and atmospheric composition. Here, only the polar ionosphere-magnetosphere coupling and auroral dynamic processes are mentioned in detail. High Latitude Ionosphere -- ionospheric properties by using digisonde DPS-4 ionograms. -- plasma convection by using digisonde DPS-4 drift measurement, response of dayside cusp and polar convection to the orientation and strength of the IMF. -- ionospheric absorption by using riometers (-30MHz and-50MHz) and an imaging riometer (8• elements) -- ionospheric signatures of cusp processes. Aurora and Particle Precipitation -- cooperative research on upper atmospheric physics between the National Institute of Polar Research, Japan, and the Polar Research Institute of China during the period from 1994 to 1999. Scientific Objectives: Auroral particles and auroral emissions in the cusp region, Ionospheric disturbances in the polar cap region associated with energetic particle precipitations, Correlative studies of high latitude ionosphere with HF radar observations from Syowa Station and ground based auroral and ionospheric observations at Zhongshan Station. The system consists of an all sky TV camera, a scanning photometer, a surface ozone detector, an imaging riometer, a monochromatic auroral TV canaera and an auroral spectrophotometer. In summer season, 1 or 2 Japanese scientists visit Zhongshan Station and establish observation systems in cooperation with Chinese scientists. In winter time, auroral and ionospheric observation instruments are operated by Chinese scientists. Space Plasma Wave Studies -- cooperative research between the University of Newcastle NSW Australia and the Polar Research Institute of China. The scientific aim of the cooperative project is to study the source and propagation characteristics of ultra-low frequency (ULF) hydromagnetic waves in the 0.001-1Hz band using identical induction magnetometers located at the Australian Antarctic station, Davis, and the People's Republic of China station, Zhongshan. Under the agreement the University of Newcastle, in cooperation with the Atmospheric and Space Physics Group of the Australian Antarctic Division, installed an induction magnetometer at Zhongshan Station in January, 1996, and the Polar Research Institute of China operate and maintain the instnament year round.
REFERENCES Dudeney, J. R., Antarctica, An International Platfom~ for Geospace Science, SCAR Atmospheric Science Working Group
Meeting, Invited Paper Review (1990). Dudeney, J. R., The Southern Hemisphere Auroral Radar Experiment, (SHARE), Short Note, Antarctic ,Science, 6(1), 123 (1994). Liu, R. Y., Advances in China in the study of Solar-Terrestrial Physics in Antarctica, in Geomagnetism, Atmosphere, Space Researches andApplJcations, edited by W. Y. Xu, pp. 125-136, Seismological Press (1996).
EVIDENCE FOR RESONANT ABSORPTION OF VLF WAVES OBTAINED AT ZHONGSHAN STATION, ANTARCTICA K.Y. Tang, F.L. Peng, Z.L. Ning, W.Z. Cao, Q.F. Meng, Y.H. Yang and C.M.Jiao
Institute of Geophysics, Chinese Academy of Sciences P.O.Box 9701, Beijing, 100101, P.R.China
ABSTRACT Evidence for resonant absorption of VLF waves was obtained at Zhongshan Station, Antarctica. The L value for Zhongshan Station is about 14. Due to such high geomagnetic latitude, no whistlers were received, but a lot of VLF emissions have been recorded since the station was founded in 1990. The VLF emission recorded in Zhongshan Station are usually wide-band emission, from a few hundred Hz to about 15kHz. In February and June of 1993, we recorded two sets of VLF emissions never seen before. The main feature for the emission is a blank curve around 5kHz cut from the wide-band emissions. We explain these observations in term of resonant absorption of VLF waves by ions. INTRODUCTION Many types of very low frequency plasma waves were received in Antarctica area. Whistler is one of the typical VLF waves that originates in lightning. The dynamic spectra of the whistler could help us to explore the plasma structure along the wave path inside the magnetosphere and ionosphere, and to understand the propagation processes and the excitation mechanism. Other VLF emissions, such as hiss, chorus, lion roars, periodic and quasi -- periodic emissions, originate within the plasma itself. All of those VLF waves could be easily received by a set of band pass VLF receiver with a tall enough antenna, especially in Antarctica area, due to the quiet electromagnetic environment. Besides using the VLF waves as a remote sensing tool to reveal the properties of plasma through which they travel, the high density and narrow bandwidth (a few hundred Hz to 25KHz, usually) also could indicate the presence of a new kind of wave particle interaction that converts the kinetic energy of charged particle into coherent electromagnetic radiation. This process is called coherent wave instability, or CWI. Thirdly, the energetic charged particles are precipitated into the ionosphere through resonance scattering by those VLF waves, causing enhanced thermal ionization, light and heat. Based on those reasons mentioned above, to better understand and use the CWI, a series of VLF simulation experiments were conducted by the Space, Telecommunications, and Radioscience (STAR) Laboratory of Stanford University. Controlled waves were injected from a 150-kW transmitter, through a 42 km antenna located at Siple Station, Antarctica into the magnetosphere and were received on the conjugate area in Quebec of Canada and by satellites. The key findings are 1) Coherent VLF signals often 37
38
K.Y. Tang et
al.
exhibit exponential temporal growth. 2) Temporal growth requires that the input signal exceed a threshold. 3) Narrowband triggered emissions can be entrained by Siple frequency ramps. 4)Simulated hiss shows coalescence of selected noise wavelets into longer and stronger chorus-like emissions(Helliwell, 1988). In 1994 and 1995, a further research on VLF simulation had been conducted by British Antarctic Survey, STAR Laboratory and University of San Paulo of Brazil, they used a chain of VLF receivers along the Antarctica Peninsula to record the lightning-induced electron precipitation (LED, i.e. trimpi events). The intersecting network of paths from the four northern hemisphere VLF transmitters NAA(Main, USA), NSS, NPM(Hawaii, USA) and NLK(Washington, USA) into four Antarctic receivers Faraday(UK), Ferraz(Brazil), Palmer(USA), Rothera(UK) had resulted in the recording of some interesting electron precipitation data excited by VLF wave(Smith et al., 1996). OBSERVATIONS In the year 1984-1985, the first Chinese Antarctic research expedition team started their work and founded the first Chinese Antarctic station -- the Great Wall station of China. Since then we have recorded many types of whistler waves, such as one and two-hop whistlers, multiple-component and multiple-source whistlers and echo trains at Great Wall Station. A new Antarctic geoscience station of China was founded in the winter of 1989. It has started to record geomagnetic and VLF emissions since the spring of 1990. The geographic latitude and longitude of Zhongshan Station are 69.37~ and 76.38~ the geomagnetic latitude and longitude are -77.06~ 122.12 ~ respectively, the L value for Zhongshan Station is about 14. Due to such high geomagnetic latitude, no whistlers have been received since the station was founded, but numerous VLF emissions have been recorded. The VLF emissions recorded at Zhongshan Station are usually wide-band emissions, from a few hundred Hz to about 15kHz (Tang, 1992). Fig. 1 shows a typical
Fig. 1 A typical spectrum of VLF wave received at Zhongshan Station. The horizontal axis is time in units of seconds, the vertical axis is the frequency in units of kHz, the darkness shows the spectrum density.
Evidence for Resonant Absorption of VLF Waves
39
dynamic spectrum o f VLF wave received at Zhongshan Station. In February and June o f 1993, we recorded two sets o f V L F emissions never seen before. The main feature for the new type o f emission is a blank curve cut from the wide-band emissions(Fig.2a and 2b).
Fig.2a On February 1, 1993, we recorded a set of VLF emissions never seen before, a clean blank curve around 5 kHz was cut form the wide-band emissions.
Fig.2b A consecutive part of Fig.2a, Feb. 1, 1993 From Fig.2a and Fig. 2b, we can see a clean blank curve around 5 kHz cut form the wide-band emissions.
40
K.Y. Tang et
al.
Similar in Fig. 1, in Fig.2a and 2b, the horizontal axis are time in units of seconds, the vertical axis are the frequency in units of kHz, the darkness shows the spectrum density. S. Ohnami et al. [1993] have reported a type of triggered wave caused by a nonlinear wave-wave interactions in the subauroral ionosphere observed by ISIS-2 Satellite. On January 6 of 1988, VLF signals were ejected from Siple Station. Besides the 2.34kHz of ground-based VLF signals, an additional sideband structure around Siple signals were received by the ISIS-2 Satellite, they concluded that the sideband were caused by the nonlinear interaction between Siple signals and pre-existing ELF emissions(Ohnami et al., 1993) CONCLUSION AND DISCUSSION We have discovered a new interesting phenomena. It is different from Helliwell's finding, their controlled VLF signal triggered another VLF signal. It is also different from that of Ohnami et al., in their report, the interaction between two waves produced new waves. Our discovery shows that due to the interaction between natural VLF emission waves and ionospheric particles, a band of waves were absorbed, we infer that the phenomena were caused by resonance absorption, i.e. a type of Landau damping. Under a certain condition, particles with a certain velocity along a certain direction will interact with natural VLF emission waves, while the wave frequency, wave vector and particle velocity satisfy the following equation, CO - - n O , ) c i + K . V ,
n=0, +
1, _+2, _+3, ....
Eq.(1)
where co is the frequency of VLF emission wave, o3c~is the cyclotron frequency of some species of ion, K is the wave vector, and V is the particle velocity. The wave energy will be absorbed by the particle while the wave and particle satisfies equation 1. This could help us to better understand the particle wave interaction processes and mechanisms in the ionosphere and magnetosphere. ACKNOWLEDGMENT The authors would like to thank The Chinese Academy of Sciences and The National Antarctic Exploration Committee of China for supporting this project. We would like also to express our thanks to Mr. Gao Zonggang and Mr. Wang Zelin for their great work on designing and manufacturing the VLF receiver. REFERENCES Helliwell, R.A., VLF Wave Simulation Experiments in the Magnetosphere from Siple Station, Antarctica, Reviews of Geophysics, Vol.26, No.3, pages 551-578, August 1988. Ohnami, S., M.Hayakawa, T. F. Bell and T.Ondoh, Nonlinear Wave-wave Interactions in the Subauroral on the Basis of ISIS-2 Satellite Observations of Siple Station VLF Signals, Geophysical Research letter Vol.20, No.8, pages 739-742, April 23, 1993. Smith, A.J., M.A.Clilved, U.S.Inan, S.J.Lev-Tovand L.R.Piazza, Lightning-induced Electron Precipitation Studies with a Chain of VLF Receivers along the Antarctic Peninsula, Abstracts of 25th General Assembly of International Union of Radio Science, Lille, France, 1996. Tang, K.Y., Whistler Observations during 1990 magnetic Storms, Annual Report of Chinese Geophysical Union, Nov. 1992.
GLOBAL-SCALE IMAGING: NEW APPROACHES IN MAGNETOSPHERIC RESEARCH J. L. Green 1, S. F. Fung 1, D. L. Gallagher 2, M.-C. Fok 3, G. R. Wilson 4, G. R. Gladstone 5, J. D. Perez 6, P. H. Reiff 7, J. L. Burch 5, T. E. Moore 1
1NASA Goddard Space Flight Center, Greenbelt, MD, 20771, USA 2 NASA Marshall Space Flight Center, Huntsville, AL 35812, USA 3 USRA, NASA Marshall Space Flight Center, Huntsville, AL 35812, USA 4 NRC, NASA Marshall Space Flight Center, Huntsville, AL 35812, USA 5 Southwest Research Institute, San Antonio, TX 78228, USA 6Auburn University, Physics Department, Auburn, AL 36849, USA 7Rice University, Dept. of Space Physics and Astronomy, Houston, TX 77251, USA
ABSTRACT A completely different paradigm in magnetospheric research will begin with the launch of NASA's Imager for Magnetopause-to-Aurora Global Exploration (IMAGE) in January 2000. All the instruments on IMAGE will provide global-scale measurements of various plasma regimes in the Earth's magnetosphere. The objective of the IMAGE mission is to determine the global response of the magnetosphere to changing solar wind conditions. IMAGE will address this broad objective in unique ways by using neutral atom imaging (NAI), far ultraviolet imaging (FUV), extreme ultraviolet imaging (EUV), and radio plasma imaging (RPI). The IMAGE theory and modeling team has accomplished an extensive number of simulations in order to model the response of the IMAGE instruments. INTRODUCTION The Imager for Magnetopause-to-Aurora Global Exploration (IMAGE) is NASA's first medium-sized Explorer mission and is scheduled to be launched in January 2000. The overall science objective of IMAGE is to determine the global response of the magnetosphere to changing conditions in the solar wind. The science payload for IMAGE consists of instrumentation for obtaining images of plasma regions in the Earth's magnetosphere. The four types of imaging techniques used by IMAGE are: neutral atom imaging (NAI), far ultraviolet imaging (FUV), extreme ultraviolet imaging (EUV), and radio plasma imaging (RPI). These instruments are being designed to make concurrent global-scale images on a time scale of approximately four minutes, providing researchers with an opportunity to readily observe the structure and dynamics of the plasmasphere, ring current, aurora, geocorona, and the magnetopause within a substorm. In addition, simultaneous comparison of these data should also provide a global-scale understanding of how different important plasma regimes interact and their relationships with the changing conditions in the solar wind. The purpose of this paper is to provide background information on the IMAGE mission and what it is expected to observe. This will be accomplished by briefly describing the IMAGE mission and instruments and presenting a variety of computer simulations of the IMAGE instrument observations. ORBIT AND DATA SYSTEM Figure 1 shows a schematic of the IMAGE orbit. The IMAGE spacecraft will be spin-stabilized (--0.5 rpm) in an 90~ orbit with a perigee of 1000 km and an apogee of 7 Earth radii (RE) altitude. One 41
42
J . L . Green et al.
IMAGE orbit is approximately 13 hours. The orbital evolution over the two-year nominal mission life is also shown. The precession of the orbit over the north pole of the Earth will provide the IMAGE instruments a global view of a number of important plasma regimes in the magnetosphere such as the plasmasphere, the cusp, and the ring current.
Fig. 1. The initial IMAGE orbit and the precession of its line of apsides. The IMAGE spacecraft will operate with about a 100% duty cycle with all instruments in their baseline operational modes and will generate nearly 200 gigabytes of Level-0 data per year. All IMAGE instrument and housekeeping data will be stored on-board and dumped to a Deep Space Network (NASA) ground station at a high rate approximately once per orbit. In order to support the National Space Weather program, IMAGE will has a 38 kb/s real-time link that will be monitered, on occasion, by the Space Environment Center of the National Oceanic and Atmospheric Administration in Boulder, Colorado. The IMAGE Level-0 data will be processed into Level-1 data (in the Common Data Format) from which browse images will be generated within 24 hours after their receipt in the Science and Mission Operations Control Center (SMOC) at the Goddard Space Hight Center. The SMOC will post only the latest IMAGE browse products on the World-Wide-Web (WWW) and transfer housekeeping, Level-0, and Level-1 data products to the National Space Science Data Center (NSSDC) for permanent archiving and long-term access. The NSSDC will ingest all IMAGE data into their NASA Data Archive and Distribution Service (NDADS) system for further distribution to the science community and the public. NDADS provides rapid (within minutes) access to a variety of space physics and astrophysics data and supports WWW and email requests for data. Higher level data products (ie: three dimensional images, movies, etc) will be generated by the IMAGE Science Team and archived at the NSSDC. The IMAGE Science Team will have the responsibility to generate and validate all the data products and claim no proprietary data rights. The IMAGE mission maintains a series of WWW pages which provides the latest information about all aspects of IMAGE, including the type and accessibility of IMAGE data (covered in the IMAGE Project Data Management Plan). These pages are accessible from the IMAGE homepage (see URL: http://image.gsfc.nasa.gov/). SCIENCE INSTRUMENTS There are three instruments to make NAI measurements: the Low Energy Neutral Atom (LENA) imager, the Medium Energy Neutral Atom (MENA) imager, and the High Energy Neutral Atom (HENA) imager. Each of these instruments covers a specific energy range and utilizes different instrument technologies. There are four instruments for photon imaging, the Extreme Ultraviolet (EUV) imager, the Spectrographic Imager (SI), Wideband Imaging Camera (WIC), and the Geocoronal (GEO) imager. The SI, WIC, and GEO instruments operate in the Far Ultraviolet (FUV) wavelength range. Finally, very long wavelength remote sounding will be accomplished with the Radio Plasma Imager (RPI). The performance requirements for the IMAGE instruments are listed in Table 1.
Global-Scale Imaging: New Approaches in Magnetospheric Research
43
The minimum time resolution for images from all instruments, except the RPI, is the spacecraft spin period of two minutes. The RPI will have modes which will allow density profile determination on shorter time scales. Images can also be constructed over time scales of multiple spin periods. Table 1. Instrumentation Required for the IMAGE Science Objectives Image Measurement NAI
EUV
Neutral atom composition and energy-resolved images over three energy ranges: 10-500 eV (LENA) 1-30 keV (MENA) 10-500 keV (HENA) 30.4 nm imaging of plasmasphere He + column densities.
FUV
Far ultraviolet imaging of the geocorona at 121.6 nm (GEO) and ',he aurora at 140-190 nm (WIC) and 121.6 and 135.6 nm (SI)
RPI
Remote sensing of electron densities and magnetospheric boundary locations using radio sounding.
Critical Measurement Requirements
FOV: 90~ 90~ (image ring current at apogee). AngularResolution: 8~ 8~(LENA),4~ 8~(MENA),4~ 4~(HENA), Energy Resolution (AF_./E): 0.8 Composition: distinguish H and O in magnetospheric and ionospheric sources,interstellar neutrals and solar wind. Image Time: 4 minutes (resolve substorm development). Sensitivity: effective area 1 cm2 for each sensor. FOV: 90~ 90~ (image plasmasphere from apogee). SpatialResolution: 0.1 Earth radius from apogee. Image Time: several minutes to hours (resolve plasmaspheric processes). FOV: 15~ 15~(SI) for aurora (image full Earth from apogee), l~ 360~for geocorona, and 22.4~ 30~OVlC) Spatial Resolution: 70 km (WlC),90 km (SI) Spectral Resolution: separate cold geocorona H from hot proton precipitation (Z,~0.2 nm near 121.6 nm); reject 130.4 nm and select 135.6 nm electron aurora emissions. Image Time: 2 minutes (resolve auroral activity). Density range: 0.1-105 cm-3 (determine electron density from inner plasmasphereto magnetopause). Spatial resolution: 500 km (resolve density structures at the magnetopauseand plasmapause). Image Time: 1 minute (resolve changes in boundary locations). .
Neutral Atom Imagers The NAI technique takes advantage of charge exchange between cold neutral geocorona hydrogen atoms and energetic magnetospheric and ionospheric ions. The hot neutral atoms created through the charge exchange process then travel in a straight line trajectory from the interaction region and can be measured at a remote location. The hot neutrals preserve the energy and direction of travel of the original ions. This allows the original ion distribution to be deconvolved from the observed energetic neutrals. Within the magnetosphere energetic neutral atoms have even been observed by instruments designed primarily for energetic charged particles on the IMP, ISEE, and POLAR spacecraft (see for example: Roelof et al., 1985; Spence et al., 1996). The science requirements driving the NAI instrumentation for IMAGE are to image the inner magnetosphere and to resolve the major species contributing to neutral atom fluxes. To meet these requirements a suite of three NAI instruments will provide angle, energy, and composition-resolved images at energies from 10 eV to 500 keV. The three NAI instruments are necessary because of the different techniques that apply to observing low (0.01 to 0.3 keV), medium (1 to 30 keV), and high (10 to 500 keV) energy neutral atoms. Angular information is obtained over 90 ~ fans with angular resolution between 4 ~ x 4 ~ and 8 ~ x 8 ~ depending on species and energy. Spacecraft spin is used to obtain angular information in the orthogonal (azimuthal) direction. All three instruments have collimators that consist of serrated, blackened surfaces to reduce internal scattering. The collimators contain deflection potentials of 10 kV that deflect and absorb charged particles below 100 keV/e. Small broom magnets remove electrons with energies 60%) in the FUV but has minimum (
DRIFT
-10
-~
-15 t-
~
!
CON
~
-:,o i j~
'9 I
!
uoo ION
5~:
-25 1 o ~-
.~E
(~)
":"
-150 i
r
055:00 423 20.54 70.82
057.00 396 20 67 63 4 2
059:00 372 20.73 56 0 6
0
,
~-
-2
_ 30 keV/e) of the lunar fly by Jan. 1,5-20, 1996 (above) and of a later high speed stream (below) without a lunar fly by. C, O, A1, Si and S-ions were detected during the moon fly by. Only few O ions could be detected during the high speed stream without moon fly by (below).
67
Search for Lunar Pickup Ions MASS
SWICS
STICS
y
Table 3. Ions
x
9 Noon
-
-
~
t
~
-
H20 Na A1 Si S K Ti Mn Fe
Midnight
Dusk
Rigidities of 100 keV Ions Mass m m n] m ill in in m nl
= = = = = = = = --
18 23 27 28 32 40 48 55 56
Gauss H H H H II H H H H
9
cln
R = 1.93 R = 2.18 R = 2.37 R = 2.41 R=2.58 R = 2.88 R=3.16 R = 3.38 It. = 3.41
105 105 105 105 105 105 10 .5 105 105
H = magnetic field strength ill Gauss R = gyroradius ill cm nl
--
nlass
ill
anlu
Fig. 1. Sectorisation (0-15) of the SWICS and STICS lneasurements and the field-of-view pointing directions of the three sensors MASS, SWICS, STICS. The S/C is spilming clockwise. DISCUSSION From December 1994 until March 1996 seven WIND lunar flybys took place. Pickup ion measurenlents were shown for two selected examples which were characterised by high solar wind velocities. Flux increases in the azinmthal sectors 4-8 were detected for in > 10 alnu/e particles which we illterprete as hmar pickup ions ill the mass range 10-42 alnu/e. However, singly charged Fe-pickup ions were not found. In Table 3 are compiled the rigidities for some typical lunar ions (E = 100 keV). For an interplanetary magnetic field lnagnitude of ,5 nT follows that the gyroradii are then of the order of 20-40 lunar radii for in = 18-56 a mu particles. Thus it ca.n be expected that a part of the pickup ions have energies well below 100 keV near the moon and do not exceed the detection lilnit of the STICS sensor. Generally three processes contribute to the release of gaseous material from the lunar surface: 1) photon sputtering, 2) solar wind sputtering and 3) impact of lnicrometeoroids. A review of model calculations was recently published by Smyth and Marconi (1995). It is worth noting that the results of the various authors diifer by few orders of magnitude. Ground based observations of the photon sputtering effect especially for the sodiuln content of the lunar atmosphere were made by Mendillo and Baumgardner (1995) during a lunar eclipse period during which the impact of the solar wind was shielded by the tail of the earth magnetosphere. The authors found spectroscopically that the sodium atmosphere extends up to ~ 10 luuar radii. As mentioned earlier Elphic et al. (1991) lnade laboratory measurements with lunar soil salnples which can be considered representative for the surface composition of the moon. Irradiation of that material with 5 keV Argon ions, 4 keV Heliuln ions and 1.5 keV protons showed that solar wind ions are able to release ions from the lunar surface which contribute then to the pickup ion flux. For example 1 keV protons generate about 10 -5 secondary particles. Thus a solar wind ion flux of typical 3 x l0 s ions/cm2s generates about 103 secondary ions/cm2s. The generation of pickup ions by photon sputtering, solar wind and inicrometeoroid ilnpacts was also studied by Morgan and Shemansky (1991) and Ip (1991). The authors Morgan and Shemansky applied an equation proposed by Zel'dovich and Raizier (1967) and found a maximum ejection velocity of 5.1 km/s. Thus at least part of the released atoms is able to leave the gravitational field of the Moon which has v= 2.38 km/s escape velovity. Ip (1991) calculated the extension of the lunar exosphere under the influence of the solar wind for various release velocities and concluded that meteoroid impact and solar wind sputtering are the major sources for the lunar exosphere. Froln the presented study it is generally concluded that during the lunar flybys in December 1994 and January 1996 pickup ions, generated by solar photons, solar wind ions and microlneteoroid impacts were detected by the SMS eperiment onboard the W I N D - S / C , although a contribution of singly charged lnagnetospheric ions calmot be excluded. ACKNOWLEDGMENTS We thank J. Cain, R. Lundgren, S. Lasely and Ed. Turns fi'oln the University of Maryland for their work during tile development of the SMS experiment and S. Chotoo for the analysis of tile STICS calibration data,. The Gerlnan contribution to the SMS-Experiment was funded by DARA-Contract FKZ 50 OC 89149.
58
E. Kirsch et al. WIND/STICS . . ~ I O ~ ,
Iu'''
....
T-r-r~q
ec 25,1200 don
30
-
7 ' -
Feb
>
m/q
Dec 29,
'
10
I ' ' '
omu/e I
'
'
r-~-w-,--q
i ' '
'~
1994
WlND/STICS 10000 ] i
.5, 1 9 9 5
I~ 2
.~
Q) 1
1
94/,De,c/,25-,94,,(Dec/29
tdoy 3597363) '
"~
. . . . . 2
o
I 4
,
,
,
I 6
,
,
I
, I , , , I , , , I , , ,
,
8
10
12
14
sun sect. 1000
100
16
Sector ,
~Io
2
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
. . . . ~:~:;_~:~. . . . . . . _:
i .....
i ,
i
,
,
i
i
,
,
i
i
i
i
h
I
2
3
4
5
6
7
8 9
mass/charge
IO-I-~
i
i
I 2
L
,
,I
L,
4
,
L~
6
ii
I 8
,
I,
l t, 10
, I L l ,
:;
I, 14
12
16 Sector
Fig. 2. Sector measurements (> 10 ainu/e)of the 2 WIND-moon fly bys (dashed lines) and of later high speed streams measured far away from the moon (full lines). The enhancements in sectors 4-8 (dashed lines) are interpreted as lunar pick-up ions. Such ions do not appear in the later measured high speed streams (full lines). Sectors 9-10 measure solar wind or singly charged lunar ions, whereas sectors 0-3 detect most likely magnetospheric ions (dashed l!nes).
I0
20
30
40 50 60
[ ainu/e]
Fig. 3. Tile additiol}al ion fluxes (Dec 25-29, 1994) measured by STICS (E/e = 6.5- 226 keV/e) in the mass range 10-42 amu/e should be caused by lunar pick-up ions and for < 10 amu/e by solar wind and interstellar He + ions. Only the sunward directed sectors 6-13 are used here (see Fig. 1).
, 96/Jan/15-96fJan/20
WIND/STICS
tdoy 015T020) ~ sun sect.
o 100.0
WIND/STIC, S 10000~
96~Jan/,15-,96,/Jan(/20
10.0
tdoy 015702j0) sun sect. ~ o
~ ,ooo! 0
1.o
100
1
10.0
2
WIND/STICS ~
L
3
,
t
4 5 6 7 8 910 mass/charge [am./~] 96/(Feb/10-96/Feb/15 t
~
~
I
i
~ J
20
30
40 50 60
tdoy 041704,6I ) sun sect
i 1
2
3
4
5
6
mass/charge
7
8 9 10
,
1416
,
i
20
30
40 50 60
[ amu/e]
Fig. 4a. The additional ion fluxes (Jan 15-20, 1996) measured in the mass range 10-42 amu/e (E/e = 6.5226 keV/e)are caused most likely by lunar pick-up ions (further details see Fig. 3).
1.0
0.1
1
2
3
4 5 6 7 8910 mass/charge [amu/e]
I
I
20
30
I
I
40 50 60
Fig. 4b. Triple coincidence count rate of the lunar fly by Jan 15-20, 1996 (above) and of a later high speed stream (below). C, O, A1, Si and S-ions of E/e = 30 - 226 keV/e were detectable during the moon fly by (above) but only few O-ions during the high speed stream (below) without moon fly by.
