Why Did The Texas Power Grid Fail?

IRA FLATOW: This is Science Friday. I’m Ira Flatow. Later in the hour, Dr. Anthony Fauci answers your questions about the pandemic and what the future might still hold. But first, the state of Texas has been in crisis this week after an historic cold snap and series of ice storms knocked power plants offline and isolated residents in unheated homes.

Millions were without electricity for days. Outages have affected everything from hospitals to water treatment facilities and dozens have died. How could this have happened? Here to talk about the Texas power grid and other stories, Amy Nordrum, Commissioning Editor for the MIT Technology Review. Welcome back, Amy.

AMY NORDRUM: Thank you, Ira.

IRA FLATOW: The stories coming out of Texas have been so hard to hear. Can you fill us in on why this electric grid was so hard hit by the winter storm this year?

AMY NORDRUM: It has been an incredible story this week, and the first thing to know is that Texas does operate its own electric grid which isn’t subject to federal regulations and connects to the rest of the US in only a few places. The rest of the US is much more interconnected and is able to import and export power across state lines much more easily.

These ice storms also happened at a time when some of the facilities in Texas that are usually generating power were already scheduled to be offline due to their normal winter maintenance, and that created more of a problem when other plants began to fail. Many coal and gas facilities, a few nuclear plants and wind turbines also went offline. There was so much demand for power from people trying to turn up their thermostats and deal with these cold temperatures and then many fewer facilities online to generate it caused a major problem.

IRA FLATOW: What about this concern that the grid could go down completely if it got too stressed? We saw people who regulate the grid saying that.

AMY NORDRUM: Yeah ERCOT, the state’s electricity distributor, that was the main problem that they were up against all week. A lot of the blackouts that you saw and the outages, they were actually created by ERCOT. They were intended to keep the grid up as best they could, just to maintain the reliability of the grid.

Even though some people didn’t have power during those times, the idea was to keep the whole grid still up and operating because a catastrophic scenario would be if the entire grid went offline, and then it would take weeks or potentially months to bring the grid back online. And fortunately, that didn’t happen, but hopefully this is a wake-up call to power regulators there and power facilities that invested in the winterization and invested in insulating their facilities for future winter storms like this, is a major priority going forward

IRA FLATOW: Yes, so they’ll have to change the way they do things and prepare themselves for this happening again.

AMY NORDRUM: Absolutely. I mean, it’s happened before, it will happen again. It’s hard to tell exactly when, but these facilities– it’s not that they can’t operate in cold temperatures. They do operate in much colder temperatures in many parts of the US and many parts of the world.

It’s just that the facilities in Texas weren’t winterized or insulated to handle such low temperatures, as they typically would be. And winterizing these means not leaving pipes or gauges exposed to the air. Just putting them inside of a building, and particularly invested in doing that for the state’s natural gas plants, where about half the state’s electricity generation comes from.

IRA FLATOW: Is there any forecast when people in Texas will be able to stop worrying about the power grid? Do they just need it to get warm enough this week?

AMY NORDRUM: Well, yesterday, there was power restored for millions of Texans, which was wonderful news. There are still some without power, and there are many still under boil advisories. There’s now concerns about burst pipes that had happened during the past week that may be contaminating water. So this isn’t a super quick recovery, but temperatures are supposed to rise this weekend and into next week, and hopefully stay that way for the near-term future. So hopefully the worst of this is over.

IRA FLATOW: Let’s go from something that’s worse to much better, and I’m talking about NASA’s Perseverance Rover making a spectacular landing on Mars yesterday. Here’s Dr. Swati Mohan in Mission Control yesterday as the Rover safely reached the surface of Mars.

DR. SWATI MOHAN: Touchdown confirmed. Perseverance safely on the surface of Mars. ready to begin seeking the signs of past life.

IRA FLATOW: Yeah, that was some spectacular landing, wasn’t it, Amy?

AMY NORDRUM: Yes, it really was. So Perseverance landed safely in its intended location, Jezero Crater, yesterday afternoon. And really exciting to see. It’s the fifth rover we’ve had there.

IRA FLATOW: And it’s got some really cool stuff on it, doesn’t it?

AMY NORDRUM: Yeah, I mean, even the systems that it used to touch down were– several of them were new, which is always kind of a stressful thing, testing several new things at once. So it used the advanced visual camera system that took images and analyzed them in real time to kind of spot hazards and find its way to an ideal landing spot.

And now that it’s there, it will be doing a number of experiments on the surface of Mars. One called MOXIE that will test whether it’s possible to convert the carbon dioxide in Mars’s atmosphere into oxygen, which could be useful for future Martian colonies. And then another one that I’m excited about, a tiny helicopter called Ingenuity that’s going to try to do the first self-powered flight on another planet.

IRA FLATOW: And what makes radio guys like me excited, it does have a microphone on it. Although, I’m not sure who’s going to say, “Testing, one, two, three,” before they actually try it out.

AMY NORDRUM: Yeah, we could have the first audio clip from Mars, which is super cool. You’d be able to potentially hear, if everything went well, what the descent itself sounded like, what the landing sounded like, and maybe even what the surface of Mars sounds like, the winds and the atmosphere there.

IRA FLATOW: Yeah. How soon does it start collecting data? I would imagine it’s already setting itself up to do that.

AMY NORDRUM: Yeah, it is. I know that now, NASA is going to be doing several weeks of testing on the rover, just to make sure that everything is in order and that the systems are working properly. And then it’ll really begin to explore the crater in earnest in a couple of weeks, or certainly by this summer.

