published in WGN, the Journal of IMO 27:6 (December 1999), pp. 267-285
The enormous number of 473 observers reported their Leonid observations to the VMDB enabling us to perform the most detailed activity analysis of a meteor shower ever. We are most indebted to the following observers for their contributions:
Ghazalaha Al-Abed (ABEGH, 5.95), Jasmel Acosta (ACOJA, 0.97), Ilidio Afonso (AFOIL, 3.00), Iyad Ahmad (AHMIY, 1.83), Seishi Akagi (AKASE, 17.50), Stephen Alden (ALDST, 4.00), Ahmad Al-Niamat (ALNAH, 5.00), Kohta Aoyama (AOYKO, 0.83), Rainer Arlt (ARLRA, 0.78), Kaori Asahara (ASHKA, 0.83), Joseph D. Assmus (ASSJO, 3.11), Zaid Ata (ATAZA, 5.00), Juan Alberto Aveledo (AVEJU, 1.20), Julia Babina (BABJL, 3.28), Pierre Bader (BADPI, 22.85), Halim Baituk (BAIHA, 2.30), Moshe Bain (BAIMO, 2.00), Lars Bakmann (BAKLA, 1.00), Igor Baluk (BALIG, 6.00), Petra Rendtel (BALPE, 10.71), Ana Bankovic (BANAN, 4.12), Rony Barry (BARRO, 0.53), Luc Bastiaens (BASLU, 5.96), Rizlane Bechar (BECRI, 1.67), Sanae Bechar (BECSA, 1.67), Luis R. Bellot Rubio (BELLU, 4.97), Pavel Belov (BELPA, 2.15), Vladimir Belchenko (BELVL, 2.55), Abdelaziz Bennouna (BENAB, 1.08), Orlando Benítez Sanchez (BENOR, 2.72), Felix Bettonvil (BETFE, 7.22), Stephen Binks (BINST, 2.98), Nicolas Biver (BIVNI, 2.22), Jim Blanksby (BLAJI, 9.33), Miroslav Blaho (BLAMI, 6.81), Boncho Bonev (BONBO, 1.00), Neil Bone (BONNE, 1.97), Michael Boschat (BOSMI, 4.00), Louisa Bowman (BOWLO, 10.11), Chris Briggs (BRICH, 10.13), Iwan Brukhanov (BRUIW, 1.98), Joana M. Brunet (BRUJO, 4.99), William Burton (BURWL, 0.97), Marija Cajetinac (CAJMA, 5.75), Arturo Carvajal R. (CARAR, 0.50), Jens J. Carlsen (CARJE, 7.15), Tal Carmon (CARTA, 0.04), Andrew Casely (CASAN, 1.00), Neophite Chanev (CHANE, 4.34), Yuk-lun Chan (CHAYU, 5.22), Qu Chengxu (CHEQU, 1.63), Koen Clement (CLEKO, 1.25), Antonio Coelho (COEAN, 3.50), Claudia Colonnelo (COLCL, 2.00), Matthew Collier (COLMA, 0.24), Tim Cooper (COOTI, 1.00), Pedro Correa (CORPD, 1.00), Uros Cotar (COTUR, 1.03), Camilla Coverty (COVCA, 9.14), Stefano Crivello (CRIST, 5.78), Hani Dalee (DALHA, 4.00), Luigi d'Argliano (DARLU, 2.64), Mark Davis (DAVMA, 7.50), Johan de Hert (DE JO, 1.50), Marc de Lignie (DE MA, 15.13), Goedele Deconinck (DECGO, 4.58), Sergey Dedik (DEDSE, 3.75), Benoit Dejust (DEJBE, 2.00), Kurt Dequick (DEQKU, 3.00), Vincent Desmarais (DESVI, 2.20), Peter Detterline (DETPE, 5.06), Asdai Díaz Rodriguez (DIAAS, 2.00), Anton Dimitrov (DIMAN, 2.14), Elena Dimovski (DIMEL, 4.35), Zhao Dongjuan (DONZH, 2.51), Dariusz Dorosz (DORDA, 2.08), Juan Jose Downes (DOWJU, 0.87), John Drummond (DRUJO, 2.50), Sergey Dubrowsky (DUBSE, 1.02), Jaroslaw Dygos (DYGJA, 20.00), Tonis Eenmae (EENTO, 3.07), Maurizio Eltri (ELTMA, 4.82), Frank Enzlein (ENZFR, 2.69), Frantisek Erben (ERBFR, 2.59), Bert Everaert (EVEBE, 5.16), Tomasz Fajfer (FAJTO, 1.00), Juan Gabriel Fernandez (FERJU, 2.00), Sharon Fletcher (FLESH, 3.00), Tamás Fodor (FODTA, 1.93), Anneleen Fransen (FRAAN, 4.95), Keiiti Fukui (FUKKE, 17.51), Nobuyuki Fukuda (FUKNO, 6.64), Siniti Fukuhara (FUKSI, 4.16), Ofer Gabzo (GABOF, 0.24), Atanas Gavrailov (GAVAT, 3.02), Christoph Gerber (GERCH, 17.17), Jaroslav Gerbos (GERJA, 8.50), Ivanka Getsova (GETIV, 3.52), Suchitra Ghosh (GHOSU, 5.00), Tom Giguere (GIGTO, 3.28), George Gilbart-Smith (GILGE, 9.08), Maarten Gillis (GILMA, 2.92), David Girling (GIRDA, 13.25), George W. Gliba (GLIGE, 3.25), Orly Gnat (GNAOR, 0.17), Shelagh Godwin (GODSH, 5.74), Amit Gokhale (GOKAM, 2.05), Sagar Gokhale (GOKSA, 1.03), Yeshodhan Gokhle (GOKYE, 3.68), Alexandra Golova (GOLAL, 3.28), Dennis Goodman (GOODE, 