Advanced science.  Applied technology.

Technology Today Podcast
Artist concept of CYGNSS in orbit above planet earth

Episode 10: Unraveling a Hurricane

How to Listen

Listen on Google Podcasts, Apple Podcasts, or via the SoundCloud media player above.


In this Episode

Collecting data from a hurricane before it makes landfall is key to saving lives and property. While current hurricane models can predict the path, there are often gaps in accurately predicting a storm’s intensity or strength. NASA’s Cyclone Global Navigation Satellite System (CYGNSS) is made up of eight microsatellites designed and built by Southwest Research Institute. The satellites use GPS signals to cut through the world’s strongest storms. Our guest in this episode is an SwRI engineer who is part of a team operating CYGNSS. Hurricane season and year-round, they are taking on tropical weather, measuring winds and working to better understand and predict hurricane intensity.

Listen now as we discuss a new era in hurricane hunting.


Below is a transcript of the episode, modified for clarity.

Lisa Peña (LP): Unraveling a hurricane—a constellation of microsatellites is sweeping Earth's hurricane zone, penetrating thick clouds and heavy rains to gather crucial weather data. Our guest today is part of a team that operates this cutting-edge constellation known as the Cyclone Global Navigation Satellite System. He tells us what it takes to track trouble in the tropics. We're on hurricane watch on this episode of Technology Today.


We live with technology, science, engineering, and the results of innovative research every day. Now, let's understand it better. You're listening to the Technology Today Podcast presented by Southwest Research Institute. Hello, and welcome to Technology Today. I'm your host Lisa Peña. 

We are in the midst of the Atlantic hurricane season right now, the time of year when hurricanes tend to form in the Atlantic Ocean. Hurricane season runs from June 1 to November 30.

When a hurricane is churning, collecting data is crucial to understanding the storm's intensity and path, to ultimately keep people safe. The Cyclone Global Navigation Satellite System, or CYGNSS, as it's called, represents a new era in hurricane hunting. The satellite system can measure a storm's rapidly changing intensity. And this system was designed and built right here at Southwest Research Institute.

Our guest today is CYGNSS systems engineer, William Wells. Thanks for joining us, Will.

William Wells (WW): Hi, Lisa. Thanks.

LP: Hi. So let's start with a description of the Cyclone Global Navigation Satellite System, CYGNSS. It's a constellation of microsatellites. But how many are there? How big are they?

WW: Sure. CYGNSS is eight microsatellites, as you said. Microsatellites typically are in the range of 10 to 100 kilograms or so. We're about 29 kilograms, about 60 pounds each, and there's eight of them. And they're all spread around the same orbit, but they're spread around the circumference of the orbit, so that we get great global coverage in the tropics.

LP: OK. So who are you working with on the CYGNSS mission, and what role does each entity play?

WW: This is a NASA mission. It is a Earth Venture-Mission, the first Earth Venture-Mission. And we've partnered with the University of Michigan in Ann Arbor. Our principal investigator is Dr. Chris Ruf there. He's a leader in the field of this type of research.

And we really did everything else for this. We procured some commercial components, but we did all the design, test, integration, and operations of the satellite out of Southwest Research.

LP: OK. So as you mentioned, this is a NASA mission, but these satellites were designed and built here at Southwest Research Institute. So what was your starting point with all this? How did you even start building them?

WW: That's a great question. So it all goes back to, I guess, who you know. There was Dr. Chris Ruf at Michigan. He was interested in GNSSR research. That's Global Navigation Satellite System Reflectometry. It's using GPS or other global satellite system reflections off of the surface of the earth to do science. So he was interested in that technology.

And we had another, there was a guy, Aaron Ridley, at the University Michigan. He was interested in CubeSats, and how to deploy many satellites into a constellation, and what type of science you could do. He actually used to work at Southwest Research before he was at Michigan. And we have Greg Fletcher. He's an employee here at Southwest Research. He actually used to work at University of Michigan. So these guys knew each other. And Greg keeps a pulse on the NASA Announcements of Opportunity [AO].

So when we saw this Earth Venture-Mission opportunity come up, they had pretty much decided that the CubeSats weren't big enough to host the type of science payload that was needed to do the science.

LP: OK, wait. The CubeSats were not big enough to host the type of science payloads needed? Did I get that right?

WW: Yes.

LP: OK. What does that mean?

WW: So CubeSats are small, even smaller than the CYGNSS spacecraft. So you're talking about something that's maybe, you know, 10 centimeters by 20 or 30, or maybe a little bit bigger. But it's just not a big enough satellite to host the science instrument that we need to do the science. So we needed, you would need a little bit bigger spacecraft, like what we have for CYGNSS.

And so these guys started talking, got together, and decided that, hey, there's this NASA Announcement of Opportunity. You know, we could launch 8 or 12 small satellites. We could put this type of science instrument on it, and we could do some great science. And that's really where it all started.

LP: So once it was, or let me back up to the actual mission description. What was it that NASA said they needed?

WW: Well, I'd have to go back to the exact language in the AO it's been so long. So Earth Venture is intended to study the Earth. A lot of the work that we've done here at Southwest in the past has been in response to missions that study other planets or astrophysics or heliophysics, interaction of the earth and the sun, etc. Not as much science instrument builds out of the Space Science Engineering Division for Earth sensing missions.

And so this Earth Venture Announcement of Opportunity was calling for earth science activities. And this filled a particular need in an earth science field. And so it was a great candidate for that AO.

LP: So how was it decided that this team could hone in on hurricanes and analyze that type of data?

WW: Well, I think the larger tropical cyclone community had a need for the type of data that CYGNSS ultimately was able to produce. I think it really goes back to the fact that in the last few decades, there's been a tremendous amount of improvement in forecasting the track of the storm, where it's going to make landfall, but almost no real improvement in being able to forecast the intensity of the storm when it does make landfall. And that's critical to deciding whether to evacuate coastal regions when the storms are approaching.