Search for Lunar Pickup Ions
69
REFERENCES
Acuna, M.H., K.W. Ogilvie, D.N. Baker, S.A. Curtis, D.H. Fairfield, and W.It. Mish, The Global Geospace Science Program and its Investigations, Space Sci. Rev., 71, 5 (1995). Cladis, J.B., W.E. Francis, and R.R. Vondrak, Transport toward Earth of Ions Sputtered from the Moon's Surface by tile Solar Wind, J. Geophys. Res., 99, 53 (1994). Elphic, R.C., H.O. Funsten, B.L. Barraclough, D.J. McComas, M.T. Paffett, D.T. Va.ninlan, and G. Heiken, Lunar Surface Composition and Solar Wind Induced Secondary Ion Mass Spect.rolnegry, Geophys. Res. Left., 18, 2165 (1991). Gloeckler, G., H. Balsiger, A. Biirgi, P. Bochsler, L.A. Fisk, A.B. Galvin, J. Geiss, F. Glieln, D.C. Halnilton, T.E. Holzer, D. Hovesgadt, F.M. Ipavich, E. Kitsch, R.A. Lundgren, K.W. Ogilvie, R.B. Sheldon, and B. Wilken, The Solar Wind and Supratherlnal Ion Composition Investigation on the WIND Spacecraft, Space Sci. Rev., 71, 79 (1995). Hilchenba.ch, M., D. Hovestadt, B. Klecker, and E. M6bius, Detection of Singly Ionised Energetic Lunar Pickup Ions Upstrealn of Earth's Bowshock, in Solar Wind Seven, ed. E. Marsch and R. Schwenn, COSPAR Coll. 3, pp. 349-355 (1991). Ip, W.-H., The Atomic Sodium Exosphere/CoIna of the Moon, Geophys. Res. Lett., 18, 2093 (1991). Mendillo, M., and J. Baumgardner, Constraints oil the Origin of the Moon's Atmosphere from Observations During a Lunar Eclipse, Nature, 377, 404 (1995). Morgan, T.H., and D.E. Shelnansky, Lilnits of tile Lunar Atmosphere, J. Geophys. Res., 96, 1351 (1991). Ogilvie, K.W., and G.K. Parks, First Results from WIND Spacecraft: An Introduction, Geophys. Res. Lett., 23, 1179 (1996). Smyth, W.H., and M.L. Marconi, Theoretical Overview and Modelling of the Sodiuln and Potassiuln Atmospheres of the Moon, Astrophys. J., 443,371 (1995). Zel'dovich, Ia.B., and Yu.P. Raizer, Physics of Shock Waves and High Temperature Gases, Academic, San Diego, Calif. (1967).
This Page Intentionally Left Blank
THE TETHER SYSTEM USED IN THE MAGNETOSPHERIC STUDIES
S. I. Klimov l, F.L.Doudkin 2, V.E.Korepanov 2, A.A.Petrukovich l, M.L.Pivovarov l, and A.V.Prudkoglyad 1
1Space Research Institute (IKI), Profsoyuznaja 84/32, 117810 Moscow, Russia 2Lviv Center of Institute of Space Research, Naukova 5-a, 290601 Lviv, Ukraine
ABSTRACT The Electromagnetic remote sounding of the low density plasma boundaries is discussed. Using a tether system, with transmitting and receiving antennas in the frequency range 9-200 kHz, the bow shock sounding is possible. INTRODUCTION Tethered systems carried by spacecraft provide the unique opportunity to perform the remote sensing of the specific boundaries in space plasma (shock wave, bow shock, magnetopause etc.). The signal reflected from the iboundary has frequency fpe.,./ RI=R2-zkR). The geometry used in the simulations presented here can be defined using equation l~used "- l~idea I -- (~l~as sin 2(20),
(2) where ARused is the distance between the hemispheres used in the simulations, ARidea I is the distance between the hemispheres for an ideal instrument, 8Z~as is a maximum deviation from the ideal distance and 0 is an angle defined in Figure 2. In the simulations presented here a value of 50~m for the parameter ~Z~as is used. The correction term 8Agassin2(20) has its maximum when 0 =45 ~ and it has its minimum 0 when 0 =0 ~ or 0 =90 ~ A view of the geometry used in the simulations is presented in Figure 3.
Z
Ideal"
Asym."
Z
z ~ -
/
/ /
]
-.
I
L
,y
'X Fig. 3. The exaggerated view of the asymmetric inner hemisphere used in the simulations.
\
Y
Simulations of a Spherical Section Electrostatic Analyzer
79
Eq. 1. and 2. yield ion orbits which have to be solved numerically. For our purposes the fourth order Runge-Kutta method was used to obtain the solutions (see Press et al. (1986)). For more details about solving ion trajectories inside the IBS instrument see Paper I. Ion Trajectories Outside the IBS A stray electric field accelerates incoming ions and thus it also changes the ion trajectories. The geometry of the front aperture plate has dramatic influence to the stray electric field. With the front aperture plate installed, about 90% of the stray electric field can be confined to the volume between the front aperture plate and the hemispheres. Removal of the ions which collide with the front aperture plate, but otherwise could traverse the sensor, improves the energy resoiution as well as the angular resolution of the instrument. The geometry of the front aperture plate of the IBS instrument is shown in the Figure 4. The geometry is identical for all of the three apertures of the IBS instrument. A cylindrical (3D) simultaneous overrelaxation (SOR) algorithm is used to solve the stray electric field (see Press et al. (1986)).
,B
Cross-section A-A
el ~2.5 mm I~ ~)~~~ I
1 2 mm
_
"~ \ ~ \ 15 m m ~
mm . "B
~
hv
Cross-section B-B
~+_j175~ j~jJ-
!
15mm 1
1
-
Fig. 4. The geometry of the IBS entrance aperture. On the left side of the figure the aperture is seen from straight above the front aperture plate. The cross-sections A-A and B-B are marked with dashed lines. The numerical values of the parameters of the geometry of the IBS entrance aperture are marked in the figure. The figure has been published earlier in Paper II.
The equations of motions are written for cylindrical aperture geometry and the ion trajectories in the stray field area have been solved using the fourth order Runge-Kutta method (see Press et al. (1986)). The initial values for a hemispherical routine, introduced in the previous section, are calculated using the output values of the stray field routine given above. For more details about solving the ion trajectories outside the IBS instrument and combining the stray field routines to the hemispherical ones see Paper II.
80
J.H. Vilppola et al.
RESULTS OF SIMULATIONS In Papers I and II, we have studied the ideal case corresponding to an instrument with perfect geometry, and in cases where linear misalignments in each main coordinate axis x, y, and z occurred. Also the effects of indentations of the hemispheres were considered in Paper I. These results are not presented here. Thus, only the results of the simulated new asymmetric case introduced in the theory section are included in this report. The values of the parameters for the laboratory ion beam used in the simulations are given in Paper II. The requirements of the stray field calculations used in the simulations are fulfilled using the potential mesh parameters shown in Paper II. Also other parameters used in the simulations are given in Papers I and II. The symbols of the IBS instrument parameters as well as their values used in the simulations are presented in Figures 2, 3 and 4. In the actual IBS, the center of the hemispheres is 2 mm above the plate tangential to the edges of the hemispheres. Thus, the real reverse angle of the IBS flight model is only about 178 ~ instead of 180 ~ which is the definition of hemispherical analyzer. This means that the [3-angles used in our simulations can be used for the actual IBS only after a correction of 2mm/100mm rad has been subtracted. The simulations were made for the ideal case, and for the case of an asymmetric inner hemisphere where the value of 50~m for the parameter 5&Ras introduced in Eq. 2. is used. In every simulation for each ion beam setting (a,~,E) 250 ions were injected. Since there are 51 different values for each of these variables, a total of 33162750 (=51x51x51x250) ions per simulation were injected. An exception had to
Fig. 5. The energy/alpha (E/o~ in text) distribution for an asymmetric case. )~ = number of particles transmitted per bin square. The key shows the gray scales for the g-numbers. The bin widths are ~5E=0.12% and 80~=3.0~ The energy is (E-Eo)/Eo where Eo is the nominal kinetic energy. An ion with nominal kinetic energy has a spherical orbit exactly in the middle of the hemispheres (r=-0.100 m). The dashed cross shows the bin square where the transmission has its maximum. The dashed lines show the bins, which have been used to produce the profile plots at the upper and right sides of the plot. The gray scales in the plot have been chosen so, that the inner black ring shows the bin squares with the full width at half maximum (FWHM) values.
Simulations of a Spherical Section Electrostatic Analyzer
81
Fig. 6. The energy/beta (E/f5 in text) distribution for an asymmetric case. The bin widths are 5E=0.12% and 513=0.12~ Energy, )~, key, profile plots and dashed lines are similar to the ones shown in figure 5.
be made when the cut through at IBS response function was studied for the parameters ~, ~ and E. To avoid excess random noise the beam intensity was increased by a factor of 10. The simulation was made once for each parameter, yielding a total number of 19507500 (=3x51x51x2500) injected ions per each studied case. In Figure 5 a two-dimensional E/o~ distribution for the asymmetric case is presented. The corresponding E/f5 distribution is shown in Figure 6. In Figures 5 and 6 a cut through for invisible parameters are used.
The values of all cut through parameters are proportional to the indices which give maximum transmission using statistical averaging instead of the real maximum. Otherwise the random noise might effect the cut selection as can be seen in profile plot selections where the actual maximum has been used. The definition of the cut through for an invisible parameter using statistical averaging is later called a cut through method. An other way to present the distributions is that the invisible parameter is summed over all its values from the three-dimensional data matrix. This method of handling the invisible parameters is called the white spectrum method. In this article the white spectrum method figures are not presented since they do not give any significant new addition to the results attained using only the cut through method. RESULTS OF CALIBRATIONS In the calibrations commandable laboratory high voltage power supply was used to control the voltage between the hemispheres. All other functions were accomplished using the flight electronics installed on the IBS instrument. In the calibrations the proton beam was accelerated using a constant 10 kV voltage (10006V to be exact) source. Energy variations result from different voltages between the hemispheres. For each set of the parameters (E,a,[3) the transmitted ions were counted. The counts measured have to be corrected for of the detector dead time given by CM c r = l_cu T , (3)
82
J.H. Vilppola et al.
Fig. 7. Energy/alpha (E/a) distribution obtained for the calibration data. The bin widths are 8E=I.0V and 8a=5.0 ~ The energy is given as voltage between the hemispheres. 10 volts in the figure is about 1.9% when comparing energies here to the ones shown in figures 5 and 6. g, key and profile plots are similar to the ones in figures 5 and 6. Note that all g-numbers are corrected using a dead time correction given in equation 3.
where cr is the true counts per second, cM is the measured counts per second and T is detector dead time. T marks the time during which the detector is inoperative after each single event. The dead time correction T for flight electronics used in the calibrations is 0.30 ps. In Figure 7 a two-dimensional E/o~ distribution calculated using calibration data is presented. The corresponding simulation distribution is presented in Figure 5. A two-dimensional E/f5 distribution
Fig. 8. Energy/beta (E/f3) distribution composed using the calibration data. The bin widths are 8E=I.0V and ~513=0.2~ Parameters g, key, energy and profile plots are similar to the ones in figure 7.
Simulations of a Spherical Section Electrostatic Analyzer
83
calculated using calibration data is presented in Figure 8. The corresponding simulation distribution is shown in Figure 6. Cut throughs for invisible parameters are used in Figures 7 and 8. The values of all cut through parameters are chosen by defining the maximum of the parameters using separate measurements before the actual calibration measurements. Thus, the results of the simulations and the calibrations are comparable. The white spectrum method cannot be used in the case of the calibrations. The measurements needed to collect the required data set for the white spectrum method cannot be accomplished in a reasonable time. DISCUSSION AND CONCLUSION The Hemispheres of the IBS Instrument The hemispheres of the IBS flight model were manufactured at VTT Automation in Espoo, Finland. The hemispheres were built to meet the 25pm tolerance limit given in Papers I and II, and they were measured using a touch sensor before assembly in order to check that the quality requirements were fulfilled. The hemispheres cannot be measured after they have been assembled, and therefore the errors arising during the assembly phase cannot be detected directly. There is a slight possibility that the hemispheres are tensed and some asymmetry is present in the hemispheres of the IBS flight model for that reason.
E/t~ Distributions The double curved shape of the E/c~ distribution shown in the calibration data plot of Figure 7 was unexpected. The simulation result shown in figure 5 confirms the opinion that there might exist some asymmetry in the hemispheres of the IBS flight model. The asymmetry in the hemispheres is not necessary exactly the same as described in the theory section but definitely there are some similarities between the geometry of the hemispheres of the IBS flight model and the assumed geometry given in the theory section. The similarities in figures 5 and 7 support this conclusion. If the assumption is made that the asymmetry is similar in the simulations and in the IBS flight model, then the magnitude of the asymmetric factor ~)Z~asintroduced in Eq. 2. is approximately the same 50pm as used in the simulations. The small counter-clockwise rotation appearing in the E/a distribution shown in the calibration data plot of Figure 7 can be explained using the methods given in Paper II. A similar rotation in the E/a distributions of the simulations occurs as a result of misalignments along the direction of y-coordinate. The magnitude of the misalignment responsible for the rotation in Figure 7 is about 10pm. This is below the limit given in Papers I and II. This conclusion can be reached by comparing Figure 7 here to Figure 7 of Paper II. Other Distributions The similarities in other distributions when comparing simulations of an ideal and an asymmetric case with the calibration results supports the conclusions given above. The butterfly effect given in Figure 8 of Paper II can be seen in the calibration results, as well as in the new simulation results. The E/f5 distributions given in Figures 6 and 8 also supports the conclusions given above. The energy resolution is approximately the same for the simulations and calibrations. The same conclusion holds for the angle 13 resolution. In the calibrations the measurements are made for E/f5 distributions with several cut through values for the angle c~ which varied from-80 ~ to 80 ~ with 10 ~ steps. Other cases than the one given in Figure 6 are not yet simulated. Because of the lack of calibration data discussed earlier which prevented the usage of the white spectrum method a new simulation might be required in the verification of the calibration data. At least some minor checks with specific cut through values for the angle c~ should be
84
J.H. Vilppola et al.
accomplished. Also new calibrations for the two side apertures are required before the IBS response function is fully understood. ACKNOWLEDGEMENT The authors would like to acknowledge the Academy of Finland,' University of Oulu and NASA for financial support of the project. The authors would like to thank Dr. J. T. Gosling, Los Alamos National Laboratory for his valuable comments and suggestions. REFERENCES G. Arfken, Mathematical Methods for Physicists, Academic Press, Inc., San Diego 1985. S. J. Bame, D. J. McComas, D. T. Young, and R. D. Belian, Diagnostics of Space Plasmas, Rev. Sci. Instrum., 57, 1711 (1986). J. T. Gosling, J. R. Asbridge, S. J. Bame and W. C. Feldman, Effects of a Long Entrance Aperture Upon the Azimuthal Response of Spherical Section Electrostatic Analyzers, Rev. Sci. Instrum., 49, 1260, 1978. J. T. Gosling, M. F. Thomsen and R. C. Anderson, A Cookbook for Determining Essential Transmission Characteristics of Spherical Section Electrostatic Analyzers, Los Alamos National Laboratory Report LA- 12962-MS, 1984. I. S. Grant and W. R. Phillips, Electromagnetism, John Wiley & Sons Ltd., 1976. D. J. McComas and S. J. Bame, Channel Multiplier Compatible Materials and Lifetime Tests, Rev. Sci. Instrum., gg, 463 (1984). D. J. McComas, J. R. Baldonado, S. J. Bame, and B. L. Barraclough, Channel Electron Multiplier Compatibility with Viton and Apiezon-L Vacuum Grease, Rev. Sci. Instrum., 58, 2331 (1987). NASA, A Proposal for the Plasma Science (PLS) Investigation for the Cassini Orbiter Spacecraft, Volume 1: Investigation and Technical Plan. W. H. Press, B. P. Flannery, S. A. Teukolsky and W. T. Vetterling, Numerical Recipes, The Art of Scientific Computing, Cambridge University Press, 1986. M. R. Spiegel, Theory and Problems of Vector Analysis and an Introduction to Tensor Analysis, McGraw-Hill, New York 1959. J. H. Vilppola, J. T. Keisala, P. J. Tanskanen and H. Huomo, Optimization of Hemispherical Electrostatic Analyzer Manufacturing With Respect to Resolution Requirements, Rev. Sci. Instrum., 64, 2190 (1993). J. H. Vilppola, P. J. Tanskanen, H. Huomo and B. L. Barraclough, Simulations of the Response Function of a Plasma Ion Beam Spectrometer for the Cassini Mission to Saturn, Rev. Sci. Instrum., 67, 1494 (1996).
A NUMERICAL, DOUBLE-LAYER SOLUTION FOR PLASMA CONTACTORS COLLECTING ELECTRONS IN ELECTRODYNAMIC TETHER APPLICATIONS A. J. Li and D. W. You
Center for Space Science & AppfedResearch, Academia Sinica, Beijing 100080, P.R. CHINA
ABSTRACI"
Tethers in space can be used for a wide variety of applications. In this case, plasma contactors play an important role. A plasma contactor is a special device which can provide electric contact of the ends of a tether with the ambient plasma. Plasma contactors based on the hollow cathode technology have been used as electron emitters or collectors for tethers, and neutralizers for spacecraft charging control. Our work has been placed an emphasis on the laboratory simulation of the double layer that exists between two plasmas, the plasma plume generated by a hollow cathode and the simulated ionosphere plasma. The plasma contactor used as a current collector was numerically computed according to the space charge limited double layer model. When the plasma contactor is used in electro-dynamic tether application, where a substantial current can be collected, the ignited mode of electron collection is needed. In this case, We numerically solve the Poisson's equation in one dimension considering the thermal electrons in the double layer and taking streaming electron ionisation into account. The calculated resuks are presented in comparison with the collisionless double layer theory and the experimental data. INTRODUCTION The electrodynamic tether system(Martinez and Hastings, 1987) has a great potential to provide great electric power for space applications. A tether can be used as a power generator to deliver the power to a load, however, if the load on the tether is replaced by a DC power supply, the tether will act as a thruster that can increase the orbital energy of the system, so the whole system can be applied as an effective energy storage system. In either case, a complete circuit that includes the space plasma, the tether and the contactors is needed and the contactors acting as space plasma brushes should improve the electrical efficiency by reducing the impedance. Therefore, plasma contactor(Williams and Wilbur, 1990) is a plasma producing device which is conceived to enhance the electron collection and emission between the tether and ambient plasma. Hollow cathodes(Sigfried,1982; Deninger et al., 1987) have been found to be particularly well suited to the plasma contactor use because of the lower parasitic power consumption and better feasibility. 85
A. J. Li and D. W. You
86
A key issue in the plasma contactor research is to enhance the electron current collection and reduce the impedance of the contactors for improving the power generation efficiency. Because in the electron collection mode plasma contactors usually has high voltage drops, so only this mode attracts more attention. Our work will focus on the electron collection mode. Langmuir(Langmuir, 1929) pointed out that between two quasineutral plasmas at different potentials, a double- layer(DL) occurs. The DL consists of a layer of positive charge and a layer of negative charge. The DL region is governed by a space-charge-limited process. Wei and Wilbur(Wei and Wilbur, 1986) introduced the first model of DL produced by the expansion of a high-density plasma into another plasma. This model describes spherical DL as spherical diodes. A current Ji of ions flows outward from the inner sphere having radius ri, and a current Jo of electrons flows inward from the outer sphere having radius ro. The outer surface is at zero potential, and the inner radius is at Vi. The particles leave each surface in a space-charge-limited manner. All components are assumed to have very low temperature compared with the voltage drop. Given the ratio ri/ro and the voltage drop Vi, the currents Jo and Ji can be calculated. This model, which treats the electrons as collisionless and unmagnetized, is able to describe many of the experimental observations. However, to further fit the experimental data, an improved model that includes ionization and thermal electrons is proposed in this paper. THEORY Following the discussion of Wei and Wilbur(1986) and using the same definitions, we can write the Poisson's equation to describe the potential variation in the DL region V 2V =- e--(n+ -n ) *0
-
,
(1)
where the space charge density is determined by the density of both accelerated and trapped charged particles. For the spherically symmetric case, Eq.(1) can be simplified to 1 d
r 2 dr
(r 2 &~_~)_
__~_o ( n + _ n
).
(2)
In Eq.(1) and (2), V is the potential at a spherical surface of radius r, where the ions from the inner surface have a density ni and velocity ui, and the electrons from the outer surface have a density no and velocity Uo. Then the ion current Ji and the electron current Jo are J o = 4 r c r 2 n o e U o and di - 4 r c r 2 n i e u i 9 (3) Because the particles are assumed to have been drawn from the boundaries at zero initial velocity, these velocities are given by u o - x/(2e / m o ) V and u i - x/(2e / mi)(Vi - V), (4) where mo and mi are the masses of the charged particles drawn from the outer and inner surfaces, respectively. From Eq.(3) and (4), we can write the stream charged particle densities as no =
do
a n d r/i =
4nr 2e4(ze / m o ) v for electron and ion densities.
(5)
di
4nr 2e4(2e / m i ) ( V i
- V) '
In addition to the above two terms that were used in Wei and Wilbur' model, now we consider the thermal electrons trapped on the inner side, which can be written as _
?leth
r"2
-'~?leth
(Vi) exp[
e(V-Vi)
-kTe
]"
(6)
Numerical, Double-Layer Solution for Plasma Contactors Collecting Electrons
87
Since the density of ions trapped on the outer side is low it can be neglected. Finally the density of accelerated ions created with zero velocity by ionization due to the streaming electrons (thermal electrons are not assumed to ionize) can be given by
@ niz =
,, ,Z t , Vl ]
,
(7)
r 2 x/2e[V(rl) - V(r)] / m i
where the rate of ion creation is given by niz (r,V) - cr(V)n n (r)n e (r)v e (V), (8) with vo(V) the electron velocity and cy the ionization cross section(Rapp and Englander-Golden, 1965) depending on the kinetic energy of the accelerated electrons. The density of neutral particles depends on the atoms expanding from the plasma contactor because the ambient pressure is very low. When a gas atom is ionized, a new electron is also created. This electron is supposed to be thermalised quickly and contributes to r~th. Then, we can rewrite Eq.(1) as r 21 drd (r2 --h-g-,-U-,,, ~ 0 d) V= - e (n i _ no) _ -~o (niz-neth) "
(9)
Using the definitions given by Wei and Wilbur(1986): qk- V / V i , p - ln(r / ro) , ]o - (Jo / 4rccoVi3/2)x/mo / 2e ,and a - (Jo / Ji)x/mo / mi , Eq.(9) can be rewritten as
(10)
d2qk dO 1 --~p2 + -~p - jo ( x[-~
(11)
1 )+aexp(_b(l_q~))_CjoS(q~), a 4 1 - q~
eri e eVi 3 where a - ~oVi net h (V i), b - -~e ' c - n n (r i )• % ,
l. l n%.. ~r and s(~b)-ri
(12)
cr(rl)
(13)
~oo r~r124~k(r1)_qk(r) dr1
We use boundary conditions ~b= 1 at p = p; = ln(~ / ro), and ~b= 0 at p = 0 = ln(ro / ro), d ~ = 0 at p - Pi where ~b- 1, and d__~_ 0 at p - 0 where r dp
(14) 0.
(15)
dp
Now, if To, Vi, ri, ro, r~(Vi) and nn(Vi) are given, we can solve the DL problem based on our model. RESULTS AND DISCUSSION Using Wilbur's experimental data(Wilbur, 1987) xenon in the plasma contactor, Vi=55V, the xenon atom density of n, r2=3.6• the thermal electron density of 1.0x1013/m3, and 4eV of the electron temperature of inner plasma, the problem is solved by obtaining the numerical solution of the Eq.(11) combined with the boundary conditions Eq.(14) and Eq.(15). Both the normalized current from outer surface jo and the normalized current ratio tx are calculated and shown in Figure 1 and Figure 2. Using Wilbur's experimental data, for ro--0.128m and ri/ro=0.61, JCE =-250mA, corresponding to jo=4.48, our model gives jo=3.60 while Wei and Wilbur's model gives jo---3.07. So our model is more coincide with the experiment data.
88
A. J. Li and D. W. You o
"~
2
1
- ~
Wilbur's Model
?
"~ 1.5
ca 0.5
0
Z:
~
Wilbur's
Model
.~08
I.. =
.~
T
0 9 --Jr
~
0.6
0
~
.5
-0.5
~
0.4 0.3
-1 0
r
'~"
~0
O0
0
o
0
0
Sphere radius ratio (ri/ro) Fig. 1 Effect of radius ratio on current magnitude
I
l
I
I
I
I
I
I
I
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.2t
Sphere radius ratio hlro
Fig.2 Effect of radius ratio on counterflowing current ratio
The at is the normalized current ratio(ct=l is the Langmuir condition). In both models(Wei and Wilbur's and this paper), cx h, where h ~ 300 km is the MIR space station mean height. Since Al > 1, and EM wave undergoes strong frequency dispersion of "whistler" type. Taking into account the relations kl = kl, R, k2 = ko, klh2 >> 1, from Eq.(16) we obtain that
EZx, n E-1 2z,n ~ h2x -1 >> 1, fl 2y, n H"-1 2z, n ~ (koy(3-2Wkoh2))-I
from Eq.(16) with I ,yl > l. When EDS is placed above the reception point it is obvious that E2:,, = 0, H2:.,, = 0, and the axes of electric and magnetic antennas must be situated horizontally and oriented along and perpendicularly to the magnetic field lines respectively. For this case, the values of the field components are shown in Table 3. Table 3.
The values of EM field components in the reception point f, (Hz) Ez~,, (V'm -l) Hzy,, (A'm l) 103 1.2x10 l~ 1.5X10"14 1.2xl 04 4.3xl 01~ 4.5xl 0in 5xl 0 4 8.7xl 0-l~ 2.2xl 0 "14 The dependence of the signal/interference ratio limits for Ez~,n and H2y.,, on f, is shown in Figures 3, 4 respectively with the same conditions as Figure 9. The signal/interference ratio, it is obvious from Figures 3, 4, in the range 1...5 kHz is 1...2 orders lower for E and H field components respectively to free space. For P - h = 300 km in the range 103...1.2x104 Hz the intermediate radiation zone conditions are used when k2p > 5 a n d Ik23k12p[ < 0.2. In i
|
this case, the EM field component ratios of the horizontal EDS may be written in the following form (Bafios, 1966): 1~2p' n E-1 2z, n "~'fp,-1e(fc,ef) 0"5 5.0. Substit~,ting Eq.(6) into Eq.(1) and Eq.(2), and using numerical calculation, we can find the distribution of the radiance in the observation plane for the different view angle ~ and KN. Figure 2c represents the radiance contour for three different optical thicknesses when the sunlight is from the top of Figure 2 and the line of sight is perpendicular to Figure 2. For an optically thick barium cloud, as shown in the lower figure of Figure 2c, we find two crescent structures. The large one is in the dayside of the cloud and the smaller one is in the nightside. CONCLUSIONS The main results can be summarized as follows: According to Eq.(1) and (2), we can find the radiance distribution of the spherically symmetric neutral barium cloud for the different optical thickness, view angel and spatial distribution using numerical method. The radiance distribution can be expressed analytically only in some simple cases. The radiance isograms of the neutral barium cloud when the sunlight is perpendicular to the line of sight for the different optical thickness or KN are shown in Figure 2, where Figure 2a, 2b, and 2c corresponds to the uniform distribution, Gaussia, distribution and shell distribution, respectively. When the barium cloud is optically thick, due to the absorption of the sunlight, a crescent structure in the radiance of the cloud is found in the dayside. For a shell distribution neutral barium cloud, we find two crescent structures, where the smaller one is in the nightside. For an optically thin neutral barium cloud, the sun light can go through the sphere without absorption, and the contours are symmetric. REFERENCES Belikov, Y. E. and A. V. Gurvich 1995, Images of Optically Thick Artificial Aerosol Clouds in the Near-Earth Space,
Adv. Space Res., 15, 103(1995). Horak, H. G., D. M. Kerr and M. S. Tierney, Resonance Radiation in Artificial Strontium Clouds, Planet. Space Sci. 20, 165(1972). Lloyd, K. H. , Theoretical models for the Radiance of Contaminant Glow Clouds in the Upper Atmosphere, Aust. J. Phys., 18, 349(1965).
