☀ + ☃ = ♨

April 18, 2017 6:34 PM

An Ice Scientist’s Worst Nightmare [The New York Times] “On April 2, the temperature of a storage freezer in the Canadian Ice Core Archive rose to about 100 degrees — some part of the cooling system failed, “then tried to get itself back into action and in the process, piped hot air back into the room,” according to Martin Sharp, the director of the archive. The freezer became so hot that it tripped the fire alarm, Dr. Sharp said, and partially or fully melted 180 ice cores collected by government scientists since the mid-1970s from the snowy expanse of the Canadian Arctic.”
posted by Fizz (53 comments total)

Wow that must suck. Can't imagine so much work going down the drain (as it were).
posted by paladin at 6:38 PM on April 18, 2017 [2 favorites]

As a teenager I worked at a Haagen-Dazs scoop shop. One night I apparently left ajar the door of the tempering cabinet, allowing thirty-odd three-gallon tubs of ice cream to melt out on the floor by morning.

The expense must have been fairly enormous but my totally chill manager Emmie just told me to be more careful.

It's kind of a major milestone in my personal cycle of "good judgement comes from experience, experience comes from mistakes, and mistakes come from bad judgement."
posted by wenestvedt at 6:44 PM on April 18, 2017 [52 favorites]

After watching the first new MST3K episode, this is pretty much how you end up with a Reptilicus
posted by thecjm at 6:48 PM on April 18, 2017 [47 favorites]

This sucks, but at least, at great expense and probably some danger, you can just go drill most of them again right?

...or did some of those cores come from places that have now melted?
posted by trackofalljades at 6:51 PM on April 18, 2017 [1 favorite]

...or did some of those cores come from places that have now melted?

From the article:

“For the moment, Dr. Sharp does not yet know if he and his team will be able to go back to the Arctic to take more cores and replace the samples that were lost. “Some of these ice caps are disappearing,” he said, “and we’re going to lose this record, in some cases sooner rather than later.””

posted by Fizz at 6:57 PM on April 18, 2017 [3 favorites]

Huh, just when the US Government is diligently trying to delete any historical record of climate change. Nothing to see here...
posted by nicwolff at 7:00 PM on April 18, 2017 [15 favorites]

You spend eons knocking about so that you may one day tell the story of a past long forgotten only to meet your sudden and undignified demise in a man made grave of steel and fire, drowning in your own water and that of your kin.

.
posted by Foci for Analysis at 7:02 PM on April 18, 2017 [17 favorites]

Backups. Always make backups.
posted by clawsoon at 7:09 PM on April 18, 2017 [5 favorites]

Ouch. Irreplaceable samples damaged. Ouch.

I wouldn't say this is every scientist's worst nightmare (that's probably discovering your coworker faked all your data), but it's pretty close.

Ouch.
posted by steady-state strawberry at 7:15 PM on April 18, 2017 [8 favorites]

It may only be almost pure melted water, but this particular freezer accident smells to high heaven.

I hope they get some sophisticated forensic analysis of exactly what happened.
posted by jamjam at 7:20 PM on April 18, 2017 [4 favorites]

This just happened to us. Luckily there were no icecores in there, just discrete samples, and only one of the bags burst. But this, oh god, worst nightmare. I'm currently researching redundant alarms.
posted by kjs4 at 7:23 PM on April 18, 2017 [4 favorites]

Even if they can go back and re-drill, getting funding for this sort of work is not trivial, and it takes years of planning to go down more than about 30m. I've seen estimates that ice-cores cost about $10,000 per metre to collect.
posted by kjs4 at 7:27 PM on April 18, 2017 [5 favorites]

In some ways it is even worse than faked data: the stuff we're able to do with these cores NOW was basically unimaginable in my youth, so who even knows what potential future information has already been lost. Furthermore, yeah, a collection that took my entire lifetime to gather is not quickly, easily, or cheaply replaced.

Throw it on the tragedy pile. I can't fully processes the bad news at the rate it's been coming in.
posted by absalom at 8:06 PM on April 18, 2017 [16 favorites]

This is terrible. I feel so bad for the scientists and staff.

