Five wild years - Reminiscences of the Leonids experience 1998-2002

By Daniel Fischer, Königswinter, Germany

This is not a scientific review of the surprises, discoveries and sensations the Leonids brought from 1998 to 2002, but a look back by one observer (and science writer) who often witnessed first-hand what went on in the sky - and also how science finally got its grip on the elusive and striking phenomenon of meteor storms. It was a truly an experience with a deep impact (no pun intended) that is not likely to repeat ...

I admit it: Before about ten years ago I had considered meteor observing, especially with your own eye, as one of the most useless branches of (amateur) astronomy, with no serious data reduction possible and dubious results. This was not true, at least at that time, of course, but the progress made by the IMO had not been widely known to the outside world. Not interested at all in systematic meteor watching, I was struck by the idea of meteor storms nonetheless: I had read a lot about the fabulous Leonids storm of 1966 and eagerly devoured two big reviews of meteor storms in British journals. Beech et al. [1995] had described these rare phenomena as both spectacular and little understood, while Mason [1995] predicted - using a crude model - storms of the Leonids with ZHRs around 5000 in 1998 and 1999.

Would it be worth to chase after those possible storms? I was already travelling a lot for astronomy's sake, both in the quest for eclipses (1983, 1988, 1990, 1991, 1992, 1994, 1995 and so on) and comets (Halley in 1986), but for meteors? Then again, Brad Schaefer, the American astronomer famous for his professional work on astronomy with the naked eye, had told me in 1988 that for him one meteor storm would be "worth ten eclipses." And I had been going on a moderately extreme expedition for a meteor outburst once before, to France for the Perseids of 1993: When they approached their marvellous peak at dawn with a ZHR of almost 500, there were clusters of several meteors within just one or two seconds, and they caused a fleeting "storm feeling" indeed. To witness that, going on for many minutes, might actually be worth a substantial effort. Eight time zones towards the East, in fact ...

The Fireballs of Mongolia

That was to be the place for the astronomical event of the century? Whole window panes were missing. The single light bulb on the ceiling was dangling freely, all shades long gone. Instead of a light switch there were two pieces of metal carrying the full 220 Volts which you had to bring into contact with your bare hands. And there was no liquid water whatsoever. So this was the famous "Hotel Edelweiss", the guest quarters of Mongolia's only professional astronomical observatory "Khurel Togoot", some 11 km Southeast of the capital Ulaanbaatar. It must have been a state-of-the-art site once - back in the late 1950's and early 1960's, that is, when the socialist "brother nations" from the Comecon had donated the buildings and half a dozen telescopes to their poor Far-Eastern partner. [...]

It had been evident from the first minute we had arrived here on Nov. 15, 1998, that this was a strange place. We, that was "ALEX'98", 14 amateur astronomers from all over Germany, led by Jürgen Rendtel, president of the International Meteor Organization (IMO), among us many veteran visual, photographic and video meteor observers. We were to join a similarly big professional Canadian expedition here, which in turn had teamed up with the United States Air Force. Upon our arrival at the place we were immediately shown the way to the bar and the disco - but no one ever bothered to show us even one of the astronomical instruments.

These were some of the first impressions of the place where we were waiting for the first meteor storm in decades [Fischer 1998]. The Canadians as well as our group (in which I had ended up more or less by accident, after chatting with an IMO member at a conference) and other amateurs that joined in later were mainly interested in the phenomenon as such while the USAF group was part of the first-ever real-time network that would warn satellite operators in case the meteor rate would reach dramatic levels. The weather conditions were mostly fine around the critical day, the equipment - both ALEX and USAF were using numerous intensified video cameras but doing visual counting at the same time - was set up around the "Edelweiss". And so it began :

[T]he first night of observations - Nov. 15 to 16 Mongolian Time = the evening of Nov. 15 UTC - lived up to expectations in that hardly any meteors were sighted. This was what everyone had predicted, and actually most also believed that there wouldn't be many meteors the next night. That would be exactly 24 hours before nodal crossing, and the general expectations were that any background activity from older Leonid particles would only start to ramp up 12 hours before the crossing. How wrong the experts could be sometimes ...

Usually we would go to bed around 8 p.m. to get up again around 2 a.m. when Leo and the radiant would already have risen (the Lion's head came above the horizon almost exactly at midnight local time, which is 8 hours ahead of UTC). But tonight there was no need for an alarm clock: Those observers who had gotten up early made sure no one overslept tonight. Something totally unexpected was going on: We were literally under fire from a hail of large meteoroid particles. There was at least one meteor visible every minute - normally no big deal, but all of them were very bright. Not one was faint.

This was a most unusual brightness distribution: Normally you have many more fainter than bright ones, and we would have seen the faint ones easily (the limiting magnitude was always better than +6m). But now they were all brilliant - and some strikingly so. One, for example, had about the brightness of the full moon and exploded right in the zenith: Its train (the ionized trail of atmosphere where the particle burnt up) kept on glowing for many minutes, and you could see it being distorted by the winds in the upper atmosphere. At times there would be several meteor trains visible in the sky at the same time - something exceedingly rare, as you normally have only a few significant trains during a whole night. Who cared that it was -30° Celsius or below?