ACOUSTIC GRAVITY WAVE (AGW) GENERATION D U R I N G THE BARIUM INJECTION EXPERIMENTS Yu.Ya.Ruzhin, A.Kh.Depueva and Lurdes Palasio
IZMIRAN, Troitsk, Moscow region, 142092 Russia; E-mail:
[email protected] ABSTRACT The main objective of the recent wavelike ionosphere disturbance dynamics studies is to obtain information about the source of the disturbance generation. As a rule in practice either generator type or its precise location are unknown. There is a reason to believe that active space experiments with artificial injection are of interest in respect to this problem. The results of spectral analysis of artificial wavelike ionosphere disturbances generated by barium orbital injection in the CRRES - Caribbean campaign experiments are presented here. Plasma density variations at the F2 peak height were registered by an ionosonde over Havana (Cuba), a long way up to 2500 km off the injection point. Exact coordinates of the injection and the time delay between the injection moment and 10 min period of the spectral component of the density variation have been used to estimate a wavelike disturbance propagation velocity of 323-390 m/s, which depends on the particular injection conditions for all five experiments. In each experiment the effective propagation velocity in the terminator zone was correlated with the length of the sunlit part of the signal trace. Energy estimation was made and the conformity of space-time scales of the barium cloud development to the wavelike disturbance parameters registered over Havana was obtained. In this case the orbital CRRES injections were the most likely the reason for the appearance of wavelike ionospheric disturbances. Terminator passage can produce wavelike disturbances of the same parameters but they usually propagate from west to east (in our experiment they propagate in opposite direction, from east to west). Natural wavelike disturbances of such periods are connected with acoustic-gravity waves propagating in neutral atmosphere. Orbital injection of artificial barium clouds appears to be the effective source of acoustic gravity waves generation at upper ionosphere heights. If so, we have to deal with unique case of wave generation and propagation when the parameters of acoustic gravity waves source and its location are accurately known and situated at high altitudes of the ionosphere. INTRODUCTION Plasma density irregularities in the ionosphere with dimensions of approximately 100 - 500 km, horizontal velocities of approximately 200 - 1000 m/s and periods from a few minutes to one hour manifest themselves as so called wavelike disturbances (WD), primarily observed by means of vertical sounding ionosondes (Hines, 1960). Their appearance is often explained by acoustic gravity waves (AGW) propagation in neutral atmosphere. AGW in turn are generated due to tidal waves, typhoons, terminator passage, earthquakes, explosions, etc. Our current knowledge about WD sources, their structure and evolution is not sufficient to change WD from the object for studies to the tool of physical process diagnostics in atmosphere. Theoretically the question of WD generation mechanism still exists. In this connection, observations of WD generation during CRRES artificial injections (Bernhardt, 1992) is of great importance due to accurate knowledge of the unusual location of WD source (at the payload heights of 400-500 km). Ionospheric observations which were held simultaneously with five barium (Ba § ) ::eleases 111
112
Yu. Ya. Ruzhin et al.
during CRRES campaign in 1991 allowed us to find some reasons to connect WD with aforesaid artificial plasma injections. It is necessary to empasize that obtained data base is not large enough for making a final conclusion but some arguments in favour of this assumption are proposed. Besides, the orbital injection of artificial ion cloud can be a source of AGW, which in turn generate wavelike structures far apart from the injection point. CRRES EXPERIMENTS AND MAN-MADE AGW From June 5 till August 12, 1991 ground-based ionospheric measurements in flames of the CRRES satellite scientific programme with artificial cloud injections in Caribbean region were earned out. Injections were made during local sunrise time with the satellite moving from west to east approximately equatorial plane (satellite CRRES orbit inclination was--18 ~ (Bernhardt, 1992). There were five experiments with Ba injections. It is obvious from the Table 1 that the injections were made in evidently different circumstances: injection height, solar dip angle, quantity of injected material and so on. The analysis (Ruzhin et.al., 1995) of the injection place, solar terminator line position at the injection height and vertical sounding station at Havana showed that only the trace of experiment G-1 lb is completely nocturnal, which is the shortest one. We have analysed the critical frequency f0F2 fluctuations measured by the ionosonde operated in Havana in continuous regime. All injections were made to the east of Havana. Only three of them ((3-09, G-11 a and G-1 l b) were made under quiet ionospheric and geomagnetic conditions and they were analyzed. Spectra of f0F2 variations for these experiments are presented in Figure 1. It is obvious from the Figure that spectral maxima with period To-~ 10 min exist delayed relative to the injection moment. The experiments were held during a few consequent days and practically at the same local time, therefore they were selected for analysis.
Fig. 1. Spectograms of F2 -layer critical frequency variations (over Havana) for three experiments: G-09, G-1 la and G-11b. Pointed by arrows spectra maxima with period To of about 10 min appears with time-delay relative to the injection moment which is.marked by solid horizontal line. Terminator line was located in all presented cases at equal distance from Havana, nevertheless distance measured along Earth's surface between Havana (place of disturbance registration) and the injection
Acoustic Gravity Wave (AGW) Generation during the Barium Injection Experiments
113
place (disturbance source) varies from 1450 km (near the shadow boundary) to 2410 km (see distance DF2 and corresponding central angle a in the Table 1.). Shadow height over Havana for three choosen experiments therefore stayed constant and is more than or equal to 1000 km at the injection moment in the night ionosphere. Further analysis considering the previously discussed conditions showed that discovered plasma density variations (mean period To _=10 rain) at maximum F2 layer height over Havana have a time-delay (AT) relative to the injection moment which corresponds to the disturbance propagation speed of 320-390 m/s (depends on the length of sunlit part of trace Dlit for particular injection). Calculated values of disturbance horizontal propagation speed from injection point to Havana Vcf are presented in the Table 1. Table 1. The Main Characteristics of the CRESS Experiments Date 13.07 19.07 22.07 25.07 12.08
Time UT (h:m) 8:35 8:37 8:38 8:37 9:31
Height (km) 495 441 411 478 507
(z
(deg) 19,0 19,3 21,7 13,1 22,9
AT (min) 99 103 75 -
OF2
Dlit
gef
(km) 2136 2142 2409 1454 2544
(km)
(m/s)
642 909 -
361 390 323 -
Experiment 6-01 G-09 G-11a G-1 lb G- 12
Our estimations of barium cloud kinetic energy and its dimentions (Ruzhin et.al., 1995) are not at variance with the supposition that Ba § cloud injections are the sources of the observed WD. DISCUSSION A typical image of the artificial cloud in visible band of solar radiation spectra during first minute after injection is a bright shining sphere (front of barium stream) moving at the ionosphere above the maximum F2 layer height (hm F2) with speed of about 10 km/s and practically perpendicularly to the geomagnetic field lines. Intensive ionization process occurs, and generated plasma is carried away by the geomagnetic field and a bright tail along satellite orbit is produced. The effectiveness of injected material interaction in practically the collisionless regime with ionospheric plasma mainly depends on interaction rate connected with barium photoionization rate. Plasma ions produced in the ionosphere moving normally to environment magnetic field will generate electric polarization field which supports the cloud movement at the same direction (Bernhardt, 1992; Ruzhin et.al., 1995). Cloud breaking takes place during transformation of the polarization electric field impulse to background plasma contained in the magnetic field tube. Electric field can propagate down in field-aligned direction and then dissipate due to the large conductivity of ionospheric E-region. Parameters which control injected ion cloud breaking are as follows: mass, normal, component of the cloud velocity, plasma tube content and the Pedersen conductivity of E-region. The increasing of the interaction volume and the ionization rate allow "soft" transmission of artificial cloud kinetic energy to the background ionospheric plasma. The effectiveness of this energy transformation to WD may be up to 100% (Ruzhin et.al., 1995).. Alternately for any explosion (instantaneous energy release) at the same height effectiveness is less than a tenth fraction of a percent. The spectral analysis of F2 -layer critical frequency fluctuations (foF2) has shown spectral maxima with periods approximately 10 min and horizontal dimentions of 200-500 km that appeared on ionogram records delayed for nearly 75-103 min relative to the moment of the injection. Exactly this kind of WD are usually related to AGW. Thus there is a reason to believe that the artificial ion cloud could be a source of AGW, which in turn cause wavelike structures far apart from the injection place. Distance between the injection place and the observation location may reach two thousand kilometers or more. If so, we have dealt with a unique case of propagation when the AGW source location and commencement time are accurately known and it is situated at high altitudes (400-500 km) of the ionosphere.
114
Yu. Ya. Ruzhin et al.
In contrast usual AGW source location and period are unknown ahead of time and they are situated a little higher or lower than Eath's surface. It is necessary to note that CRRES injections were held in local sunrise time, so the question is if the observed phenomenon caused by coupled activity of terminator and Ba + release. Terminator passage can produce WD of the same parameters but it is usually propagating from west to east (in our experiment WD propagates in opposite direction, from east to west). We know that at present time our Chinese colleagues make preparations to active experiment (Xu et.al., 1995) over China territory. So, we would like to propose organization of ground-based experiments. WD parameters (direction and propagation velocity, period and wavelength) are usually studied by spatial net of measuring instruments. In the horizontal plane these are as a rule three ionosondes distributed in space and measuring one or another parameter of reflected signal. In vertical plane measurements are made by a number of frequencies reflected from various ionospheric heights. WD cause signal parameter fluctuations and similar fluctuations that are recorded with some time delays. By such kind of delay it is possible to estimate the phase velocities and wavelengths of WD. Statistical errors of time-delay estimates force somebody to increase dimension of the measurement base for increasing delays and, as a result, improve the accuracy of WD estimations. But it is known that unfortunately increasing distanses between observation points leads to coherent decrease of the records and consequently leads to less accurate time delay estimates. For instance in practice the distance between observation points has to be not more than 250 km for middlescale WD. At minimum, a triangle of ionosondes is necessary for the estimation of velocity and direction of arrival. CONCLUSION Active experiments not only give the possibility to study a number of problems, but also put some new questions. It shown for CRRES experiments that orbital barium injections cause WD appearance at extra (up to 2500 km) distances. In spite of the fact that the source location in active experiments is known, ionospheric disturbance phase and amplitude spatial distribution in horizontal plane have to be given for AGW model construction. During future campaigns, it would be possible not only to repeat similar experiments but also to continue them. Even simple confirmation of CRRES results could be very useful. Organization of simultaneous ionospheric observations on a triangle of ionosondes will allow us to easily get unique information about magnitude and direction of WD velocities. The results during various times (evening, morning, midday) will help us to estimate the terminator role in WD generation. Planned in China, the active experiment (Xu et.al., 1995) concerning the CIV problem give such opportunity. An optimal system of groundbased ionospheric observations by standard ionosonde net could be proposed for this aim. In addition it would be fruiteful to use other methods of AGW data collection: Doppler shift measurements, radioastronomic observations, etc. We think that organization of such observations would not be too expensive but would allow us to obtain new results concerning AGW properties connected with the artificial plasma injections with orbital velocities. We also propose to use a new method (Oraevsky et. al., 1995) of tomography which give us the possibility to measure directly such kind of wavelike structure. REFERENCES Hines, C.O., Intemal Atmospheric Gravity Waves at Ionospheric Heights, Can. J. Phys., 38, pp. 1441-1481 (1960). Bernhardt, P.A., Probing the Magnetosphere Using Chemical Releases from CRRES satellite, Phys. Fluids, B4(7), pp. 2249-2256 (1992). Ruzhin, Yu.Ya., V.N. Oraevsky, A.Kh. Depueva, J. Perez, and Lurdes Palasio, Orbital Barium CRESS Injection Effective Source of Ionospheric Wavelike Disturbances, Adv. Space Res.,15, 12, pp.(12) 127-(12) 130 (1995i. Xu, R.-L., D.-C.Lin, and F. Wu,.Theoretical Studies on the Space CIV Experiment over China, Adv. Space Res., 15, 12, pp.(12)139-(12)142 (1995). Oraevsky, V.N., Yu.Ya. Ruzhin, V.E. Kunitsyn et.al., Radiotomographic Sections of Subauroral Ionosphere along Moscow-Arkhangelsk trace, Geomagn. i Aeronomia, 35, 1, pp. 117-122, (in Rus.) (1995).
Minimizing The Adverse Effect Of The Photosheath Around A Spacecraft Hua
Zhao 1
K. Torkar 2
R. Schmidt 3
C.P. Escoubet 3
W. Riedler 2
1 Center for Space Science and Applied Research, Chinese Academy of Sciences, Beijing, P. R. China 2 Space Research Institute, Austrian Academy of Sciences, Inffeldgasse 12, A-8010 Graz, Austria 3 Solar System Division, Space Science Department ofESA, ESTEC, NL-2200 AG Noordwo'k, The Netherlands
ABSTRACT A small potential barrier (< 2V) might emerge in the photoelectron sheath during the active potential control operation using energetic (some keV) ion beams. Although the measurement conditions are improved considerably, some very low energy electrons of the ambient plasma cannot overcome the potential barrier and return to the plasma. It is designated as the screen effect of the potential barrier. In this paper, the relationship between the "'hidden" low energy electron population and the parameters of the photosheath potential barrier, such as the height (< 2V) and position of the potential barrier is analyzed. The domains of the "'hidden" and "'detectable" ambient plasma electrons in the velocity space are presented. The distribution function (in velocity space) of the undisturbed passing ambient plasma electrons can be derived from the measured data, if the photoelectron sheath electrostatic potential is assumed to be with spherical symmetry. INTRUDUCTION A spacecraft operating in the outer magnetosphere is often charged positively on its outer surface. The ambient plasma electrons as detected by a sensor on the spacecraft will have an energy corresponding to the potential added to their natural energy. If they are accelerated too much the distribution becomes highly compressed in energy and difficult to resolve in an analyzer (Johnstone et al., 1993). The detection of cold plasma electrons is also difficult because the lowenergy part of the ambient electron population is usually contaminated by photoelectrons from the satellite surface. Schmidt et al. (1995) have shown that active potential control operation decreases the electrostatic potential on the outer surface of the spacecraft. The surface potential decrease of the positively charged spacecraft will allow more photoelectrons emitted from the surface of the spacecraft to move into the ambient plasma. These photoelectrons are designated as the passing photoelectrons. If the controlled surface potential drops below a critical value, the passing photoelectrons will dominate the region near the spacecraft. A space charge dominated by photoelectrons near a conductive spacecraft surface may develop a potential minimum in the photoelectron sheath resulting in a nonmonotonic potential profile (Whipple, 1976). The height of the potential barrier has the tendency to increase as the ion emitter current becomes larger (Zhao et al., 1996). In this work, the influence of the potential barrier on the low energy electron measurement is described. The ambient plasma electrons with kinetic energy smaller than the height of the potential barrier cannot reach the instruments mounted on the spacecraft. The domain of the detectable ambient plasma electrons in the radial and transverse velocity, (vr, v T), phase space is shown in Section 2. In section 3, the distribution function of the passing plasma electrons in the undisturbed plasma region is derived analytically from the measured distribution function at the surface of the spacecraft. Some discussion and conclusion is presented in Section 4. D E T E C T A B L E E L E C T R O N IN (v r ,VT) PHASE SPACE The spacecraft velocity is small compared to the plasma electron thermal velocity, therefore the electron distribution 115
H . Z h a o et al.
116
function is assumed to be isotropic in the region out of the photoelectron sheath area. The ambient plasma electrons around the spacercraft are divided into passing and undetectable types. A plasma electron located with kinetic energy angular momentum L will hit on the surface of the spacecraft, if it is located within the passing region in (E, L 2) phase space, and the bounderies of the region are given as (zhao, et al., 1996): (2) L2 = E - us (1) L2 = XB2 (E m UB) where u s = - e ~ b s / k T ,
and u B = - e q } B / k T ,
~bs is the surface potential of the spacecraft, ~bB, the potential
minimum of the photosheath, which has been shown by Zhao et al. (1996) in Figure 1. XB = rB / R , r B is the position of the potential minimum and R is the radius of the spacecraft, k is Boltzmann's constant, and T is the ambient plasma electron temperature. The domain of the passing ambient electrons can be mapped from (E, L 2) phase space into (Vr, v T) phase space, where v r and v T are radial and transverse velocities, respectively. It should be pointed out that
the undetectable and passing domains in the velocity space are dependent on x, which is the radial position of the undisturbed plasma electrons. Any electron which can hit the surface of the spacecraft from the ambient plasma will pass through the edge of the photoelectron sheath around the spacecraft. The radius of the photoelectron sheath is assumed to be x o . x o is the order of the plasma Debye length. In the velocity space, the passing domain of the plasma electrons which are originally located at x o is analyzed in this section. The passing region in the ( E , L 2) phase space can be mapped into a passing domain in the velocity space (Vr, v T ) by 2 substituting E = m (v 2 + v 2 ) a n d L 2 = x o m v 2 into the Eq.(1) and Eq. (2): 2kT
2kT 2kT
2
(Xo2 - 1)v 2 - v r = - ~ u , m 2
(1
x82 . 2 2 k T --5-)vr ~ u xo
(3)
--21 B
(4)
- (Xo2
3
m
In the velocity space, the passing region of the plasma electrons located at x o is shown in Figure 1. The dashed line 1 in Figure 1 is corresponding to the case of flat eletrostatic potential profile in the plasma sheath around the spacecraft, and therefore the value of the potential is equal to the potential of undisturbed plasma. It means that the potential inside the plasma sheath is constant and equal to zero. The equation of the dashed line 1 is: vr = ~
o,,.i
v~ - 1) 1/2
(5)
A passing plasma electron originally located at Xo with velocity components vr and vr (corresponding to the total energy E and ngular momentum L ), will changes its velocity components to v' r and v' T respectively,when it moves onto the surface of the spacecraft. 2kT v' r = [V2r - - ( X 2 -- 1)@ - - ~ U s ]
1/2
~ 0 ,,,-?-! ,
radial velocity v r Figure 1. The region of the passing ambient plasma electrons, located at x o , in the ( V o , V T ) V e l o c i t y space. The passing region is limited by line 2 (given by Eq.(3) and line 3 (given by Eq. (4). The dashed line 1 is corresponding to the situation that the sheath potential profile is flat and therefore its value is equal to the potential of the undisturbed plasma.
(6)
V ! 7"---
(7)
XoVr
m
At the surface of the spacecraft, the velocity components of a passing plasma electron v' r and v' T can only take values in a special region. The region can be obtained by substituting Eq. (6) and Eq. (7) into Eq. (3) and Eq. (4): Vr>O
(8)
V, r -2 >
2
x~ 2--1
[2kT (
m
uB-us)-v
w 2 ]1/2
r
(9)
XB
The region in which v' r and v' T take values is schematically shown by region B in Figure 2. The detected plasma electrons cannot exist inside region A. Region A is filled with returned photoelectrons.
Minimizing the Adverse Effect of the Photosheath Around a Spacecraft
117
It is assumed that an electron detector is installed at the edge of the spacecraft, where x = 1. An electron which impacts on the sensor through a collimator with viewing angle 0, should satisfy the following condition: t a n ( 0 + A0) < v'---r-r< t a n 0
(10)
Vtr
where A 0 is the opening angle of the collimator, The viewing angle is defined as the angle between the viewing direction and the radial direction. It is easy to find that the maximum energy, e max , of the returned photoelectrons moving into the electron instrument through the channel 0 can be given by: 2
=
~
XB
2 __ s i n 2 0
(u B - u s )
(11)
XB
From the above analysis, one can obtain that emax, the maximum kinetic energy of the retumed photoelectrons goes up, as the viewing angle 0 becomes larger. The instruments looking along the surface of the spacecraft are more disturbed by photoelectrons than instruments looking radially away from the spacecraft. This effect is caused by the existence of the potential barrier near the spacecraft. For the thick sheath assumption, the position of the potential minimum XB >> 1, and it is easy to find that the maximum kinetic energy of the returned photoelectrons for this case: O~ x ~ b/B -- b/s (12) From Eq. (12), it is understood that the disturbance of the returned photoelectrons to the measurements of the ambient plasma electrons is not dependent on the viewing angle of the collimator, if Xa >> 1. It should be pointed out that the effect of the magnetic field has been neglected here.
THE DISTRIBUTION FUNCTION OF THE PLASMA ELECTRONS The measured passing ambient electron distribution function, f ( v ' r ,v' T )at the surface of the spacecraft is distorted from the original distribution, F ( V r , V T ) i n the undisturbed plasma region by the existence of the plasma sheath electrostatic
potential
around
the
spacecraft.
The
distribution
function
F ( v r , V T) can
be
obtained
by
transforming f ( v ' r , v' T ) into (Vr, v T ) phase space, if the spherical symmetry of the plasma sheath potential is assumed. Using the continuity equation: d
0
(13)
where V is a spherical volume, which coincides with the centre of the spacecraft, and S is the surface of V. p is the plasma electron density, and J e , the plasma electron number current density vector, one can gain the following equation for the spherical symmetry: 3N
---4mC2o dt
~vrF(Vr,Vv,x)dvrdv r = 0 v
(14)
passing ambient plasma electrons at the position x o . where N is the total number of the passing plasma electrons inside V , x o is the radius of V. F ( v r , v T , x o ) is the distribution function of the For a steady-state solution, the condition cyTV / d t = A = c o n s t is valid, and it is easy to set up the relationship between the measured distribution function f ( v ' r ,v' T )and the original
5 [-
a
-5
RegionB
b radial velocity v' r
Figure 2. The regions of the electrons located at the surface of the spacecraft, in the (v' r ,v' T ) velocity space. Region A is fulfilled with the returned photoelectrons. Region B is for passing ambient plasma electrons. a = [ xZ---~B 2k_~T(u a _ Us)]l/2, x2-1 m
b = [ 2 k T (u B _ Us)]l/2 m
distribution function at position x o , F ( V r , V T ) : Xo2 I v r F ( v r D~
, v,
)dvrdv r = Iv, r f ( v , r , v, r ) d v , r dr, r D:
(15)
H. Zhao et al.
118
where D 1is the passing domain of the ambient plasma electrons located at x o, which shown in Figure 1, D 2 is the domain of the detected ambient plasma electrons, which shown by region B in Figure 2. From Eq. (15), the distribution function of the undisturbed passing ambient plasma electrons can be expressed as
F(vr,Vr) = v'~ d(V'r ,V' r) f (v,v ,
2
Xo Yr
O(Vr V~)
V'~
)
(16)
Eq. (16) is correct for the plasma sheath electrostatic potential with the spherical symmetry. This equation can be used to minimize the adverse effect that the plasma sheath electrostatic potential has upon the distribution measurement of the ambient plasma electrons. For the case of the active potential control operation, the relations between (Vr,VT) and (v' r , v' T ) are given by Eq. (6) and (7), and substituting them into Eq. (16)
F(vr,vr) = l-~-f (V'r (Yr,YT),Y'T(Yr,YT))
(17)
Xo
DISCUSSION AND CONCLUSION From Figure 1 it can be observed that the minimum kinetic energy of the detectable ambient plasma electrons is Z/B, which is the height of the potential barrier.Figure 2 shows that the photoelectrons emitted from the surface of the spacecraft with kinetic energy smaller than lUB l+ lUs ]will return to the surface of the spacecraft. The maximum kinetic energy of the returned photoelectrons is dependent on the viewing angle. The larger the viewing angle, the larger the maximum energy of the returned photoelectrons detected by an electron instrument. The instruments looking along the surface of the spacecraft are more disturbed by photoelectrons than instruments looking radially away from the spacecraft. If the position of the potential barrier is far from the spacecraft, x B >> 1, this adverse effect can be neglected. The minimum kinetic energy of the detected ambient plasma electrons is ]uB 1+ lUs l" Zhao et al. (1996) have analyzed the height of the potential barrier around the Geotail satellite during its active potential control operation which was carried out on 1 September, 1992 (Schmidt, et al., 1995) It has been shown that the height of the potential barrier was about 2 V corresponding to the ion emitter current, 38.3/.tA, and the surface potential of Geotail satellite was about 4 V. In this case, the undisturbed ambient plasma electrons can be measured above the energy of 2 eV. The minimum kinetic energy of the detected ambient plasma electrons is about 6 eV. A continuation of this study will deal with methods to minimize the height of the potential barrier. The position of the potential barrier X B, estimated in the previous study (Zhao, et al. 1996), is about 25R. The dependence of the maximum kinetic energy of the returned photoelectrons on the viewing angle can be neglected for the situation analyzed previously. The distribution function of the passing ambient plasma electrons in the undisturbed plasma region can be derived analytically from the measured distribution function, which is distorted by the plasma sheath electrostatic potential, if the surface potential of the spacecraft is known.This is very useful when the ambient plasma electron distribution cannot be assumed as Maxwellian. ACKNOWLEDGMENT This work was partly supported by the National Nature Science Foundation of China. REFERENCES Johnstone A. D., C. Alsop, P. J. Carter, A. J. Coates, A. J. Coker, et al., PEACE: A plasma electron and current experiment, in: Cluster: mission, payload and supporting activities, Noordwijk, The Netherlands, ESA SP-1159, 165, 1993. Schmidt, R., H. Arends, A. Pedersen, M. Fehringer, F. Rudenauer, et al., Results from active spacecraft potential control on the Geotail spacecraft, J. Geophys. Res., 100, 17253, 1995. Whipple, E. C. Jr., Theory of the spherically symmetric photoelectron sheath: a thick sheath approximation and comparison with the ATS 6 observation of a potential barrier, J. Geophys. Res. 81, 601, 1976. Zhao H, R. Schmidt, C. P. Escoubet, K. Torkar, W. Riedler, Self-consistent Determination of the Electrostatic Potential Barrier due to the Photoelectron Sheath near a Spacecraft, J. Geophys. Res., 101, 15653, 1996.