And, because I was recently watching Fortitude (about a nasty prehistoric Arctic bug that thaws out and tries to kill everyone), well...This is also the premise of every "Hot Zone"-type movie, right? Sleep tight, everybody!
posted by rtha at 8:17 PM on April 18, 2017 [3 favorites]

.
posted by Going To Maine at 8:33 PM on April 18, 2017

This is indeed very disappointing. I was at a public talk at the University of Alberta just a month ago where Martin Sharp introduced the new home of the Canadian ice core archive and Kurt Cuffey spoke about the science of ice cores. It was both fascinating and impressive in how much they can infer about the paleoclimatic record from a block of ice. I hope this malfunction was a one-off.
posted by piyushnz at 8:36 PM on April 18, 2017 [2 favorites]

Be prepared for the reactions of people who deny climate change.
posted by amtho at 8:50 PM on April 18, 2017

Meanwhile, the U.S. National Ice Core Lab in Colorado is very concerned about funding. This local public radio piece I heard a few weeks back suggested the nightmare scenario of having to dump cores collected at great expense and maintained for future research projects. They had been hoping to expand the facility, and now they're worrying over whether they'll have to stop accepting new samples altogether or toss painstakingly collected ice that's thousands of years old out into the parking lot to melt.
posted by deludingmyself at 9:46 PM on April 18, 2017 [4 favorites]

Whoops, that's an old CPR piece. This is the one I heard, with an article as well as the audio.
posted by deludingmyself at 9:48 PM on April 18, 2017 [3 favorites]

Not trying to derail my own thread, but in a similar vain, are there similar libraries/labs for tree-ring maintenance and research of ancient trees/plants? I guess its obvious that there are all kinds of dangers similar to this one that impact the preservation of lab samples. Just wondering about how those are stored?
posted by Fizz at 9:52 PM on April 18, 2017

I was a chemist and never had to deal with freezer outages. It's possible in chemistry but my biologist friends until the present day from undergrad up until now always have some horror story, ranging from going into the lab at 3 AM on a Sunday to losing hundreds of cell lines a few weeks after getting them all set up in a centralized inventory system.

This is a new one though.

Be prepared for the reactions of people who deny climate change.

This. Unlike some posters hints I find nothing suspicious about this--for one thing there's no motive for the deniers to cause harm. It's not like these were going to give us anything fundamentally new. Climate change is proven, the effect of CO2 on climate change is well qualified, the CO2 levels at these periods are already measured to far greater detail than any denier cares about.

But yes, the some deniers for sure will now be convinced that this was going to disprove climate change, even though the period under question is quite small really and as I said we have tons of data.
posted by mark k at 9:53 PM on April 18, 2017 [6 favorites]

some part of the cooling system failed, “then tried to get itself back into action and in the process, piped hot air back into the room

I've had this happen on our cold rooms. It's called the hot gas return. In our case, it's used as a defrost protection to keep the coils from icing up when they're below freezing. When the freezer fails, the hot gas can flip on in error. I don't understand how the control system works in detail (it's 20 years old and all analog and has three generations of patches on it), but in our case this drove a normally sub-zero freezer up to 40C. Fortunately, we didn't lose much. But if it had happened to the fridge beside it, we might have lost decades worth of irreplaceable samples. I feel their pain.

I've just had our "sample library" fridge serviced and renovated. It's taken almost ten years to get the funding in place for it.
posted by bonehead at 10:09 PM on April 18, 2017 [9 favorites]

I'm currently researching redundant alarms.

Wait! What??? So am I!!
posted by sylvanshine at 10:32 PM on April 18, 2017 [12 favorites]

This hurts extra because of the fight to stop stephen Harper dismantling the ice core library. It was one of few wins during the anti science era-- we lost a fungi library and government access to a wealth of fisheries research was maintained by a renegade bureaucrat who saved the books from the dumpster and ran a covert ops library for colleagues from his garage.

Anyway, to lose anything after the epic struggle science just fought in Canada....uuuuugh!

(March for science, I'll be there!!)
posted by chapps at 11:22 PM on April 18, 2017 [11 favorites]

How about some offsite backups this time around? Can you split cores up like that?
posted by Tell Me No Lies at 11:44 PM on April 18, 2017 [1 favorite]

Instead of keeping the samples frozen, could they reduce a sample to multiple slices and store each slice in its own room-temperature container? Or is there a lot of value in retaining the ice as ice?
posted by pracowity at 11:53 PM on April 18, 2017

[I]n our case this drove a normally sub-zero freezer up to 40C.

We had it happen to our 4ºC fridge/deli case thing. We lost some chemicals but thankfully it wasn't all that serious - nothing irreplaceable like this. We actually just got some alarm-things for our -80ºC freezers that are supposed to contact us if the temperature goes above certain set values. Seems worth it to me - a few hundred bucks to give us enough warning to prevent catastrophic sample loss? Fair trade!

(March for science, I'll be there!!)

Me too! Hoping enough other people turn out...
posted by ubersturm at 1:20 AM on April 19, 2017 [2 favorites]

The air in bubbles in the ice is also super valuable. It's how they know the preindustrial CO2 conc. It's got to stay below -5C to be unchanged by dispersion.