3.00), Prerana Gore (GORPA, 2.67), Roberto Gorelli (GORRO, 8.20), Lew Gramer (GRALE, 11.22), Valentin Grigore (GRIVA, 6.00), Matthias Growe (GROMA, 3.16), Gong Guanghui (GUAGO, 2.50), Monica de la Guardia (GUAMO, 4.36), Qin Guoming (GUOQI, 1.51), Cathy Hall (HALCA, 3.87), Michal Haltuf (HALMI, 0.17), Torsten Hansen (HANTO, 1.98), Hiromi Harada (HARHI, 7.83), Takema Hashimoto (HASTA, 23.32), Roberto Haver (HAVRO, 5.12), Alana Hawkens (HAWAL, 8.50), Kim Hay (HAYKI, 2.73), Steven M. Hayward (HAYST, 0.83), Amera Hemsy (HEMAM, 5.33), Santiago Hernández (HERSA, 0.56), Veerle Herrygers (HERVE, 1.18), Motoyasu Higuchi (HGCMT, 2.00), Nathalie Hontelé (HONNA, 2.81), Kamil Hornoch (HORKM, 3.39), Mitsuhiro Igarashi (IGRMT, 2.50), Isamu Iidsuka (IIDIS, 0.83), Oomi Iiyama (IIYOO, 3.50), Hiromi Imai (IMAHR, 0.83), Osamu Imamura (IMAOS, 2.17), Hikaru Ishida (ISDHI, 4.30), Iori Ishiyama (ISHIO, 1.66), Kaoru Ishii (ISHKA, 2.34), Megumi Isii (ISIMG, 3.00), Masaharu Ishizaki (ISZMS, 2.96), Akira Ito (ITOAK, 4.16), Daiyu Ito (ITODA, 9.58), Nobuhisa Itou (ITONB, 0.83), Shigeharu Ito (ITOSG, 1.67), Rositsa Ivanova (IVARO, 2.02), Shun-ichi Iwamoto (IWAS , 7.50), Yumi Iwasaki (IWSYU, 4.17), Hiroki Izumoto (IZMHI, 1.00), Kiyoshi Izumi (IZUKI, 14.96), Yumi Izuhara (IZUYU, 1.58), Helle Jaaniste (JAAHE, 3.35), Jan Janssens (JANJA, 5.50), Steve Jaworiwsky (JAWST, 1.50), Vibor Jelic (JELVI, 4.52), Ilhame Jemmah (JEMIL, 0.50), Simon Jenner (JENSI, 2.00), Carl Johannink (JOHCA, 16.35), Ivan Jokic (JOKIV, 2.20), Kevin Jones (JONKE, 6.35), Kurt Jonckheere (JONKU, 2.34), Wojciech Jonderko (JONWO, 1.09), Javor Kac (KACJA, 9.09), Primoz Kajdic (KAJPR, 2.09), D. Kalayda (KALDU, 3.33), Toshio Kamimura (KAMTO, 6.50), Dmitrij Karkach (KARDM, 3.28), Niladri Kar (KARNI, 3.72), Junichi Kasai (KASJU, 1.50), Jun Kataoka (KATJU, 1.00), Kenya Kawabata (KAWKE, 3.04), Satosi Kaya (KAYSA, 2.33), Katsuyuki Kobayashi (KBYKT, 0.50), Srdjan Keca (KECSR, 3.70), Ákos Kereszturi (KERAK, 3.57), Katarina Kerekesova (KERKT, 8.78), Stephen Kerr (KERST, 1.85), Noor Al-Khateeb (KHANO, 4.44), Mark Kidger (KIDMA, 1.50), Nobuya Kikuchi (KIKNO, 0.50), Kevin Kilkenny (KILKE, 3.21), Timo Kinnunen (KINTI, 1.00), Warwick Kissling (KISWA, 5.00), Shigemi Kanbara (KNBSG, 4.50), Toshiaki Kon-no (KNNTO, 1.65), André Knöfel (KNOAN, 17.53), Masata Kobayashi (KOBMA, 2.06), Wakaba Kobayashi (KOBWA, 13.00), Antoon Koegels (KOEAN, 0.83), Daniel Köhn (KOHDA, 1.46), Hideki Koide (KOIHI, 2.58), Khalil Konsul (KONKH, 5.50), Marcin Konopka (KONMA, 2.00), Detlef Koschny (KOSDE, 1.88), Nobuyuki Kosiyama (KOSNO, 3.00), Marija Kotur (KOTMA, 0.91), Jakub Koukal (KOUJA, 22.35), Zoran Kraljevic (KRAZO, 2.23), Nikola Kresojevic (KRENI, 5.55), W. Gary Kronk (KROGA, 6.50), Hideo Kusakai (KSKHI, 5.50), Tom Kucharski (KUCTO, 2.42), Brigitte Kuneth (KUNBR, 1.00), Werfried Kuneth (KUNWE, 1.00), Karimu Kuragaki (KURKA, 9.75), Takehisa Kuwabara (KWBTA, 0.83), Maciej Kwinta (KWIMA, 10.00), Sylvio Lachmann (LACSY, 1.15), Marco Langbroek (LANMA, 18.71), Zsolt Lantos (LANZS, 2.98), Alberto Latini (LATAL, 4.49), Trevor Law (LAWTR, 3.00), Anne-Laure Lebacq (LEBAN, 1.71), Adrian Lelyen (LELAD, 1.00), Anna S. Levina (LEVAN, 6.57), Semion Levin (LEVSE, 3.66), Mihir Limaye (LIMMH, 1.18), Alister Ling (LINAL, 1.38), Romulo Liporacci (LIPRO, 0.83), Vladimir Lukic (LUKVL, 9.20), Robert Lunsford (LUNRO, 16.40), Hartwig Luthen (LUTHA, 10.08), Kathy Machin (MACKA, 7.84), Kimio Maegawa (MAEKI, 2.17), Kouji Maeda (MAEKO, 0.77), Shane Majoros (MAJSH, 5.96), Mirjana Malaric (MALMR, 5.00), Katuhiko Mameta (MAMKA, 54.67), Roman Manak (MANRO, 4.47), Adam Marsh (MARAD, 18.85), David