SciTech Now Episode 339

In this episode of SciTech Now, taking farming to a whole new level; a look into fossil data reveals that vision prompted fish to make the leap onto land; cleaning up nuclear waste; and Science Reporter Dave Mosher tells us about his first space shuttle mission story.


[ Theme music plays ]

Coming up, farming of the future...

There in front of me was this rotating three-dimensional model of a root.

...the evolutionary leap of fish from water to land...

So, there is this transitional period where you go from a finlike thing to a limblike thing.

The eyes triple in size, and then we see complete limbs. up nuclear waste...

The basic idea is to pump up and treat the groundwater, get the waste out of the tanks, and process it into glass logs.

...launching into space.

So, if you can recover that booster or other parts, which they're actively doing, you're saving millions and millions and millions of dollars.

That's what lowers the cost ultimately.

It's all ahead.

Funding for this program is made possible by the Corporation for Public Broadcasting, Sue and Edgar Wachenheim III, and contributions to this station.


I'm Hari Sreenivasan.

Welcome to 'SciTech Now,' our weekly program bringing you the latest breakthroughs in science, technology, and innovation.

Let's get started.

For more than 10,000 years, humans have been farming on this planet.

Over time, farmers have developed tools such as the tractor and fertilizer to make the practice more efficient.

Now a team at the Pennsylvania State University Applied Research Laboratory is using 3-D imaging, machine-learning algorithms, and the power of a laser to take farming to a whole new level.

Here's a look.

A laser isn't a typical tool for the agriculture industry, but Ben Hall and his State College-based company, L4IS, don't have a typical origin story.

I'm not a biologist.

I actually abhorred biology in high school, although I'm very ashamed of that now.

I think it's fascinating.

Hall was working with lasers at the Penn State Applied Research Lab when crop scientists came to him with a question.

Could lasers help them cut samples from a root?

What they were doing over in the Horticulture Department was they were cutting these under a microscope -- really, really thin, little slices, about 100 microns, which is about the thickness of a piece of paper, you know, with a razor blade.

You have a lot of human error, crushing it, tearing it, cutting it crooked, too thick.

But it's a very slow process, and they would end up on the usable slices, which they'd get about three to five per hour.

Not only did the laser work, it could churn out 11 samples a second.

The collaboration led to a grant from the National Science Foundation.

That led to another opportunity.

The grant had me cutting like 2,000 of these roots and putting the little pieces into a vial and then trudging across campus to give it to them.

But I was just thinking, 'If I could just image it here, then I wouldn't have to walk across campus, and I could e-mail the files.'

So instead of cutting samples, Hall decided to destroy one by slowly pushing it through the laser sheet and recording it with a high-magnification lens.

And it was the next morning.

I was thinking, like, 'Well, each frame in this video represents essentially a slice of this object.'

If I could just find a way to take out the frames and stack those frames up, you know, be able to reconstruct the three-dimensional root.

By the time I was finished eating breakfast, I found two programs to do it that were free, and there in front of me was this rotating three-dimensional model of a root.

The images you are looking at were all created the same way.

Hall and his partners patented the process.

It's called Laser Ablation Tomography.

It creates a high-resolution 3-D digital structure by photographing the original sample as a laser vaporizes it, slice by slice.

This is a scan created from the back half of a yellow jacket.

It shows subtle textures in connective tissue.

You can even peel off individual layers for a virtual dissection, revealing internal structures and organs.

This is the largest image Hall has produced -- a hummingbird from beak to tail.

These are pretty pictures, and videos are interesting and scientifically valuable a little bit.

But what's more important is making sense of this, is extracting the data out of it.

The speed and accuracy of Laser Ablation is already helping scientists analyze plant structure to interpret DNA.

Default program.

They'll grow 500 plants with gene 'A' turned off and one with gene 'A' turned on, and it'll scan through those.

It will get the data, and then we apply machine-learning algorithms and pattern recognition to then map out these different things that they want to see.

The technology is already being used by researchers at Penn State to help develop crops that can grow in dry, low-nutrient soils.

It's also helping them understand complicated interactions.

This sample shows the root, the soil, and everything in it.