If any is interested in ice core science, there's a great documentary about Claude Lorius called Ice and Sky. Not sure if it's widely available, I saw it at a conference.
posted by kjs4 at 1:30 AM on April 19, 2017 [5 favorites]

Also, things grow in room temp samples.
posted by kjs4 at 1:31 AM on April 19, 2017 [2 favorites]

I once worked in a university IT department where somehow, one of my co-workers got it into his head that it would be ok to migrate some data on to a server running software RAID on, I think, NT 4.0, when this was experimental and very much not supported. And this data was 30 years of demographic data that New Zealand's foremost demographer had been lovingly maintaining from punch card days onwards. It was his life's work. And that server's RAID crapped itself, and then it was discovered that backups had not been running properly for months, and nothing was to be done.

Amazingly no one involved in this incident was fired, which tells you a lot about New Zealand universities in the 1990s.

But the good news is that $15000 US dollars later, a specialist disk recovery firm was able to get all the data back. No such thing as reconstructing your ice cores from dirty water, is there.
posted by i_am_joe's_spleen at 1:34 AM on April 19, 2017 [9 favorites]

Ice and Sky, or Antarctica: Ice and Sky appears to be available for rent in the US and UK on ITunes, Amazon and a few other services.

Freezers that are this important obviously have alarms and systems that are meant to automatically alert people if they fail. Like ours, something obviously went wrong with it. It may have been simply that they had less time than they needed because it was pumping hot air in. I should find out if ours can fail like that....
posted by kjs4 at 1:38 AM on April 19, 2017 [1 favorite]

The tree ring guy I know mostly seems to store his slices of tree in amongst his paperwork on his desk.

I'm not sure what ice cores they had stored, but splitting them up for redundancy isn't done, as far as I've ever heard. Cutting up an ice core takes many people a really long time, and it greatly reduces your options for future analyses. Many of the measurements need a specific size of sample or a specific method of sample prep. Resolution and sample size varies from site to site, and between analytes, and even depends on depth (they get more dense as you go down, also radioactive species decay, so you need more sample). You don't cut up an ice core without a massive amounts of planning. Also, labelling ice cores is essentially impossible, and barring the occasional melt layer or drilling mishap, they all look the fucking same. It's really, really easy to mix them up or turn them round when your cutting them up, especially when you're spending multiple hours a day in a freezer for weeks on end.
posted by kjs4 at 1:57 AM on April 19, 2017 [4 favorites]

An Ice Scientist’s Worst Nightmare

No competition from the worst case climate change scenario panning out?
posted by fairmettle at 2:46 AM on April 19, 2017

i_am_joe's_spleen: No such thing as reconstructing your ice cores from dirty water, is there.

Pour them into popsicle molds and nobody will be the wiser.
posted by dr_dank at 3:37 AM on April 19, 2017 [2 favorites]

A detail which makes the story a little less depressing:

But much of the collection was unaffected by the malfunction, thanks to a stroke of luck. A television crew had been documenting the ice core move and had asked that the samples be put in a second freezer because the lighting was better. The university complied, storing nearly 90% of the collection in an unaffected freezer.

“That’s basically what saved us,” said Sharp.

posted by A Thousand Baited Hooks at 4:22 AM on April 19, 2017 [35 favorites]

I wouldn't say this is every scientist's worst nightmare (that's probably discovering your coworker faked all your data), but it's pretty close.

Walking into your freezer to find only puddles is a bad nightmare. Your coworker faking data is a bad nightmare.

Walking into your freezer to find Paul Ryan eating your children would be worse than either of those. Your coworker -- who is a thirty foot tall circus clown made entirely out of writhing loops of human intestine and who speaks with your mother's most disappointed condemnatory voice -- faking your data is worse than that.
posted by ROU_Xenophobe at 4:22 AM on April 19, 2017 [2 favorites]

Wouldn't it be possible to slice each core lengthwise and keep the two halves in different facilities?

If not, wouldn't the bulk of the expense in collecting cores come from planning and logistics rather than drilling? If so, wouldn't drilling multiple cores from each site, and distributing them to different facilities, ensure redundancy at very little additional cost?
posted by acb at 4:51 AM on April 19, 2017

I'm going to Kluane National Park with a different group of climate scientists next month to drill ice cores (same general area as the melted cores and no I'm not a scientist - doing an artist-in-residence with the scientists). We emailed about the melted cores and they told me it was possible on a future trip that they'd try to re-collect those data but everything they will do this trip has been long planned out.

The logistics for this stuff are crazy - fly into Whitehorse, drive a couple hundred kilometers to the Kluane Lake Field Station and then get flown into the back country with a lot of gear in order to get around on skis while collecting cores.