Martinez Delgado (MARDA, 2.01), José Alfonso dos Reis Martins (MARJO, 2.88), Khalid Marwat (MARKH, 2.48), Pierre Martin (MARPI, 4.65), Takuya Maruyama (MARTA, 0.67), Antonio Martinez (MARTI, 4.42), Tony Markham (MARTO, 6.49), Edgardo Ruben Masa Martín (MASED, 3.96), Yukihisa Matumoto (MATYU, 2.50), Alastair McBeath (MCBAL, 7.05), Bruce McCurdy (MCCBR, 1.38), Stephen McCann (MCCST, 0.23), Tom McEwan (MCETO, 0.69), Kieron McGrath (MCGKI, 3.00), Kevin McKeown (MCKKE, 1.00), Norman McLeod (MCLNO, 4.85), Lukas Mecir (MECLU, 0.13), Mark Mikutis (MIKMR, 15.20), Ana Milovanovic (MILAA, 0.91), Iris Miljacki (MILIR, 2.65), Vjera Miovic (MIOVJ, 1.17), Koen Miskotte (MISKO, 23.01), Shigeo Mitsuma (MITSH, 4.22), Hidekatu Mizoguchi (MIZHI, 4.29), Naoko Minigawa (MNGNA, 2.00), Amruta Modani (MODAM, 2.66), Marilena Molaco (MOLMA, 3.36), Sirko Molau (MOLSI, 14.17), Michael Morrow (MORMI, 4.00), Sigehiro Mori (MORSI, 1.58), Takako Mori (MORTK, 0.83), William Morgan (MORWI, 1.16), Erick Mota Perez (MOTER, 2.90), Ken-ichi Mohri (MOUKE, 1.55), Krzysztof Mularczyk (MULKR, 3.00), Darshan Mundada (MUNDA, 1.28), Minoru Muraki (MURMI, 21.39), Yoshikane Murakami (MURYO, 4.00), Joanne Muskarovsky (MUSJO, 13.38), Sin Nakayama (NAKSI, 7.76), Koji Naniwada (NANKO, 3.33), Tasso Napoleao (NAPTA, 3.00), Tatsunori Naruke (NARTA, 0.92), Toru Naruse (NARTO, 4.49), Sven Näther (NATSV, 12.35), Jonathon Newton (NEWJH, 6.00), John Newton (NEWJO, 9.00), Kevin Nicasi (NICKE, 3.67), Dalibor Nikolic (NIKDA, 2.46), Markku Nissinen (NISMA, 7.01), Prakash Nitsure (NITPR, 4.05), Yumiko Nobukiyo (NOBYU, 6.17), Jaroslaw M. Nocon (NOCJA, 1.00), Tooru Nishino (NSNTO, 0.50), Mohammad Odeh (ODEMO, 4.15), Ibrahim Odwan (ODWIB, 4.75), Eran Ofek (OFEER, 3.45), Hiroshi Ogawa (OGAHI, 1.50), Masamichi Ohno (OHNMA, 1.66), Hiroyuki Okayasu (OKAHI, 4.98), Masayuki Oka (OKAMA, 5.34), Dragana Okolic (OKODR, 5.38), Marcelo Oliveira (OLIMA, 2.50), Matt Orsie (ORSMA, 6.00), Elke Ortmans (ORTEL, 2.82), Kazuhiro Osada (OSAKA, 66.99), Kazuhiko Osaki (OSKKA, 6.67), Jesus Otero (OTEJE, 1.00), Fumio Oyama (OYMFM, 4.77), Ketan Pendse (PENKE, 1.33), Miroslav Penev (PENMI, 2.15), Alfredo Pereira (PERAF, 5.71), Dusan Perovic (PERDU, 3.59), Radame Perez (PERRA, 1.00), Suyin Perret-Gentil (PERSU, 1.52), Natasa Petelin (PETNA, 3.22), Furio Pieri (PIEFU, 4.73), Mila Popovic (POPMI, 1.20), Dubravko Potkrajac (POTDU, 1.08), Tushar Purohit (PURTU, 2.85), William Pyke (PYKWI, 4.75), Zhao Qingshan (QINZH, 1.92), Daniela Rapava (RAPDA, 7.09), Pavol Rapavy (RAPPA, 8.34), Simona Rapava (RAPSI, 5.85), Ina Rendtel (RENIN, 1.08), Jürgen Rendtel (RENJU, 21.68), Francisco Reyes Andrés (REYFR, 1.03), Janko Richter (RICJA, 0.88), Ian Rigney (RIGIA, 2.88), José Ripero (RIPJO, 2.68), Mileny Roche Lamas (ROCMI, 1.00), Beriozka Rodriguez (RODBE, 0.68), Francisco Rodriguez Ramirez (RODFR, 5.05), Juan Rodríguez (RODJU, 4.86), Dirk Rombouts (ROMDI, 0.83), Martin Rudd (RUDMR, 12.29), Jelyl Rufat (RUFJE, 1.00), Victor Ruiz Ruiz (RUIVI, 3.91), Etsuko Saito (SAIET, 4.00), Mitsue Sakaguchi (SAKMI, 14.53), Andy Salmon (SALAN, 1.00), Jeffery Sandel (SANJE, 9.53), K.V. Sankaranarayanan (SANKV, 2.50), Masanori Sano (SANMK, 0.83), Koetu Sato (SATKO, 6.45), Mikiya Sato (SATMK, 1.00), Tatuo Sato (SATTA, 6.75), Richard Schmude (SCHRI, 2.75), René Scurbecq (SCURE, 4.22), Harald Seifert (SEIHA, 1.07), Abderazak Sersouri (SERAB, 1.67), Miguel Serra Martin (SERMI, 2.66), Shashank Shalgar (SHASH, 4.03), Masumi Shimizu (SHIMA, 2.33), Brian Shulist (SHUBR, 3.10), Yasuo Shiba (SIBYA, 2.33), Hendrik Sielaff (SIEHE, 5.65), Hiroyuki Sioi (SIOHI, 6.16), Kazuaki Siotani (SIOKA, 