With Leo and the radiant climbing higher and higher, the number of meteors per hour climbed, too. This was not simply due to the improving geometrical conditions: The meteor rate was really rising fast, as the IMO analysis would later show - the peak rate wouldn't be reached until 1:30 UTC = 7:30 Mongolian time. But we were in a good spot nonetheless: A single observer in Mongolia could easily see 100 Leonids, all rather to very bright, in one hour around 6 a.m. on Nov. 17th (22:00 UTC on Nov. 16th).The ZHR at the time was around 200, according to the global analysis. As usual, there would be lulls in activity and then rapid fire from the sky, like 5 meteors within less than half a minute.

Now imagine 5 meteors within a second - some optimistic models had thought that possible 24 hours from now. But would it happen after all? The models had been dead wrong about the ramping up of the background particle component we had seen so spectacularly tonight (and that the Europeans were just now starting to experience - if they had clear skies and had bothered to observe one night "early"). There was a distinct possibility now that the whole prediction business was off and that the "real" maximum the next night could be a dud. There was only one way to find out...

How it went on is well known now, of course: only a moderate sprinkle of faint meteors was seen from Mongolia or anywhere else during the supposed night of the storm. Numerous guests as well as journalists had come up to the Khurel Togoot Observatory, only to learn - after standing around in the cold for many hours or rather having some drinks, courtesy of the American embassy, that the storm apparently wasn't coming at all:

At 2:20 Mongolian time (18:20 UTC) Col. Worden broke the bad news to the press: "We're not seeing any increase that would indicate we'll have a major storm in the next few hours," he stated and suggested that one might as well go home. "Do you have any idea what happened?" someone inquired - Worden: "No!" But let there be no mistake: Last night's "very very strong bright shower of fireballs," according to Worden, "was probably one of the more impressive fireball shows on record." For him personally "that's the most impressive thing I've ever seen in the sky."

Even without the storm, the expeditions were clearly a success, and the Night of the Fireballs in particular had yielded some of the best persistent train videos ever recorded. The attempt to determine the meteoroid flux in real-time by the Canadians and USAF was also in full swing, with the video signals being scrutinized by both a rudimentary automatic meteor detection system and a human observer, while another observer was doing visual counts directly in the sky. Every 15 minutes the results were summarized and phoned through to a data center halfway around the world:

A typical message going out via satellite telephone would sound like this (an actual example from the morning of Nov. 18): "Mongolia Observatory. We have counts of 11 for automated detection, 4 for TV and the human observer, limiting magnitude was 8.0, the visual observer outside was 22, and the limiting magnitude now is 6.0 - over." Canada would read back the numbers, and that would be it for the next 15 minutes - satellite time is expensive. From the Canadian center the information - now transformed to ZHRs - would eventually be passed on to the 55th Space Weather Squadron at Schriever AFB in Colorado Springs, Colorado, U.S.A., which would have warned satellite operators in case of a real storm brewing. None had been, of course, and no obvious satellite anomalies were reported either.

The 12 months between the fireballs and the non-storm of 1998 and the 1999 return of the Leonids were anything but dull: In a fast pace several key discoveries were announced. During the fireballs Croation physicists - also in Mongolia but in an even more remote location - had for the first time recorded electrophonic sound during two meteors. Just after the meteors had peaked, a diffuse cloud of glowing sodium was discovered in the sky which turned out to be a sodium tail of the Moon, enhanced by the Leonids impacts on the Moon and seen from the inside, looking towards the tail end. And, most importantly, papers came out that explained where the fireballs had come from, why there had never been a chance for a storm in 1998 - and which predicted that there would be storms in 1999, 2001 and 2002 [McNaught & Asher 1999].

Drawn to Jordan by a model

I cannot remember when I started to really trust these calculations, but while I was still pondering the evidence (mainly in the most readily available papers by David Asher and Rob McNaught), a surprise invitation arrived from Amman, Jordan. As an e-mail, a fax and a letter: The Jordanian Astronomical Society (JAS) would be a Leonids conference in Jordan it said, plus a dedicated observing camp in the desert - and all expenses (except for the airfare) would be paid! I had been invited, I learned, because JAS web expert had liked by account of the Mongolian experience. And Jordan would be the natural choice if one believed in the dust trail models that gave the Middle East the highest possible storm rates, an hour before dawn, with the radiant high in the sky.