Also relevant - the Eclipse Icefield - where those cores came from - has warmed 4 degrees C mean temp from 2002-2016. This article details how ice cores are collected and is by the guys I'm going with from last year's field season. And the glaciers at Kluane are retreating in a big way, leading to the river that feeds Kluane Lake disappearing.
posted by leslies at 5:05 AM on April 19, 2017 [7 favorites]

Oh Metafilter, diverting into a semantics discussion of what constitutes proper english usage of the title, you never disappoint.
posted by RolandOfEld at 5:26 AM on April 19, 2017 [1 favorite]

Would it be possible to put the storage facility in Alaska, underground, so they could rely on passive cooling if necessary?
posted by Joe in Australia at 5:34 AM on April 19, 2017 [2 favorites]

Would it be possible to put the storage facility in Alaska, underground, so they could rely on passive cooling if necessary?
posted by Joe in Australia

That was my thought as well. Sort of like the Svalbard Global Seed Vault which is totally on my list of super neat places to check out someday.
posted by lazaruslong at 5:42 AM on April 19, 2017

It was one of those IoT fridges, wasn't it?
posted by scruss at 5:43 AM on April 19, 2017 [3 favorites]

Well, about Svalbard...
posted by d. z. wang at 5:43 AM on April 19, 2017

Well, about Svalbard...
posted by d. z. wang

Yep, that sucks - but to my understanding the Seed Vault was designed with exactly that kind of problem in mind, among others.

"Spitsbergen was considered ideal because it lacked tectonic activity and had permafrost, which aids preservation. Its being 130 metres (430 ft) above sea level will keep the site dry even if the ice caps melt. Locally mined coal provides power for refrigeration units that further cool the seeds to the internationally recommended standard of −18 °C (−0.4 °F). If the equipment fails, at least several weeks will elapse before the facility rises to the surrounding sandstone bedrock's temperature of −3 °C (27 °F)."
posted by lazaruslong at 5:47 AM on April 19, 2017 [2 favorites]

Hey, look! It's one of my anxiety dreams come to life!

Fabulous!

I'll just be over here in the corner, twitching and occasionally letting out little "eep" noises.
posted by Harvey Jerkwater at 6:01 AM on April 19, 2017 [2 favorites]

I'm currently researching redundant alarms.

--Wait! What??? So am I!!

Well we can't only have one person doing it, can we?
posted by EndsOfInvention at 6:53 AM on April 19, 2017 [16 favorites]

>>It may only be almost pure melted water, but this particular freezer accident smells to high heaven.

>>Huh, just when the US Government is diligently trying to delete any historical record of climate change. Nothing to see here...

Trump may yet support Paris climate accord.
posted by My Dad at 7:15 AM on April 19, 2017

The best place to store an ice core is in the ice i.e. don't drill it until you know what you're going to do with it. From the second article, it sounds like the affected cores were relatively new and had yet to be sliced and diced. Sometimes this happens in the field, but not all analyses can be done outside a lab, so the ice is shipped to a facility like this. Not that older ice cores aren't stored - good luck finding a scientist who's willing to through out samples, but they've generally already been processed for something.

Unfortunately taking two cores is pretty much double the work, getting in and out is a logistical headache, but it doesn't take that long. Some drilling campaigns take years, and some do involve more than one core, as they can't squeeze all of the analysis they want out of one.
posted by kjs4 at 7:24 AM on April 19, 2017

Information is inherently fragile.
.
posted by sexyrobot at 7:44 AM on April 19, 2017 [1 favorite]

The best place to store an ice core is in the ice ...

Well, except for the part about global warming and your continental ice shelf or glacier has just melted.
posted by BlueHorse at 7:53 AM on April 19, 2017 [3 favorites]

The glaciers they're taking those cores from are rapidly changing. I doubt it's really possible to simply regard any ice environment as an unchanging reservoir any more. I'm not an ice guy as such (though I do do some ice work), but I do cross path with some on a regular basis.

Besides there's a huge cost and time difference between an environmental sample in your freezer and one still uncollected. By far and away my highest project costs are field campaigns, winter and Arctic doubly so (or more). You have to drill cores when you can, and keep them preserved for as long as you can. Some young person with a brilliant idea will do a really fantastic experiment with them, maybe, twenty years from now. She might not have had access to such samples if they weren't gathered in the 1990s or 2000s. Well tended archives have huge value.
posted by bonehead at 8:45 AM on April 19, 2017 [3 favorites]

I'm not an ice guy as such (though I do do some ice work), but I do cross path with some on a regular basis. […] Some young person with a brilliant idea will do a really fantastic experiment with them, maybe, twenty years from now.

So it's really true. Ice guys finish last.
posted by Joe in Australia at 2:12 AM on April 20, 2017 [6 favorites]

could they reduce a sample to multiple slices and store each slice in its own room-temperature container? Or is there a lot of value in retaining the ice as ice?

So they have just now figured out how to use x-rays to read the charred scrolls from Herculaneum that were discovered 200 years ago. Of course before now they have fucked up a bunch of them, trying to unwrap them.

Data is fragile, and you never know what will be possible in the future.
posted by Meatbomb at 12:44 AM on May 7, 2017 [2 favorites]