2.42), Andrzej Skoczewski (SKOAN, 1.17), Vesna Slavkovic (SLAVE, 2.65), Keiko Shimada (SMDKE, 0.83), Alton Smith (SMIAL, 2.00), Amanda Smith (SMIAM, 9.76), James N. Smith (SMIJN, 15.26), Steven Smith (SMIST, 4.58), Makoto Shimamune (SMMMA, 0.83), Kiko Soares (SOAKI, 3.50), Andrey Solodovnik (SOLAD, 3.05), Manuel Solano Ruiz (SOLMA, 1.25), Je Soub-park (SOUJE, 2.59), Willian Souza (SOUWI, 0.50), George Spalding (SPAGE, 4.92), Ulrich Sperberg (SPEUL, 5.76), Yoshihisa Seshimo (SSMYS, 5.50), Mark Stafford (STAMA, 1.82), Octaaf Steen (STEOC, 4.39), Enrico Stomeo (STOEN, 2.12), Niko Stritof (STRNI, 3.04), Moomi Suen (SUEMO, 3.54), Kazuhiro Sumie (SUMKA, 11.33), Paul Sutherland (SUTPA, 1.67), Masafumi Suzuki (SUZMA, 1.00), Máximo Svárez Tejera (SVAMX, 4.21), David Swann (SWADA, 1.92), Eva Szabados (SZAEV, 0.90), Konrad Szaruga (SZAKO, 2.30), Richard Taibi (TAIRI, 4.42), Akihiro Takeda (TAKAK, 2.64), Hiromi Takai (TAKHR, 0.83), Masaaki Takanasi (TAKMA, 0.67), Mika Takanasi (TAKMI, 4.49), Keiko Tanaka (TANKE, 1.77), Sizuka Tanaka (TANSI, 0.50), Syoiti Tanaka (TANSY, 5.38), Lance Taylor (TAYLA, 1.40), Khaled Tell (TELKH, 10.47), István Tepliczky (TEPIS, 1.53), Kazumi Terakubo (TERKA, 1.00), Michiaki Taguchi (TGCMI, 0.67), Neelima Thatte (THANE, 5.90), Axel Thomas (THOAX, 5.66), Chiro Takahashi (TKHCH, 0.83), Ysohinori Takahashi (TKHYS, 1.66), David Todd (TODDA, 1.92), Masayuki Toda (TODMA, 5.58), Robert Togni (TOGRO, 4.95), Marko Toivonen (TOIMA, 3.50), Danilo Tomic (TOMDA, 2.40), Yasuhiro Tonomura (TONYA, 8.08), Michael Toomey (TOOMI, 2.49), Tamas Tordai (TORTA, 4.10), Hamid Touma (TOUHA, 1.17), Daniel Trainor (TRADA, 15.79), Gabrijela Triglav (TRIGA, 0.93), Josep M. Trigo Rodriguez (TRIJO, 1.07), Mihaela Triglav (TRIMI, 5.33), Aleksander Trofimowicz (TROAL, 3.00), Pawel Trybus (TRYPA, 1.00), Arnold Tukkers (TUKAR, 15.41), Megan Turner (TURME, 11.14), Shigeo Uchiyama (UCHSH, 4.58), Toshihiko Ueno (UENTO, 3.83), Akira Umetsu (UMEAK, 1.66), Tadasi Usui (USUTA, 1.25), Brigit van Opstal (VANBI, 3.64), Caroline van Dissel (VANCA, 8.00), Erwin van Ballegoy (VANER, 4.92), Glenn van Olmen (VANGL, 2.97), Hendrik Vandenbruaene (VANHE, 1.93), Jan Vansteelandt (VANJJ, 1.97), Michel Vandeputte (VANMC, 5.51), Steven van Impe (VANST, 4.11), Vishnu Vardhan (VARVI, 10.50), Trent Veitch (VEITR, 2.71), Cis Verbeeck (VERCI, 2.08), Daniel Verde (VERDA, 9.81), Jan Verbert (VERJN, 6.43), Ivaylo Videv (VIDIV, 2.30), Miquel A. Villalonga Vidal (VILMQ, 1.36), Myriam Vingerhoets (VINMY, 4.30), Catarina Vitorino (VITCA, 3.35), Helio Vital (VITHE, 3.00), Marija Vlajic (VLAMA, 2.40), Roger Vodicka (VODRO, 2.25), Björn Voß (VOSBJ, 8.60), Maja Vuckovic (VUCMJ, 1.17), Song Wanfang (WANSO, 7.80), Richard Ward (WARRI, 3.53), Anne van Weerden (WEEAN, 9.03), Thomas Weiland (WEITH, 4.41), Barbara Wilson (WILBA, 5.24), Vaya Willemen (WILVA, 1.50), Jean-Marc Wislez (WISJE, 3.46), Jeff Wood (WOOJE, 12.18), Noriko Watanabe (WTNNR, 2.66), Liu Xiaoyan (XIALI, 9.43), Zhou Xingming (XINZH, 0.90), Yasuo Yabu (YABYA, 0.99), Masayuki Yamamoto (YAMMA, 2.67), Yoko Yamashita (YAMYK, 0.83), Yoshio Yamada (YMDYS, 2.33), Hisamoto Yamaguchi (YMGHI, 5.50), Miyoko Yamamoto (YMMMI, 0.83), Motohiro Yamanaka (YMNMO, 10.00), Ruka Yamashita (YMSRK, 4.58), Yasuyuki Yonekura (YONYA, 1.67), Kazuko Yosino (YOSKA, 22.98), Noriko Yosimura (YOSNO, 0.50), Yosihito Yosida (YOSYO, 5.00), Kim S. Youmans (YOUKI, 4.00), Muneo Yoshizawa (YSZMU, 5.50), Chen Yu (YU CH, 2.37), George Zay (ZAYGE, 20.64), Su Zhiping (ZHISU, 2.00), Zhao Zhiheng (ZHIZH, 3.38), Jin Zhu (ZHUJI, 1.75), Florian Zschage (ZSCFL, 2.98), Tomasz Zywczak (ZYWTO, 15.00).
