Multicolor Molecules and New Horizons’ Data Dump
In science, a picture is worth a thousand data points. And recently, our glimpses at two very different worlds got much, much clearer.
The problem with color, Stockton tells Science Friday’s Ira Flatow, is that some things are just too tiny for photons, the smallest bits of light, to resolve. “[The photon] breaks down at about 0.2 micrometers, which is about 1/1,000 the breadth of a strand of hair,” Stockton says. “Beyond that, light really doesn’t work, it doesn’t diffract.”
And with no light, there are no bright hues. Stockton explains that to get an image on a molecular level, scientists have to use electrons, which are much smaller and don’t pick up the spectral information that includes color. “They only pick up shape and texture and contours,” he adds.
The regular electron imaging process involves treating specimens with heavy metals like lead. “Electrons like to bounce off of that stuff,” Stockton says. Then scientists bombard the specimens with electrons. “The resulting image is more like a shadow casting than a photograph,” Stockton writes.
To add color to electron microscopic images, scientists first create a normal, grayscale “base layer” image. Then, Stockton says, they clean the specimen and repeat the process, this time treating it with “rare earth metals” called lanthanides (which are also in lithium ion batteries). Lanthanides are “picky,” he notes, and only stick to certain types of molecules — allowing scientists to focus the new image on objects of interest. “Say they want to look at a certain peptide going through a neuronal synapse … or a type of virus,” Stockton says. “They will concoct a lanthanide that will only stick to those.”
When scientists process the lanthanide image, they can assign a color to the molecules that appear. It won’t be their true color, but when the lanthanide image is layered back on top of the monochrome base image, the effect is like a highlighter.
“It helps them do detective work on the on the images they’re looking at, so they can pick out, ‘OK. This is where all this virus is,’” Stockton says.
And while some scientists are busy peering at newly colored molecules, others are celebrating images of the dwarf planet, Pluto, that have taken over 15 months to appear in the metaphorical darkroom wash. On Oct. 25, the final packet of data from New Horizons’ flyby of Pluto arrived safely on Earth, for a total of 6.25 gigabytes collected.
The images, transmitted across 3 billion miles of deep space — at the achingly slow speed of 1,200 bits per second — are by far the best pictures of Pluto that we’ve ever had. But Stockton says that getting them was a chore. When he spoke with Alice Bowman, the operations manager for New Horizons, she told him it took 500 data drops to score those six-odd precious gigabytes. Each one involved about eight weeks of planning because every piece of data on New Horizons is indexed using information about when and how it was collected.
“They had to say, ‘OK, we want the piece of data that was collected at, you know, 342,265,863 seconds after New Horizons left Earth, and the probe was tilted ‘x’ degrees this way, ‘y’ degrees that way, and the Ralph imager was on or the LORRI imager was on,” Stockton says. (Ralph and LORRI are two of New Horizon’s three cameras.)
What’s more, waiting up to 15 months for any kind of image seems a Herculean test of fortitude, but Stockton adds that mission scientists weren’t even sure what the images were going to be. Their patience was rewarded.
“They just knew it was going to be aimed at what Newtonian physics told them where Pluto would be, and then they got them back, and we get these wonderful things like ice volcanoes and you know, a giant beautiful heart,” Stockton says.
Sounds well worth the wait.
Nick Stockton is a science reporter at Wired, based in San Francisco, California.
IRA FLATOW: This is Science Friday. I’m Ira Flatow. We’ve all seen images created by an electron microscope. Could be a closeup of the honeycomb texture of a fly’s eye magnified millions of times, or perhaps the curved and spiraled bacteria that inhabit your gut. You know the photos draw you in with their minute details.
But they all have one thing missing. What’s that color? Yeah, they’re always– have you noticed? They’re always black and white images. Makes you wonder what those photos might look like in color. Well, researchers wondered that, too.
Nick Stockton is here to fill us in on that story and other selected short subjects in science. He’s a science reporter for Wired, based out of San Francisco. He joins us from KQED studios. Welcome to Science Friday.
NICK STOCKTON: Hi, Ira.
IRA FLATOW: So Nick, answer the question. Why are electron microscope images always in black and white?
NICK STOCKTON: This is really fascinating. So I didn’t know this before I started reporting this story, but these things get so small that they are too small for light to resolve. So photons are just too big, either the photon itself or the wave, whatever you want to think of it as. Because you know they’re just basically the same thing. It’s just too big to resolve that image at that small. It kind of breaks down at about 0.2 micrometers, which is about 1/1,000 the breadth of a strand of hair. So beyond that, light really doesn’t work. It doesn’t diffract. So you have to use electrons which are much smaller. But those don’t pick up spectral information, which is what we get color from. They only pick up like shape and texture and contours and stuff like that.
IRA FLATOW: You know we have images from planets, when we take pictures of planets that have sort of false color images. Can’t we fill in the color on an electron microscope the same sort of way?