The bright colors you see correspond to different chemicals.

Yeah, to the legs.

Analysis like this could help farmers cut back on pesticides and fertilizer, but Hall thinks there are even deeper discoveries on the horizon -- molecular discoveries that could unlock clues about the building blocks of life.

It's just like letters of the alphabet.

There's 26 of them, and there's probably a similar amount of chemical elements that we're constructed out of.

But then the words that they form, you know, there are 10,000 words.

We have 26 letters, but the way they're put together is important.

So looking at and trying to determine which one of those 10,000 you're looking at, that's the harder part, but it's, you know, it's a frontier.

It's really cool to try to figure that out.

[ Theme music plays ] [ Computer keys clacking ]

For many years, scientists believed it was the development of limbs that led water-dwelling creatures onto land.

However, old fossil data may now reveal that it was, in fact, vision that prompted fish to make the leap onto land hundreds of millions of years ago.

Here's the story.

Almost 400 million years ago, some adventurous fish made a huge leap that eventually led to the evolution of humankind.

They decided to come up on land.

We humans probably would never have evolved the intelligence we have today if not for that move onto land.


Because it vastly improved our eyesight.

According to neuroscientist Malcolm MacIver, the way we think about the world is closely tied to what we can see, and fish can't see a whole heck of a lot.

That's mainly because light doesn't travel very far in water.

Our ancient fish ancestors lived their lives in a constant fog, so their brains evolved to react quickly to whatever suddenly appeared on the horizon.

MacIver had a theory that the move to land expanded our aquatic ancestors' vision and, in turn, their brains.

To test his theory, he and paleontologist Lars Schmitz spent a year running simulations with fossil data.

Their research revealed new clues about why fish came onto land in the first place.

[ Fossil breaks ] It all seemed to have started when the first fish peeked above the water's surface.

Suddenly, it was able to see 70 times further.

And behold, a smorgasbord of tasty land-dwellers.

To capitalize on this discovery, the fish would have to evolve.

Its eyes soon moved to the top of its head and tripled in size, and its fins began evolving into limbs so that it could stalk its new prey like a crocodile.

Hunting on land was a mental game-changer for the early tetrapods.

Their total sensory volume increased a millionfold, giving them a much bigger window into the future.

The first animal that figured out how to plan accordingly, instead of just reacting, would have had a huge evolutionary advantage.

Iterate that kind of natural selection a million times, and eventually, we have something called prospective cognition.

That's our ability to weigh different options for the future and choose strategically.

To this day, fish have not evolved those kinds of complex planning skills, but many land animals have.

Understanding the evolutionary roots of intelligence helps explain how humans got so smart but also why we are so dumb.

We've evolved to deal with the things we can see in the here and now.

We still don't plan well for things that are too far away in time or space.

Will our vision ever evolve so that we can see the faraway consequences of our actions more clearly?

Evolution won't make that happen anytime soon, but understanding the relationship between vision and planning may help us engineer solutions like using technology to bring faraway things closer.

That just might give us the evolutionary advantage we need to survive the next 400 million years.

Here with us to discuss his research is Malcolm MacIver, professor of mechanical engineering and biomedical engineering at Northwestern University.

It's kind of an interesting idea when you looked at this fossil data.

I mean, this is something that we've thought about forever and ever.

We thought that fish grew limbs.

And they're... What's the vision connection?

Yeah, so the vision connection is that, well, 10 years ago, I predicted that animals coming up onto land, given that air was much more transparent, that would favor bigger eyes because air allows vision to work much better.

So myself and Lars Schmitz and some collaborators went and checked the fossil record, and what we found was actually much more interesting, which is that there was a tripling of the sockets that hold eyes just before we came up onto land.

So the idea is that this might have actually had something to do with coming up onto land.

So, how do I know which came first -- the eye sockets or the limbs or the -- You know what I mean?

Right. Yeah.

So, one thing is that it's clear that this tripling occurred before limbs became digited, before fingers and toes emerged.

So, up to that time, you can imagine a pectoral fin of a fish, kind of stubby, limblike thing, and they presumably paddled up onto land to get what they were interested in there, using those limbs.