The observers come from the following 43 countries:
Australia, Austria, Belarus, Belgium, Brazil, Bulgaria, Canada, China, Croatia, Cuba, Czech Republic, Denmark, Ecuador, Estonia, Finland, France, Germany, Hong Kong, Hungary, India, Israel, Italy, Japan, Jordan, Kazakhstan, Marocco, the Netherlands, New Zealand, Pakistan, Papua New Guinea, Poland, Portugal, Romania, Slovakia, Slovenia, South Africa, South Korea, Spain, USA, UK, Ukraine, Venezuela, Yugoslavia.
The meteoroid stream of the Leonids is connected with the periodic comet 55P/Tempel-Tuttle which returned to perihelion in February 1998. The close encounter with particles directly behind the comet led to the prediction of a short-lived activity peak near the time of Earth's passage at the descending node of the comet's orbit. Forecasts were thus based on very young material ejected only two or three cometary revolutions ago. Indeed, such an activity peak occurred, almost unnoticed by the observers who expected a much stronger outburst, and thus "overlooked" the enhancement revealed by more detailed analysis. We will call this activity peak near the passage of the descending node of the parent Comet 55P/Tempel-Tuttle the nodal peak.
These two features will be superimposed by a very broad background component which is observable for at least 10 years near the Comet's perihelion passage. In the following, we will refer to these terms:
Newspapers reported that astronomers had miscalculated the peak time. In fact, it was 'bad luck' in a sense, that particles ejected many orbital revolutions ago caused the fireball peak which were not considered for predictions before the 1998 Leonids. Meteor activity from a stream component as young as the expected one is supposed to have a large mass index, (as well as a large population index), that is, a large proportion of smaller particles, and hence faint meteors, is expected.
In order to derive some population indices at the far ends of the activity graph, we reduced the last criteria to a total meteor number of 5 and a minimum true meteor number of 1.0 per magnitude class for the periods before November 16 and after November 18.
Figure 1 shows the overall profile of r in the period between November 14, 7h and November 20, 1h UT. Too few Leonids were reported outside these limits to yield meaningful population indices; in fact, these dates are beyond the activity period suggested by the IMO for visual observations. The two main features of this graph are the abrupt dip in r near lambda=234.5° and the characteristic peak of r near lambda=235.3°, close to the nodal passage.
The small error bars near these features suggest that even smaller bins could give information about the small-scale structure in the population index. Figure 2 shows the profile between November 16, 18h00m and November 18, 0h50m UT with finer bins.
Figure 1: Entire profile of the population index r of the 1998 Leonids.
Table 1:Bin sizes for the population index profile in Figure 2.
| Range in lambda | Bin width | Shift |
|---|---|---|
| 234.18°-234.60° | 0.04° | 0.02° |
| 234.60°-234.86° | 0.08° | 0.04° |
| 234.86°-235.12° | 0.20° | 0.10° |
| 235.12°-235.50° | 0.04° | 0.02° |
All these population indices have been calculated with the regression-line method which fits a linear function through the logarithmic, extrapolated true meteor numbers as a function of magnitude. Another method provides an even finer resolution of the profile and has been introduced in [2]. The average magnitude difference between the meteors and the stellar limiting magnitude is a unique function of the population index and can be converted into r, thereby enabling a lookup table to be constructed which involves the numerical integration of sample magnitude distributions. We present the population index profile near the nodal peak with a fine binning in Figure 3. Both methods provide comparable results, the method of mean-magnitude distances delivering slightly higher population indices. The method runs into problems during the bright-meteor maximum, since a substantial number of meteors brighter than magnitude -6 are all archived in the -6 class of the VMDB, since -6 is the brightest class stored. The average magnitude difference will be affected, whereas the regression line will not, unless class -6 is used. Therefore, we only give a highest-resolution graph for the nodal peak when r>1.5.
A significant triple maximum in the population index is evident: one of the maxima - the highest - coincides with the maximum in ZHR activity at lambda=235.338°±0.010°. The times, however, do not match exactly: The population index maximum time is 0.027° or 38 minutes after the ZHR maximum. A population index as high as r=2.3 is unusual for major-shower maxima (which generally consist of older ejections) indicating that very recently ejected material dominates the particle mass distribution. A considerable number of meteors from the bright-meteor component might also be incorporated into this measure of the population index; thus if we could separate the bright-meteor component from the nodal component, the population index might be even higher.
Figure 2: High-resolution profile of the population index r of the 1998 Leonids covering the period of highest activity.
Figure 3: Profile of the population index with even higher resolution than in Figure 2 as obtained from the average distance of meteor magnitudes from the limiting magnitude.
Figure 4: Entire ZHR-profile of the 1998 Leonids. Left and right margins of the graph correspond to November 12.6 and 21.6 respectively.
where lm is the stellar limiting magnitude, F is the correction for possible obstructions of the field of view, and h_R is the radiant elevation above the horizon.
It is non-trivial to find the optimum bin size for averaging a quantity like the ZHR. Large bins may suppress short-lived structures in the time series, while smaller bins may produce much larger error bars than the fluctuations they reveal, and the profile will be less reliable. Some examples of bin-size estimates include that of Brooks and Carruther (cf. [3]) with
k <= 5 log n,
where n is the number in individual ZHR values and k is the suggested number of bins. We may apply this relation to the part of the activity graph between lambda=234.46° and 234.62° (November 17, 0h20m-4h10m UT) where we obtained 544 individual ZHR estimates and plot our highest resolution in Figure 4. The bin-size relation allows a maximum of 14 classes. Sturges [4] gives
k = 1 + 3.32 log n
giving only 10 classes. Heinold and Gaede [5] suggest
k ~ sqrt(n)
giving the largest number of 23 bins. The ZHR profile in Figure 4 consists of 16 non-overlapping bins in the respective period which is certainly near the upper limit for resolving the small-scale structure of the shower. We find the clear maximum of the bright-meteor peak at lambda=234.528°±0.006° corresponding to November 17, 1h55m UT. This time and the peak ZHR of 357± 11 agrees well with the preliminary analysis in [1]. Given the fairly small error margins for the ZHR values, resulting from the large numbers of Leonids involved in each average, the variability of the activity appears significant. Clear sub-peaks can be spotted, such as at lambda=234.281°±0.010°, lambda=234.398°±0.010°, lambda=234.481°±0.005°, lambda~234.63°±0.01°, and lambda=234.700°±0.010°. Due to the variability of the ZHR-profile, the full width at half maximum (FWHM) is difficult to measure. It is found to be in the range 0.440°-0.565° corresponding to 10.5-13.5 hours. The individual sub-peaks in the ZHR-profile exhibit much smaller time-scale variations. The additional upper panel in Figure 5 shows the average limiting magnitude for each ZHR average; the absence of a clear correlation of individual ZHR peaks with sky conditions supports the physical reality of the variations. In 12 hours, the Earth travels 1.3 million km, but not perpendicularly through the stream. We thus get a smaller perpendicular extent of the bright-meteor component of about 380000 km - the distance of the Moon.