IRA FLATOW: Let’s move from the Red Planet to a planet that may or may not exist. We’ve been talking about this for years on this show. You’ve got an update on the debate over whether there is a Planet Nine in the Solar System.

AMY NORDRUM: That’s right. Yes, back in 2016, researchers at Caltech had said they thought there might be a ninth planet orbiting far out there in our solar system. And their theory was based on an observation of these far-off objects called extreme trans-Neptunian objects, or ETNOs. And they detected six of these objects that seemed to be orbiting in a peculiar way. They had these elongated orbits around the Sun instead of a circle. And they appeared to all be kind of clustered in one direction, which led them to think that there was a giant planet out there whose gravitational pull had made all these smaller objects line up in that fashion.

But last week, a team from the University of Michigan posted a pre-print paper that challenged this theory, saying it could have just been a product of where astronomers had been looking at that time. So for their new study, they looked at 14 other objects like this that weren’t included in the original analysis. And they said they didn’t find any evidence of clustering and that it could just be a random distribution of ETNOs throughout space.

And since these ETNOs have these long orbits, it’s really only possible to detect them when they’re close to Earth. So if you look at any particular time, you might see a few that happen to be clustered or look like they’re clustered in one direction, but if you look again later, you’d see a lot more and they might be in a different area. So this evidence does seem to challenge the Planet Nine theorists. And it’s still a small sample size. This new one just had 14 of these objects, the original had six so there might be still many more out there that we could detect over time and give us a better picture.

IRA FLATOW: The next story is about artificial intelligence tools used in hiring and other kinds of assessments. We have talked about the very legitimate concerns about racial, gender, and other bias these algorithms can embody, and this story looks at one of the most commonly proposed solutions for bias. Tell us more about that.

AMY NORDRUM: Yes, as you said, AI bias, it’s a well-known problem in AI, in automated systems producing biased results when the data that they’re trained on is biased in any number of ways. And the solution that’s often proposed is AI audits, and there’s actually a rule being debated right now in New York City that would make these kinds of audits mandatory for companies using AI. So an audit would kind of interrogate or examine the algorithm used in an AI tool and determine whether or not it’s fair, whether or not it’s biased.

And so reporter Hilke Schellman looked into these audits for us to see what they could really achieve. And in her report, she found that they’re really not well-defined, and they’re often much more limited in scope than the average person might think. While they’re a good step, they’re certainly worth doing, they’re not a final fix. They can’t say with certainty that there’s not the chance or risk of bias in any AI tool.

IRA FLATOW: So what we need is some standardization, is what you’re saying.

AMY NORDRUM: Yeah, there’s many AI auditing firms right now just kind of doing their own thing, operating AI audits independently with no real common definition of what that means or how it examines fairness or what it tests for. And so that would be a good step, is standardizing this across the industry and across firms that are working on this. And it might also be helpful to have a third party involved, whether it’s a government regulator or somebody who isn’t involved with the company, isn’t paid for by the company like many of these auditors are, to just avoid any conflict of interest.

IRA FLATOW: And we end today with the Beatles! Oh, the insect kind, and learning more about how they smell. Wow. Tell us about that.

AMY NORDRUM: Yes, in this case, bark beetles. These are a huge problem throughout the West. They’re native to the US, but their populations have exploded as temperatures have increased and they’ve killed millions of spruce trees in Colorado and California and Alaska. And they’re also a big problem in Sweden and Asia. So researchers at Lund University have been trying to figure out how these beetles find their way around and what might be done to stop them, basically.

And they knew that these beetles rely on smell because they have tiny eyes and really bad eyesight. So they’ve done some work to figure out what smells the beetles are attracted to. And they used a genetic technique to examine the antenna of these beetles and found that each beetle has 73 different olfactory receptors on its antenna. And then they took a panel of different smells and pheromones and tested them across the receptors to find which ones they responded to.

And on their first go, they found a number of pheromones that come from beetles themselves that the receptors responded to, so it does seem like the beetles are able to detect and move toward other beetles, find their friends. And they’re continuing this work for the rest of the receptors. They’re starting to find evidence that the beetles can also respond to pheromones that come from competitor species, and also potentially compounds in the spruce trees themselves, which might be how they navigate to trees.

So they plan to continue this work until they’ve figured out what all the receptors respond to and what all smells are involved. And they’re linked up with another team working on a kind of biosensor that would be able to use some of these pheromones to maybe detect where beetles are present in trees, when it’s not yet obvious just from looking at the tree that it’s infested with beetles, and perhaps do some early treatment or remove that tree if it seems like it’s a lost cause.

IRA FLATOW: Do you think if you find what pheromones the beetles are attracted to, you could lure them away from the tree using that smell?

AMY NORDRUM: That’s another possibility. There’s lots of different ways that this information might be used for or against the beetles. Another technique would be to create a kind of synthetic pheromone that mimics the smell of other beetles that gets there first and kind of blocks their ability to detect and find their way to another tree or to another beetle. That’s certainly a possibility.

IRA FLATOW: Yeah, maybe you can find something the beetles hate to smell, and then you could spray that on the tree and they won’t go there. All kinds of stuff, once you get to know what they like to smell or don’t like to smell. Thank you, Amy.

AMY NORDRUM: Thanks, Ira.

IRA FLATOW: Always interesting stuff. Amy Nordrum, Commissioning Editor for the MIT Technology Review.

Copyright © 2021 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/