Previous: Chapter 2: Détente
Chapter 3: The actors, Part I
In which fedi offers help, and we get to know the players in the drama
A few days later, there were two responses to kagetsuko’s post. One was from klimagalka, who by her bio was Verena Galina, a climate researcher at the Siberian Branch of the Russian Academy of Sciences in Novosibirsk. Stylised female-looking avatar, like something drawn by Möbius.
@klimagalka:
@kagetsuko I work at the Siberian SuperComputer Center of SB-RAS. I will try and find out what happened to the archives of that project. It’s from the Soviet times, and a lot has changed, but the Center has been there continuously since then, so maybe some materials will still be here.
Another was from silverstacks, whose profile only said SilverStacksBooks, they/them. They ran an indie second hand bookshop in Honolulu.
@silverstacks:
Hi @kagetsuko, I am a librarian at Hawaii University in Honolulu. I’ll be happy to ask around to see if there’s anything left from that time. The JCRE was closed down when Bernard Miller died in 1979, so the archives might be scattered all around the University and NOAA.
@kagetsuko:
@klimagalka @silverstacks @lores That’s wonderful, thank you all so very much for wanting to help me!
A few slow weeks followed. Interminable dark February ended after all, March came in like a lion. I trailed through the BESM-6 code repositories but could not unearth anything of interest. I also researched the scientists involved, in particular Marchuk, Dymnikov, Chinchuluun, Lightman and Miller.
Dr. Alexey Ilun Chinchuluun was a bit of an enigma. There were no pictures of him. His name was clearly Mongolian, with a Russian first name which was not uncommon at the time. He got his PhD from Tomsk University, in Siberia, in 1973. I found his PhD dissertation, and it was brilliant. He was part of the hand-picked team led by the renowned Valentin Dymnikov, so he seemed destined for a great career. He was already co-author on several papers and technical reports of the Center. But after his 1976 visit to Hawaii, nothing. No trace of him to be found anywhere, neither in publications nor in the CIA’s documents.
Gury Marchuk, the Direktor of the Computer Center in Novosibirsk, was no mystery at all. He was by all accounts a great scientist and leader, Hero of Socialist Labour, recipient of no less than four Orders of Lenin, the Chebyshev Gold Medal and much more besides, and his career was an open book, if a very voluminous one. A very giant amongst men, and yet by all accounts an affable person. He looked the part: distinguished, well groomed, bright-eyed. He had practically single-handedly built up climate research in the USSR, and was instrumental in starting the general circulation model project at the Siberian Computer Center. At the time of the events, he was no longer involved in the technical side of the project.
That part was left to Dr. Valentin P. Dymnikov, who also had an illustrious and public career and was, as far as I could make it, something of a genius, both in modelling and programming. Results obtained with the climate model he developed were included in the IPCC’s 4th Assessment Report, from 2007, the one that finally put the responsibility for global warming squarely and unequivocally on the shoulders of humanity. He looked a bit like the young Yuri Gagarin, but was ethnically Mari, a Finnic people who had an Autonomous Soviet Socialist Republic. He was credited with one of the great breakthroughs in numerical simulation of nonlinear atmospheric dynamics. That sounded promising. And yet, he had not seemed to be involved in the project a the time. According to Lightman’s report of his visit, Dymnikov was notionally the project lead but he was on sabbatical at the time, and Chinchuluun was the acting lead. Based on some documents I found, Dymnikov was at the time at the Joint Institute for Nuclear Research at Dubna, near Moscow, working on a compiler for the Minsk-32 machine, another one of those Soviet behemoths.
Lt. (JG) Dr. Nathan Lightman was Chinchuluun’s US counterpart. He was a Lieutenant Junior Grade in NOAA’s Commissioned Officer Corps and I found a picture of him looking dapper and charming in his uniform. His career before 1976 was very easy to follow, and was that of a model atmospheric scientist who excelled at computer simulation. The visit to the USSR, which he got on the strength of those skills, should have given it a considerable boost as well. But just like Chinchuluun, after 1976, nothing. It seemed like they had both gone up in smoke. I found an interview with him organised by NOAA in 2020 as part of their ongoing oral history project. He talks about his Jewish family background, how his family emigrated from Vilnius and ended up in Cincinnati, about his education, and how he became an atmospheric scientist, but of his visit to Siberia he only tells a brief anecdote, how he found out that the apartment he was staying in was bugged. He says nothing about why he left NOAA, and the interviewer does not broach the subject either.
Finally, there was Dr. Bernard Miller. His mother was Dutch; she had fled to the US at the start of the war and there married his father. After the war, the family moved to Brussels. As a result, young Bernard was fluent in Dutch and French as well as English. He got his PhD in 1965 from the Institute of Geophysics and Planetary Physics in San Diego on the topic of climate physics and he joined the newly created Joint Climate Research Effort right away. In 1972, he became its Director. In pictures from that time he looks like a French movie star. During the interview, Lightman talked a lot about Miller. I got the clear impression that he was an exceptional and rather flamboyant person with a knack for bringing out the best in people, and a brilliant organiser with a win-win style of management, but no longer an active scientist. It is also clear from the interview that Miller was not only Lightman’s boss but also his mentor, and that they were very close. Sadly, Miller died young, in 1979.