Ti-C REACTION IN LABORATORY AS A HEATING TECHNIQUE IN SPACE CHEMICAL RELEASE EXPERIMENTS F. Wu, R. L. Xu, L. Li, Z. G. Zhang, and Y. B. Liu
Center for Space Science and Applied Research, P.O.Box 8701, Beij'ing 100080, China
ABSTRACT A small sample of Titanium-Carbon (Ti-C) reaction is studied in laboratory as a barium (Ba) heating and vaporization technique in space chemical release experiments. Instead of Ba, we use ferrum (Fe) and stibium (Sb) in the laboratory experiment, the time profile of the reaction temperature, the effect of the different weight percentage (wt%) of Fe as a diluent on the maximum reaction temperature, the combustion velocity, and the wt% of vaporized Sb have been studied in the laboratory. The experimental results are compared with some theoretical estimat'on of the effect of the different wt% of Fe, Ba and Sb as diluents on the adiabatic temperature. INTRODUCTION In the chemical release experiments of the Combined Release and Radiation Effects Satellite (Reasoner, 1992), the reaction of Titanium and Boron (Ti-B) was used to vaporize the release substances Ba and strontium(Sr). The adiabatic temperature of Ti-B reaction is higher than 3000K. When the other substance is added as a diluent, the reaction temperature will decrease, and depends on the wt% of the diluent. With 40 wt% of Ba, the reaction temperature is about 2500K, which is more than 500K higher than the boiling point of Ba. In our Chemical Release Experiment Program (Xu et al. 1995), we use Ti-C reaction, Ti + C --" TiC + 44 kcal/mole
(1)
which is similar to Ti-B reaction, but cheap, stable and easy to handle. Instead of Ba, we use Fe and Sb as diluents. The melting and boiling point of Sb is similar to Ba's, but Sb is easier to handle in the air. The relations between the adiabatic temperature and the wt% of Ba, Fe and Sb as diluents have also been studied theoretically. LABORATORY EXPERIMENT The general arrangement of the apparatus in the laboratory experiment is shown in Figure 1. A cylinder sample with 20 mm diameter and 40mm height [2] is placed in a graphite disk [6] on an adjustable platform [8] at the center of the vacuum chamber. The sample consist of Ti(-200 mesh) and C(-360 mesh) mixed in the ratio of 1 mol Ti to 1 mol C. Fe (-200 mesh) or Sb is added as a diluent with the different wt%. The mixture is combusted in loose form. The electric current for the ignition tungsten [ 1] is supplied by a 24V 250A power source. The distance between the sample and the ignition tungsten can be adjusted by the adjustable platform. The Ti-C reaction can occur in atmospheric pressure, in vacuum, and in argon. An Infrared Thermometer [5], with a measurement range up to 3000~ target scale of 1.5mm and response time of 0.1-0.2 sec, is used to measure the temperature of the sample of Ti-C reaction. 119
F. Wu et al.
120
To es!~,aate the combustion front velocity of the reaction, three solar battery sensors [4] are employed. The viewing window of the first sensor is 2n solid angle, which measures the total radiation of the combustion and the ignition tungsten. The other two sensors have optical systems with viewing windows of a few degrees. The distance of the extra-axial points of these two optical systems is equal to 2cm. According to the shift time of the output of these two sensors, we can estimate the combustion front velocity of the reaction. High-speed multi-channel data sampling is achieved by the A/D board and the computer, program. The data sampling system has 16 channels, the time interval between two sampling points ranges from 10 las to 0.1s.
1
F:~.2 5
1. Ignition Tungsten 2. Sample of Ti-C 3. Temperature sampling point 4. Radio meter 5. Infrared high temperature meter 6. Graphite disk 7. Heat sink 8. Adjustable platform 9. Base disk 10. Current conductor
Fig. 1 Schematic drawing of the experiment apparatus The main results of the experiment are as follows: The Reaction Temperature The variation of the temperature of the Ti-C reaction with time is shown in Figure 2 by curve A. The horizontal axis corresponds to time in second (s), and the vertical axis represents temperature in K. The time profile of the temperature shows that the temperature increases suddenly in one to two seconds to its maximum value about 2500K at time t = 10s, then is followed by a slow decay for more than 40s. The small raise before the sudden large increase is expected due to the influence of the tungsten heating. The maximum temperature becomes lower when the other substance is added as a diluent. Curve C in Figure 3 displayed the variation of the maximum temperature of Ti-C reaction with different wt% of Fe. Steady combustion with temperature higher than 1800K has been observed when the wt% of Fe is 50, which is similar to the Ti-B-Ba experiment of Wescott et al. (1994), where the wt% of Ba is equal to 40. 0
10
20
30
40
50
60
2500
2500
2OO0
2OO0
1500
1500
1000
1000
3600
3200
I-.
-
~. 2400 2OOO
500
500 1600
0
0 0
10
20
3o
40
50
60
t(s)
Fig. 2 The variation of temperature(A) and radiometer output (B,C,D) with time
1200
0
......... 10
~ .... ' .... ' ......... 2O 3O 4O 5O
6O
wt%
Fig.3 The effect of Fe as a diluent on (A)adiabatic, (B) corrected and (C) measured temperatures
Radiation of the Reaction and the Combustion Front Velocity Curve B in Figure 2 illustrates the time variation of the total radiation measured by the 2n field of view solar
Ti-C Reaction in Laboratory as a Heating Technique in Space Chemical Release Experiment
121
battery sensor. It presents the logarithm of the illuminance of the whole sample surface. The first large increase of curve B is expected due to the heating of the ignition tungsten, while the second increase is the start of the Ti-C reaction. Curve C and D are the time profiles of the output of the other two solar battery sensors with optical systems. From the shift time of curve C and D, and the distance between these two sensors, we can estimate the combustion front velocity, U. The variation of U with the wt% of Fe as a diluent is shown in Figure 4. We find that U is different for the different wt% of Fe, and the variation of U is not so obvious when the wt% of Fe is larger than 20, which is expected since U is effected both by the thermal diffusivity and the reaction temperature (Holt, 1986).
2224I 6
/t~"%%
eo 201 //I
5 ~'4
"~ 16
E E
9
~ ' 32
9
9
9
12 10
1 0
0
10
20
30 wt%
40
50
. . . .
8
60
Fig. 4 The variation of the combustion front velocity with the wt% of Fe
2
4
6 8 weight of Sb (g)
10
12
Fig.5 The relation between the total weight of Sb and wt% of vaporized Sb
The Vaporization of the Metal The relation of the wt% of the vaporized Sb and the total weight of Sb, measured in open air, is shown in Figure 5, the maximum wt% is about 23.5. In vacuum, the maximum wt% is higher, and is equal to 30. THEORETICAL ESTIMATION The experimental results of the effect of Fe as diluent on the temperature can be compared with the theoretical estimation in a thermally isolated system. For Ti-C reaction with Fe, Ba or Sb as diluents, the adiabatic temperature Tad can be find from /~'/208 = f 298
Cp,TtcdT+x[ l~tr+ f Cp'dT+Z~m+ ffp"dT+ Z~Ib+ ffp"'dt+ fCpdT ]-dduent 298
T~
T,,,
(2)
To
0
where Z~k/298 is the standard heat of formation of TiC, Cp. T,c is the solid phase mole heat capacity of TiC, ,
,,
.,
Cp, Cp , Cp and Cp are the heat capacities of the diluent in solid phase 1, solid phase 2, liquid phase and gaseous phase, d applicable, x is the amount of the diluent in mole. A/-/tr, Ann and AHb are the heat of transition, heat of fusion and heat of vaporization respectively. Tr ,Tm, and Tb are the temperatures of the diluent at the transition point, melting point and boiling point respectively. All these parameters can be found in Ye(1981). Using Eq.(2), we can estimate the effect of the diluent on the adiabatic temperature of Ti-C reaction. Figure 6 shows the variation of the adiabatic temperature for the different wt% of Ba and Sb as diluents. These curves are similar, but the fiat region for Ba is wider than Sb. Since the heat capacity and density of Fe is higher than Ba, the effect of Fe as a diluent on the reaction temperature may be more obvious than Ba (Wu et al., 1995). The relation between Tad and the wt% of Fe is shown with curve A in Figure 3. Due to the vaporization of Fe, there is a flat section on the
122
F. Wu et al.
curve, where the temperature keeps constant. The temperature measured in our experiment shown with curve C in Figure 3 has the same trend with the theoretical prediction, but is much lower than the theoretical prediction. There are several reasons for this difference, one of which is the emissivity coefficient correction of the Infrared Thermometer. The temperature measured by the Infrared Thermometer corresponds to the brightness temperature and should be corrected according to the surface emissivity coefficient of the reaction sample(Liu, 1985). After the correction of the emissivity coefficient and the response time of the Infrared Thermometer, the final result is shown with curve B in Figure 3. There is still a difference of 200K between curve A and B. The difference is expected due to the influence of heat loses by radiation and conduction. CONCLUSIONS
3500 3O00
"~-.
2500
Sb
~2000 I--
1500
\
\
1000 . . . .
0.1
I
. . . .
0.2
1
. . . .
0.3
I
. . . .
0.4
I
,
0.5
i
i
i
I
i
0.6
i
L
i
I
i
0.7
i
i
i
I
,
0.8
i
L
i
0.9
wt%
Fig. 6 The effect of the wt% of Ba and Sb as diluents on the adiabatic temperature
A small sample of Ti-C reaction with the different wt% of Fe or Sb as a diluent has been tested in the laboratory. The maximum temperature and the combustion velocity of the sample for different the wt% of Fe, and the vaporized wt% of Sb have been studied. The effect of the different wt% of Ba and Sb as a diluent on the adiabatic temperature of the reaction is also estimated theoretically. All the foregoing results lead us to conclude that the TiC reaction can be used as the chemical material heating and vaporization technique in the space active experiment. ACKNOWLEDGMENT We arc greatly indebted to Professor Zhengzhi Fang, Center for Space Science and Applied Research, Chinese Academy of Sciences and Professors Ho-Yi Lai, Sheng Yin and Dr. Qing Tang, University of Science and Technology of Beijing, for their support and inspiring discussions. Finally we are grateful to Ms. Haiwen Xiao for her appreciated assistance. REFERENCES Holt, J. B., D. D. Kingman and G. M. Bianchini, Kinetics of the combustion synthesis of Ti B 2 , Materials Science
and Engineering, 71, 321 (1985). Holt, J. B., Z. A. Munir, Combustion synthesis of titanium carbide: Theory and experiment, J. Mater. Sci. 21, 251(1986). Liu, Dexian, Basic lnfrared System Design, Hangzhou Institute of Technology press, Wuhan(1985) (in Chinese). Reasoner, D.L, Chemical-release mission of CRRES, J. Spacecraft and Rockets, 29, 580 (1992). Wescott, E. M., H. C. Stenbaek-Nielsen, D. L. Hampton, and P. A. Delamere, Results of critical velocity experiments with barium, strontium, and Calcium release from CRRES satellite, J. Geophys. Res., 99, 2145 (1994) Wu, Feng, Lei Li, Ronglan Xu, Zhiguang Zhang, Dechun Lin and Haiwen Xiao, A heat source by Ti-B, Ti-C reaction, Proceedings of 94' Symposium on Chinese Material Sciences (1995) (in Chinese). Xu, Ronglan, Dechun Lin and Feng Wu, Theoretical studies on the space CIV experiment over China, Adv. Space Res., 12, 139 (1995). Ye, Dalun, Practical Thermodynamic data Handbook of Inorganic Substance, Metallurgy press, Beijing(1981) (in Chinese).
Section III Numerical Simulation and Theoretical Modeling
This Page Intentionally Left Blank
ION BEAM VELOCITY DISTRIBUTIONS SHEET BOUNDARY LAYER
IN PLASMA
L.M. Zelenyi x, A.L.Taktakishvili 1, E.M. Dubinin a, E.Yu. Budnik 1 and I. Sandahl 2
1Space Research Institute, 84/32 Profsojznaja str., Moscow, 117810, Russia 2Swedish Institute of Space Physics, S-981 28 Kiruna, Sweden
ABSTRACT Using quasi-adiabatic description of the nonlinear dynamics of ions in two-dimensional current sheet with constant electric field in the direction of the current, we investigate new, strong effect of additional acceleration (it is mostly betatron acceleration), during the convection of plasma in such a configuration. This allows us to explain some important features of ion velocity distributions in the plasma sheet boundary layer of the Earth's magnetotail, observed by INTERBALL-TAIL satellite. INTRODUCTION Since the first reports of Lui et al. (1977) and De Coster and Frank (1979), the study of high speed ion flows in plasma sheet boundary layer (PSBL) of the Earth's magnetotail is subject of intensive theoretical and experimental studies. The interest to this problem has raised recently due to new observational data on the features of fast ion flows, obtained by instruments on board of number of spacecrafts - GALILEO, GEOTAIL, INTERBALL-TAIL. Recent measurements of 3-D velocity distributions have revealed different types of plasma populations in PSBL. In this paper we focus on the problem of velocity distribution of the ions jetting in PSBL. v
Theoretically this question was addressed before by many authors. De Coster and Frank (1979) considered acceleration of ions by an electric field parallel to the_magnetic field. Birn et al. (1981) investigated the adiabatic deformation of the flowing Maxwellian distribution. Lyons and Speiser (1982) studied numerically the velocity distributions of ions in a simple model of tail magnetic field with the constant normal component, Bz = const, plus electric field in the direction of the plasma sheet current,, E u = const. Acceleration occurs during particle motion along "Speiser orbits" (Speiser, 1965). They obtained the distribution functions of streaming ions resembling observed "kidney bean" shapes in velocity space. Recently B/ichner and Kuska (1996) developed a quasi-adiabatic theory of acceleratign in two-dimensional (Bz(z) ~ const) current sheet, in the presence of Eu = const, to explain new features revealed by the 3-D distribution measurements. They took into account nonlinear dynamics of the particle motion, but particles are accelerated only non-adiabatically, similar to Speiser acceleration. 125
126
L.M. Zelenyi et al.
In the present paper we try to develop simple, but more comprehensive qualitative model of formation of ion velocity distribution also on the bases of the quasi-adiabatic description of the nonlinear dynamics of particles in two-dimensional current sheet with constant electric field Eu = coast: But, unlike B/ichner and Kuska (1996), where only non-adiabatic effects were taken into account, in our model quasi-adiabatic character of particle motion plays an important and very often dominant role in the final energy gained by the particles. NONADIABATIC AND QUASIADIABATIC ACCELERATION OF PARTICLES Quasiadiabatic particle trajectories in one-dimensional current sheet (Bz = coast and E = 0) were first studied by Sonnerup (1971), where the conservation of adiabatic invariant relevant to particle a fast oscillations across the current sheet, I, = ~ ~ pz dz = coast, was assumed. We will use below the dimensionless expression for adiabatic invariant, introduced first by Bufihner and Zelenyi (1989), I ' = I,/Io, where I0 = ~(~'-mL/12o)l/2(2W) 3/4, L is Bx magnetic field scale length, 120 is gyrofrequency in the lobe field Box, m and W are particle mass and energy respectively. Motion in x and y direction is much slower. If there exists a nonzero electric field Eu -~ 0 (e.g. dawn-dusk quasistationary magnetotail field), it can be removed by De Hoffman and Teller [1950] transformation to the coordinate frame which moves in the direction of the x-axes (in GSM coordinates towards the Earth) with the velocity vc = cEy/B, (we will refer to this frame as HT below). B/ichner and Zelenyi (1989) developed more detailed theory of particle motion in such configurations in application to magnetotail. In HT frame particle trajectory projection on X Y plane resembles a cucumber prolonged toward the Earth. It's wider part is close to the fragment of a circle due to Larmour gyration in B~. In a weakly two-dimensional geometry, when Bz(x) ~ coast, but Bz is a slowly changing function of x, approximate conservation of an additional, "longitudinal" adiabatic invariant, Ix = ~ px dx = coast, can be assumed (Zelenyi et al., 1990b; Vainshtein et al., 1995). Though HT frame is not inertial due to x dependence of B~(x), by taking into account force of inertia F in = - m d v c ( x ) / d t , we can move to local HT frame (see Ashour-Abdalla et al., 1994). The drift of particle in y direction caused by F in, is not significant, because of weak dependence of geometry on X. In this case, together with slow E x B drift in X direction, particle exhibits additional, quasiadiabatic drift in y direction, due to nonuniformity of Bz field in X direction. In the rest frame, where Eu -~ 0, this implies additional acceleration of particles. Trajectories of particles launched from a mantle sourceto the Tsyganenko (1989) model tail magnetic field, in the presence of constant Eu field, were studied in Ashour-Abdalla et al. (1994). The X Z projection of the typical trajectory exhibits fast Z-oscillations together with quasiadiabatic bouncing in X-direction. Figure 1 shows the schematic of X Y projection of the typical trajectory. Cold mantle particles coming from the regions placed above (or below) the plane of the plot, hit the current sheet and the first, relatively big "semi-circle" corresponds to "Speiser" orbit and appropriate nonadiabatic acceleration. After this, particle motion driven by quasiadiabatic mechanism, results in evident X and Y drifts. One could imagine the Z-averaged particle motion as the combination of the drifts in X and Y of the "cucumber" orbit. Finally, following the character of particle trajectory, the process of acceleration could be divided into three main stages: 1) Nonadiabatic acceleration, when particle hits the current sheet for the first time- mechanism of a single, rather large energy gain; 2) Quasiadiabatic acceleration, during repeated interaction of the particle with the current sheet in the course of the convection of the plasma. The acceleration at each step in this case, is smaller then at the first one, but due to multiple steps, total energy gain can be even larger; 3) The last interaction
Ion Beam Velocity Distributions in Plasma Sheet Boundary Layer
127
of the particle with the current sheet before it's registration by the instrument. Particle orbit is not closed anymore, and for the energy gain on this stage it is important where has the particle left the "cucumber" orbit. Gained energy depends on the value of adiabatic momentum I' at this site and I' in it's turn is related with pitch-angle (see Eqs.(6-7) below). As it is clear from the Figure 1, in the magnetospheric frame the particle energy oscillates. That is why, in HT frame, where particle trajectory is closed, we will define particle motion energy on the "cucumber" orbit relative to the center of the "cucumber" y t o c a t H T "-- O. We will assume first that the source particles coming from the mantle, are cold at the beginning, and thus the condition v0 n,,'
5 E)
S_=..3:..VoN~=O.3
4
0 s
Z LU 13..
n," ILl 13.
~'-.,
-~::~: ......... .
.............
,,=3, v~Nc=o.l"%~'.,, "" ~
\,,
\', \ .., ',
"\
",
\\
",,\
,,
2
-
0 0
I
} 2
I
,__ S 1 . 2
i
4
'
i
~i
Il 6
PARELLEL VELOCITY /Vc
! !
8
Fig. 2. The shapes of "backbones" for different s and different initial velocities (velocities are normalized to the value of v~(x2) =-- v~ at the observation site). Figure 2 shows a scheme which illustrates shape of the proton distribution function in PSBL. Distribution functions are pulled on backbones (as it is shown schematically in Figure 3). In Figure 2 we plotted the "backbones" for s = 1,2, 3, 4 and initial values of pitch-angle and velocity 90=0, v0 = 0 (solid lines). We also plotted "backbones" for different initial velocities, two for each of the cases 8 = 3 and s = 4 (dashed lines, appropriately marked in the Figure). Velocities are normalized to v~(x2). It is clear that the shape does not depend substantially on the initial velocity. The dependence on initial pitch-angle (not shown) is even less. "Backbones" change the shape with the increasing of the distance between xl and x2: the more is this distance, more they resemble fragment
L.M. Zelenyi et al.
130
of a circle. While for xx = x2 (s = 1), the structure is prolonged in parallel direction. This result becomes clear from Eqs.(ll-12), because for large s, the influence of the pitch angle on the final energy is less, which makes the distribution in velocity space more isotropic. Thus, one could say that backbones for different observational points x2, very roughly have circular shape with radius R shifted along the VII axis on the value of a speed gained during ion motion in the current sheet V~(x2). Radius R determines the locus of energy conservation points R
=
-
In Figure 3 we plot the case when at the observation site vc(x2) = 100, and initial v0 = 00 = 0. The last curve, s = 9, could be considered as a "backbone" for the beamlet distribution function, presented in Figure 4.
Fig. 3. "Backbones" of PSBL pitch-angle distribution for different observation sites.
Fig. 4. Proton distribution function demonstrating a beamlet structure.
From Eqs.(ll-12) it is obvious that there exists a maximum pitch angle, 0m~, for each v, as the expression under the square root shall not be negative. This can explain large pitch angle cut off in the ion distribution function which is often observed for the beamlet like structure in PSBL (see Figure 4). OBSERVATIONS The PROMICS-3 instrument (Prognoz Magnetospheric Ion Composition Spectrometer) was installed onboard the INTERBALL-TAIL satellite launched in the August 1995. With the apogee near 30 RE and the inclination about 65 ~ the INTERBALL probes main magnetospherical regions including the PSBL. PROMICS-3 performed simultaneous measurements of ion and electron distribution functions, moments and plasma composition of the magnetospheric plasma. The experiment consists of the ion composition spectrometer TRICS for ion measurements with mass separation in the energy range 4 e V - 70 keV and electron spectrometer MEPS (12 e V - 35 keV) (Sandahl et al., 1996). The TRICS unit is divided into 3 subunits with energy ranges 4 e V - 1.5 keV, 1.1 k e V - 30 keV, 5 k e V - 70 keV to provide a high sensitivity at all energies. TRICS-1 and TRICS-2 subunits contain 5 detectors each and TRICS-3 contains one detector. Each subunit consists of a toroidal analyzer
Ion Beam Velocity Distributions in Plasma Sheet Boundary Layer
131
followed by Wien velocity filter. Ions enter the instrument with a total opening angle 180~ Behind the analyzers TRICS 1,2 five detectors are arranged in a fan-like fashion. The PROMICS-3 measures 3-D distributions of charged particles due to a spinning of the satellite.. The satellite spin axis is directed to the Sun. Spin period is 2 min. The instrument operates in different modes of measurements and telemetry transmission. A complete measurement cycle for all ions takes 9.6 s giving 12 cycles per spin. Each cycle includes 32 energy steps at each subunit for different ion species. In the "FAST" mode energy spectra at 16 energy steps are measured during 0.8 s. Figure 4 shows the example of proton distribution functions in the PSBL (November 20, 1995). Distribution functions are presented in the ~ 1 - V• plane (k~l and V• are velocity components parallel and perpendicular to the magnetic field respectively) ' Positive ~1 correspond to earthward proton fluxes. The phase space density is given in s 3 cm -6. We have restricted the analysis by the data from the TRICS-2 unit which covered the typical range of energies for CPS and PSBL ions. Beamlet-like structure with a shape, which well resembles a distribution function discussed in the previous section, is clearly observed. The value of vr at the point of location of the spacecraft can be estimated as .-~ 100 km/s. From the expression for R and Figure 4 we can estimate s ~ 9. This means that the instrument records particles mostly accelerated in the distant tail (B,(x2) .~ 0.8 nT). ACKNOWLEDGMENTS This work was supported by the grant 96-05-645334 of the Russian Science Foundation and INTAS grants 94-2638 and 94-2031. REFERENCES Ashour-Abdalla, M., L.M. Zelenyi, V. Peroomian, ana R.L. Richard, Consequences of Magnetotail Dynamics, J. Geophys. Res., 99, 14891, (1994). Birn, J., T.G. Forbes, E.W. Hones, Jr., S.J. Bame, and G. Paschmann, On the Velocity Distribution of Ion Jets During Substorm, J. Geophys. Res., 86, 9001, (1981). Bfichner, J. and J.-P. Kuska, On the Formation of Cup-like Ion Beam Distributions in the Plasma Sheet Boundary Layer, J. Geomag. Geoelectr., 48, 781, (1996). Bfichner, J. and L.M. Zelenyi, Regular and Chaotic Charged Particle Motion in Magnetotaillike Field Reversals, 1. Basic Theory of Trapped Motion, J. Geophys. Res., 89, 11821, (1989). De Coster, R.J. and L.A. Frank, Observations Pertaining to the Dynamics of the Plasma Sheet, J. Geophys. Res., 84, 5099, (1979). De Hoffman, F. and E. Teller, Magnetohydrodynamic Shock, Phys. Rev., 80, 692, (1950). Lui, A.T.Y., E.W. Hones, Jr., F. Yasahura, S.-I. Akasufu, and S.J. Bame, Magnetotail Plasma Flow During Plasma Sheet Expansions: VELA 5 and 6 and IMP 6 observations, J. Geophys. Res., 82, 1235, (1977). Lyons, L.R. and T.W. Speiser, Evidence for Current Sheet Acceleration in the Geomagnetic Tail, J. Geophys. Res., 87, 2276, (1982). Sandahl, I., S: Barabash, H. Borg, E.Yu. Budnik, E.M. Dubinin, Eklund, H. Johansson, H. Koskinen, K. Lundin, R. Lundin, A. Mostrom, R. Pellinen, N.F. Pissarenko, T. Pulkinen, P. Toivanen and A.V. Zakharov, Annales Geophysicae, 1997 (in press). Sonnerup, B..U.O., Adiabatic Particle Orbits in a Magnetic Null Sheet, J. Geophys. Res., 76, 8211, (1971). Speiser, T.W., Particle Trajectories in Model Current Sheets, 1. Analytical Solutions, J. Geophys. Res., 70, 4219, (1965). Tsyganenko, N.A., A magnetospheric magnetic field model with a warped tail current sheet, Planet.
132
L.M. Zelenyiet al.
Space. Sci., 37, 5, (1989). Vainshtein, D.L., L.M. Zelenyi, and A.I. Neishtadt, Quasi-Adaiabatic Description of the Motion of Charged Particles in Configurations with a Reversed Magnetic Field, Plasma Physics Reports, 21,457, (1995). Zelenyi. L.M., A.A. Galeev, and C.F.Kermel, Ion Precipitation from the Inner Plasma Sheet due to Stochastic Diffusion, J. Geophys. Res., 95, 3871, (1990a). Zelenyi, L.M., D.V. Zogin, and J.Buchner, Quasiadiabatic Dynamics of Charged Particle in the Tail of the Magnetosphere, Cosmic Res., 28, 369, (1990b).