So, there is this transitional period where you go from a finlike thing to a limblike thing.

The eyes triple in size, and then we see complete limbs in the record.

And there's also a correlation between our ability to see and how our brains develop.

Yeah, well, so, this is -- Sort of the origin of the idea was with this idea that animals in water, because they can only see about a body length in front of them, have to react to sensory input reflexively, kind of like what happens when somebody taps your knee, that they have to take sensory input indicating a looming predator or a prey, and they have to immediately react.

There's no advantage to strategic thinking in that kind of environment.

Now, once you're on land, you can see 100 times further out.

So now you could still react reflexively to what you see, but the first animal that can acquire the set of mutations that lets you sort of disconnect sensory input from motor output and think of possible futures and then pick the best one, that animal will have a massive evolutionary advantage.

What is this idea of this connection between how we think, our environment -- What does that mean for human evolution, going forward?

Right. So, you know, this is a really good question.

And what we think, studying this process or studying this possible evolutionary trajectory whereby we became analyzers of strategic futures, might help us see why we have certain holes in our ability to think strategically.

And in particular, a lot of people have noted that we have a very hard time thinking on geological time scales.

Thinking even three or four generations forward is very tough for people.

And one of the ideas we've suggested in our work is that this might trace back to why it is we became cognizant of multiple futures in the first place, which was to act very strategically in a fairly short range, like, for immediate benefit.

And we can bootstrap that with language and with, you know, plans over 5 to 10 years, but ultimately, what we're pointing to is it might be a difficult fit, and we might need to think very carefully about how we might augment our ability to think strategically to deal with problems like climate change and other things that happen over vast time scales.

So, how would we augment that?

Well, one of the things that is -- One of the things we could do is to think about, 'How do we make the distant consequences of our action, either spacially distant, as in affecting someplace else on Earth, or temporally distant, far away in the future -- How do we make those -- How do we put those into our sensory bubble?'

When you think about something like a nest or you think about a Prius car with those indicators showing what you're doing, what are the environmental or energetic consequences of what you're doing right now, if we can take long-range consequences of our behaviors, of our policies, and put those in and sort of draw back the conclusions of that into our current space, that may help us think more strategically at the current time.

Malcolm MacIver of Northwestern University, thanks for joining us.

Thank you.

Located in Washington State, Hanford is the nation's largest nuclear cleanup site, with 56 million gallons of radioactive waste sitting in old, leaky underground tanks.

There's a plan to clean it all up, and our environmental reporting partner, EarthFix, has the story.

This is a gallon, and this is nuclear waste.

And this is 56 million gallons of nuclear waste.

It's toxic, radioactive, and is sitting in old, leaky tanks just a few hours upriver from Portland.

It's here at Hanford, where all that waste is polluting groundwater, which is seeping toward the Columbia River.

And after more than 20 years and $19 billion spent, none of it has been treated.

For decades, Hanford was America's plutonium factory.

It was one of the original Manhattan Project sites supplying the radioactive material at the heart of America's nuclear arsenal.

Over the course of Hanford's military lifetime, it was home to nine nuclear reactors used to irradiate uranium fuel rods creating plutonium.

That plutonium was then extracted with chemicals, processed, and shipped off to weapons factories.

Each step of that process produced radioactive waste, some of it liquid, some of it solid, some of it somewhere in between.

For decades, workers poured much of the liquid waste, hundreds of billions of gallons' worth, into the ground.

They dumped the rest, including solids and sludges, into underground tanks to be dealt with later.

And now it's later.

A patch of contaminated groundwater the size of Seattle lies under the site.

Those old tanks are deteriorating, and the official plan to clean it all up has serious problems.

The basic idea is to pump up and treat the groundwater, get the waste out of the tanks, and process it into glass logs.

The first big problem is that radioactive wastes can generate flammable gasses, which can build up in the tanks and treatment facilities and, if ignited... [ Explosion ] On top of that, some of the waste in the tanks still contains plutonium, which is heavy.

And as the waste is moved around, the plutonium could settle out, clump together, and start an uncontrolled chain reaction.