The picture was clear: it was with Lightman and Chinchuluun that our quest lay. There could be no doubt that they were the key authors of the model code. And very inconveniently, they seemed to both have disappeared into thin air.
I shared these thumbnail sketches with my co-conspirators. To my surprise, silverstacks replied
@silverstacks:
@lores Here’s how I imagine Lightman getting his assignment from Miller @kagetsuko @klimagalka
One fine morning — but of course in Hawaii, most mornings are fine — the director of the Joint Climate Research Effort breezed into my office. Bernard was an easygoing guy and never stood on protocol, even though he was one of the key decision makers at NOAA. “Nathan,” he said with a mischievous smile, “how would you like to go to Siberia?”.
Talk about a bolt from the blue, but I wasn’t going to let it show. I shrugged, “What gives?”
“Well,” he said, as always cutting straight to the chase, “as part of Kissinger’s ‘détente’, we’ve been having talks with our Soviet counterparts to set up a joint program on simulation of ocean-atmosphere circulation models.”
Now he really had my attention. This was right up my street, and the Soviets had the best theoretical models, even if we had the better computers. “So what’s the deal?”
“For the first phase, we’ve agreed to sending someone to their new Computing Centre at Novosibirsk this summer. I haven’t given them a name yet, but I’m sure I could find somebody if — “ “No way,” I interrupted. “I’m definitely in!”
Bernard beamed. “Wonderful! Let’s start planning.” And without further ado, he sketched out a travel plan and itinerary. It was uncanny, so perfectly was it tailored to me, to the last detail. He clearly knew me much better than I had assumed. He was a handsome guy and I liked him a lot, but I didn’t fancy him. I wondered if maybe he was more interested in me than I had realised. I hoped not. He was married, and a big shot, and it would definitely complicate our working relationship.
“Don’t look so surprised,” he said, “I took the itinerary of my trip to the USSR of five years ago. I figured it would work for you too.”
How absurdly simple. I was relieved. Overthinking is one my faults.
So apart from being a librarian and bookshop owner, silverstacks was also a writer.
@klimagalka:
@silverstacks Amazing! You’ll have to do one on Chinchuluun and Marchuk as well!
@kagetsuko:
@silverstacks Yes please!!! @klimagalka
klimagalka, silverstacks and kagetsuko really hit it off. klimagalka has a wicked sense of humour, kagetsuko loves leaning towards the absurd and silverstacks was just rolling with it. They sprinkled my timeline with brief, witty exchanges peppered with obscure emoji like :neocatangelpleading: , :nekowave: or :blobmeltsoblove: . I joined in the fun occasionally but was mostly happy to sit back and enjoy the fireworks.
March had nearly gone, if not like a lamb, its weather having been more volatile than ever. The days had finally started to lengthen when klimagalka got some news.
@klimagalka:
@kagetsuko Unfortunately the material was moved to the Hydrometeorological Center in Moscow in the early 90s, when the USSR broke up. According to the personnel records, Alexey Chinchuluun left the Center in 1976. I couldn’t find anything else about him. I have a friend at the Center in Moscow, I will ask them. @silverstacks @lores
Shortly after, silverstacks managed to unearth a very interesting historical document.
@silverstacks:
@kagetsuko I found an internal JCRE technical report from 1976 about the work by Lightman and Chinchuluun. I will DM you a link. Apart from that there is no material left at Hawaii University as far as I can find, and I dug really deep. Records show that Nathan Lightman left in that year. He did not publish anything after that date. @klimagalka @lores
The report confirmed the performance and resolution of the model. The authors noted that their old CDC 3600 could not match that performance but they did apparently not consider it exceptional. This wasn’t all that surprising as they did not have the expertise to judge the capabilities of the BESM-6. And the much-publicised performance of that machine on the Apollo–Soyuz Test Project, where it was said to have beaten NASA’s computers by a large margin, probably added to the myth of its speed.
The report further mentioned that, because the code was completely written in FORTRAN, it would be possible to run it on JCRE’s soon-to-be-installed CDC Cyber 175. But it did not describe the model kernels in any detail, and there was no code attached.
Enlightening as it was, it didn’t get us much further. But at least we knew the results were not a fluke, and that we could rule out hand-written assembly, which would have been very hard to port to any modern machine.
A further update came from klimagalka, she’d hit another obstacle:
@klimagalka:
@kagetsuko Just to update you, my friend no longer works there but they have promised to ask one of their ex-colleagues. @silverstacks @lores
Meanwhile, kagetsuko had tried another angle. She told me about it later, when I had moved to Japan. All she posted at the time was:
@kagetsuko:
@silverstacks @klimagalka @lores I have an aunt in Hawaii and she just solved a part of the puzzle. In 1976, Chinchuluun and Lightman started a surf shop in Honolulu! Apparently it still exists. They both must have quit their jobs. Chinchuluun must have defected so he could be with Lightman. That’s why they never published.
This is the story as kagetsuko told it to me one hot summer afternoon in Kyōto, while we were drinking cold barley tea and eating watermelon.
Next: Chapter 4. Mieko’s story