In early November 1999, I summarized the insights gathered so far in an article for my German space newsletter [Fischer 1999] - unfortunately most other journals had not grasped the theoretical progress in time to advise their readers properly. "The theoreticians seem to have made their homework," I wrote: "The origin of the fireballs is now understood precisely - and there are now detailled models for the spatial distribution of comet Tempel- Tuttle's dust, which seem to have transformed Leonids prognosis into an almost exact science." Soon I would learn whether these rather bold assessment would hold true. And for this experience, I also found out, I would be joined by the two best-known theorists themselves, as David Asher and Rob McNaught were also following JAS' invitation! Our choice was a good one:

Some had stopped counting alltogether now and were just gazing mesmerized at the sky on fire, others kept shouting frantically into their tape recorders, trying to preserve a permanent scientific record. But how do you cope with seven meteors flashing over all parts of the sky at the same instant, then another burst and another one? Whereever you looked now, be it close to the radiant in the sickle of Leo, be it at the North Celestial Pole or anywhere around the horizon, you didn't have to wait for more than a second, and there was another streak of light flashing through the sky.

Looking close to Leo, the constellation already high up in the sky, you could now really see the meteors streaming towards you from one spot - now you knew why scientists call it the "radiant". At times there would be several meteors 'radiating' there within a fraction of a second in different directions, forming a cosmic fireworks display. If you, on the other hand, looked far away from the radiant and close to the horizon, e.g. towards the Hyades, Pleijades and the setting planet Saturn, the meteors were flying downwards in parallel, almost like a curtain in front of the stellar canvas.

This was not a dream or a planetarium simulation. This was the real thing - seen, photographed and videotaped on the morning of November 18th, 1999, at the First International Astronomical Camp of the Jordanian Astronomial Society or JAS, at the former Hamzeh oil field deep inside Jordan's vast Eastern Desert near Al Azraq. [...] Initially the JAS had just planned to invite maybe 3 or 4 foreigners to join them for the occasion, but somehow the event had grown and grown - and now Jordan and the Al Azraq camp in particular had become one of the centers of the 'Leonid Universe', mentioned directly or indirectly ('50 astronomers freezing in the Jordanian desert') in most news reports about the meteor storm.

Thus begins the still-fresh account [Fischer 1999a] of the Great Meteor Storm of 1999 - which had peaked within 5 minutes of the predicted time, but with a rate 5 to 10 times greater than most had anticipated (McNaught e.g. had reduced his prediction to just 500 per hours just a few weeks ago). With a ZHR near 5000 at least in the Middle East where the observing conditions were perfect, this storm would remain the most impressive one of all during the 1998-2002 return of the Leonids, and just watching the rate climb and climb had been simply breathtaking:

[T]he Leonids rate in the sky had grown so much that one started to get the feeling of actually racing towards the constellation Leo. Towards 3 a.m. the shower was already far more impressive than the strong Perseids of 1993 had been. The meteors now often came in pairs or triplets, travelling on nearly identical tracks, within sometimes less than one second. Soon the meteor rate for a single observer would surpass 5, then 10 a minute. The meteors were everywhere, regardless of whether you looked towards the radiant or anywhere else in the sky. [...]

Soon there are 7, then 13, then 27 meteors a minute - all counts for a single observer, of course, with his 110 degrees or so field of vision. Of course the temporal distribution of the meteors is irregular: Statistical fluctuations and perhaps real 'gusts' of activity bring several meteors per second in one time interval and leave gaps of several seconds without any meteor. But with the rate climbing on, those intervals get rarer. By now many are just marvelling at the rain of meteors while others count on - and Moh'd Odeh alone records up to 91 meteors in one minute (followed by 86 in the next minute, while before and after these 2 minutes the rate was in the 50s). [...]

The numbers alone, 5000 or 3500 per hour, cannot convey the visual drama, however. It was better expressed by the frenetic shouting from some of the more enthusiastic Jordanian meteor counters - who still tried to record the brightness of individual meteors as well. Calls of "Wahat!" (1st magnitude), "Arba'a!" (4th) or "Chamzeh!" (5th) filled the air - the fainter meteors were now in the majority, any fireballs were missing, and there were only a few persistent trains of note. But the sheer number of meteors made more than up for their faintness (that was no problem anyhow for our dark site), and the maximum rate held for several minutes, with a marked decrease evident only after 4:15 a.m. (while the percentage of bright meteors increased somewhat).

Apart from the stunningly high peak rate (about which no one complained, of course :-) the Leonids of 1999 were a total triumph - for a science that was less than one year old (although the general concept of dust trails had been mentioned in isolated papers as early as in 1899, one would learn eventually):

It was a night that made history: On the morning of November 18th, the Earth experienced the first meteor storm since 1966, one of only four this century - and the peak struck exactly on time, just as predicted by the new dust trail model by David Asher and Rob McNaught. Both astronomers (and yours truly as well) were among the guests of the Jordanian Astronomical Society for a unique meteor conference and an observing camp deep in the desert and could experience the triumph of their work with their own eyes: Meteor stream prognosis has matured into an exact science!