GENERATION MECHANISM OF THE FIELD-ALIGNED CURRENT SYSTEM DEDUCED FROM A 3-D MHD SIMULATION OF THE SOLAR WIND-MAGNETOSPHERE-IONOSPHERE COUPLING
T. T a n a k a
Communications Research Laboratory, Koganei-shi, Tokyo 184, Japan ABSTRACT The d r i v i n g m e c h a n i s m o f t h e f i e l d - a l i g n e d c u r r e n t (FAC) s y s t e m is i n v e s t i g a t e d u s i n g a t h r e e - d i m e n s i o n a l (3-D) m a g n e t o h y d r o d y n a m i c (MHD) s i m u l a t i o n . The s i m u l a t i o n is f u l l y self-consistent in t h e s o l a r w i n d - m a g n e t o s p h e r e - i o n o s p h e r e (S-M-I) c o u p l i n g scheme. The r e s u l t o f t h e c a l c u l a t i o n r e p r o d u c e s t h e r e g i o n 1 a n d 2 c u r r e n t s y s t e m s in t h e polar region. Additionally, the result for the northward interplanetary magnetic f i e l d (IMF) e x h i b i t s t h e f o r m a t i o n o f t h e n o r t h w a r d I M F - a s s o c i a t e d (NBZ) c u r r e n t . The d i s t r i b u t i o n o f - J 9 E in t h e d a w n - d u s k m e r i d i a n p l a n e shows t h a t t h e p o s i t i o n o f t h e region 1 current d r i v e r o v e r l a p s with a n t i - s u n w a r d f l o w s in t h e m a g n e t o s p h e r e , s h o w i n g t h a t t h e r e g i o n 1 c u r r e n t and a n t i - s u n w a r d f l o w s in t h e m a g n e t o s p h e r e a r e g e n e r a t e d t h r o u g h t h e same p r o c e s s . M a g n e t i c f i e l d l i n e s p a s s i n g t h e o u t m o s t p a r t s o f t h e m a g n e t o s p h e r e c o n t a i n l a r g e momentum. T h e s e m a g n e t i c f i e l d l i n e s a r e e x t e n d i n g f r o m m e r g i n g r e g i o n s . Thus t h e c a u s e o f a n t i - s u n w a r d f l o w s a n d r e g i o n 1 c u r r e n t is t h o u g h t t o be t h e c a p t u r i n g o f t h e s o l a r wind momentum b y t h e m e r g i n g p r o c e s s and subsequent deceleration o f t h e c o n v e c t i o n flow. This s o l a r wind c a p t u r i n g p r o c e s s a l s o r e s u l t s in t h e f o r m a t i o n o f t h e c u s p a t t h e f o o t p o i n t o f t h e m e r g e d m a g n e t i c field lines. INTRODUCTION The b a s i c f i e l d - a l i g n e d c u r r e n t (FAC) s y s t e m s t h a t c o n n e c t t h e m a g n e t o s p h e r i c a n d t h e a u r o r a l i o n o s p h e r i c c u r r e n t s y s t e m h a v e b e e n p r e s e n t e d b y I i j i m a a n d P o t e m r a (1976), who r e f e r r e d t o t h e m as t h e r e g i o n 1 and r e g i o n 2 c u r r e n t s y s t e m s . D u r i n g t h e p e r i o d of northward interplanetary m a g n e t i c f i e l d (IYlF), an a d d i t i o n a l l a r g e - s c a l e s t a b l e FAC s y s t e m c a l l e d t h e NBZ c u r r e n t d e v e l o p s a t h i g h e r l a t i t u d e p o l e w a r d o f t h e r e g i o n 1 s y s t e m [ I i j i m a a n d S h i b a j i , 1987]. The g e n e r a t i o n o f FAC is s t r o n g l y r e l a t e d t o t h e f a c t t h a t t h e m a g n e t o s p h e r i c c o n v e c t i o n c a n n o t d e v e l o p i f i t is n o t f o l l o w e d by t h e i o n o s p h e r i c c o n v e c t i o n . The m a g n e t o s p h e r e and t h e i o n o s p h e r e must a d j u s t t h e i r m o t i o n in s u c h a way t h a t m a g n e t i c f l u x c a n b e i n t e r c h a n g e d w i t h o u t a c c u m u l a t i o n a t a l l h e i g h t . Not a c h i e v i n g t h i s a d j u s t m e n t , t h e m a g n e t o s p h e r i c p e r p e n d i c u l a r motion, c o n v e c t i o n , is t r a n s m i t t e d t o t h e i o n o s p h e r e by an A l f v e n wave. The p o l a r i z a t i o n c u r r e n t p e r p e n d i c u l a r t o t h e m a g n e t i c f i e l d which a p p e a r s in f r o n t o f t h i s A l f v e n wave a n d is c o n n e c t e d t o t h e f i e l d a l i g n e d c u r r e n t a t t h e c o n v e c t i o n s h e a r [Kan and Sun, 1985] s h r i n k s i n t o t h e i o n o s p h e r i c c u r r e n t t h r o u g h t h e wave r e f l e c t i o n p r o c e s s a n d a c t s t o a c c e l e r a t e t h e i o n o s p h e r i c c o n v e c t i o n by t h e J • B f o r c e . This f a c t means t h a t - J 9 E is n e g a t i v e in t h e i o n o s p h e r e c a u s i n g t h e d i s s i p a t i o n o f e l e c t r o m a g n e t i c e n e r g y . The FAC which f l o w s o u t o f a n d i n t o t h e p o l a r i o n o s p h e r e is t h u s c o n n e c t e d w i t h t h e i o n o p s h e r i c c u r r e n t on o n e end. In o r d e r t o m a i n t a i n a s t e a d y c u r r e n t s y s t e m , t h e o t h e r e n d o f t h e FAC m u s t b e c o n n e c t e d t o t h e m a g n e t o s p h e r i c p e r p e n d i c u l a r c u r r e n t in t h e dynamo r e g i o n t o f o r m a s t e a d y c u r r e n t l o o p . P o s i t i v e - J 9 E must e x i s t s o m e w h e r e in t h e c u r r e n t s y s t e m . In o r d e r t o c l a r i f y t h e m e c h a n i s m o f r e g i o n 1 c u r r e n t 133
134
T. Tanaka
generation,
therefore,
positive
-J
9E
distribution
must
be
investigated
in
the
magnetosphere. NUMERICAL MODEL
A numerical magnetohydrodyamic (MHD) simulation is adopted for the study of the present problem. The FAC and plasma convection plays a central role in the magnetosphere-ionosphere coupling, whereas the state of the energy source for these current system depends on the solar wind-magnetosphere interaction. Thus, a self consistent treatment of the solar wind-magnetosphere-ionosphere (S-M-I) coupling process is required for the investigation of FAC system. The calculation employs the finite volume (FV) t o t a l variation diminishing (TVD) scheme with an unstructured grid to obtain a sufficient resolution in the ionosphere [Tanaka, 1994]. The outer and inner boundaries for the simulation are at 60 Re and 3 Re. A uniform solar wind with its speed of 350 km/sec is assumed at the upstream boundary. Dependent variables are projected along the field line from the inner boundary to the ionosphere. In the ionosphere, Ohm's law is solved to match the divergence of the Pedersen and Hall currents with the FAC. For the ionospheric conductivity, a uniform value of 0.8 mho is selected.
Results
southward
IMFs (B~ - +
of
the
simulation
are
obtained
for
two
cases
of
northward
and
5 nT).
RESULTS OF CALCULATION
Figures 1 and 2 show magnetic field configurations seen from dusk side for northward and southward IMF cases, respectively. These figures show only the magnetic field lines that have at least one end on the Earth. The spheres at the center of the figures show the size of the Earth. The meaning of the color codes on field lines will be shown later. Sharp bends of newly merged field lines are seen in Figure 1 on the tailward side of the c u s p , whereas they are seen in Figure 2 in the dayside magnetopause. It is easily observable from Figures 1 and 2 that tail magnetic field is stretched more severely for the southward IMF. -J.
E distribution
Figures 3 and 4 show the color coded distributions o f - J 9E in the noon-midnight meridian (xz) plane and the equatorial (xy) plane for northward and southward IMF cases, respectively. Dashed lines in Figures 3 and 4 show positions of the bow shock and magnetopause. In Figures 3 and 4, red lines show the three-dimensional region 1 current loop configurations. The evening region 1 currents flow outward from the Earth to the magnetopause, and change direction toward high latitudes near the low-latitude boundary layer (LLBL), where plasma flows are sheared between the magnetosheath and the LLBL. Then the dayside region 1 currents f l o w up the magnetopause to high latitudes. In the noon-midnight meridian (xz) plane, the main part of the region 1 currents passes the tailward side of the cusp [Tanaka, 1995]. While the region 1 current is closed through the flank magnetosphere, the region 2 current is closed in the inner magnetosphere. The evening region 2 current s t a r t s from the inner edge of the plasma sheet, flows longitudinally along the equatorial plane, and then turns earthward in the ring current region where strong current drivers due to the azimuthal pressure gradients are distributed. These results coincide with former theories [Harel et al., 1981]. It was already shown by Tanaka [1995] that the FAC and convection patterns in the polar ionosphere corresponding to
Generation Mechanism of the Field-Aligned Current System
135
Fig.1. T h r e e - d i m e n s i o n a l c o n f i g u r a t i o n of m a g n e t i c f i e l d f o r t h e n o r t h w a r d IMF.
Fig.2. T h r e e - d i m e n s i o n a l c o n f i g u r a t i o n of m a g n e t i c f i e l d f o r t h e s o u t h w a r d IMF.
136
T. Tanaka
Fig. 3. T h r e e - d i m e n s i o n a l c o n f i g u r a t i o n s o f c u r r e n t l o o p s a n d d i s t r i b u t i o n s o f - J t h e n o o n - m i d n i g h t a n d e q u a t o r i a l p l a n e s , f o r t h e n o r t h w a r d IMF c a s e .
9 E in
Fig. 4. T h r e e - d i m e n s i o n a l c o n f i g u r a t i o n s o f c u r r e n t l o o p s a n d d i s t r i b u t i o n s o f - J 9 E in t h e n o o n - m i d n i g h t m e r i d i a n a n d e q u a t o r i a l p l a n e s , f o r t h e s o u t h w a r d IMF c a s e .
Generation Mechanism of the Field-Aligned Current System
137
F i g u r e s 3 a n d 4 show a g o o d a g r e e m e n t with o b s e r v a t i o n s . In F i g u r e 3, t h e NBZ c u r r e n t f l o w i n g i n t o t h e p o l a r c a p i o n o s p h e r e is s e e n t o be m a p p e d t o t h e t a i l r e g i o n . The e v e n i n g NBZ c u r r e n t f l o w s in t h e t a i l f r o m dawn t o d u s k , a n d n e a r t h e d u s k m a g n e t o p a u s e i t t u r n s t o w a r d h i g h l a t i t u d e . F l o w i n g up a l o n g t h e m a g n e t o p a u s e , t h e e v e n i n g NBZ c u r r e n t g r a d u a l l y t u r n s d i r e c t i o n t o w a r d h i g h l a t i t u d e a n d c h a n g e s t o t h e e a r t h w a r d FAC. In F i g u r e 3, a w e a k c u r r e n t d r i v e r f o r t h e NBZ c u r r e n t is s e e n in t h e l o b e r e g i o n n e i g h b o u r i n g t h e p l a s m a s h e e t b o u n d a r y . In t h i s r e g i o n t h e NBZ c u r r e n t f l o w s f r o m dawn t o d u s k . Then t h e e l e c t r i c f i e l d must be f r o m d u s k t o dawn in o r d e r t o g i v e a dynamo a c t i o n . Thus p e n e t r a t i o n o f d u s k t o dawn e l e c t r i c f i e l d a s s o c i a t e d w i t h t h e n o r t h w a r d IMF is t h e main c a u s e o f t h e NBZ c u r r e n t . In t h e p l a s m a s h e e t , -J. E is n e g a t i v e b o t h f o r n o r t h w a r d a n d s o u t h w a r d IYIF c o n d i t i o n s b e c a u s e m a g n e t i c t e n s i o n a c t s t o e n e r g i z e p l a s m a t h e r e . This r e s u l t means that electromagnetic e n e r g y is d i s s i p a t e d in t h e p l a s m a s h e e t t h r o u g h t h e d i r e c t energization o f p l a s m a . Thus, t h e p l a s m a s h e e t p a r t of tail 0 current acts energetically in a s i m i l a r r o l e t o t h e i o n o s p h e r i c p a r t o f t h e r e g i o n 1 c u r r e n t l o o p . These c o n f i g u r a t i o n s s u g g e s t t h a t the r e g i o n 1 and t a i l 8 c u r r e n t systems a r e of a similar origin. Comparing the distributions of-J 9 E in F i g u r e s 3 a n d 4, t h e most remarkable differences are observed in t h e c u s p a n d HLBL r e g i o n s . On t h e high-latitude s i d e o f t h e c u s p , - J . E is n e g a t i v e f o r t h e n o r t h w a r d IYIF, w h i l e i t is p o s i t i v e f o r t h e s o u t h w a r d IYIF. On t h e l o w - l a t i t u d e s i d e o f t h e c u s p , t h e r e v e r s e t e n d e n c y is o b s e r v e d . In t h e n o r t h w a r d IYIF c a s e , n e g a t i v e - J . E r e g i o n on t h e h i g h - l a t i t u d e s i d e o f t h e c u s p is c o n n e c t e d t o t h e n e g a t i v e - J 9 E r e g i o n in t h e HLBL. On t h e o t h e r h a n d , t h e n e g a t i v e - J 9 E r e g i o n which e x i s t s on t h e l o w - l a t i t u d e s i d e o f t h e c u s p in t h e s o u t h w a r d IYIF c a s e is c o n n e c t e d t o t h e d a y s i d e m a g n e t o p a u s e . T h e s e d i f f e r e n c e s in t h e e n e r g y s t a t u s a r e c l o s e l y r e l a t e d w i t h t h e m e c h a n i s m o f r e g i o n 1 c u r r e n t g e n e r a t i o n , as w i l l b e shown in t h e f o l l o w i n g p a r t o f t h i s p a p e r . Region 1 - northward
IMF c a s e
The u p p e r p a n e l o f F i g u r e 5 shows t h e d i s t r i b u t i o n o f t h e p l a s m a f l o w on t h e d a w n - d u s k m e r i d i a n p l a n e f o r t h e n o r t h w a r d IMF. In t h i s p a n e l , t h e p r o j e c t i o n o f t h e velocity component perpendicular to the magnetic field, the convection flow c o m p o n e n t , is shown b y a r r o w s , a n d a n t i - s u n w a r d v e l o c i t y (-Vx) is shown b y c o l o r . N o t e t h a t t h e r a d i a l c o o r d i n a t e is p l o t t e d n o t l i n e a l l y b u t d e p e n d i n g on t h e o r d e r o f t h e s i m u l a t i o n g r i d . From t h i s p a n e l o n e c a n o b s e r v e t h a t a r r o w s a r e v e r y s h o r t in t h e undisturbed solar wind o u t s i d e t h e bow s h o c k , while the convection in t h e m a g n e t o s h e a t h is d i r e c t e d t o w a r d t h e e q u a t o r . In b o t h r e g i o n s , t h e c o l o r o f a r r o w s ( w h i t e c o l o r ) shows l a r g e a n t i - s u n w a r d v e l o c i t y . The e q u a t o r w a r d c o n v e c t i o n in t h e magnetosheath is c a u s e d b y t h e t e n s i o n f o r c e d u e t o d r a p e d IYIF a r o u n d t h e m a g n e t o s p h e r e . In t h e m a g n e t o s p h e r e , on t h e o t h e r h a n d , t h e p l a s m a f l o w is r e l a t i v e l y s m a l l a n d t h e s u n w a r d f l o w d o m i n a t e s in two r e g i o n s , o n e in t h e r e g i o n a b o v e t h e p o l e a n d t h e o t h e r in t h e r i n g c u r r e n t r e g i o n . The f o r m e r is t o w a r d t h e h i g h - l a t i t u d e m e r g i n g r e g i o n , a n d t h e l a t t e r is t h e r e t u r n f l o w e x t e n d i n g f r o m t h e p l a s m a s h e e t . M a g n e t o s p h e r i c a n t i - s u n w a r d f l o w is s e e n b e t w e e n t h e s e two s u n w a r d f l o w s . T h e s e structures a r e p r o j e c t e d o n t o t h e p o l a r i o n o s p h e r e as t h e 4 - c e l l c o n v e c t i o n p a t t e r n . N e a r t h e m a g n e t o p a u s e in t h e a n t i - s u n w a r d f l o w r e g i o n , a r r o w s e x h i b i t s a c o n v e c t i o n s h e a r b e t w e e n t h e m a g n e t o s h e a t h and t h e m a g n e t o s p h e r e , i n d i c a t i n g t h a t a n t i - s u n w a r d f l o w s in t h e m a g n e t o s p h e r e a r e n o t c a u s e d b y t h e v i s c o u s - l i k e p r o c e s s [ F e d d e r a n d L y o n , 1995].
138
T. Tanaka
Fig. 5. D a w n - d u s k d i s t r i b u t i o n s
o f f l o w a n d - J - E f o r t h e n o r t h w a r d IMF.
The l o w e r p a n e l o f F i g u r e 5 r e p r o d u c e s t h e u p p e r p a n e l o f F i g u r e 5 c h a n g i n g t h e d a t a shown by c o l o r f r o m a n t i - s u n w a r d v e l o c i t y t o - J 9 E. In t h i s p a n e l , p o s i t i v e v a l u e f o r - J 9 E which i n d i c a t e s dynamo a c t i o n is s e e n on t h e bow s h o c k a n d in t h e m a g n e t o s p h e r i c a n t i - s u n w a r d flow r e g i o n n e a r t h e m a g n e t o p a u s e . On t h e bow s h o c k , dynamo a c t i o n is p r o v i d e d b e c a u s e s o l a r wind f l o w is s e v e r e l y b r a k e d t h e r e a n d e n e r g y is c o n v e r t e d f r o m t h e p l a s m a e n e r g y t o e l e c t r o m a g n e t i c e n e r g y . In t h e m a g n e t o s p h e r e , t h e d i s t r i b u t i o n s of p o s i t i v e - J - E o v e r l a p s with a n t i - s u n w a r d flows i n d i c a t i n g t h a t t h e b r a k i n g of a n t i - s u n w a r d flow p r o v i d e s a r e g i o n 1 c u r r e n t d r i v e r . In F i g u r e 1 t h e c o l o r c o d e s on m a g n e t i c f i e l d l i n e s t h a t h a v e a t l e a s t o n e e n d on t h e E a r t h c o r r e s p o n d t o m a g n i t u d e s o f momentum on m a g n e t i c f i e l d l i n e s . Where t h e o r d e r o f c o l o r s is same t o t h o s e u s e d In F i g u r e s 3-6. I t is s e e n f r o m t h i s f i g u r e t h a t magnetic field lines passlng the outmost part of the magnetosphere contain large momentum. T h e s e m a g n e t i c f i e l d l i n e s a r e e x t e n d i n g from t h e n e i g h b o u r o f t h e m e r g i n g r e g i o n a t h i g h l a t i t u d e and a r e c o n n e c t e d t o t h e a n t i - s u n w a r d f l o w r e g i o n in F i g u r e 3.
Generation Mechanism of the Field-Aligned Current System
Fig. 6. D a w n - d u s k d i s t r i b u t i o n s
139
o f flow and - J - E f o r t h e s o u t h w a r d IMF.
T h e r e f o r e , t h e LLBL p a r t o f t h e s e m a g n e t i c f i e l d l i n e s is t h o u g h t t o be c o n s t r u c t e d b y m e r g e d IMF. The f o r m a t i o n mechanism o f t h e LLBL u n d e r t h e n o r t h w a r d IMF c o n d i t i o n c a n be u n d e r s t o o d v e r y n a t u r a l l y f r o m t h i s c o n f i g u r a t i o n . This mechanism c a n w e l l explain the observational r e s u l t t h a t t h e LLBL grows in w i d t h f o r n o r t h w a r d IMF [ M i t c h e l l e t a l . , 1987]. From F i g u r e s 5 a n d 1, i t can be e s t i m a t e d t h a t t h e p r i m a r y s o u r c e o f a n t i - s u n w a r d f l o w and r e g i o n 1 c u r r e n t is t h e c a p t u r i n g o f t h e s o l a r wind momentum t h r o u g h t h e m e r g i n g p r o c e s s . This p r o c e s s , a t t h e same time, r e s u l t s in t h e f o r m a t i o n o f t h e c u s p in t h e l o w - a l t i t u d e r e g i o n . A f t e r s h o r t e n i n g a n d s i n k i n g i n t o t h e LLBL, momentum o n t h e m e r g e d f i e l d l i n e s is l o s t in t h e c o u r s e o f a n t i - s u n w a r d c o n v e c t i o n . L o s i n g e n e r g y , t h i s c o n v e c t i o n flow g e n e r a t e s e l e c t r o m a g n e t i c e n e r g y b y a polarization current and p r o v i d e s a dynamo a c t i o n . Thus, n o t o n l y t h e m a g n e t o s p h e r i c c o n v e c t i o n [ F e d d e r a n d Lyon, 1995] b u t a l s o t h e r e g i o n 1 c u r r e n t c a n b e d r i v e n m e c h a n i c a l l y [ J o h n s o n , 1978]. F i e l d l i n e s t h a t h a v e j u s t m e r g e d in F i g u r e 1 a r e d i s t o r t e d
severely. These field lines
140
T. Tanaka
e x e r t t e n s i o n f o r c e t o p l a s m a a n d s e r v e as a m o t o r a c t i o n b e f o r e s i n k i n g i n t o t h e LLBL. T h u s , - J 9 E b e c o m e s n e g a t i v e on t h e h i g h - l a t i t u d e side of the cusp and the HLBL. I t is a p p a r e n t from F i g u r e 3 t h a t t h e p r e s s u r e g r a d i e n t f o r c e is n o t as i m p o r t a n t as t h e r e g i o n 1 c u r r e n t d r i v e r . On t h e t a i l w a r d s i d e o f t h e c u s p , t h e pressure gradient is v e r y s t r o n g f r o m t h e t a i l l o b e t o t h e m a g n e t o s h e a t h in t h e n o r t h w a r d IMF c a s e [ T a n a k a , 1995]. H o w e v e r , F i g u r e 3 shows t h a t no c u r r e n t d r i v e r is distributed in t h i s r e g i o n . M o r e o v e r h i g h p r e s s u r e in t h e c u s p d o e s n o t s e r v e a c u r r e n t d r i v e r in t h e n o r t h w a r d INF c a s e . Region 1 - southward
INF c a s e
Results corresponding t o F i g u r e s 5 a n d 1 f o r t h e s o u t h w a r d INF c a s e a r e s h o w n in F i g u r e s 6 a n d 2. S i m i l a r t o t h e n o r t h w a r d INF c a s e , t h e e q u a t o r w a r d c o n v e c t i o n in t h e magnetosheath and a convection shear between the magnetosheath and the magnetosphere are observed in t h e u p p e r panel of Figure 6. In t h i s p a n e l , magnetospheric anti-sunward flow prevails over the entire polar cap. This configuration of the magnetospheric convection corresponds to t h e two c e l l c o n v e c t i o n in t h e p o l a r cap. I t is s e e n f r o m F i g u r e 6 t h a t d i s t r i b u t i o n o f p o s i t i v e - J 9 E in t h e m a g n e t o s p h e r e o v e r l a p s w i t h a n t i - s u n w a r d flow. Thus, t h e p r i m a r y s o u r c e o f t h e r e g i o n 1 c u r r e n t f o r t h e s o u t h w a r d IY[F c a s e c a n a l s o be t h e c a p t u r i n g o f t h e s o l a r wind momentum t h r o u g h t h e m e r g i n g p r o c e s s . In t h e s o u t h w a r d INF c a s e , s h a r p b e n d s o f n e w l y m e r g e d f i e l d l i n e s w h i c h e x e r t a m o t o r a c t i o n t o p l a s m a a p p e a r in t h e d a y s i d e r e g i o n . From t h e s e f i e l d l i n e s , we c a n w e l l u n d e r s t a n d why n e g a t i v e - J 9 E in F i g u r e 4 is d i s t r i b u t e d on t h e d a y s i d e m a g n e t o p a u s e and the low-latitude side of t h e cusp. Merged f i e l d l i n e s a r e t h e n a c c u m u l a t e d i n t o t h e m a n t l e in t h e c o u r s e o f a n t i - s u n w a r d c o n v e c t i o n . T h e s e f i e l d l i n e s c o n t a i n l a r g e s o l a r wind momentum a n d c a n p r o v i d e a d y n a m o a c t i o n , t h r o u g h t h e f l o w d e c e l e r a t i o n . These processes a r e w e l l r e f l e c t e d in t h e d i s t r i b u t i o n of-J. E. The a b o v e p i c t u r e c o i n c i d e s w i t h t h e o b s e r v a t i o n a l r e s u l t t h a t t h e m a n t l e g r o w s in w i d t h f o r s o u t h w a r d INF [ S c k o p k e e t a l . , 1976]. DISCUSSION AND CONCLUSION In t h e c u r r e n t l o o p which c o n n e c t s t h e m a g n e t o s p h e r e and the ionosphere, the i o n o s p h e r i c p a r t s e r v e s as an e n e r g y s i n k . The m a g n e t o s p h e r i c c o u n t e r p a r t o f t h e current c l o s u r e must i n c l u d e a dynamo a c t i o n which c o n v e r t s k i n e t i c or i n t e r n a l energy to electromagnetic e n e r g y . Without a dynamo a c t i o n , t h e c o n v e c t i o n will s t o p s h o r t . A p p a r e n t l y t h i s k i n d o f e n e r g y c o n v e r s i o n r e q u i r e s a c o n d i t i o n - J - E > 0. With t h i s c o n d i t i o n , p l a s m a b u l k motion o r p r e s s u r e g r a d i e n t p r o v i d e a dynamo a c t i o n . It m u s t b e n o t e d h e r e t h a t a c o n d i t i o n - J 9 E > 0 i n d i c a t e s t h a t FAC c a n n o t be g e n e r a t e d w i t h o u t t h e c o n v e c t i o n e l e c t r i c f i e l d . In g e n e r a l , t h e r e l a t i o n - J . E r 0 means t h a t energy conversions take place between electromagnetic energy, kinetic energy, and i n t e r n a l e n e r g y . F o r e x a m p l e , f i e l d s t r e t c h i n g d u e t o d e c e l e r a t i n g p l a s m a f l o w means a dynamo a c t i o n which c o n v e r t s k i n e t i c e n e r g y to e l e c t r o m a g n e t i c energy, and plasma acceleration due to magnetic tension force indicates a motor action which converts electromagnetic e n e r g y t o k i n e t i c e n e r g y . A l l o f t h e e n e r g y c o n v e r s i o n s in t h e magnetosphere are promoted under the control of the magnetospheric convection. In t h e p l a s m a s h e e t , p l a s m a a n d m a g n e t i c f i e l d s a t i s f y t h e e q u a t i o n J x B = V P in t h e 1 - s t o r d e r a p p r o x i m a t i o n . Where, t h e r a d i a l c o m p o n e n t o f p l a s m a p r e s s u r e g r a d i e n t
Generation Mechanism of the Field-Aligned Current System
141
prevents the collapse of tail-like flux against magnetic tension force. Although it is a n a t u r a l expectation that plasma flow directs from high-pressure to low-pressure regions, the sunward convection flow in the plasma sheet is occurring against V P force. The magnetic tension is the driving force for this flow. Therefore, - J . E < 0 in this region (see figures 3 and 4). As a consequence, inner edge of the plasma sheet contains high-pressure plasma. T h i s relation means that energy conversion from electromagnetic energy to internal energy is taking place there. Thus the plasma sheet is acting as an electromagnetic pump which generates high-pressure region in the inner edge of the plasma sheet. This high-pressure region serves as the region 2 c u r r e n t driver. In the course of convection, the sunward flow from the plasma sheet diverts to both sides of the Earth to the dayside. In this region, VP force drives flow against the line tying effect of the auroral oval. Thus, the region 2 FAC is considered to be generated from azimuthal pressure gradient formed in the inner magnetosphere through plasma redistribution in the course of convection. In this configuration, flow due to ~ P force generates region 2 current, converting internal energy to electromagnetic energy. This picture coincides with a well known theory presented previously [Harel et al., 1981], and can be confirmed from Figures 3 and 4. While the region 2 current is closed in the inner edge of the plasma sheet, the region 1 c u r r e n t is closed through the boundary layer and the cusp. Contrary to the case of the region 2 current, -J 9E distribution concerning the generation of the region 1 c u r r e n t s t r o n g l y depends on the IMF orientation. On the other hand, pressure distribution in the boundary layer and the cusp weakly depends on the IMF orientation [Tanaka, 1995]. From this fact, it seems difficult to consider here the FAC generation due to the pressure driven convection. Alternately, we propose the FAC generation due to the conversion of kinetic energy to electromagnetic energy. This picture coincides with the previous expectation that the region 1 FAC is generated in the magnetosphere from inertia current due to the braking of plasma flow in the boundary layer [Stern, 1983; Tanaka, 1995]. Figures 5 and 6 give a f u r t h e r confirmation for this mechanism. Thus, we can reach the conclusion that the merging process is the main c o n t r o l l i n g mechanism for both the magnetospheric convection and dynamo action of the region 1 c u r r e n t in both the northward and southward IMF cases. It is concluded in this paper t h a t not only the magnetospheric convection but also the region 1 current is driven mechanically. REFERENCES
Feddar, J.A., and J.G. Lyon, The Earth's magnetosphere is 165 RE long: Self-consistent currents, convection, magnetospheric structure, and processes for northward i n t e r p l a n e t a r y magnetic field, J. Geophys. Res., 100, 3623 (1995). Harel, M., R. A. Wolf, P. H. Reiff, R. W. Spiro, W. J. Burke, F. J. Rich, and M. Smiddy, Quantitative simulation of a magnetospheric substorm, 1, Model logic and overview, J. Geophys. Res., 86, 2217 (1981). lijima, T. and T. A. Potemra, The amplitude distribution of field-aligned currents at northern high latitudes observed by Triad, J. Geophys. Res., 81, 2165 (1976). lijima, T. and T. Shibaji, Global characteristics of northward IMF-associated (NBZ) field-aligned currents, J. Geophys. Res., 92, 2408 (1987). Johnson, F.J., The driving force for magnetospheric convection, Rev. Geophys. Space Phys., 16, 16 (1978). Kan, J.R., and W. Sun, Simulation of the westward traveling surge and Pi2 pulsations during substorms, J. Geophys. Res., 90, 10911 (1985).