[ Siren blares ] That might not be such a big deal if workers could monitor and step in to prevent the accident, but the treatment process -- at least for the most hazardous waste -- has to happen in special rooms, called black cells, that are too radioactive for humans to enter.

According to the official plan, the machinery in those black cells has to work smoothly for 40 years with no human intervention.

If something goes wrong, there would be little workers could do to prevent hazardous waste from spreading around the site.

But the cleanup has made some progress.

The liquid waste in the oldest, leakiest tanks was transferred to newer tanks, and in 2015, workers treated 2 billion gallons of groundwater.

But the vast majority of contaminated groundwater is still there, and they haven't even started treating the actual waste in the tanks.

In fact, they haven't even finished building the treatment plant.

Construction was supposed to be done by 2007.

That slipped to 2011, then 2019, and recently, the deadline was extended again, to 2036, for the most dangerous waste.

Humans have never cleaned up a mess quite like the waste at Hanford.

And after years of delays, we don't know if the current plan will be up to the task.

That means that, without some breakthrough technology or new plan, those 56 million gallons of nuclear waste will likely continue to sit just hours upriver from Portland.

[ Theme music plays ] [ Computer keys clacking ]

Dave Mosher is a science reporter, who has written for 'Scientific American,' 'Popular Mechanics,' 'National Geographic' news, and

Throughout his career, he's watched humans and robots launch into space, flown over the North Pole to catch a total solar eclipse, and toured a cutting-edge nuclear reactor.

He joins me now to discuss his first space-shuttle-mission story.

You've covered a few of them.

Yeah, so, I've covered four space-shuttle missions, and obviously, the space shuttle is retired now...

Right. I have a sort of nostalgia for that.

But it was very scary [Chuckles] because I'd never covered something like that before.

And one of the things I remember, I was at Cape Canaveral.

You don't -- On TV, if you watch a rocket launch, you don't get the feeling in your chest of all those thousands of little explosions that are happening as this thing takes off.

You just -- The TV sound or whatever just sort of says... [Whooshes] -- right? -- and sort of wipes it out.

But it's a powerful event.

Yeah. So, the press mound, when you're in Cape Canaveral is about 3, 4 miles from the launchpad.

And the whole experience of covering one of these launches is it takes it to, like, multiple levels above what you would see on TV, because you meet the astronauts before they go.


So you remember there's people on there.

You get to see it launch and feel that reverberation in your chest, so there's that, too.

And you get to see all the equipment and the mission, like, what they're trying to do and why they're doing this.

So you become part of it, in a way.

And when you see it launch, it just, like -- it's incredible.

I know.

Of course, at the same time, you're also fearing for the lives of the astronauts because space flight is very risky.

It's very hard, and there are a thousand things that can go wrong.

And you know...

Usually, something does, and they have to fix it.


That's right.

So, fast-forward now to basically private companies that are benefiting and profiting from the science at NASA and literally all of the rocket science that's happened.

But you see the launches -- the successful launches and some of the failures of SpaceX that's happened recently.

They've just started to get into the reusable rocket phase.

What's the next step?

When do they get to a point where they are putting astronauts back up?

Yeah, so, this is the long slog here.

Elon Musk, the tech billionaire, has been working for 15 years to get to this moment, which happened in March.

He launched a rocket in April.

He landed it.

Well, it landed itself.

And then they relaunched it.

This is the dream.

This is the holy grail as a space expert.

John Logsdon, who's at George Washington University, told me this is the holy grail of space flight.

If you can do this, reusability, you can drastically lower the cost of access to space.

And they believe, within about 2 years, this is going to be the norm for the industry, because right now, when you launch, the booster and other parts of the rocket just sort of fall back to Earth, and they break up and crash in the ocean and sink to the bottom, and nobody ever sees them again.

So if you can recover that booster or other parts, which they're actively doing, you're saving millions and millions and millions of dollars.


And that's what lowers the cost ultimately.

So, you know, they're due for a launch with NASA in 2018 with people on top, and that is going to be a big moment.

There's been a lot of digging into that.

And you know, is SpaceX operating safely, given the accidents?

It seems like they're doing a very good job with that in their investigation.