But while the time of the maximum was hit spot-on by the model and the expectation that most meteors would be faint became true also, the surprising strength of the outburst was not foreseen - the zenithal hourly rate (ZHR) grew to roughly 10 times the anticipated value, peaking at more than 5000 in 5-minute intervals around 2:04 UTC. Also the time profile of the storm and the shower in general from Nov. 17 til 19 was more complex than expected, with a secondary storm peak of about ZHR 3500 ten minutes before the main peak and a broad shallow maximum with a ZHR in the low 100's 15 hours later (the latter best seen from China). [...]

For the 2nd year in a row a major international observing campaign had been put together to a) provide "situational awareness" for satellite operators (who would have taken action to safeguard their spacecraft in the case of a major storm) and to b) get the most science out of the rare Leonids activity. While in 1998 NASA and a joint Canadian astronomers/USAF group had gone different ways, this time they had joined their forces. Observers with video and radar equipment on more than half a dozen ground sites around the globe and an two aircraft reported their observations to a "Leonids Environmental Operations Center" at NASA's MSFC in Huntsville.

The scientific part of the operation, esp. the observations from the aircraft, seems to have gone very well, and lead scientist Peter Jenniskens was elated. But something went seriously wrong with the situational awareness part: The number for the maximum meteor rate the LEOC provided, 1688, was too low by at least a factor of 3, and the value of 2200 provided by the European Space Agency was too small as well. The Cosmic Mirror has learned from several sources involved in the campaign that the meteor counting had gone smoothly, be it visually or via intensified video feeds (as on the planes) - but that the conversion of these data into a precise ZHR value has apparently failed.

A science matures

These initial impressions [Fischer 1999b] were followed by months of detailled analysis - as well as by my personal transformation into a serious meteor "scholar," if there is such a thing. At least I tried to look beyond the sheer visual drama of the 1999 storm by attending three major meteor conferences in 2000: the German AKM Seminar in Radebeul, the MAC Workshop in Tel Aviv, Israel, and the ICM in Pucioasa, Romania, where I presented our initial raw results (in the latter case together with JAS' Khaled Tell) - and learned of work from many others, sometimes employing techniques never before used to study meteors:

The systematic observation of meteors with the naked eye, photographic and especially image-intensified video cameras has become one of the rare fields in astronomy in which amateurs can not only contribute to science - but where the science produced from the amateur data can be crucial to advance the whole field. This has become clear again at the annual meeting of the German Working Group for Meteors (AKM) at the hospitable Sternwarte Radebeul on March 17-19, where both new insights into the workings of the Leonids were revealed but also the high state of 'routine' observations these days.

The main discoveries about the Leonids, as derived from a torrent of data from the 1999 storm presented at Radebeul were:

  • There is an enormous fine structure in the activity profile, i.e. the rate of meteors seen as a function of time, during the hour- long storm - but it becomes evident only when one looks at observations (visual and esp. by video) from specific locations in the world. If one adds up the profiles from all places (Tenerife to Jordan), the details average out. The video data from the Jordan camp in particular reveal a strong 'early' peak of activity around 1:45 UTC, 20 minutes before the sharp main peak, plus enhanced activity around 2:30 UTC - all these features are considered significant now. Confirmation by other (non-visual) methods could be forthcoming.
  • Since observers at other sites (Spain was covered particularly well) saw and recorded a rather, different profile than Jordan or France, it is even possible to generate a 'tomographic picture' of the dust trail(s) that made the meteor rate explode. The 1:45 UTC peak, e.g. was probably due to Earth's distant encounter with a dust trail from Tempel-Tuttle's 1932 perihelion passage, though a significant effect on the meteor rate had not been predicted. The main peak has resulted from the 1899 dust trail, of course, confirming brilliantly the model calculations by D. Asher and R. McNaught.
  • Other surprises were the lack of faint meteors - video cameras with better limiting magnitudes but smaller fields of view saw far fewer meteors than those with worse sensitivity but larger fields - and a possible breakdown of the geometrical ZHR correction formula. Since decades the influence of the elevation (h) of the radiant on the number of meteors seen has been corrected geometrically into the Zenithal Hourly Rate (ZHR), dividing the seen number of meteors by sin(h). (Other corrections, such as for obstructions in the field of view and the sky quality, apply as well.) The data from the 1999 Leonid storm cast a doubt on that simple formula: Those with a low h got ZHRs of only 2000-3000 for the peak despite the correction formula, while those with the highest h got 5000 as the peak rate - it seems that the sin(h) effect must be replaced by a (sin(h))**gamma correction, with gamma other than one.