Table 2: Bin sizes for the ZHR profile in Figures 4 and 5.
| Range in lambda | Bin width | Shift |
|---|---|---|
| -232.00° | 2.00° | 1.00° |
| 232.00°-234.10° | 0.20° | 0.10° |
| 234.10°-234.46° | 0.04° | 0.02° |
| 234.46°-234.62° | 0.01° | 0.005° |
| 234.62°-235.35° | 0.02° | 0.01° |
| 235.35°-236.00° | 0.04° | 0.02° |
| 236.00°- | 1.00° | 0.50° |
Figure 5: Magnification of the 1998 ZHR-profile of the Leonids near the maximum characterized by low population indices. The upper panel shows the average limiting magnitude for each ZHR value.
Figure 6: Magnification of the 1998 ZHR-profile of the Leonids near the nodal maximum.
In order to produce a stand-alone profile of the nodal peak structure, we try to subtract the influence of the bright-meteor maximum (index 1) and the relatively weak background component (index 2) of more than a day duration. We compute a combination of two Gaussian profiles
ZHR = C_1 exp (-(lambda - lambda_1)2 / (2 sigma_12) )
+ C_2 exp (-(lambda - lambda_2)2 / (2 sigma_22) ),
where C_1 and C_2 are the amplitude of the two profiles, lambda_1 and lambda_2 are their centers, and sigma_1 and sigma_2 are the Gaussian standard deviations representing the width of the individual component. Without the background (index 2), we would not obtain a reasonable fit with only one Gaussian. Note that the two Gaussian profiles do not refer to the two maxima.
The fit ran from lambda=233° to lambda=236°, excluding the period of the nodal peak, lambda=235.00°-235.45°. We obtained C_1=225, C_2=66, lambda_1=234.567°, lambda_2=234.841°, sigma_1=0.216°, and sigma_2=0.764°. Figure 7 shows the ZHR profile in the respective period with the fit in the upper panel and the reduced ZHR profile near the nodal passage in the lower panel. The peak ZHR of the reduced nodal maximum is 80, and the skewness is retained. Neglecting the shoulder before lambda=235.25° we get a FWHM of 0.11° or 2.6 h. This time corresponds to a traveling distance of 280000 km and a perpendicular extent of the nodal peak - the thickness - of 82000 km.
Figure 7: Best fit of a sum of two Gaussians, representing the bright-meteor component and a weak background component, to the ZHR profile excluding the nodal-peak range lambda=235.0°-235.45° (top). Reduced nodal peak profile with the fit function subtracted (bottom).
We should recall that the maximum population index occurred 38 minutes after the ZHR maximum. High r-values of r>2.0 were observed up to 4 hours before the ZHR maximum, whereas a sudden decrease in r was observed less than 2 hours after the r-peak and 2.6 h after the ZHR maximum.
Statistical means to study the shape of distributions are the momenta
m_q = {sumiN (lambda_i-<lambda>)q ZHR_i / (sigmaq sumiN ZHR_i),
where lambda_i is the independent distribution variable, ZHR_i are the corresponding rates, <lambda> is the average solar longitude (not necessarily the highest value), and sigma is the standard deviation of the distribution belonging to <lambda>. The value of q determines the type of momentum.
The skewness of the nodal peak can be estimated through the third momentum (q=3) of the ZHR distribution versus solar longitude. The nodal maximum delivers a skewness of m_3=-0.15<0 confirming the right-weighted asymmetry in the profile; m_3>0 means a left-weighted profile. The bright-meteor maximum has m_3=+0.02 indicating a highly symmetric distribution of rates. The fourth momentum (q=4) measures the "excess" of the profile compared with a Gaussian distribution. A positive value means the profile is too steep, the maximum is too high, and the wings are under-represented. A negative value indicates a bump-like profile with too low a maximum and strong wings. We get m_4=-0.9 for the nodal peak indicating an excess of the wings. A triangular distribution has m_4=-0.6, a rectangular distribution has m_4=-1.2. These momenta were applied to the reduced ZHR profile, after subtraction of the bright-meteor background affecting the nodal peak. The bright-meteor component has m_4=-0.9, probably because of its roughly bimodal structure at maximum.
The error bars were obtained using error propagation of the Poissonian error of meteor numbers and the error of the population index which affects the flux density not only through the ZHR, but also through the the standard collecting area and the extrapolation to the true number of meteors. The striking feature of the flux density graph in Figure 8 is the completely reversed amplitude of the peaks: The fireball maximum is less pronounced than the nodal peak. This is an observational effect. The monitored area increases with zenith distance of the field, and thus an observer could see more meteors at low elevations; but the distance to these meteors increases with the zenith distance of the field, and the meteor intensity decreases with the square of this distance. The question whether the increased monitoring area balances with the magnitude loss is controlled by the population index. During the 1998 fireball maximum these effects combined to produce a huge excess of meteors in fields not towards the zenith - the actual number of particles per unit area, however, was lower than during the nodal peak. In fact, many observers reported that it was their impression that looking towards low elevations was best for detecting the greatest number of meteors.