NICK STOCKTON: Not in the same sort of way. But there are work arounds. And what these researchers did is they kind of created like a hack in some sort of way. Where so in order to understand what they did you kind of got to understand what the normal electron microscope process looks like.
And what they do at first is they treat whatever specimen they want to look like with a type of heavy metal, like a lead or something like that. Because electrons like to bounce off that stuff. But then they found by using these rare earth exotic metals like lanthanides, which are what they also used to make lithium-ion batteries out of, they can make special formulations of lanthanides that will only stick to certain molecules.
And so say they want to look at a certain peptide going through a neuronal synapse or something, or a type of virus. They will concoct a lanthanide that will only stick to those. And so they first make that grayscale base layer using heavy metals, clean that off, and then treat the sample once again with this special picky lanthanide that’s only going to stick to this one kind of molecule. And then they treat it again. And then they come back and they just kind of assign a color to that. It’s not the actual color of the thing that they’re looking at, but it helps them do detective work on the images they’re looking at. So they can pick out, oh, OK, this is where all this virus is or oh, these are all the neuronal synapses we want to track.
IRA FLATOW: So it helps them pick out the detail that they couldn’t see before?
NICK STOCKTON: Exactly.
IRA FLATOW: Well, let’s move on to the New Horizons fly by Pluto made that last year. And sent some amazing images, speaking about images. It’s still buffering though? It’s still downloading, sending back to Earth some of those images?
NICK STOCKTON: It’s actually just finished. Last week on October 25th they sent home their last little batch of data. And it took 15 months. They were downloading at about a rate of 1,200 bits per second, which is really slow.
IRA FLATOW: It’s the old modem days.
NICK STOCKTON: It’s the modem days, exactly. But they had to do it over three billion miles of you know deep space, so cut them a little slack on that.
IRA FLATOW: And so they’re finished? Everything is done?
NICK STOCKTON: It’s all in, yeah. They got home 6.25 gigabytes. And it was a pretty interesting process. Like they had to– I was speaking with Alice Bowman about it. She’s the Operations Manager for the New Horizons project.
And she explained to me that there were 500 data drops they did over the course of 15 months. And each one was an eight week process of setting up, trying to figure out or– not trying to figure out. But they knew where this stuff was.
But they had to index every little bit of data by the time code from when the New Horizons spacecraft became active. So they had to say, like OK, we want the piece of data that was collected at 342,265,863 seconds after New Horizons left the Earth, and the probe was tilted x degrees this way, y degrees that way, and the Ralph imager was on or the Lorri imager was on. And they’d get a piece of data back. And at the time they didn’t know what these things were going to have.
They just knew it was going to be aimed at what Newtonian physics told them where Pluto would be. And then they got them back. And we get these wonderful things like ice volcanoes, and a giant beautiful heart on the front of New Horizons.
And they did this process of figuring out what these things were and running simulations on it. And having engineers double check and make sure that this is going to get the data that they want. And this is the priority data they need. And they got it all home safely, which is pretty remarkable.
IRA FLATOW: Yeah, that’s good. Let’s move on to California, where you are now. California is known for earthquakes– of course, you know that. But researchers are reexamining, I understand, a few historic ones. Tell us about that.
NICK STOCKTON: Right, so this pair of USGS geologists based in Pasadena came across a bunch of old drilling data from the early 1900s, back when LA wasn’t the sprawling metropolis you know it now. It was still kind of a city, but it was one of the heaviest oil producing regions in the world at the time. Especially around Huntington Beach– if you see photographs in the era, they’re just like a forest of derricks.
And they found these records that people had forgotten about that had exact drilling data, like depth and location for these. And they started correlating that to some big, relatively big earthquakes that happened around that time. Particularly one that happened in March of 1933, which killed 120 people. And by correlating the drill depth and stuff like that, they came to the hypothesis that these oil drilling operations might have caused that earthquake, and not just normal fault line pressures.
IRA FLATOW: So we’re talking about fracking 100 years ago almost, causing earthquakes?
NICK STOCKTON: Yes, similar. So the reason they started doing this is because of the earthquakes that are happening in Oklahoma. And those aren’t from fracking, they’re from wastewater injection. But same kind of process. You’re putting extra stress on the fault line that didn’t exist before.
IRA FLATOW: And so this has been going on longer than we thought, these earthquakes caused by re-injecting that water back down there?
NICK STOCKTON: Right, well, this is a little bit different. What they did differently was they were actually just sucking a bunch of oil out without putting anything back in there. And it was creating this what’s called subsidence. And it creates this kind of pressure on the fault and that causes the faults to slip.
And in the decades since that happened, they will put more water down into the well to create kind of a balance in the fault. Well, actually to get more oil out of it. But the benefit is it balances out the faults. So we don’t get these kinds of earthquakes anymore in southern California.
IRA FLATOW: It’s kind of interesting. That’s really good. Thanks, Nick. Nick Stockton joining us for Wired magazine, based out of San Francisco. Thanks for taking time to be with us today.