And this was just what could be learned from the German amateur observations, at numerous sites in Europe and in Jordan! The Radebeul conference [Fischer 2000] was not dealing solely with Leonids but also the more "routine" meteor work which is possible nowadays - and this, especially in the video field, convinced me with a bang that meteor studies are not just worth the effort but also that amateur astronomy is probably contributing more to this field than to most others. That this was not my impression alone became evident at the MAC Workshop one month later [Fischer 2000a], where professionals and amateurs alike from many countries presented their insights about the Leonids:

It was one of those Woodstock-style events that punctuate the steady flow of scientific progress now and then: The first scientific conference after a rare space event can sometimes turn into a fast-paced presentation of one mysterious and/or spectacular observation after the other, with often little theoretical understanding - but everyone present feels that something special has happened and that years of scientific work will be initiated by the encounter with the unknown. The sessions on the Great Comet Crash of 1994 at the IAU General Assembly in The Hague were such a special event - and the Leonid MAC 2000 Workshop in mid-April in Tel Aviv was no less exciting.

"MAC" stands for the Multi Instrument Aircraft Campaign that the U.S. Air Force, NASA and other agencies had run during the 1999 Leonid storm, and the meeting at Tel Aviv University was the first occasion for a joint presentation of the observations made from the two aircraft that had flown "through" the storm near Italy. But many groundbased teams, associated with MAC'99 or independent, were represented, too, as were amateur astronomers from Israel (who had observed the storm in the Negev desert) and Germany (who had been in Tenerife and Jordan).

The theme running through most of the conference could be described as: "The meteors are coming - let's try something new!" A number of speakers had used instruments during the meteor storm that had rarely or never been used for meteor work before: big radar dishes, networks of radio antennae, HDTV cameras etc. Almost every other speaker had a video cassette to show with sometimes strange but more often breathtaking views of the celestial spectacle. The "best picture" award would clearly go to the wide-angle HDTV tape shot by Japanese researchers from one of the airplanes which really showed the rain of several meteors per second during the peak of the storm (H. Yano). Almost as captivating was a Czech video of meteor spectra in realtime (J. Borovicka), also shot from the air.

Here are a few of the highlights of the meeting and some early discoveries - a collection of refereed papers should be published as a special issue of Earth, Moon, and Planets as well as a book this November:

  • You see 7 times as many meteors from the air than from the ground! This surprising 'discovery' by the airborne observers has been analyzed theoretically and found to be perfectly logical (D. Koschny): From, say, 10 km altitude you can look through a much larger volume of the atmosphere with less extinction than from the ground, so the number of meteors near the horizon rises dramatically. This fact was also exploited by the above-mentioned Japanese HDTV system.
  • How real is the fine structure in the ZHR profile of the peak? There was considerable debate about the reality of several peaks of the meteor rate before and after the main peak, with P. Jenniskens arguing for a remarkably smooth ZHR profile (with a Lorentzian shape) as generated from multiple airborne video cameras, but several others (among them I. Manulis, H. Yano, S. Molau and the author) saw clear evidence for at least one peak 20 minutes before the maximum that shows up in the airborne video as well as in the Israeli and Jordanian visual and video data.
  • Are there two layers in the atmosphere where the meteoroids burn up? Puzzling Israeli radar data (N. Brosch) could indicate that one class of dust particles decays at 250 km and the other at 120 km altitude, but the result remained controversial and the technical details of the (military) radar system unknown. The Leonid activity had also been monitored by a worldwide network of identical mobile radar antennae (W. Hocking), but many of the detailled recordings have still to be made public (and could prove the fine structure mentioned above).
  • What can we learn from the radio emission of meteors? An array of radio antennae normally used for lightning research had been placed in the Israeli desert - and recorded up to 18 000 VLF radio signatures an hour during the Leonid storm that are characteristic for meteors (C. Price). The tons of data collected have hardly been looked at, but the tentative activity profile from the meteor emissions is puzzling: There are both the main and the above-mentioned pre-peak 1/2 hour earlier - but also a huge number of signatures 1 1/2 hours before the peak that have no visual counterparts whatsoever...
  • Nebulous meteors are real! A few researchers had obtained video recordings of meteors with high angular resolution - and sometimes meteors would show an extended V-shaped halo (I. Murray, M. Taylor). The detailled hydrodynamics of how meteoroids decay in the upper atmosphere, one could learn at the conference, are not that well understood, as are the subtleties of the emission spectra from the meteors. Those, in turn, might eventually even teach us about how cosmic dust could bring organic molecules to Earth - a reason why even "astrobiologists" show some interest in the Leonid data.

Not all of the results presented at the conference made it into the proceedings eventually (e.g. a poster based on IMO video data that showed a periodic signal in the ZHR from some places), but the thick book [Jenniskens et al. 2000] is still a treasure trove of exciting details - and some theoretical insights which drove on the young field of Leonid studies. The 2000 Leonids were the next test, of course, and while the three expected ZHR peaks duly materialized (all staying well below 1000, as predicted by most), some of the timings were off - by less than one hour, but still somewhat disturbing. And what about the rates one could hope for during the other two storms that now practically everyone expected in 2001 and 2002? By the time the Meteoroids 2001 conference in Kiruna, Sweden, came along in August 2001, the picture [Fischer 2001] was more confused than ever:

Although no one doubts the general physics of meteor storms anymore, there is considerable confusion right now about the details - which translates to great uncertainty about whether there will be a big Leonids storm this November at all and if so, whether it will be best in East Asia & Australia or rather in North America. All models published so far are based on the famous dust trails that worked well in 1999 and 2000 and apparently in the past centuries as well. But among the four major groups or individual theorists working on predictions for 2001, three markedly different results have been achieved:
  • The classical result, basically the same for D. Asher & R. McNaught as well as E. Lyytinen, has two strong peaks of the meteor rate on Nov. 18, 2001, one around 10 UTC with about 2000 meteors/hour and the bigger one around 18 UTC with 6000 to 15,000 meteors per hour. As far as the CM has learned, none of these authors have revised their predictions which remain basically the same after the 2000 observations (which were forecast correctly in time, though not the maximum ZHRs).
  • A very different result has been obtained by B. Cooke of NASA's MSFC who - taking into account worldwide observations of the Leonids in 1999 and 2001 - comes to the conclusion that there will be only one veeery broad and shallow maximum, peaking at a mere 1300 meteors/hour around 13 UTC.
  • Again completely different is a model presented by P. Jenniskens at the Meteoroids 2001 conference in Kiruna earlier this month: He sees again two peaks, at the times predicted by Asher, McNaught & Lyytinen, but with the intensities the other way around. He expects a whopping 32,000 meteors/hr at the 'American' 10 UTC peak and only some 2000/hr at the 'Asian' 18 UTC one. This model is work in progress, however, the final paper hasn't been released yet, and it is already being criticized, as the CM has learned.
How can the same basic model yield three very different results? The problem is that there has never been a situation like in 2001 (or 2002) when the Earth is coming rather close to certain dust trails, but several years after the parent comet Tempel-Tuttle came by. Thus there is no way of telling beforehand how much dust there is in the trails so far behind. And there are also several very poorly understood mechanisms that can shift the trails away from the Sun (Cooke) or towards it (Jenniskens). While the observations of the Leonids activity from 1998 to 2000 are excellent, the data for the decades and centuries before are often poor or contradictory, and the 3D shape and location of the trails are just not constrained enough.

Another storm in Asia

By the time November 2001 had approached, Jenniskens had revised his predictions yet again (from no storm at all in 2000 to a ZHR of 32,000 in mid-2001 to now 4200 at best), and six of the eight best-known theorists were betting on two distinct storms with peak ZHRs ranging from 2000 to 8000; only Cooke (joined by Peter Brown who had once 'lured' us to Mongolia) was still expecting a single and very shallow broad maximum. The others disagreed on whether the first peak (best seen in North America) or the 2nd one (in East Asia) would be higher: This time I confidently went with an expedition to Korea [Fischer 2001a] - where the last good storm in moon-free skies became quite a media event:

It was an amazing sight: A newsstand at a rest area, near one of South Korea's crowded highways, on November 20, 2001 - and each of the five national newspapers had an astronomical photograph on the front page above the fold! Pictured were the night sky over whereever (deciphering the Korean alphabet was one thing, understanding the text was another ...) - and several bright meteors. Those were among the best multi-meteor photographs I have seen, and to find them featured so prominently on Korean broadsheets was no accident: It had been in this country, just two nights ago, that the second Leonids storm of 2001 could be observed under near-ideal geometrical conditions. In the end, the weather had cooperated as well, and we had been there for the show, as guests of the Bohyunsan Optical Astronomy Observatory (BOAO), the most modern astronomical facility in South Korea.

It had been an amazing confluence of events that had brought more than 20 amateur astronomers from several countries in Europe and Asia to this spot, many of them top visual and video meteor specialists and/or veterans of the legendary desert camp in Jordan where we had, two years ago, witnessed the first meteor storm since 1966. A site with such a perfect viewing geometry for the predicted main peak, combined with good climate statistics, had not been evident for 2001, and for a long time most German amateurs had been pondering expeditions to Eastern China and Australia. But at the national meeting of the meteor observers in early 2001 suddenly Korea was on the map: decent geometry, good cloud stats, perfect infrastructure - and there would be a German-Korean couple of meteor observers taking care of all the details.

It is to Mr Wiechell and Mrs Lee-Wiechell that we - the 17 from Germany and Singapore who eventually signed up for the tour - owe much of the success as well as ease of our expedition to: They had made crucial contacts to the BOAO and gotten us the unexpected and highy appreciated invitation to stay there for three nights. And they had organized a highly concentrated tour around much of South-Eastern S. Korea, exposing us to quite a percentage of the most important cultural sights of this country. [...] On November 17th the time had come to retreat to the BOAO, located on the top of Mount Bohyun (the 'san' in the observatory name means mountain).