142
T. Tanaka
Mitchell, D.G., F. Kutchko, D.J. Williams, T.E. Eastman, L.A. Frank, and C.T. Russell, An extended study of the low-latitude boundary layer on dawn and dusk flanks of the magnetosphere, J. Geophys. Res., 92, 7394 (1987). Sckopke, N., G. Paschmann, H. Rosenbauer, and D.H. Fairfield, Influence of the interplanetary magnetic field on the occurrence and thickness of the plasma mantle, J. Geophys. Res., 81, 2687 (1976). Stern, D. P., The origins of Birkeland currents, Rev. Geophys., 21, 125, 1983. Tanaka, T., Generation mechanisms for magnetosphere-ionosphere current systems deduced from a three-dimensional MHD simulation of the solar wind-magnetosphere-ionosphere coupling processes, J. Geophys. Res., 100, 12057 (1995). Tanaka, T., Finite volume TVD scheme on an unstructured grid system for three dimensional MHD simulation of inhomogeneous systems including strong background magnetic fields, J. Comput. Phys., 111, 381 (1994).
CONFIGURATION INSTABILITY OF THE N E A R - E A R T H MAGNETOTAIL IN THE PRESENCE OF AN E A R T H W A R D PLASMA FLOW AND SUBSTORM ONSET M.H.Hongl
Z.y.pu 2
X.M.Wang3
Z.X.Chen 2
Z.X.Liu 4
A.Korth 5
R.H.W.Friedel 5
1 Institute of Geophysics, Academia Sinica, Beijing 100101, China 2 Department of Geophysics, Peking University, Beijing 100871, China 3 Institute of Space Medico-Engineering, Beijing 100094, China 4 Center of Space Science & Applied Research, Academia Sinica, Beijing 100080, China 5 Max-Planck-Institutftlr Aeronomie, Katlenburg-lindau, Germany
ABSTRACT The one-fluid MHD approach is used to investigate configuration instability in the near-Earth magnetotail in the presence of plasma flows prior to onset of the substorm expansion phase. The results show that drift ballooning modes become absolutely unstable and grow more rapidly in the presence of an earthward plasma flow. On the basis of these results and the configuration instability model of substorm expansion onsets, we suggest that magnetic reconnection in the midtail and resulting plasma flows may lead to rapid development of dipolarizaton of the magnetic field in the nearEarth tail (Pu et al., 1996, 1997), causing intense substorm expansion onsets. This model integrates the NECD and NENL models, giving a physical picture for the initiation of intense substorms. Comparison of the theory with GEOS 2 measurements indicates that a number of substorm observations are consistent with this substorm expansion scenario.
INTRODUCTION
The major task in the substorm research is to understand the mechanism and effects of magnetospheric substorms. The sudden initiation of the substorm expansion phase indicates that "there must exist an instability as the direct cause of a substorm.. Expansion waves, current disruption and other dynamic phenomena may well be the nonlinear effect (but not the cause) of the growth of an unstable mode"(Siscoe, 1993). So far, a great number of substorm expansion models have been proposed in which the near-Earth neutral line model (NENL) and the near-Earth current disruption model (NECD) are the most popular ones. The NENL model emphasizes that it is magnetic reconnection in the magnetotail that is the primary mechanism responsible for the release of energy stored in the magnetotail in the growth phase (Baker et al., 1996). Recent GEOTAIL measurements show that during substorm periods magnetic reconnection does seem to occur in the midtail and distant tail (Mukai et al., 1996). However, observations fail to find evidence that the NENL might reside typically inside 20 R E. On the other hand, in the NECD model the expansion phase starts when the cross-tail currents are disrupted and diverted into the ionosphere, which in many cases, begins in regions near and earthward of 8 R E (Lopez and Lui, 1990). Nevertheless, at the present stage, "current disruption" seems to be an idea seeking a physical mechanism (Siscoe, 1993). Recently, with a careful theoretical study and comparison of the theory with GEOS 2 observations, Pu et al. (1996, 1997) have shown that the drift ballooning mode instability (DBM) in the transition region 143
144
M.H. Hong et al.
between the taillike and dipolar-like configuration may play an important role in leading to current disruption and dipolarization of magnetic field at the substorm onset. There is no doubt that magnetospheric substorms should be regarded as global processes involving various dynamical phenomena in the near-Earth tail, midtail and distant tail, and energy coupling between the magnetosphere, the ionosphere and the thermosphere. While studying substorm expansion mechnisms, one must take this global aspect of substorms into consideration, and make one's model consistent with
observations in different regions in geospace. Pu
et al. (1996, 1997) found that the configuration instability (i.e., the DBM) in the near-Earth tail can only be generated
when the preonset particle precipitation makes the ionospheric conductance jump to a critical value, and that the waves excited by the instability may propagate tailward, earthward, and radially, These results are consistent with observations made by ISEE 1/2, GEOS 2 and AMPTE (Lopez and Lui, 1990; Ohtani et al., 1992; Jacquey et al., 1991, 1993). A lot of satellite data show that earthward high speed plasma flows often appear during substorms in regions of 10--20 R E in the magnetotail. These plasma flows are commonly thought to be associated with magnetic reconnection down tail of 20 R E and are often found to be correlated well to the dipolarization of background magnetic field. In order to understand the relationship between the occurrence of earthward flows and magnetic field dipolarization in the near-Earth tail, in this paper we use the one-fluid MHD approximation to extend the DBM instability model for substorm expansion onsets of Pu et al. (1996, 1997). The dispersion equation of the near-Earth DBM in the presence of a plasma flow is derived. The necessary conditions for the instability are discussed. It is found that the earthward plasma flows make both the DBM 1 and DBM2 absolutely unstable and grow much more rapidly than without flows. On the basis of these results we suggest that magnetic reconnection in the midtail and resulting earthward flows may lead to rapid development of dipolarization of magnetic field in the near-Earth tail, causing intense substorm expansion onsets. Comparison of the theory with GEOS 2 measurements shows that a number of substorm observations are consistent with this substorm expansion scenario. CONFIGURATION INSTABILITY IN THE PRESENCE OF A PLASMA FLOW Dispersion Equation It has long been noticed that the onset region of the substorm expansion phase is located near the magnetic equator in the nightside magnetosphere where a strong earthward pressure gradient appears and the magnetic field lines get stretched out toward the tail during the growth phase (Korth et al., 1991; Pu et al., 1992). In this area the ion gyroradius r i ~ 200 km, the ion gyroperiod fi "-~ 1 s, while the scale lengths of the magnetic field and the pressure gradient ~ 0.5-1R E , and the time scale of substorm dipolarization and low-frequency waves--~ 1 min. Therefore the one-fluid MHD approximation can be roughly used to investigate the dynamical processes there. In the present study a two-dimensional coordinate system of (b,n,y) is adopted where b is the unit vector parallel to the unperturbed magnetic field B, n lies in the direction of the field line curvature, and y = bxn directs westward. At the equator, b is northward, and n is sunward. For simplicity we assume that in the initial state the plasma pressure PO is isotropic, the plasma flow is earthward with velocity Vo_l_B0 and parallel to n , and all background quantities are uniform in the y direction (O/3y= 0 ) and are symmetric or antisymmetric with respect to the equator. Note that in the near-Earth tail V0 ~ 102 km/s and that the Alfven speed VA ~ 103 km/s, hence VO/VA = e 0 ( < 0). No matter whether the perturbed cross-tail current is westward or eastward, the perturbed electric field 6Ey is always eastward (westward), hence
the perturbations will grow further,
(see Figure3). When the background flow exists,magnetic disturbances bring electric field 6Ey t
___
about an Von•
additional disturbed o 6B//y. It has
been known that for the DBM the phase of disturbed pressure 6 P is opposite to that of magnetic field6B//. (Miura et al.,, 1989; Ohtani et al., 1989). Therefore in
x.._.__~~ , a~'
the region at the magnetic equator where the isobaric surface is convex tailward (earthward), one has 6B//O), hence 6 E'y is in
the
same
BZ
is eastward (westward) which
direction
as
aE', laE, . ' ~ .
that of 6 E y . The t
.,-'9 VB
.,,--vp,
appearance of 6 EythUs makes the DBM develop more r
rapidly. This is the
reason
as X B B2
why the earthward flows
destabilize and reinforce the near-Earth instability (see also Figure 3).
configuration Fig.3 DBM destabilizing process in the presence of earthward flows
PLASMA FLOWS FROM THE MIDTAIL AND ONSETS OF THE EXPANSION PHASE IN THE NEAR-EARTH PLASMA SHEET We have shown in the previous section that the occurrence of earthward plasma flows makes the near-Earth plasma sheet much more unstable than without the plasma flows, causing the near-Earth configuration instability to dramatically develop. High speed flows in the near-Earth central plasma sheet have often been reported in the literature in recent years (Baumjohann et al., 1989; Angelopoulos et al., 1992, 1994). Baumjohann et al. found from IRM data that the flow bursts are directed predominantly earthward in the near-Earth magnetotail (]xJ-
," ~.4
.... , ..-
"L--
;:-:
- ..........................
: :1..s . . . . .
-=1..2",
. _.-
...........
-
":
-"
:
..................... . . . . . . . .
--
t#z ~.
0
z
O O z
-5o
~ - - ~ = -
~
4
.......
_
:::-
"-(
-100 -150 DAWN
-200 . 0
. 300
.
. 600
. 900
. 1200
. 1500
. 1800
.
. 2100
2400
2700
3000
X (kin)
Fig. 3. Asymptotic field-aligned current distribution showing the occurrence of multiple polar cap arcs due to the effect of the background ionospheric conductance.
162
L. Zhu et al.
Figure 2 shows how the initial magnetospheric shear flow is modified by the M-I coupling, leading to the multiple structures of the arcs. The top panel in Figure 2 shows the dawn-dusk (Y direction) profile of the field-aligned current distribution associated with the initial shear flow carried by a downward propagating Alfv4n wave. This field-aligned current distribution has an upward field-aligned current in the center and downward field-aligned currents at the two edges. If the ionosphere is just a passive load, a single arc should be expected in the ionosphere due to such a magnetospheric driver with a single precipitation channel. The bottom panel in Figure 2 shows the dawn-dusk profile of the field-aligned currents associated with the polar cap arcs at an asymptotic state. The appearance of multiple structures, with a morphology significantly different from the initial magnetospheric driver, indicates that an M-I coupling process has been involved in which the ionosphere is not just a passive load. The next question is what are the key parameters controlling the occurrence of multiple polar cap arcs in the M-I system 9 Simulations using the M-I coupling model show that the occurrence of multiple polar cap arcs depends strongly on the magnitude of the large-scale ionospheric background convection (not shown). With the same initial magnetospheric driver, the number of the arcs increases with enhanced ionospheric background convection. The second key parameter controlling the occurrence of multiple arc structures is the background ionospheric conductivity. Figure 3 shows the asymptotic field-aligned current distribution for the case in which the ionospheric background conductance decreases sharply from 2.5 mho on the dayside to 0.5 mho on the nightside. It can be seen that in the regions where the ionospheric background conductance is above 2 mho, only a single bright polar cap arc exists. This can be explained by the fact that the high ionospheric conductance allows the magnetospheric current to close freely in the ionosphere, which acts to smooth localized discrete structures. In the regions where the ionospheric conductance ranges from 0.5 to 1.5 mho, there are multiple polar cap arcs. These results further indicate that the ionosphere plays an active role in the formation of multiple polar cap arcs and that it dynamically responds to the magnetospheric driving, which is significantly different from the situation of simple mapping from the magnetosphere to the ionosphere.
HORIZONTAL CURRtSNT. NOON-MIDNIGHT COMPONENT (mA,q')t)
HORIZONTAL CURRENT. NOON-MIDNIGHT COMPONENT (mA/M}
200
200 _.
150
150
.. -~.0
. . . . . . . .
.
.~176.....
lO0
. . . . . . . . . . . . . . . . . . . . . . .
_ . . . . . . .
~ ......................
-.8.0"
- ....
_ .....
_ ............. ~ .
9
~
100-
_-
.
~
.-~0.
_
r ~.~ >-
o
,.,
~
50-
5o - ......................... ..................................................
-__'_-_--_
--------_--:
0-
d,..08.,.0.
c.
-50
-50"
-100
........
-100
a2.9..
-150
-150
T = 8 rain
T = 0 rain -200
-200
300
600
900
1200
1500
x (kin)
1800
2100
2400
2700
3000
0
xo~)
Fig. 4. 2-D distributions of the horizontal currents in the noon-midnight direction at T = 0 min (left) and at the asymptotic state (T=8 min, right).
Model Study of Polar Cap Arcs
163
Another important feature of polar cap arcs found in the theoretical modeling is the existence of a horizontal electrojet associated with the arcs. Figure 4 shows 2-D distributions of the horizontal currents in the noonmidnight direction at T = 0 min (left panel) and at an asymptotic state (T=8 min, right panel) for the case shown in Figure 1. The solid lines in Figure 4 denote the currents flowing toward the noon. It can be seen that in the left panel of Figure 4 there is no net current flowing in the noon-midnight direction since the sunward and antisunward currents have a symmetric distribution. By comparing the current loaded by the initial magnetospheric shear flow (left panel) with the current reflecting the M-I coupling at the asymptotic steady state (right panel), it can be seen that a strong net horizontal current flowing along the arcs has developed. The intensity of the current at the asymptotic state is almost one order stronger than the loaded current from the initial magnetospheric shear flow. This indicate that the strong electrojet along the polar cap arcs is not externally imposed from the magnetosphere, instead, it is self-consistently generated in the M-I coupling processes associated with the polar cap arcs. 3. SUMMARY Model studies of M-I coupling effects on the polar cap arc formation have been conducted. It is found that the appearance of multiple arc structures and strong electrojet flowing along the arcs is due to the M-I coupling processes associated with polar cap arcs and is not externally imposed by the magnetosphere. The model results indicate that a strong ionospheric background convection and a low ionospheric background conductivity favor the appearance of multiple polar cap arcs. ACKNOWLEDGMENTS This research was supported by NASA grant NAG5-1484, NSF grants ATM-96-12638 and ATM-96-12835, and ONR grant N00014-95-1-0652 to Utah State University. REFERENCES Burke, W. J., M. S. Gussenhoven, M. C. Kelley, D. A. Hardy, and F. J. Rich, Electric and magnetic characteristics of discrete arcs in the polar cap, J. Geophys. Res., 87, 2431 (1982). Chiu, Y. T., Formation of polar cap arcs, Geophys. Res. Left., 16, 743 (1989). Davis, T. N., The morphology of the polar aurora, J. Geophys. Res., 48, 4447 (1960). Frank, L. A., J. D. Craven, D. A. Gurnett, S. D. Shawhan, D. R. Weimer, et al., The theta aurora, J. Geophys. Res., 91, 3177 (1986). Weber, E. J., and J. Buchau, Polar cap F-layer auroras, Geophys. Res. Left., 8, 125 (1981). Zhu., L., J. J. Sojka, R. W. Schunk, and D. J. Crain, A time-dependent model of polar cap arcs, J. Geophys. Res., 98, 6139 (1993).
This Page Intentionally Left Blank
PRECIPITATION OF HOT PROTONS FROM A STRETCHED NEAREARTH CURRENT SHEET W. W. Liu, G. Rostoker and J. C. Samson
Department of Physics, University of Alberta, Edmonton, AB T6G 2J1 ABSTRACT Recent observational evidences indicate that the near-Earth tail current sheet at approximately ten earth radii often experiences substantial and sudden changes which some researchers consider to be related to the substorm expansive phase onset. Observations of the ionosphere have revealed that auroral substorm activity is often preceded by an enhancement of precipitation of >20 keV protons causing H 13 emissions. Protons in this energy range are most likely to be found in the near-Earth plasma sheet, giving support to the interpretation of a near-Earth source of substorm onset. However, the reason why proton precipitation intensifies during the auroral substorm has not yet been clarified. We propose that a possible explanation of this observation lies in the nonadiabatic dynamics of energetic protons in the stretched near-Earth current sheet formed in the late growth phase of the substorm. While pitch-angle scattering associated with such nonadiabatic dynamics has been considered before, there have been few studies which determine quantitatively how this process might contribute to proton precipitation. We have carried out numerical integrations of proton orbits in a magnetic field consisting of the Earth's main field and that due to a Harris current sheet. We find that, in the regime of highly nonadiabatic dynamics, the proton precipitation rate can exceed the already strong limit of isotropically filled loss-cone precipitation by an order of magnitude. This finding both contributes to our understanding of nonadiabatic effects on precipitation and fits nicely into the overall scheme of near-Earth onset theory. INTRODUCTION Precipitation of magnetospheric particles is responsible for auroral emissions at different wavelengths. The morphology of aurora in turn can be used as a diagnostic tool to probe magnetospheric processes. Photometric studies of aurora have been instrumental in understanding such problems as field-line resonances [Samson et al., 1992a; Xu et al., 1993; Liu et al., 1995] and the location of substorm onset [Samson et al., 1992b]. Samson et al. [1992b] reported cases where an auroral substorm was accompanied by significant intensification of auroral HI] emission (4861 A). These authors inferred that this behavior implies that the source region of the onset must be fairly close to Earth, in the inner plasma sheet where adiabatically heated protons of >10 keV are most abundant. How the protons manage to precipitate at an increased rate, however, was not specified. Here we relate the enhanced proton precipitation to the formation of a thin cross-tail current identified as the precursor to substorm eruption [Lui, 1996 and references therein]. Ohtani et al. [1992] inferred the formation of such a current sheet from in-situ magnetic field observations. Lui et al. [ 1992] gave characteristic parameters inside the structure. Obviously, in linking the thin current sheet with H [3 enhancement, we are interested in the current disruption model, and our results will have enhanced and expanded this concept by: 1) assimilating another observation into this paradigm and 2) providing a diagnostic tool to take advantage of ground-based observations in characterizing magnetospheric processes. The preonset current sheet is observed to have a structure in which ion motions are likely chaotic [Mitchell et al., 1990; Lui et al., 1992; Sergeev et al., 1993]. Many theoretical and computational works have been carried out in this respect [Chen, 1992, and references therein], but few were concerned with precipitation per se. Speiser [1965], in his pioneering investigation on ion orbits through a current sheet, used a localized field representation where the dipolar component of the Earth field was ignored; many other authors followed suit [Lee and Gray, 1982; Lyons and Speiser, 1982; Chen and Palmadesso, 1986; Buchner and Zelenyi, 1989]. Since the exit of ions from the current sheet is not equivalent to precipitation into the ionosphere (there is a large magnetic mirror force yet to overcome), it is clear that a global representation of magnetic field, with the dipolar component fully incorporated, is necessary. The preliminary results of computation reported here has theoretical significance in giving us a first quantitative look at the potential of nonadiabatic precipitation, namely how chaotic motion might hasten the entry of protons into the loss cone. On this objective we are reminded that protons slowly diffuse into the loss cone though interaction with ion165
W.W. Liu et al.
166
cyclotron waves [e.g., Kennel and Petschek, 1966]. Nonadiabatic precipitation is a process that is entirely unrelated to diffusion and occurs on a single particle basis. When proton motion is sufficiently chaotic, each encounter with the current sheet will on average produce a large pitch angle change. This situation, in theory, could lead to a faster entry in the loss cone. We are interested if nonadiabatic precipitation can dominate diffusive precipitation and, if so, under what conditions. In answering this question, we shall first derive a theoretical limit of strong pitch angle diffusion as a standard of comparison. Then we shall develop a computational model to determine numerically the nonadiabatic precipitation rate relative to the strong diffusion limit. We shall conclude with a discussion on the application of two-dimensional results in 3D cases, among other things. LIMIT OF STRONG PITCH-ANGLE DIFFUSION In a perfectly adiabatic plasma free of waves, the loss cone would be empty in a steady state. On a collective level, however, such a sharp void in phase space is unstable to electromagnetic waves [Kennel and Petschek, 1966; Krall and Trivelpiece, 1973]. Self-limitation considerations argue that the amplitude of scattering waves cannot be large because the free energy derivable from an unfilled loss cone is small. Therefore, each encounter with the wave will most likely change the pitch angle of proton by a small amount, and collectively the process can be described by a quasilinear pitch angle diffusion theory. One can imagine that there is an upper limit to the diffusive equilibrium. The limit occurs when the diffusion timescale 1/D,~ a is short compared to the timescale of ion loss through the loss cone. When this condition is fulfilled, the loss cone is nearly filled to isotropic limit ~f/3ct = 0. We call this situation the strong pitch-angle diffusion limit (SPADL). We can estimate the precipitation rate of a plasma in SPADL by considering a Maxwellian extending into the loss cone, mpV2 mp
2kT
(1)
f ( v ) = n ( 2 r t k T )e
where k. is Boltzmann's constant, m_F is .the proton mass, and T is the proton temperature. Assuming that there is no , field-aligned electnc field, the Maxwelhan f ( v ) is constant along a field line. For a unit flux tube (Figure 1), there is an downward unidirectional flux of precipitation oo
F =
j.
v,lf(v)
kr
d3v = n (2rtmp)
1/2
(2)
0
Since the flux tube has two ionospheric openings with area Bo I , where B o is the polar magnetic field, the rate of proton precipitation is then
n 2kT 1/2 R = ~o(~e)
(3)
The number of protons in the flux tube, on the other hand, is N=
Ids n V = n -~-
(4)
where V is the volume of a unit flux tube. We measure the strength of precipitation in terms of the average lifetime of protons,
N rtmp to = ~ = ( 2 ~ 1
1/2_
(B 0V)
(5/
The flux tube volume V is usually related to the equatorial distance r e and equatorial magnetic field B e as V = are/B e, where a is a number of order unity. The form of (5) suggests that the average proton lifetime is greater than the typical bounce time by a factor of Bol B e, which is usually large in the region of interest.
Precipitation of Hot Protons from a Stretched Near-Earth Current Sheet
167
Fig. 1. Geometry of proton precipitation in a dipole-like field. The loss cone is filled. Equation (5) establishes the standard against which we characterize a nonadiabatic precipitation as strong or weak; that is, nonadiabatic precipitation is called strong if the average lifetime of protons is short compared to (5). NONADIABATIC PRECIPITATION The basic tenet of our view concerning the H 13 intensification is that it is due partly to the thinning of the equatorial current sheet and the associated enhancement of proton precipitation over and above the SPADL value given in (5). Obviously, to test this thesis, a computational model must be developed to track a statistically significant and carefully initiated ensemble of protons. Further, as pointed out in the introduction, the model magnetic field must possess a thin equatorial current sheet (so that there is a strong source of scattering action) and asymptote to a dipolar field sufficiently close to Earth so that the precipitation will not be exaggerated because of lack of low-altitude mirroring. In the actual magnetosphere, a proton follows a complicated trajectory in a thin current sheet. Because the adiabatic condition breaks down, one can no longer use the guiding center approximation, and details on the gyroscopic scale must be followed especially in the current sheet. On the other hand, because of strong mirror confinement by the dipole field, it takes a long time for an average particle to precipitate into the ionosphere, even with strong scattering in the current sheet. Hence, in theory one has to follow a proton for a time comparable to the expected lifetime and with a time step fine enough to resolve the nonadiabatic details in the current sheet. For the current sheet, this corresponds to tracking a proton for many hours at time steps of a fraction of a second. Then two additional difficulties complicate the computational task. First, many different protons must be followed to render a statistically meaningful picture. Second, in a three-dimensionally geometry, the thin current sheet is longitudinally limited. A proton will eventually exit the current sheet and embark on an adiabatic drift around Earth. While outside the thin current sheet, the proton has a much reduced probability of precipitation, making the outside journey of no particular interest to us. One way to overcome the latter problem is to create a periodic boundary condition in the longitudinal range of the thin current sheet. When a particle exits the system, the computation ceases to follow it but replaces it with a new particle injected at the other end of the longitudinal range with proper correspondence to the lost particle in terms of location, energy, and pitch angle. In essence, this strategy makes every particle in the simulation constantly exposed to current sheet scattering, and it essentially allows a further simplification into a twodimensional model. This consideration leads to the two-dimensional model described below. Field Model and Equations of Motion The 2D magnetic field is represented by the vector potential
a, xz,
=
2 x+z
I
z1
2 + BrAin cosh ( ~ )
(6)
where the coordinate system has the same orientation and axis designation as the GSM but with the x and y axes reversed. Separated into components, the magnetic field is given by
W. W. Liu et al.
168 ~Ay
lL(x,z) = - - ~
2xzBoR ~
z (x 2+ Z2 ) 2 - Brtanh ( S )
=
(7)
By=0 c)Ay _
Bz (x, z)
0x
B 0 (x 2 -- Z 2)
(x 2 + z2) 2
where A is the thickness of the current sheet and B r the associated magnetic field. The magnetic field is assumed to be steady-state, so that there is no inductive electric field. This is a reasonably good approximation for the period preceding the substorm onset. For precipitation associated with inductively generated precipitation during the dipolarization, the reader is referred to Delcourt and Moore [ 1992]. For simplicity, we further neglect any steady-state electric field. It is known that nonadiabatic dynamics are chiefly a consequence of reduced scale length of the current sheet; the local electric field per se has a limited impact in deciding whether nonadiabatic dynamics are important. However, the electric field can play an important role in the precipitation problem because of its ability to transport particles spatially. If a particle is on an open convection streamline (with a lifetime of a few hours in the system), it obviously will have a minimal chance of being lost to a precipitation with a timescale of many hours. However, this does not obviate the fact that a thin current sheet can increase the probability density of a proton to enter the loss cone, and it is the probability density that determines the precipitation flux on a field line. Moreover, the precipitating protons reported by Samson et al. [ 1992b] had an average energy of --20 keV and, moving on fairly low-latitude field lines, are likely to originate from the trapped ring current population. Therefore, we restrict our model to the earthcircling trapped proton population, to justify the neglect of steady-state electric field. The equations of motion for a proton in a 2D magnetic field retain an important invariant of previous 1D models, namely the canonical momentum Pv = mvv + eA ( x ) . This allows the xz motion to be separated from the y motion. The resulting Lorentz equation is written in the dimensionless component form as
d2~ _ bTr2{C+2~-br~ilnlcosh(~)l} 2~ d, 2 +
(~2_~.)
(~2 +
(8)
~212
~-b~r2{C+~2~~' -+----~-b,'81nEcosh( )3} (~,,,+ ~2)2+br tanh( dT.z
)
(9)
where etBr
x
mv
RE
Z
Py
= -R-E' C -
eBpR E
BT
(10)
A
br= ~ , 8 - RE Equations (8) and (9) are solved numerically with a fourth-order Runge-Kutta method. We use a variable time step proportional to the inverse of B = IBI so that each gyroperiod is resolved in an equal number of parts regardless of particle's spatial location. A particle is regarded as lost when it approaches within a distance 1.05 R E from the Earth's center; theextra 0.05 R E is to account for the height of the atmosphere. Figure 2 shows two typical proton trajectories. Plot a represents a rare case of loss upon first exit of the current sheet. The orbit in plot b, much more typical, involves many re-entries into the current sheet before the particle's final precipitation.