And the latest accident, which was on a launchpad, nobody was on top, was very thorough.

So I think there's a lot of confidence, going into this, but, you know, we'll have to wait and see.

We'll have to see what their upcoming launches go like and how NASA deals with those if there's any incidents, you know.

Putting people on top of rockets, that's what we used to do, decades ago, right?

The space shuttle now, in the longer arc of history, is going to look like a little bit of a blip of this thing that glided back down gently to the earth while carrying these people and carrying this kind of payload in this kind of way.


Yeah, so, there's a lot of critique about, you know, the 'reusability' aspect.

And you mentioned SpaceX getting a lot of science out -- leaning on a lot of science that NASA and other companies have done.

Well, their contractors.

And that's true because they're using some of these engineers to help design the next generation.

But the problem with the space shuttle and its reusability was, you had to scrape down the rockets, like, all the way to the bone basically and completely rebuild them, essentially.

It wasn't truly reusable.

It didn't get the funding it needed at the time it needed, and so it wasn't able to achieve that vision.

So we're going into this new era, and it actually looks like we can do this now.

The rockets are liquid-fueled.

They can land themselves.

It looks like there's going to be very short turnaround time -- the turnaround time that the shuttle wanted and then some.

Musk has said he wants to do a launch every 24 hours.

That would be incredible.

I don't know if he can do it, but...

Well, some of this is also just the pace of technology and how much things have improved, right?

I mean, the space shuttle, some people have said that my smartphone has more tech in it than the space shuttle when it first launched.

Yeah, that's true.

You know, once you make a system that works and it doesn't blow up your space fliers, you're going to stick with that system.

So the computers are very old, and they did some minor upgrades to them, but, yeah, we've come a long way.

I mean, just looking at the booster of a SpaceX, that is all driven by artificial intelligence by software.

It's constantly correcting when it's coming down.

It knows exactly where the drone ship is.

It's making minute course adjustments, like...

Hundreds of times a second.

...maybe thousands or even millions of times per second to try to land that booster on the pad, and that's all done with automation.

There's even a robot that's going to slip under the rocket and secure it if the seas are very wavy.

So all these new technologies have really helped enable what SpaceX is trying to do now.

They've done the heavy lifting of figuring out the solutions and what works and what doesn't.

But we're at a point in history where this is actually really possible, and it's not laughable because it's being down right before our very eyes.

What does the landscape of these private companies look like?

I mean, SpaceX gets a lot of the press because you see these rocket launches.

And at Virgin Galactic, kind of this idea that tourists will be able to go just out of reach of gravity, right?

What is Jeff Bezos working on?

Where are they all at?

Yeah, so, Blue Origin, which is owned by Amazon founder Jeff Bezos, is probably the second farthest along, I would say, but there are other companies out there, too, who, like Bezos and Blue Origin, have been keeping very quiet.

So there may be a big announcement from something, a start-up in Washington, I believe, that Paul Allen is behind, the name of which escapes me right now.

But they're doing something like Virgin Galactic is doing, but they're doing it in an orbital way.

So for now, Blue Origin has focused on suborbital flight, where you just go in a big arc, and so has Virgin Galactic.

Now they're talking about getting stuff into orbit.

So Blue Origin is building a bigger reusable rocket, and now that they've demonstrated the basic technology in their own style, Paul Allen and his start-up are doing what Virgin Galactic did.

They're dropping a giant rocket from two 747s linked together, and that's going to launch into space.

So they'll save some fuel by lifting it part of the way up in the sky.

So there's all these, like, different players, and we haven't really heard too much from a lot of them, because they're very secretive.

They want, you know, to maintain their trade secrets and market edge.

Yeah. All right.

Dave Mosher, thanks so much for joining us.

Yeah, thanks for having me.

And that wraps it up for this time.

For more on science, technology, and innovation, visit our website, check us out on Facebook and Instagram, and join the conversation on Twitter.

You can also subscribe to our YouTube channel.

Until next time, I'm Hari Sreenivasan.

Thanks for watching.

Funding for this program is made possible by the Corporation for Public Broadcasting, Sue and Edgar Wachenheim III, and contributions to this station.

[ Theme music plays ]