It is the most modern and the largest of the three observatories operated by the Korea Astronomy Observatory, founded only in 1996: The two main instruments are a 1.8 meter reflector and an unusual Solar Optical Flare Telescope (the control building of which would later serve as the logistical center for the meteor video cameras). Add to that a visitor center with a well-stocked gift shop (they have tiny working planispheres as key chains!) that draws an astonishing number of day visitors, despite the remote location of the BOAO and the winding road leading up Bohyunsan. Of even more interest to us was the main building, though: He we would sleep (Korean-style on the heavily heated floor), eat (mainly the Korean instant 'Ramen' soups we had brought along) - and have around-the-clock access to the Internet through a really fast connection.

Through this access we could, for the first time, experience the Leonids as a global phenomenon in near-real-time: While the first, 'American' peak could not be seen from Korea at all, meteor counts from the U.S. as well as from an airplane were available on a special website that added data points continuously:

Sometimes the software seemed stuck but then the plot advanced on our computer screens, half an Earth away from Arizona. And, boy, did it take off, to eventually peak around a ZHR (or whatever the computer was calculating) of over 2000. The peak seemed a bit late (more like 11 UTC than the 10 or 10:30 UTC predicted), but otherwise it was reason to rejoice: Even though more than 3 years had gone by since the parent comet Tempel- Tuttle has passed the Sun, there were still highly concentrated dust trails meeting the Earth even now. [...] One of the four models that had not predicted any distinct peak over the U.S. could already be discounted. And, even more thrilling for us at BOAO, there was now every chance for an even bigger storm than the Americans had just seen over our place.

But would we see it? Watching the weather trends over the last week, be it on Korean or foreign internet sites or on TV, we had always remained optimistic - all cloud patterns seems to stay clear enough of the BOAO site. But, just hours before the Leo radiant would rise and while we were still digesting the observations from the U.S. (the first e-mails were now in, confirming the general impression from the Mt. Lemon plot), we had suddenly lost sight of the stars! The top of Mount Bohyun was in the middle of a localized cloud, too small to show up on the satellite pictures. At times we could see only for a few meters. The cloud hovered around, sometimes letting thru a few stars, then covering up everything again. It was a most bizarre atmosphere now, almost comical ...

The fog would not go away tonight: That were the firm words of Young-Beom Jeon, an observer with the 1.8 meter telescope, an experienced amateur astro (and art) photographer and our personal host at the BOAO. But, he said, in his experience one would just have to drive down the mountain a few hundred meters (there was no 'up' left anyway) to find clear skies, and he confirmed his suspicion with a small scouting trip. Frantically (more or less) we started to pack our gear, willing to go any extra mile required for clear skies, and the engine of our bus was already running warm. But at our set decision time of 23:30 KST the cloud had disappeared mysteriously, and after some confused looking around we decided to stay for the time being. The skies above us were as perfect as before (with a limiting magnitude of 6.3 for most), while near the horizon some light pollution - from squid fisihing boats offshore! - was inevitable.

With midnight approaching phase one of our Leonids spectacle could begin: As 8 hours earlier in the U.S. the first indication of coming wonders was a barrage of so-called Earthgrazers, wonderfully slow meteors travelling over much of the sky. Those are Leonid meteoroids just scratching the upper layers of the EarthUs atmosphere and not burning up in a fraction of a second as ordinary meteors do. Some of them displayed intermittent or terminal explosions, others just faded away after having sped 100 degrees and more over the sky. Soon thereafter the Lion's head appeared in the East and with it the more ordinary meteors. The 2nd, our, peak was predicted for a little after 3 a.m. KST (it was now already November 19th here), and the meteor rate soon started to take off. There were two effects driving it up simultaneousy: We (i.e. the Earth) were heading more and more into a dust trail shed by Tempel-Tuttle in 1866, and the radiant was climbing higher and higher, so that we could actually see more and more of the particles entering the atmosphere.

At first I was counting the meteors in 10-minute intervals (around 1 a.m. KST = 16:00 UTC there were some 15 per 10 minutes), but at 2 a.m. I switched to 5-minute blocks. Not being a seasoned visual meteor observer I refrained from trying to estimate the magnitude of every meteor (as enough others on the mountain were doing) but concentrated on counting. At 2:15 there were already 30 Leonids in every 5-minute block, at 2:30 it were 45, at 2:45 more than 70, at 3 a.m. 85. Thus my personal, uncorrected hourly rate (HR) was now at 1000: The Leonids were storming again! And that was just the beginning: From 2:15-2:20 I counted 118, from 2:20-2:25 another 91 and from 2:25-2:30 again 108 Leonids. The personal HR thus peaked at about 1500 near 18:22 UTC, but with the proper corrections for limiting magnitude and esp. radiant elevation a Zenithal Hourly Rate of 3000 and beyond is likely. [...]