Precipitation of Hot Protons from a Stretched Near-Earth Current Sheet
169
Speiser orbit 0.5
/
0.0
/
-O.fi
-1.0
/
/
-1 .fi
-2.0
/
/
f/ -2.5
l
0
2
4
i
t
t
8
6
i
t
,
,
1~o
i
x (RE)
:E '
-3 0
,
Mirroring orbit '
,
'
i
,
i
2
,
,
,
,
,
i
,
I
4
,
,
,
'
,
I
,
I
6
,
,
,
,
,
i
,
l
8
,
,
,
,
J
i
,
110
,
,
,
j
,
2
x (RE) Fig. 2. Two examples of particle trajectories. The upper panel shows a rapid precipitation. The bottom panel shows the much more prevalent case of repeated entires into the current sheet before precipitation.
Results and Statistical Interpretation In order to achieve statistical significance, we repeat the numerical calculation portrayed in Figure 2 for a large number of particles. In this study, we set the number of particles at 2,000. Without loss of generality, all test particles are initiated in the equatorial plane, with randomly chosen initial conditions. All particles start from the equatorial distance L = 10. The particles are tracked to a time limit XM which is long enough to allow meaningful statistical information to be deduced; specifically, the condition xM > 1:N (where "r.N the nonadiabatic precipitation timescale) is in effect. If a particle precipitates before T.u , we decrement the total number of particles N by 1. Repeating this simple procedure over time, we form a profile N (t) as the number of surviving particles at time t. The ratio N~ N can then be used as a statistical measure of the average particle lifetime. The loading of N randomly distributed protons is accomplished through three independent sets of N random numbers, n ( N ) , m ( N ) , and l ( N ) , all between 0 and 1. First, the gyrophase (~) distribution is assumed to be uniform; thus, ~ , ~ 2rtl (i) for the ith proton. Second, the pitch angle c~ is specified according to the rule sin 2c~i = 1 - t:n': (i) ; this corresponds to an isotropic pitch angle distribution for the starting particles. Third, the
W. W. Liu et al.
170
particle energy E i is determined through inversion from n (i) = eft [ ( El/k T) 1,'2] _ 2 ( El/k T) t/2e-e'/kr where erf denotes the error function and T is the proton temperature. The above specification ensures that the N particles conform to the isotropic Maxwellian distribution. In totality, the above procedure constitutes a Monte Carlo simulation. We characterize the standard parameters of the system as follows: x = 10 Re; B r = 40 nT; B z = 8 nT at L = 10; A = 0.5 Re; and T = 20 keV. With the above parameterization, the real time is related to the normalized time as t = 0.261x(O.O5/br) s. We can define an effective ~: parameter [Buchner and Zelenyi, 1989] by using gmi n = BeAIB T and Pmax = mvrlqBe, w i t h ~i ~/2 0.05 ~ 1r = 0.954 (0-~) ( - ~ r ) (kT (keV))-1/4
(11)
Let x. be the time of precipitation of the i-th proton. A straightforward definition of precipitation timescale would se~m to be t N = 0.261 (xi)(0.05/b r) s, where the angles bracket denotes arithmetic average. This definition is of limited use. In any random ensemble of particles, there are ones whose dynamics are adiabatic and which take a very long time to precipitate, if they ever do. In a straight arithmetic average, these particles would make disproportionate contributions, and tu could easily diverge. A more sensible way of estimating average lifetime is by fitting the simulated profile N(t) with an exponential curve of decay, N ( 0 ) [a + be -~' ~ ] (a + b = 1). This fit has the ready interpretation that a represents the percentage of adiabatic particles in the distribution, which will not precipitate and b represents the percentage of nonadiabatic precipitating particles. The fitting parameter (x) is a more proper measure of precipitation timescale. The standard case is then run according to the Monte Carlo scheme outlined above. Figure 3 shows the result of a single run. The termination time is chosen to be x. = 1 0 . Figure 3 shows the particle distributions in the initial, middle, and final stages of the s~mulatlon. A sahent feature in Figure 3 is that nearly three quarters of the original particles are lost at the end of simulation; this justifies our choice of x~ above. Another feature of note is that the thermal spread of particle shrinks visibly as time progresses, indicatmg a cooling of protons. Put differently, it shows that more energetic particles precipitate earlier on average. If we normalize the initial proton temperature to 1, then the temperature is about 0.81 at x = 5 • and 0.75 at x = 104. There is, however, no suggestion that protons with smaller initial pitch angles tend to precipitate early. In fact, the pitch-angle distributions in all three panels are isotropic, within the error associated with the finite sample size. .
.
.
.
4
9
.
.
M
Precipitation
of Hot 1.0
Protons
from
~ , 7 ; ~ r
.,,.',~.~.
k,;~'Z~-.~.5~(~ :.....:.,.-§ . ,,.. ..~' "r .-:.:" "~.. :." ....
;'.'-~.r
.9. . .
.
. ..~.
:;!~i~?"" :~!!? : ; ' -
"',
,..",~
:Lr~, ~( ~, ~' ~r .",',+' .;~~,.~' .. ~ :".' ~~'"".."'""-. ,-" "
o.sF~/-~;".,~..'.....
171
.....
: ....
,
9
.
~ ..... :
~.:;,...-.~,..:-....::.:.....:.:..
..
" :.,.",.. ~:~.:" :" ."" ". '-. 9". "
0.6
"~
Stretched Near-Earth Current Sheet
a
o.4 :..
0.0
.
.
~=0
-t'
,
,
"
1
0 1
I
2
"~.4 "*'.,'r'......"."
~,....
,.'~;..
.~..,-
.:
,
,
,
4
9. ~ . ~ a c , . z . . ~ ; r - ~ - ' l . . ~ . : •0 "=:.~,:~,...... ~ ;. . . . . .
0.8
,
"
9".. .. . ..' . -.. -.
". ""
.... .
:. . . .
.;:~';."
...'...:...
:'S.-';;~'.:~.
.:
9 .
i
,
,
6
I
.
.
.
.
i
I
i
i
8 ,
,
,
i
,
,
".
.
;
I
.
.
.. 0.6
"
.,-i 0.4
.
.
.
0.2
1; = 5 x 1 0 3 0.0
,
I
0
i
i
2
~.:: 9
. . . ' : t
":t"':".. r:"
.
i
i
,
4
"..'
.
I
I
l
,
6
I
,
i
,
,
,
8 '
,
,
,
i
10
". :
...
0.8 9 "..~.
-..-. -. ...;
:
-.,.
"" . . . . 9 ...
:"9
0.6
.
9 9 ".7 .
."'" . : ' . .
o,~
9
-..
9
0.4-:"
.
9
9
"
.
."
9 . ' ~ . '
..
9.
-". 9
.
.
...
9
.. 9
.
0.2-
.
.
1;= 0.0
,
. . . . . .
0
I
,
2
,
,
I
,
4
,
,
I
6
I 8
104 ,
,
, 10
E/kT
Fig. 3. Time history of an isotropic Maxwellian ensemble of 2000 protons, as nonadiabatic precipitation depletes the population. It is instructive to plot aggregate properties of the proton distribution as a function of time, as shown in Figure 4a (total particle number) and Figure 4b (total particle energy). The temporal profiles are fit excellently by the offset exponential curve (dashed). For the particle number, the fit parameters are a = 0.18 and (xN) = 4.5x103. For the particle energy, a = 0 15 and ( x ) = 3 5x103 The difference between the two time constants is consistent with the previous observation that the d~stnbution becomes cooler over time. The offset constant a means that, in the asymptotic limit x ---->co, one expects that approximately 18% of the particles and 15% of particle energy to remain in the magnetosphere. These correspond to the adiabatic population discussed, for example, by C h e n a n d P a l madesso [1986]. Referring to our normalization scheme, we find that tu = 1.17x103 s. As a comparison, we evaluate the SPADL timescale in (5) in a dipole field. Since the thin current sheet evolves from a dipolar configuration, this comparison indicates how significantly the nonadiabatic effects enhance proton precipitation. For a 2D dipole, the unit flux volume is V = ~ r e / 2 B e. For r e = 10 R E and the standard parameters above, (5) gives 9.05x103 s. By comparing the two timescales we see that the nonadiabatic precipitation is nearly an order of magnitude stronger than the SPADL precipitation. We have repeated Monte Carlo runs for several different sets of random initiations and found no significant difference in either the profile of N (t) or the fit parameters. This suggests that the choice of N = 2000 is adequate. 9
o
"
E .
o
"
~
"
o
172
W. W. Liu et al. 2 0 0 0
,
,
,
1
simulation 1500
~"
.......
fit
looo
5OO
0
0
l
~
A
t
2000
i
I
~
L
t
|
l
i
I
I
i
i
=
4000
6000
8000
10000
4000
6000
8000
10000
3000
"\\
"\~
2000
\\\
1000
0
2000
0
T
Fig. 4. T ~ i m u l a t e d profiles of particle number (upper panel) and total energy (lower panel) fit to the offet exponential curve a+be-
,
A minor modification can be made of the above estimate of the nonadiabatic precipitation timescale. Because the precipitation profile in Figure 4 is an offset exponential curve, the ratio din N~ dt is not a constant but decreases with time. In strongly nonadiabatic situations such as the standard case (K: = 0.451 ), most particles (82%) precipitate, and using (xu) in the manner above would not result in serious errors. However, in nearly adiabatic situations (~: ~ 1 ), most particles do not precipitate and using (1:) as above would not be accurate A more proper definition of preclpltanon nmescale is therefore proposed asthe inverse of d l n N / d t at t = 0, or (I: u) = (XN)/( 1 - a ) . For the standard case, this will increase t u to 1.43• ~ s, still much shorter than the SPADL timescale. For nearly adiabatic cases a - 1, however, the modification offers a more sensible representation of the situation. It is known that the adiabaticity of particles is strongly regulated by the K-parameter defined by B u c h n e r and Zelenyi [1989]. For 1( ~ 1, the dynamics are adiabatic, and few protons will precipitate as described here. For ~: = O ( 1), the nonadiabatic effect is found to be most chaotic. For 1( ~ 1, the situation reverts to a generalized adiabaticity [Sonnerup, 1971 ], and the protons execute the meandering bi-sector motion. In this extreme case, precipitation is also limited. We have carried out an expanded study by adjusting the bar parameter between 0.001 and 0.2 to probe the effect on precipitation timescale of ~. Without changing other parameters, this creates a ~: range between 0.226 and 3.190, according to (11). Our Monte Carlo and fitting procedures above are repeated for a number of 1( parameters, and the resulting precipitation timescale is presented in Figure 5. Not surpris0!gr~ly, the timescale increases with increasing 1r and is fit reasonably well by an exponential curve tN/t o = 0.183e " . Apparently, the lower limit of ~: = 0.226 is not enough to allow us to get a view of the possible confinement by the meandering .
.
.
.
.
'
/V
~ /
Precipitation of Hot Protons from a Stretched Near-Earth Current Sheet
173
motion which is expected to arise in the extreme limit of B. ---) 0. Because we use dipolar value for B., this limiting case cannot be reproduced in our study unless we use grossly unrealistic values for B r.
tN to
_
_
/
-
00
, , , , , , , ~
, , l
. . . . . . . . .
1
l
2
. . . . . . . .
j
l
;
;
,
j
,
,
,
,
,
3
Z
4
K Fig. 5. Dependence on K of the ratio of nonadiabatic precipitation timescale to the SPADL timescale. SUMMARY AND DISCUSSION The main concern of this study was a comparison of nonadiabatically induced proton precipitation with one associated with pitch-angle scattering by cyclotron waves. The question in itself has certain theoretical significance. The primary finding was that for K parameter less than 1, nonadiabatic precipitation dominates pitch-angle scattering. The effort further led to an explanation of the H 13 intensification on the basis of the thinning current of substorm onset. With future elaborations, we hope to develop the framework into a useful diagnostic tool of the magnetosphere, i.e., mapping the location and intensity of current sheet by using ionospheric data as a proxy for precipitation. An implication of this study is strong nonadiabatic behavior will create an anisotropic pitch-angle distribution in which smaller pitch angles are overpopulated. This behavior has been noted since the work of Speiser [ 1965; 1967] but has not been quantitatively determined in a realistic setting. Two issues related to this question were considered in this study: 1. Computation of the precipitation flux in the presence of the dipole field; 2. the influence of the K parameter on the streaming flux. The statistical techniques and exponential fitting can be further extended to more sophisticated studies of this problem in three dimensions. Our model in its present form is limited to various degrees by its several idealizations. The neglect of timedependence limits its applicability to relatively slowly varying period before eruption. The neglect of steady-state electric field would hamper a detailed prediction because in reality protons in the near-Earth magnetosphere are a mixture of trapped and open-streamed populations. The most serious shortcoming of the model, in our view, lies in the two-dimensionality of our computational model. As remarked earlier, the employment of the 2D model was prompted primarily by the consideration of computational economy. It is also a significant improvement over the 1D calculations set in the Harris or parabolic sheets in studies of chaotic ion motions [Chert, 1992, and references therein]. However, since the actual magnetosphere is three dimensional, we should be cognizant of the price we have to pay to achieve computational economy. The principal drawback of the 2D model is that the angle of the loss cone is artificially inflated. For a 2D line dipole, the loss cone angle otc = L-~ , while for a 3D point dipole, o~c = L-3/2. Thus, the precipitation timescale calculated from any 2D model cannot be taken at face value. The compensation we made toward this problem was through relative comparison with the SPADL value in 2D. Since both SPADL and nonadiabatic values contain roughly the same degree of loss-cone inflation, it seems reasonable to assume that the ratio of the two values should be relatively insensitive to the dimensionality. Granting this is the case, we can extend the 2D ratio to 3D and make predictions given the K parameter and local current sheet geometry. Since the SPADL value is easy to calculate in three dimensions, this in effect gives us a way to estimate nonadiabatic precipitation flux in 3D, without actually performing 3D calculations. Obviously, this idea needs to be verified through controlled 3D modeling.
174
W.W. Liu et al.
Another important 3D effect is the limited longitudinal range of the thin current sheet. We have mentioned that a proper setting of periodic boundary condition, through a balance of exit and injection at opposite sides could lead to a computationally efficient model. We have also failed in the study to account for possible particle energization during the sudden formation of the thin current sheet. Ganguli et al. [1995] have demonstrated that during the late growth phase of current interest, higher-energy ions may undergo a Fermi acceleration, pushed by the converging lobe and associated electric field. We point out, however, that the incorporation of this effect will further increase the efficiency of scattering, as it works to reduce the ~r parameter. Therefore, the qualitative conclusion of this study, namely the thinning of equatorial current sheet can be associated with much enhanced proton precipitation and H 13 emission is not affected. Finally, we offer a qualitative consideration as to what has transpired in the nonadiabatic enhancement in Figure 5. For simplicity, we adopt the assumption that pitch angle changes effected by the traversal of current sheet are independent of each other, large (Ao~- O (1) ), and random for practical purposes. By these assumptions, one can argue statistically that the probability of a single crossing deflecting a particle into the loss cone is the ratio of the solid angle of the loss cone t2o the solid angle of a hemisphere, 2ft. A consideration of geometry indicates that this probability is 1 -coso~ c -- c~c/2. Hence it will take an average of 2/c~, = 2BolBe crossings for the proton to enter the loss cone. The average between equatorial crossings is x~ - l~ v r where I is the typical length of a bouncing path and vr is the particle thermal speed. This probabilistic argument leads to a characteristic timescale of precipitation of 21Bo/Bev r. For a dipolar field, the parallel scale length is 1- r e (the equatorial distance of the field line). The resulting estimate of 2reBolB e agrees within a factor of 2 with the SPADL value rcreBol2B e. In the thin current sheet, there is a new and much reduced spatial scale A, and most bounce paths have the length ~A, instead of r e. Suppose the great majority of crossings result from the local gradient. Then the characteristic time for the same 2Bol Be crossings is ~4BoAI B e. In comparison with the SPADL value, this represents a shortening of timescale by rtLI 8~5. For the standard case L = 10 and ~i = 0.5, the argument predicts a timescale shortening by a factor of ~8, close to the computed value presented in the text. Therefore, the emergence of a small scale length enhances precipitation in two ways; first by chaotizing the ion orbit, and second by increasing the bounce frequency. ACKNOWLEDGMENTS The work of WWL was supported by the Canadian Space Agency research contract 9F007-5-8004/01-SR. The work of GR and JCS was supported by the Natural Sciences and Engineering Research Council of Canada. REFERENCES
Buchner, J., and L. M. Zelenyi, Regular and chaotic charged particle motion in magnetotaillike field reversals, 1. Basic theory of trapped motion, J. Geophys. Res., 94, 11821, 1989. Chen, J., Nonlinear dynamics of charged particles in the magnetotail, J. Geophys. Res., 97, 15011, 1992. Chen, J., and P. J. Palmadesso, Chaos and nonlinear dynamics of single-particle orbits in a magnetotaillike magnetic field, J. Geophys. Res., 91, 1499, 1986. Delcourt, D. C., and T. E. Moore, Precipitation of ions induced by magnetotail collapse, J. Geophys. Res., 97, 6405, 1992. Ganguli, G., P. J. Palmadesso, J. Fedder, and A. T. Y. Lui, Role of Fermi acceleration in explosive enhancement of cross-tail current in late substorm growth phase, Geophys. Res. Lett., 22, 2405, 1995. Gray, P. C., and L. C. Lee, Particle pitch angle diffusion due to nonadiabatic effects in the plasma sheet, J. Geophys. Res., 87, 7445, 1982. Kennel, C. E, and H. E. Petschek, Limit on stably trapped particle fluxes, J. Geophys. Res., 717 1, 1966. Krall, N. A., and A. W. Trivelpiece, Principles of Plasma Physics, p. 497, McCraw-Hill, New York, 1973. Liu, W. W., B. L. Xu, J. C. Samson, and G. Rostoker, Theory and observation of auroral substorms: A magnetohydrodynamic approach, J. Geophys. Res., 100, 79, 1995. Lui, A. T. Y., Current disruption in the Earth's magnetosphere: Observations and models, J. Geophys. Res., 101, 13,067, 1996. Lui, A. T. Y., R. E. Lopez, B. J. Anderson, K. Takahashi, L. J. Zanetti, R. W. McEntire, T. A. Potemra, D. M. Klumpar, E. M. Greene, and R. Strangeway, Current disruption in the near-Earth neutral sheet region, J. Geophys. Res., 97, 1461, 1992. Lyons, L. R., and T. W. Speiser, Evidence of current sheet acceleration in the geomagnetic tail, J. Geophys. Res., 87, 2276, 1982. Mitchell, D. G., D. J. Williams, C. Y. Huang, L. A. Frank, and C. T. Russell, Current carriers in the near-Earth crosstail current sheet during substorm growth phase, Geophys. Res. Lett., 17, 583, 1990. Ohtani, S., K. Takahashi, L. J. Zanetti, T. A. Potemra, R. W. McEntire, and T. Iijima, Initial signatures of magnetic field and energetic particle fluxes at tial reconfiguration: Explosive growth phase, J. Geophys. Res., 97, 19,311, 1992. Samson, J. C., D. D. Wallis, T. J. Hughes, E Creutzberg, J. M. Ruohoniemi, and R. A. Greenwald, Substorm intensifications and field-line resonances in the night-side magnetosphere, J. Geophys. Res., 97, 8495, 1992a. Samson, J. C., L. R. Lyons, P. T. Newell, E Creutzberg, and B.-L. Xu, Proton aurora and substorm intensification, Geophys. Res. Lett., 19, 2167, 1992b. Sergeev, V. A., D. G. Mitchell, C. T. Russell, and D. J. Williams, Structure of the tail plasma/current sheet at -11 Re
Precipitation of Hot Protons from a Stretched Near-Earth Current Sheet
175
and its changes in the course of a substorm, J. Geophys. Res., 98, 17345, 1993. Sonnerup, B. U. O., Adiabatic particle orbits in a magnetic null sheet, J. Geophys. Res., 76, 8211, 1971. Speiser, T. W., Particle trajectories in model current sheets, 1" Analytical solutions, J. Geophys. Res., 70, 4219, 1965. Speiser, T. W., Particle trajectories in model current sheets, 2: Applications to auroras using a geomagnetic tail model, J. Geophys. Res., 72, 3919, 1967. Xu, B. L., J. C. Samson, W. W. Liu, F. Creutzberg, and T. J. Hughes, Observations of optical aurora modulated by resonant Alfv6n waves, J. Geophys. Res., 98, 11531, 1993.
This Page Intentionally Left Blank
COMPUTER SIMULATION OF THE THREE-DIMENSIONAL DECAY OF THIN COLLISIONLESS CURRENT SHEETS J. Biichner and J.-P. Kuska
Max-Planck-Institut f'dr Aeronomie, Max-Planck Str. 2, D-37191 Katlenburg-Lindau, Germany ABSTRACT Recent theoretical investigations and simulations of collisionless space plasma current sheets have claimed their stabilisation against reconnection by finite cross-sheet magnetic field components. However, all these theoretical investigations and simulations were based on two-dimensional models. Currrently we have shown that the energy variations change quite a bit as soon as one considers the problem in three dimensions- the sheet can become unstable in three dimensions where it was stable in two dimensions. Since it is difficult, however, to evaluate the three-dimensional structure of the dynamical current sheet decay analytically, numerical simulations will be helpful. Here in this paper we report results of our numerical simulations of the three-dimensional decay of a thin Harris current sheet. We demonstrate that in three dimensions the decay of a thin collisionless current sheet starts with the unstable growth of a compressional sausage-mode wave, propagating in the current direction. Afterwards reconnection begins at a time scale faster than a tearing mode instability known from the past two-dimensional considerations. INTRODUCTION Reconnection is a very efficient way to transform the energy of plasma flows and magnetic fields to plasma heating and particle acceleration. In space it is supposed to act at interfaces between magnetized plasmas (magnetopauses), during magnetospheric substorms as well as in solar flares and other astrophysical and space plasma energy release processes. From its first suggestion by Giovanelli, Hoyle, Dungey and from its pioneering investigation by Parker, Sweet and Petschek it is clear that reconnection means irreversible energy transformation (see Vasyliunas, 1975 as well as Axford, 1984 and references therein). In coronal and magnetospheric plasmas, which are essentially collisionless, wave-particle resonance processes must provide the mechanism of irreversible energy transfer. Coppi et al. (1966) developed a first appropriate kinetic theory of current sheet reconnection after a so called tearing mode instability. In their theory electron Landau damping provides irreversible wave particle energy transfer for reconnection. Later Biskamp et al. (1970) and Schindler (1974) drew the attention at the more realistic situation of current sheets, whose magnetic fields contain a finite component accross the sheet, which inhibits Landau damping on the electrons. They suggested that in this case Landau damping on ions might provide the necessary dissipation. Galeev and Zelenyi (1976) provided a kinetic analysis of this situation. They pointed out that an ion tearing mode instability might not work due to the stabilizing influence of the reacting electron gas. Lembege and Pellat (1982) claimed the total stability of a sheet which contains a finite normal magnetic field component Bn. Pellat et al. (1991) later confirmed this result on more general grounds. Brittnacher et al. (1994) and Quest et al. (1996) showed that the compression of the plasma frozen in two-dimensional magnetic flux tubes causes the stability of sheets in the presence of a finite Bn. The two-dimensional simulations of Pritchett (1994) confirmed this result (see also Pritchett and Biichner, 1995). All these past models were essentially two-dimensional. This means that the available free energy as well as the irreversible energy transfer are evenly distributed along the invariant, the current direction. And so is the compression of flux tubes, the major stumbling block, inhibiting collisionless reconnection through current sheets penetrated by a finite Bn. The transition to three-dimensional considerations may essentially change the situation. For example the energy necessary to compress a flux tube in order to allow reconnection, changes a lot if one considers a perturbation with a finite wave vector component ky in the current direction in addition to the usual sheet-tearing perturbation modulated perpendicular, in the X-direction (Biichner, 1995; 1996). Unfortunately, it is difficult to describe the three-dimensional dynamics of a current sheet and to understand the dominating mode of a three-dimensional current sheet decay just by analytical methods. In order to find the mode which determines the decay of thin current 177
178
J. Btichner and J.-P. Kuska
sheets in the real three-dimensional world we have carried out appropriate particle-in-cell (PIC) kinetic plasma simulations by using the GISMO code (cf. Bfichner and Kuska, 1996). Here in this paper we report the results of our simulations of the three-dimensional unstable decay of thin Harris (1962)- type current sheet equilibria.
Figure 1: 3-D iso-density surface close to the initial moment of simulation time (lower panel) and two crosssections, through the plane Y = 0 (upper panel) and Z = 0 (middle panel) in the first type of simulations (hot, thicker sheets) SIMULATION SETUP / PARAMETER CHOICE In collisionless space plasmas resonant interactions between particles and fields provide the irreversible energy transfer necessary for reconnection. The simulation of collisionless reconnection must, therefore, be essentially kinetic. We investigate the decay of thin current sheets toward reconnection by using the newly developed three-dimensional electromagnetic particle-in-cell (=PIC) code GISMO (see Biichner and Kuska, 1996). The code solves the wave equations for the scalar (r and vector potentials (A) as well as for higher accuracy, the Poisson equations coupling fields and charge density. The fields are determined on a threedimensional equidistant rectangular spatial mesh. The grid distance are chosen to be Ax = Au = Az = A in the X, Y and Z directions, respectively. The relativistic equations of particle motion F = e ( ~ + g x / 3 ) are integrated by means of a Runge Kutta scheme with stepsize control. For more details about the code see Biiclmer and Kuska (1996).
Three-Dimensional Decay of Thin Collisionless Current Sheets
179
Figure 2: 3-D iso-density surface after the reconnection has started (lower panel) and two cross-sections, through the plane Y = 0 (upper panel) and Z = 0 (middle panel) We have initialized thin Harris-(1962) sheet equlibria by randomly loading electrons and ions in accordance to the drift- Maxwellian distribution fj (~',~ -- 7r3/-'~-nj(Z)i.i v-a -exp { 2kBTjmJ (v2 + (v ~ - Udyj )2 + Vz2)} ,
(1)
where n j ( Z ) - n e ( Z ) = n i ( Z ) - no " cosh -2 ( ~ z )
and
Lz = ~/kB . (Te + Ti) . c ~/21r no e 2 u d i - Ude
(2)
Here j ----e, i denote electrons and ions, respectively, Te and Ti are the electron and ion temperatures and u4/ denotes the drift velocities of the particle species, kB is the Boltzmann constant. The current in a Harris-equilibrium is diamagnetic, it corresponds to the pressure gradient across the sheet. For constant temperatures it is, therefore, directly related with the density gradient in the Z direction, characterized by the length scale Lz. The resulting magnetic field can be estimated by using the Harris-sheet identity /3 = B o ' t a n h ( L ~ ) -ex "
where
Bo = ~ / 8 r . n o .
k B . (Te + Ti).
(3)
As far as the boundary conditions are concerned we have applied for both the fields and particles periodic conditions in the X and Y directions. High frequency waves are damped away before reaching a boundary.
180
J. Bfichner and J.-P. Kuska
Almost no particles reach the boundaries in the Z direction at Z = 0 and A Z = A X / 2 . potentials are put zero.