Different populations constitute different fractions of the full sample of particles. The upper panel of Figure 8 shows the flux density for particles causing meteors brighter than magnitude +6.5 which corresponds to a mass of 2.2.10-5 g according to the conversion
m = 40 - 2.5 log (2.732.1010 M0.92 V_G3.91)
given in [7] and [8] where m is the meteor magnitude, M is the mass in grams, and V_G is the geocentric velocity of the meteoroids. The fireball maximum is much more prominent in the flux if larger particles are exclusively considered. The nodal peak is still visible in the graph of the flux for meteors up to medium magnitudes. It is completely absent when examining fireball-class events alone.
A second feature is the varying peak time with varying mass limit. Large particles have their highest flux earlier than smaller particles within the fireball maximum. The difference in peak time as derived from the top and bottom panels of Figure 8 is 0.11° corresponding to about 2.6 hours. The fact that the population index has a minimum an hour before the bright-meteor maximum and climbs gradually during times of highest activity (as already presented in Section 3) foreshadowed this apparent mass sorting.
Figure 8: Flux density profiles of the 1998 Leonids for various particle populations; particles causing meteors brighter than magnitude +6.5 (>2.2.10-5 g, top), than magnitude +2.7 (>1 mg, middle), and -4.2 (>1 g, bottom).
Figure 9: Range of total ejection velocities as a function of the Comet's distance from the sun and beta, the ratio of solar radiation pressure force to the sun's gravitational force. All ejection velocities lie in the band between the two curves. Negative distances mean pre-perihelion positions. Note that the abscissa skips distances closer than the perihelion distance of the Comet.
Briefly, the basic procedure consists of generating a suite of test particles close to each perihelion passage of Tempel-Tuttle and following each of these through to the epoch of interest. The "daughter" Leonids are created through random ejection on the sunward hemisphere of Tempel-Tuttle and are distributed at random in true anomaly along the cometary arc inside 4 AU. The osculating elements for Tempel-Tuttle at perihelion are taken from Yeomans et al. [10]. A total of 10000 test meteoroids with density 0.8 g cm-3 are ejected in each decadal mass interval from 10 g - 10-5 g, for a total per perihelion passage of 70000 test particles. This corresponds to values of beta from 5.10-5 - 5.10-3 approximately separated by factors of two between each of the 7 categories. Here beta refers to the ratio of solar radiation pressure force to the sun's gravitational force, i.e. beta=F_r/F_grav. This procedure is repeated for each of the last 58 perihelion passages of the comet so that the complete "run" consists of just over 4 million test particles.
After the initial conditions are specified in this way, each test particle is numerically integrated forward from ejection to the epoch of interest and followed until it reaches its descending node (the only point along its orbit at which it might possibly be observable from the Earth) and its Keplerian elements at the time of nodal passage are stored.
For this model we adopt a bulk meteoroid density of 0.8 g cm-3 over all masses and release meteoroids with velocities following the standard Whipple ejection routine [11], modified to follow an r-0.5 heliocentric velocity dependence. More details can be found in [9]. Figure 9 shows the range of total ejection velocities as a function of distance from the sun and beta.
Figure 10:Comparison of observed ZHR with a modelled "ZHR" scaled to the maximum of the observed profile.
The final particle distributions are chosen such that the test meteoroids have nodal passage times within 1 week of the peak of the shower in 1998 (this is equivalent to taking a slice of 0.2° width in mean anomaly along the stream orbit) and nodal distances within 2.10-3 AU of Earth's orbit. All Leonid test particles which meet these two conditions are then binned in 0.02° intervals of solar longitude (approximately 30 minutes). As the majority of the recently ejected Leonid "streamlets" (over the last 3-4 revolutions) are primarily more than 0.005 A.U. inside the Earth's orbit, we can adopt such a large sieve size to enhance the statistical sample for the older ejections which dominated in 1998 (see below).
However, the total final summed particle distributions found in this manner are not a true reflection of a ZHR-like value as all masses have been weighted equally, whereas some power-law distribution at the source must be assumed. For simplicity, we have used a mass index at the source of 1.4. The choice of this value heavily influences the final distributions and we emphasize that what follows is not a unique solution. We note that such a low value for the mass index at the source implies that we are giving more weight at the source to the large meteoroids in the distribution, as was observed during the 1998 return.
The final activity found in this manner is then scaled to the same peak ZHR as observed to compute the resulting shapes of the two curves over the solar longitude range 234°-235°. This is shown, along with the ZHR values discussed previously, in Figure 10. The basic shape and profile of the two curves is remarkably similar over the range 234.2°-234.9° (where good observational data exist). The simulated results are still not of a fidelity to permit detailed comparison with all the fine structure in the ZHR curve even assuming all are real features of the stream. However, two notable features are visible in each curve - the good agreement of the location of the peak activity and a plateau visible from 234.58°-234.68° in both. As well, the theoretical profile falls off more sharply than is observed, particularly after 234.8°.
Figure 11: Ejection years contributing to the 1998 activity profile. Test particles are counted as Leonids in 1998 if they have nodal passage times within 1 week of the shower maximum (at 235.3) and nodal distances within 0.002 AU of Earth. Here only ejected meteoroids with masses greater than 0.1 g are included.
Figure 11 shows the age distribution of material composing the stream over this interval. We note that the single-most prolific ejection era contributing meteoroids in 1998 was the 1333 passage of 55P/Tempel-Tuttle. This has been previously noted by Asher, Bailey and Emel'yanenko [12] and was ascribed to trapping of large Leonids in the 5:14 resonance with Jupiter. However, unlike the conclusions from [12], we note considerable (almost equal) contributions from many passages within a few revolutions on either side of the 1333 ejection - particularly from 1167. The reason for the difference in final distributions between those found by [12] and here might be that the results from [12] were confined to ejection at perihelion, whereas we extend these ejection locations over almost all of the cometary arc where significant ejection activity might be expected. Unsurprisingly, removing the constraint that ejection must be precisely at perihelion allows a much greater range of initial meteoroid ejections to be "visible" in 1998. Using this model, we may conclude that the 1998 shower was composed of material which was 500-1000 years old, but cannot confine the exact origin more precisely than this. This suggests that the fireball peak was actually composed of a series of filaments from different ejection epochs, though it is unclear that these would be necessarily separable in terms of ZHR activity as most of these ejections end up spread over nearly the same range of solar longitudes, whereas the ZHR profile shows a variety of submaxima.