How did it measure up with the 1999 experience? The HR clearly could not top what we had seen two years earlier over Jordan's desert: There an individual observer could easily count 70 to 90 meteors per minute at the peak, corresponding to an HR of 5000. Simple geometry played a key role there: The radiant had been about as high as possible at 66 degrees elevation back then while in Korea we had about 42 degrees. Since there is a division by the sine of the elevation in the HR-to-ZHR formula (plus corrections for the limiting magnitude and obstructions), the Jordanian value had to be raised by only 10% to get the true value while the Korean number has to be raised by 50%! The factor-of-2 difference in HR also corresponded (at least to me personally) to a different 'feeling' of the meteor storm: In Korea I only rarely had the subjective impression of racing towards the radiant or of meteors raining out of that virtual spot, while this 'rain of meteors' feeling had been overwhelming in 1999.

The lower HR in 2001, however, came along with a much higher mean magnitude of the meteors - and a much higher incidence of fireballs than we had in 1999 (when there were hardly any)! Before and especially after the peak in numbers, when the radiant was higher, the fireballs took over! And while that wasn't a perfect replay of the famous global "rain of fireballs" from 1998 either, when the fireballs had been even brighter, the combination of a true meteor storm in numbers and the high number of fireballs turned the Leonids of 2001 into yet another show not to miss! With the fireballs also came back the persistent trains, those "glowworms in the sky" left by particularly bright ones, lingering at a hardly fading surface brightness for sometimes many minutes. Watching those trains in binoculars, being twisted by the upper atmosphere winds, was an out-of-this-world experience that cost me a few of my 5-minute counts (esp. between 4:20 and 4:50 KST), but what the heck ... :-)

The fireball show continued all the way to and beyond the onset of dawn at 5:50 KST - there were some still shooting around even as the sky had brightened up considerably. What a night that had been, with the first peak on the WWW, the cloud scare, the Earthgrazers, the meteor storm, the fireballs and trains. Another hour was spent by many posting first thoughts to the world, before a colorful sunrise thru distant cloud banks ended the night for good. The night before hadn't brought any advance notice of what was to come (some rare grazers and fireballs aside), and the following night would turn out even more ordinary (with more Taurids than Leonids for most of the time). Have we to be thankful to the handful of tireless theoreticians or what, whose calculations had said: yes, there will be another storm, and you'll have to travel to East Asia, sorry?!

The shows draw to a close

By now the art and science of Leonids predictions had become almost routine: There would be a handful of largely independent models published just before the event, some would win (if only with respect to certain aspects, such as the times or peak ZHR values), some would fail, and a year later the more successful theorists would give it another try. In 2002 there were five predictions in the end, without an outlier like the one in 2001 but otherwise not much closer to each other: Once again one could have a pick whether the first or the second of two expected peaks would be stronger. As the Moon was almost full, there were fewer expeditions than in 1999 or 2001, and many were clouded out, but soon a consistent picture [Fischer 2002] emerged:

None of the five predictions [...] had it all right, but Vaubaillon & Colas got the best timing while - unfortunately for the visual observers - Langbroek 'won' with respect to the meteor rates. Both predicted peaks materialized, both were some 10 minutes late, and both reached a zenithal hourly rate of only about 2500 (which was further diminished by the light from the almost full Moon). Nonetheless the Leonids did it again, even four years after the parent comet had passed by. And while the two storms of 2002 rank only as the 3rd and 4th strongest of the five that pleased us in 1999, 2001 and 2002, there were again all the features we've come to expect from the Leonids: occasional earthgrazing meteors while the radiant rose, bright meteors all over the sky (though less so during the 2nd peak) and occasional fireballs with persistent trains (though not nearly as many as in 2001, let alone 1998). The peaks were very sharp, however, with full width half maximum times of only about 45 minutes each (in this respect, by the way, Jenniskens 'won').

While I had missed the actual storm peak due to a narrow cloud band over Tenerife, the viewing conditions were rather good in the hours before and after - and for the first time I had my own video camera (a Mintron). Despite using an improvised and not very fast lens if captured more meteors than I saw with my own eyes, encouraging further experiments 'using' the Perseids, Geminids etc. As this article is coming to an end, the Leonid storms have already done so for the coming decades if not the full century. But they are still active, with more (though mostly very weak) peaks expected than ever, spread over a full ten days in 2003. With all we have learned in those "five wild years" and the new technologies at hand, meteor observing will remain an interesting activity. And you will certainly not hear from me anymore that it doesn't make sense ...


Beech et al., 1995, Quarterly Journal of the RAS, 36, 127
Fischer, 1998,
Fischer, 1999, Skyweek, 15 # 42, 2
Fischer, 1999a,
Fischer, 1999b,
Fischer, 2000,
Fischer, 2000a,
Fischer, 2001,
Fischer, 2001a,
Fischer, 2002,
Jenniskens et al. (eds.), 2000, Leonid Storm Research, Kluver, Dordrecht
Mason, 1995, Journal of the BAA, 105, 219
McNaught & Asher, 1999, WGN, 27, 85

Appeared in a similar form in Triglav-Cekada & Trayner (Eds.), Proceedings of the International Meteor Conference, Bollmannsruh, Germany, 19-21 September 2003, published by the IMO 2004, p. 29-43