There the
Let us denote the thermal Larmor radius in the asymptotic field poj - vtj/f~oj with vtj = x / k B T j / m j being the thermal velocity of the j- particle species, f~oi = e B o / m i c is the gyro-frequency in the asymptotic field, e is the elementary charge, m i and me are the ion and electron masses, respectively and c denotes the speed of light. Since we chose a frame, in which the equilibrium electric field vanishes, the electron and ion drift velocities Udye and Udui are related according to as Udue/Udyi = - T e / T i . In order to avoid additional numerical instabilities due to unequal electron and ion temperatures (Pritchett, 1994) we assume Te = Ti = T at t = 0, so that udyi -- --Ud~e = Ud. Then the sheet half-width can be expressed as 1 F
~/kB 9T Lz -
~/4r no e 2
C
•d
__
C
~td
hD
Vti " hi = 2" vti Poi Ud ~td
(4)
Equation (4) relates the half-thickness Lz via the drift and ion thermal velocity (Ud/Vti) to the ion inertial lengths hi = c/wpi as well as to the Larmor-radius in the external field poi. Notice that due to the diamagnetic nature of Harris-sheets hi = 2 Poi (cf. equation ( 4 ) ) . Hence, thin sheets in the sense of Lz --+ Poi, are automatically also thin in the sense of the ion inertial length, i.e. Lz --+ ,ki/2. We have investigated the dependence of the sheet stability on the sheet thickness in the broad range poi/Lz = 0 . 1 . . . 1. Here in this paper we present only two examples of our simulation results, for thicker sheets with Poi/Lz = 0.25 and and for very thin sheets with p o i / L z - 1. In the course of our simulation runs we took special care of possible non-physical, i.e. numerical instabilities. Usually numerical instabilities are considered ignorable, as a rough rule of thumb, if the grid distance A does not exceed the Debye length h D = x/kB T/47rnoe 2 by more than a factor of 7r, A < Ir hD (cf. Birdsall and Langdon, 1991, pp. 179). Since the main irreversible interaction is supposed to take place inside the current sheet we aim at resolving the whole sheet, by, say, twenty grid spaces, i.e. Lz = 10 A. Then, in order to meet the numerical stability condition, the sheet half-thickness Lz should not exceed 30 hD. From this condition one obtains a lower limit of Udj > 0.03 c for the drift velocity. Since we aim at the investigation of thin sheets, i.e. according to equation (4) of v_tt = 0 . 5 . . . 2, we have to take into account the limiting condition on the allowed thermal ttd speed vtj > 0.015... 0.06 c or kB2q > 140 k e V . . . 2.3 M e V and kBTe > 80 e V . . . 1.3 k e V . In the beginning we used the real proton mass and equal mass electrons, i.e. M i = me = Mp. In these rims we had to chose k B T = 5 M e V in order to resolve at least three Debye length by the grid spacing as needed according to the old rule of thumb (Birdsall and Langdon, 1991). To verify the results of this first series of runs we increased the resolution of the Debye length up to ten times. In order to achieve this high resolution we reduced the particle mass to the electron mass and the temperature to 40 k e V . In the following section we present the results of a thicker sheet simulation. It could be carried out by using the ion mass but a temperature of 5 M e V . The best results for the thin sheets were obtained, however, by using the electron mass and a temperature of 40 k e V , they are presented in the but next section. T H I C K E R (HOT IONS) S H E E T SIMULATIONS (Poi/Lz = 0.25) In a first series of rims both particle species were given the ion mass and the large initial temperature of kB T -- 5 M e V . The chosen parameters reveal vti - 2.2- 10 9 c m / s . A moderately thin sheet of Poi/Lz = 0.25, enforced by Ud/Vti = 0.5, is reached by a drift speed Ud = 1.1- 109 c m / s = 0.03 c. In order to obtain realistic field strength and time scalings we chose in these runs a density no - 1.4 10-4cm -3 which yields, together with a temperature of kB T = 5 M e V , an estimated asymptotic magnetic field 13o -- 4~/7r no kB T of ~ 24 n T . The absolute value of the Debye length is 1.4.108 cm. For a equilibrium sheet half-width Lz ~ 3.8- 109 cm the h D / L z ratio equals, therefore, 0.037. We resolved the whole sheet by about 20 grid planes, i.e. Lz ~ 10 A. Hence, hD ~,~ 0.37 A which is close to the old rule of the thumb level 7r-hD ~ A. In order to leave a safe distance to the box' upper and lower boundaries the box extends over 64 grid planes in the Z-direction, i.e. A Z -- 64A ~ 6.4Lz. Since the energetically minimum possible length of a tearing island is hx,min - 21rLz, we chose a box extending over 128 A in the X- direction, which corresponds to a box length of A X ~ 12.8 Lz in X. The box length in the Y-direction was limited to Lz = 32A, i.e. about 3.2Lz.
Three-Dimensional Decay of Thin Collisionless Current Sheets
181
Figures 1 and 2 depict 3-D iso-density surface (lower panel) and two cross-sections, through the plane Y - 0 (upper panel) as well as through Z - 0 (middle panel). Figure 1 corresponds to a moment close to t - 0 while Figure 2 relates to a moment, when reconnection already has started.
Figure 3: 3-D iso-density surfaces; Left: At ~2oit = 0 and right panel at f~oit = 6, after reconnection has started The middle panel of Figure 1, showing a cut of the density isosurfaces in the current plan, indicates that from the very beginning a tendency exists toward attraction of the current elements along the X axis, i.e. toward sheet pinching or tearing. However, from the very beginning also a density structuring starts along the current direction Y. Finally, when a tearing island is formed (Figure 2, upper and lower panel) the structuring in the current direction (Figure 2, middle and lower panel) is well expressed. In this first series of simulations the three-dimensional box accomodated more than two energeticlly minimum wavelengths, i.e. more than 2 - 2 r L z in the X direction, but only 3.2 Lz in the current direction. Nevertheless, these runs have shown convincingly that the current sheet instability is essentially threedimensional. For making sure that we have caught the correct wavelength of the perturbation in the Y direction we then have increase the relative box length in Y. When studying thinner and thinner current sheets with Poi/Lz --+ 1.0 using the same high temperature of 5 M e V and ion masses we first tried to meet always the old rule of thumb condition for the necessary resolution of the Debye length A < lrAD, which was valid for one-dimensional problems (see Birdsall and Langdon, 1991). In our three-dimensional simulations of thin current sheets we found, however, that the old rule of thumb does not suffice to suppress numerical finite-grid related instabilities. For investigating the unstable decay of thinner sheets more accurately we increased the resolution of the Debye length. Improving the resolution up to a factor of ten we found that, finally, the growth rates have settled and the numerical instabilities became weaker than the physical one. To increase the resolution of the Debye length we had to compromise by using the electron mass but we could use more realistic temperatures and densities. This way we found that the growth rates of the three-dimensional current sheet decay started to converge beginning with a 7.5 higher resolution as the old rule of thumb predicted. THINNER (WARM ELECTRONS') SHEET SIMULATIONS (Poi/Lz = 1.0) As discussed in the previous section it is, first, desirable to extend the relative box length in the Y direction and, secondly, to resolve the Debye length better. Since the potential limits for parameter variations of the Harris sheet equilibrium are tight, one can vary the resolution of the Debye length mainly by lowering particle mass and plasma temperature. In this section we report a typical result from a second series of
182
J. Bfichner and J.-P. Kuska
runs with a growth rate independent on further possible numerical refinements. For this we lowered the particle mass to the electron mass and decreased the initial temperature to k B T - 40 k e V . The simulation box spans in this series of runs over A Y - 6.2Lz ..~ 2 r Lz in the Y, the current flow direction. This was achieved at the expense of cutting it in the X-direction down to the energetically minimum length of a tearing island A X - 6.2 Lz ..~ 2 r L z . The extension of the box length in the Y direction should demonstrate the spatial structure of the instability in the current direction better. One obtains a marginally thin sheet Lz - Poi, by chosing ud/vti - 2. Electron mass and temperature reveal a characteristic thermal speed of 4- 109cm -3, i.e. ud = 8- 109 c m / s . This makes the Debye length resolved by several grid distances (cf. equation (4)). In physical units the Debye length is now about 7.4-104 cm. With an estimated equilibrium sheet half-width Lz .~ 2.6-105 cm the ~ D / L z ratio now equals 0.28. Since, again, the whole sheet is resolved by about 20 grid planes (Lz ~ 10A), the resolution of the Debye length is now ~D ~ 2.8 A, i.e. 7.5 times better than in the first type runs.
Figure 4: Cuts through the plan Y = 0.5 A u of iso-density surfaces; Left: At Ctoit = 3.5, after the current instability has developed and right panel at ~oit - 5, when reconnection has started The resulting unstable evolution of a thin Harris sheet follows qualitatively the same scenario as in all other simulation runs, although the growth rate settled down to a level of several times the tearing growth rate in the two-dimensional theory. In contrast to the two-dimensional tearing mode instability we see first a whole-sheet current instability causing a propagating density structure moving in the Y, i.e. the current direction. Finally three-dimensional reconnection starts, still much quicker than two-dimensional tearing. Let us demonstrate this sequence by showing characteristic snapshots of the density evolution. In Figure 3 three-dimensional isodensity surfaces do depict first the initial sheet (left panel) and later the resulting structure after both current instability and reconnect ion have taken place (~oit - 6, right panel of Figure 3). Notice that the modulation of the number density n, shown in Figure 3, is practically identical with the modulation of the current density jy - e n < vy > , since < vu > ~ Ud .~ const. Figures 4 and 5 depict cuts through the three-dimensional iso-density surfaces shown in Figure 3. Figure 4 shows cuts of the density distribution through the Y -- 0 plane and Figure 5 through plane Z - 0, the initial symmetry plane. Both Figures 4 and 5 illustrate the sequence of current sheet decay very clearly. As already mentioned, the current instability has started from the very beginning, i.e. well before ~2oit - 3.5, when the snapshot depicted in the left panel of Figure 4 was taken. Reconnection, however, has not yet started at ~oit = 4.5, which is illustrated by the cross-current cut (left panel of Figure 5). A comparison between the right panel of Figure 5 at ~oit = 5 with the right panel of Figure 4 corresponding to ~oit = 6, however, indicates that, finally, reconnection has taken place.
Three-Dimensional Decay of Thin Collisionless Current Sheets
183
Figure 5: Cuts through the plan Z - 0 of iso-density surfaces; Left: At ~2~t - 4.5 there is still no reconnection, while (right panel) at ~ t -- 5. reconnection, finally, has started What is not shown in our snapshot presentation is that the current instability causes a wave propagating in the Y-direction, i.e. in the current flow direction. The full time sequence of Figure 4-type plots (not shown here) allows the determination of wavelength, frequency and velocity of the excited wave. According to our simulations the wave frequency is w ~ 2.5 ~ . The wave number k~ corresponds to k~Lz ~ 1. The resulting wave velocity is, therefore, Vwave = w / k y ..~ 2.3 ~2oiLz. From equation (4) and Ud/Vti = 2 one finds ~2oiLz - vti, i.e. Vwave ,'~ 2.3 vti .~ 1.1u4. The wave propagates, therefore, at a velocity close to the drift speed of the positively charged current carriers. SUMMARY AND DISCUSSION The aim of this work was to demonstrate the mode of the three-dimensional decay of thin Harris-type collisionless current sheets. Since the decay of collisionless thin current sheets is an essentially kinetic process, we had to carry out kinetic plasma simulations. We took special care to ensure the simulated instability is due to phyical processes rather than numerical heating. First we solved the Poisson equation explicitely and then we increased the resolution of the Debye length essentially beyond the rule of thumb level given by Birdsall and Langdon (1991). Here in this paper we have presented two examples illustrating our simulation results for hotter and thicker as well as for'colder" and thinner sheets. Our main result is, first, the identification of a sausage-mode current instability preceeding the onset of reconnection through thin current sheets in three dimensions independent on the temperature. Secondly, the current instability enhances the growth rate of reconnection well above the classical two-dimensional tearing mode growth. In order to verify this result we have increased the resolution of the Debye length ten times above the level of the old rule of thumb A < ~rAD. In the occurence of a current instability our simulation results agree with those of Zhu and Winglee (1996). However, in contrast to them we do not see a kinking of the sheet but a sausage-mode compressional instability (cf. Figure 3) preceeding and enhancing reconnection. Also, the instability growes in our simulations slower than in Zhu and Winglee's (1996). We think this has to do with the solution of the Poisson equations in our simulations and with our enhanced resolution of the Debye length above the level of the old rule of thumb. We have carried out systematic investigations of the dependence of growth rates on the resolution of the Debye length by the grid spacing and ended up with a necessary improvement of the resolution by a factor of about 7.5. The new rule of thumb for three-dimensional simulations would be rather A < ~-1~ D.
184
J. Bfichner and J.-P. Kuska
As far as the interpretation of the observed compressional sausage-type wave mode is concerned we want to draw the attention to the fact that in thin current sheets the threshold of an ion acoustic instability Ud > V t i and also Ud ~> Vte (Bunemann, 1959) are met (cf. equation (4)). We conclude that in three dimensions a direct coupling between a compressional current and a sheet tearing (pinching) instability takes place, which enhances the growth of reconnection. The results, presented here, correspond to a mass ratio Mi/me unity. We have prepared a publication of our results concerning the influence of the mass ratio on three-dimensional decay of thin current sheets in (Biichner and Kuska, 1997). The corresponding results for mass ratios up to 64 are - in short - that the character of the unstable decay, with a primary development of a sausage - mode current instability, enhancing reconnection, does become even more expressed, although the growth rate is influenced. In the process of finalizing the preparation of this paper the results of similar investigations by Pritchett et al. (1996) became known to us. These authors have also simulated the stability of thin current sheets kinetically, they have found sheet kinking as Zhu and Winglee (1996) and they have analyzed the possible mode of current instability by a two-fluid approach. These authors claim that the sheet becomes kinkunstable as found in Zhu and Winglee (1996). Our different result indicating a sausage mode instability rather than a kink-mode may be due to our explicit solution of the Poisson equation, which has not been done by Zhu and Winglee (1996), from our better resolution of the Debye length, which allowed us to consider the irreversible wave-particle interaction more accurately and from the weaker influence of the boundaries which we achieved by damping away waves before they reach the boundaries. A linear dispersion analysis seems to support the sausage-mode type of the instability of the current sheet. We are aware that in the case of a finite Bn, in contrast to the pure Harris sheet, two-stream instabilities do not occur in the same way. In this case cross-field instabilities will take over, as discussed by Lui et al., (1991) (see, also, LUI, 1996). Which instabilities in occur first in the case Bn r 0 and how they couple to the macroscopic evolution of current sheet system, eventually to reconnection, has still to be investigated. Supporting results and GISMO code simulations are on their way. They will be published after cross-checking with theoretical investigations. ACKNOWLEDGEMENTS The authors gratefully acknowledges the support of Prof. G. Haerendel for this work in Berlin till the end of 1996, and Prof. Sir I. Axford at the Max-Planck-Institut t'dr Aeronomie in Lindau as well as the the DFG support for J.P.K.'s work on GISMO.
Three-Dimensional Decay of Thin Collisionless Current Sheets
185
REFERENCES Axford, I., Magnetic field reconnection, in Re.connection in space and Laboratory Plasma, ed. Hones E.W.Jr., Geophysical Monograph 30, AGU, Washington D.C., 1, (1984). Birdsall, C.K., and A.B. Langdon, Plasma Physics via Computer Simulation, IOP Publishing, Plasma Physics Series (1991). Biskamp, D., R.Z. Sagdeev, and K. Schindler, Nonlinear evolution of the tearing instability in the geomagnetic tail, Cosmic electrodyn., 1, 297 (1974). Brittnacher, M., K.B. Quest, and H. Karimabadi, On the energy principle and ion tearing in the magnetotail, Geophys. Res. Lett., 21, 1591 (1994). Biichner, J., Multiscale coupling in reconnection - Three-dimensional current sheet tearing, in Physics of space plasma, MIT Cambridge, Massachusetts, No. 14, 79 (1995). Biichner, J., Three-dimensional current sheet tearing in the Earth's magnetotail, Adv. Space Res., 18, 267 (1996). Biichner, J., and J.-P. Kuska, Three-dimensional collisionless reconnection through thin current sheets: Theory and self-consistent simulations, in Proc. 3rd International Conference on Substorms, ESA S P 389, October 1996, 778 pp, 773 (1996). Biichner, J., and J.-P. Kuska, Numerical simulation of three-dimensional reconnection due to the instability of collisionless current sheets, Adv. Space Res., in press (1997). Coppi, B.G., Laval, R. Pellat, A model for the influence of the Earth magnetic tail on geomagnetic phenomena, Phys. Rev. Lett., 16, 1207 (1966). Galeev, A.A., and L.M. Zelenyi, Tearing instability in plasma configuration, Soy. Phys. JETP, Engl. Transl., 43, 1113 (1976). Harris, E.G., On a plasma sheath separating regions of oppositely directed magnetic field, Nuovo Cimento, 23, 115 (1962). Lemb~ge B., and R. Pellat, Stability of a thick two-dimensional quasi-neutral sheet, Phys. Fluids, 25, 1995 (1982). Lui, A.T.Y., Extended consideration of a synthesis model of magnetospheric substorms, in Magnetospheric Substorms, Geophys. Monograph Set., 64, p. 43, AGU Washington, D.C. (1991). Lui, A.T.Y., Current disruption in the Earth's magnetosphere: Observations and models, J. Geophys. Res., 101, 13067 (1996). Pellat, R., F.V. Coroniti, and P.L. Pritchett, Does ion tearing exist?, Geophys. Res. Left., 18, 143 (1991). Pritchett, P.L., Effect of electron dynamics on collisionless reconnection in two-dimensional magnetotail equilibria, J. Geophys. Res.,, 99, 9935 (1994). Pritchett, P.L., and J. Biiclmer, Collisionless reconnection in configurations with a minimum in the equatorial magnetic field and with magnetic shear, J. Geophys. Res., I00, 3601 (1995). Pritchett, P.L., F.V. Coroniti and V.K. Decyk, Three-dimensional stability of thin quasi-neutral current sheets, J. Geophys. Res., 101, 27,413 (1996). Quest, K.B., H. Karimabadi, and M. Brittnacher, Consequences of particle conservation along a magnetic flux surface for magnetotail tearing, J. Geophys. Res., 101, 179 (1996). Schindler, K., A theory of substorm mechanisms, J. Geophys. Res., 79, 2803 (1974). Vasyliunas, V.M., Magnetic field line merging, Rev. Geophys. Space Phys., 13, 303 (1975). Zhu, Z., and R.M. Winglee, Tearing instability, flux ropes, and the kinetic current sheet kink instability, J. Geophys. Res., 101, 4885 (1996).
This Page Intentionally Left Blank
MAGNETIC R E C O N N E C T I O N AND PLASMOID EVENTS IN THE M A G N E T O P A U S E BOUNDARY LAYER 1
1
1
Z. X. Liu , H. Zhang , T. Chen , Z. Y. Pu 2, and S. Y. Fu 2 1
Center for Space Science and Applied Research, Chinese Academy of Sciences, Beijing 100080, China 2 Department of Geophysics, Peking University, Beijing 100871, China ABSTRACT An important role of velocity shear in producing transient magnetic reconnection in the magnetopause boundary layer region is supposed. Two kinds of velocity shears may be there in the magnetopause boundary layer: the velocity shear of field-aligned flow occuring at the high latitude boundary layer region in the dayside magnetopause outside the subsolar point region or cusp region; and the velocity shear of transverse flow near the subpolar point region. The influence of velocity shear on magnetic reconnection processes has been investigated systematically in the magnetopause boundary layer, including: vortex induced reconnection(VIR), single X line bursty reconnection(SBXR), transverse shear flow rushed reconnection, stochastic reconnection, and local transient reconnection in the high latitude boundary layer under different IMF conditions. Finally, the plamoid events in the magnetospheric boundary layer are discussed. INTRODUCTION Magnetic reconnection processes produced by the interplanetary magnetic field (IMF) and the Earth's magnetic field near the dayside magnetopause are of vital importance in the energy transfer from the solar wind to the magnetopause. It was found that both steady-state reconnection and transient reconnection are important in the flux transfer events (FIEs) in the dayside magnetopause boundary layer region. At present time, several models have been suggested to explain the occurrence of FTEs, such as the multiple X line reconnection(MXR) model (Lee and Fu, 1985), the busty single X line reconnection (BSXR) (Scholer, 1988; Southwood et al, 1988), and vortexinduced reconnection (VIR) model (Liu et al., 1988). Each of these models can explain the formation and the significant properties of the FTEs under different conditions. However, at present time the available data are not sufficient to determine which model is the most appropriate. This paper is mainly to study the influnce of velocity shear in the transient reconnection in the magnetopause boundary layer, we suppose that the velocity shears may have
significant
influence on the transient reconnection in the magnetopause boundary layer. Because the solar wind flows around the magnetosphere, the solar wind flow mainly parellel to the magnetopause except in the sub-solar point region, and a rather strong velocity shear exists across the magnetopause boundary layer. 187
188
z.x. Liu et al.
There are two kinds of velocity shear in the magnetopause boundary layer: the field-aligned velocity shear which occurs in the magnetopause outside the sub-solar point region; and the transverse velocity shear which exists in the stagnation point region. In the following sections, the effect of these two kinds of velocity shear in the transient reconnection at the magnetopause boundary layers will be discussed. INFLUENCE OF THE FIELD-ALIGNED FLOW SHEAR ON THE VIR AND SXBR We know that the BSXR may take place when there is a local enhancement of anomalous resistivity, while VIR may be produced by the K-H instability caused by the velocity shear. Which reconnection models are appropriate if the velocity shear and local resistivity enhancement both exist in the boundary layer region? We recently studied this problem by a two-dimensional MHD simulation (Fu et al., 1995). In this simulation, the local enhancement of resistivity takes the form r / = r/0{1 + a e x p ( - ( x - x ~
+(Z-Zo) 2 lo2 )exp (-t)}'to
where 7/0 is the normalized background dimensionless resistance, t o is the decay time, a is the enhancement coefficient, and t is the time. The main results are summarized as follows: 1. If the velocity shear is weak and the value of c~ is large, the BSXR dominates the system, whereas if the velocity shear is relatively strong and a is smaller, the VIR controls reconnection process. Here the level of velocity shear is proportional to the field-aligned Alfven Mach number M A 9
2. There exists a critical value (Mc) of M A . For moderate Rm (magnetic Reynolds number) and R (fluid Reynolds number), BSXR and VIR occur repectively in ranges M A 3 (if cr =0) or MA 0.6 (if a =3). 3. Flow and magnetic patterns of BSXR can evolve as the VIR type with increasing time. For example, when a =3, MA=0.8 and 7/o =0.1, the initial structure is of the BSXR type, then the structure becomes VIR type with incresing time. 4. It is expected that near the subsolar point region where M A is very small, the BSXR type transient reconnection may lead to the formation of FTEs. 5. Outside the subsolar point region the VIR may became important if the FTEs are caused by BSXR near the subsolar point region. When FTEs move toward high latitude, the flow-aligned Alfven Mach number M A increases with increasing latitude, finally the structures of magnetic field and plasma of FTEs evolve into VIR type. INFLUENCE
OF
THE
TRANSVERSE
FLOW
SHEAR
ON
THE
TRANSIENT
RECONNECTION NEAR THE SUBSOLAR POINT REGION. There may exists transverse flow shear at the dayside magnetoshealth near the subsolar point region. What happens if the transverse flow shear impose on the magnetopause boundary layer
Magnetic Reconnection and Plasmoid Events in the Magnetopause Boundary Layer
189
near the subsolar region, we recently studied this problem by two-D MHD simulation (Chen and Liu, 1995). The main conclusions are summerized as follows: 1. Transverse flow shear may have an important influence on the transient reconnection. 2. The topologies and evolution of the flow and magnetic field depend on the strength of transverse flow shear and its lasting period. 3. The reconnection is very different for variant transverse flow shear levels. 9
When a transverse flow is homogeneous, magnetic fileds will not occur bent and reconnected, and the field lines are only pushed downstream.
9 9
If the flow has a weak shear, reconnection occurs in the magnetopause region. If the flow has a moderate shear, reconnection occurs first in the magnetoshealth and then in the magnetopause region.
9
When the transverse flow shear is strong,
reconnection occurs sequently in the
magnetosheath region and may appear in the magnetosphere region, while there are many small-scale stratified structures of the magnetic field in these regions. 9
The model of transverse shear flow rush reconnection(TSFRR) is different from other models,
9
TSFRR may be a possible generation machamism of FTEs near the subsolar point region,
such as MXR model, BSXR model and VIR model. and it can be used to explain the phenomena which occur near the subsolar point region. INFLUENCE
OF
FIELD-ALIGNED
FLOW
SHEAR
ON
THE
STOCHASTIC
RECONNECTION AT THE DAYSIDE MAGNETOPAUSE. Observations of ISEE1,2 found that there are some irregular small-scale structures in the dayside magnetopause current sheet. These small-scale structures may be caused by the stochastic reconnection. Recently we have studied the stochastic reconnection by using two dimentional MHD simulation (Fu et al., 1995;Pu et al., 1995). It was found that in an open system with strong field-aligned flow shear and large Reynolds number (R) and magnetic Reynolds number (Rm), stochastic reconnection can take place in the magnetopause current sheet when the interplanetay magnetic field(IMF) is southward. The stochastic reconnection can form irregular mesoscale and smallscale structures(ISSSs). The main simulation results are in the following:
1.
The evolution of the ISSSs for R = RM=400, and M A -
V~
-
20
.
Simulation results indicate
Lo that the K-H instability is firstly excited in the system, creating a mesoscale fluid vortex. As time goes on, a number of irregular mesoscale and small scale magnetic structures are thus produced.
v0
2. The development of IMSSSs strongly depends on the velocity shear (V-~-o) and magnetic Reynolds number Rm.
190
z.x. Liu et al.
9
If the velocity shear is weaker, the development of magnetic structure is slow; However, irregular mesoscale and small structures fastly appear for stronger velocity shear.
9
For the case of small Rm (Rm=80), only mesoscale structure is formed; more small-scale structure occurs for larger Rm. (Rm=800).
It can be concluded that the field-aligned flow shear plays an important role in producing the stochastic reconnection at the dayside magnetopause boundary layer when IMF is southward; the strong velocity shear is, the faster irregular small-scale structure appear; the stochastic reconnection may be produced by eddy diffusion. TRANSIENT RECONNECTION AT THE NIGHTSIDE MAGNETOPAUSE BOUNDARY LAYER. Interplanetary magnetic field(IMF) brought by the solar wind moves around the magnetosphere. Near the magnetopause, the magnetic field lines become bent and parallel to the magnetopause. In this case, there are reverse magnetic field regions at the dayside magnetopause and the nightside magnetopause when the IMF is southward and northward, respectively, so that, there exist reconnection regions at the dayside magnetopause and nightside magnetopause when IMF is southward and northward, respectively. There is a strong field-aligned flow shear at the High Latitude Boundary Layer(HLBL), which may have an important influence on the magnetic field reconnection in the HLBL. Recently, we have studied the transient reconnection in the HLBL region for different interplanetary conditions by using a 2-D compressible model (Liu et al., 1995, Zhang and Liu, 1995). Case 1. The IMF B z C.omponent Is Very Small. ( Biz ~ 0 )
In this case, magnetic field reconnection mainly depends on the direction of IMF X component Bix, for B~x is solarwind, i. e., Bix>0, reconnection occurs in the northward hemisphere. The simulation results show that a fluid vortex is first formed in the plasma mantle, which flows tailwards with increasing time, and a new vortex will be formed before the old one flows out of the simulation domain. Corresponding to the flow field, over the fluid vortex, magnetic fields are curved, and with time increasing, magnetic reconnection takes place, magnetic island and X point are formed. Magnetic island brought by the solar wind flows down to magnetotail. We have also simulated the density and temperature structure. It is found that the density and temperature have wave structures along X axis. High density is located in the center region of the island, this kind of islands can be regarded as a plasmoid. The high temperature is located near the X point. Case 2. The IMF Has a Northward Component When IMF is northward and B~x is earthward(Bix>0), at the beginning, the direction of the IMF and geomagnetic field in the HLBL region are the same. However, under the effect of the solar
Magnetic Reconnection and Plasmoid Events in the Magnetopause Boundary Layer
191
wind, the IMF lines are curved violently, a strong reversed magnetic field region is formed, where magnetic reconnection occurs, and the magnetic island and fluid vortex are produced. When B~x is tailward (B~x