Noteable also in this Figure is the drop in the numbers of accepted Leonids prior to the 833 A.D. epoch. A partial explanation for this may lie in the fact that 55P/Tempel-Tuttle did not enter the 5:14 resonance until 700-800 A.D. Material ejected prior to this epoch would not have such easy access to the 5:14 resonance and thus show lower transfer efficiencies at the present epoch (unlike the material from 900 A.D.-1500 A.D.).
If we examine the age of material across the profile from 234°-235°, we find, in general that the material tends to become slightly older as we move to larger values of solar longitude. Figure 12 and 13 show the normalized age distribution of material over the interval from 234.2°-234.9° in steps of 0.1°. The rising portion of the curve and the region around the peak are most populated by very large meteoroids ejected at the few passages of Tempel-Tuttle on either side of 1333. At the point where the plateau occurs in the theoretical and observed profiles near 234.6°, the dominant population switches to become centered around the 1167 passage of Tempel-Tuttle and moves (relatively speaking) toward smaller meteoroids. The descending portion of the profile is richer in smaller meteoroids and material of slightly older (the 1167 A.D. and a few nearby passages of the comet) origin. This change in origin may also be responsible for the increase in r starting at 234.6°. Figure 14 shows the distribution of accepted particles as a function of solar longitude.
Figure 12: Contributions of particles from various ejection years. Each of the graphs gives the perihelion epoch contributions for a 2.4-hour period of the activity profile.
Figure 13: Contributions of particles from various ejection years for more bins of the activity profile.
Figure 14: Particle number density as a function of solar longitude and particle mass. The sum of values for all particle masses gives the activity profile. The maximum of the largest particles is found before the activity maximum.
The nodal peak observed at 235.3 is almost completely lacking in these modelled results. Previous work (see [14]) suggests that material observed near this solar longitude range is 3 revolutions old at most (older ejections would tend to peak at different solar longitudes). We note from [13] and [14] that the center of these recent streamlets were approximately 0.005 A.U. interior to the Earth's orbit at the time of the 1998 shower. As a result, any particles from these ejection epochs which reached Earth had significantly larger nodal distances than did the average Leonid with low ejection velocity and modest size (beta<10-3). From [14], ejection velocities in the range 15-20 m/s (with a significant transverse component) can produce an increase in the nodal radii (ignoring planetary perturbations) of more than 0.005 AU for ejections within a few tens of degrees in true anomaly of perihelion. As well, radiation pressure forces on very small particles (with beta ~ 10-2) can produce an increase of comparable magnitude in the nodal radii. As a result, it is reasonable to expect that the smallest meteoroids associated with the last few revolutions of the comet (namely, 1965, 1932, and 1899) would be visible (albeit in small numbers) near the current nodal longitude of the comet in 1998. From [14], the nominal ejection epoch with the greatest contribution is 1932 for this solar longitude interval (exclusively small particles), but the numbers are so small (roughly 0.01% of the initial sample) as to make the determination questionable. In summary, all we may conclude is that very small, high ejection velocity, high beta particles from some (or all) of the 1965, 1932, and 1899 ejections contributed to the nodal peak in 1998. For somewhat older trails ejected 4 to 7 revolutions ago (such as those given in [13]) with expected peaks in solar longitude between 235.6-235.9, no obvious features were visible in the 1998 ZHR profile.
[1] Arlt, R., Bulletin 13 of the International Leonid Watch: The 1998 Leonid Meteor Shower. WGN 26 (1998), pp. 239-248
[2] Arlt, R., Global Analysis of the 1998 Perseid Meteor Shower. WGN 27 (1999), pp. 237-249
[3] Sturm, R., Wahrscheinlichkeitsrechnung, mathematische Statistik und statistische Qualitätskontrolle. Fachbuchverlag Leipzig, 7th ed., 1988, p. 86
[4] Sturges, H.A., The choice of a class interval. J. Amer. Statist. Assoc. 21 (1926), pp. 65 and 66
[5] Heinold, J., Gaede, K.-W., Ingenieur-Statistik. Oldenbourg, München-Wien, 1972 (3rd edition)
[6] Arlt, R., Global Analysis of the 1997 Perseids. WGN 26 (1998), pp. 61-71
[7] Verniani, F., J. Geophys. Res. 78 (1973), p. 8429
[8] Hughes, D.W., P/Halley dust characteristics: a comparison between Orionid and Eta Aquarid meteor observations and those from the flyby spacecraft. Astron. Astronphys. 187 (1987), pp. 879-888
[9] Brown, P., Jones, J., Simulation of the Formation and Evolution of the Perseid Meteoroid Stream. Icarus 133 (1998), pp. 36-68
[10] Yeomans, D.K., Yau, K.K., Weissman, P.R., The Impending Appearance of Comet Tempel-Tuttle and the Leonid Meteors. Icarus 124 (1996), pp. 407-413
[11] Whipple, F.L., A comet model. II. Physical relations for comets and meteors. Astrophys. J. 113 (1951), pp. 464-474
[12] Asher, D., Bailey, M.E., Emel'yanenko, V.V., Resonant meteoroids from Comet Tempel-Tuttle in 1333: the cause of the unexpected Leonid outburst in 1998. Mon. Not. R. Astr. Soc. 304 (1999), pp. L53-L56
[13] McNaught, R., Asher, D.J., Leonid Dust Trails and Meteor Storms. WGN 27:2 (1999), pp. 85-102
[14] Brown, P., Evolution of two periodic meteoroid streams: The Perseids and Leonids. PhD Thesis (1999). pp. 171-258