So when that community was asked, what do you need to better predict the intensity of the storm, they said, we need wind data. We need lots of wind data. We need to incorporate that into our large models that forecast the storm intensification processes. So give us that wind data. And CYGNSS was able to do that for them.

LP: So you had the minds that were able to decide what the best solution would be. You have the problem that NASA presented in that this data that was needed, needed to be enhanced, really. So where do you start the actual physical building of microsatellites? What's the first component, I guess? Where does that come from?

WW: Yeah. So it's, you come up with a concept. And obviously, there's lots of iterations, especially early on, to develop the design. One of the biggest constraints was fitting all of these eight satellites, was what it ended up to be, onto a single launch vehicle.

It's a relatively small mission in terms of the actual cost. We built these eight satellites for right around $100 million, which may sound like a lot to some people. But in the scheme of things, this is eight satellites for that cost. And it's really not a large mission. And for this size of a mission, you don't get a very large rocket.

Actually, you get a small rocket, about the smallest rocket that you can get that'll give you access to space. So physically fitting the eight spacecraft in the fairing of the launch vehicle was definitely a design constraint.

And that kind of drove the initial shape and size of the satellites. And we took measures to really optimize the design to accommodate the science instrument, but then also to fit on the launch vehicle.
Lisa Peña and William Wells against a solid blue background

Left to right: Lisa Peña and Will Wells

LP: So you need all of the eight satellites to fit on this rocket, to blast them up through the atmosphere, so that they can take their place above the earth and do what they need to do?

WW: Yep. CYGNSS orbit is about 520 kilometers above the surface of the Earth, and 35 degrees inclination. So what that means is as we go around the Earth, we're mostly covering the regions plus or minus 35 degrees latitude. Our science measurements actually go a little bit higher than that, because our antennas are pointed off to the side of the spacecraft a little bit.

LP: So what is the main job of the satellites now that they're up there?

WW: Definitely, the primary mission is to collect wind data. So 24/7, 365, all eight spacecraft are taking these measurements. We don't actually measure the wind speed from space.

What we're able to do is we measure the ocean surface roughness. And there's a strong correlation between the roughness on the surface of the ocean, and how fast the wind is blowing at the surface.

So we make use of the GPS signal. We receive the direct signal. We also receive the reflected signal off of the ocean surface. And by comparing those two, we're able to measure the roughness, and therefore, infer the wind speed at the surface.

LP: Are you collecting data from all oceans, or specifically the Atlantic?

WW: Everywhere that we fly over. So all the way around the world plus or minus about 35 degrees latitude, we're collecting data. And the primary mission is definitely the ocean surface wind speed, but especially since launch, and after the last 2 and 1/2 years of operations, we've realized there's some other cool things that we can do with the CYGNSS data as well over land.

For example, CYGNSS is able to identify inland water bodies quite well. And the resolution that we have can pick up small bodies of water, you know, rivers and streams, and so anything that has to do with standing water or flooding, that sort of thing. Even soil moisture is another thing that we're getting involved with now. There's a lot of land applications as well.

LP: So that's cool. It extends from the oceans, over the land-- pretty much, this is a global sort of mission.

WW: Yeah. It really is. And most recently, we completed our two-year primary mission. And we're into extended mission now. And as part of that, we've hired on several new co-Is, co-investigators, to lead new investigations that are oceanographic missions, as well as land science missions as well.

LP: OK. So we get that CYGNSS is collecting this wind data, and is able to measure the intensity of big storms like hurricanes. But ultimately, what is the goal of collecting all of this data? Why do we care? Why do we want to have that?

WW: Well, like I said, we want to improve the ability to forecast the storm intensity when it makes landfall. So ultimately, that's going to result in being able to save people's lives by making sure they get evacuated when they need to if the storm is going to be strong. And also, trying to save money by not evacuating if the storm is going to fizzle out by the time that it makes landfall.

And right now, you have to be careful of waiting too long to issue an evacuation order. Because the storm may or may not intensify. You don't know. So the better we can predict how strong the storm is going to be, the better off we'll be there.

LP: So when do you think we'll have enough data, enough trips around the world, to say that this system is 100% reliable, and maybe it becomes the new mainstream for hurricane hunting?

WW: That's a tough question. I know I'd love to say that we're on the brink of being right there, and that CYGNSS was going to absolutely revolutionize it. The reality though is that the tropical cyclone science community, I mean, they've spent decades developing these models. And they take in information from all different sources.

So CYGNSS is doing lots of great stuff. But we're just providing a portion of the information that feeds into these large models that people use. So it's really a long process to become, say, an operational mission, and for everyone to absolutely agree that your data is better than everyone else's. So, you know--

LP: We see now on weather reports all those spaghetti models. And at the beginning of a tropical system, they're all over the place. And then they start to get a little bit more cohesive as the storm approaches land. Are you-- do you think in the long run, CYGNSS has the ability to replace some of these models, if not all of them?

WW: I wouldn't say replace, but definitely contribute to improving them. And they will get better over time. One of the things that we've done so far, and will continue to do, is we rerun those forecast models, those spaghetti models and intensity forecasts. And we rerun them after incorporating all the CYGNSS wind data, and compare to what actually happened. If we had used CYGNSS data, would the forecast have matched truth better?

And so we have shown that if you incorporate the CYGNSS data, yes, you both improve the track and the intensity forecast. So as we go forward and we collect more data, and we further refine the data processing techniques that we use to make the data more accurate and more consistent, it'll just continue to improve.

LP: What does that data that's beamed back look like? Are we talking about graphs or numbers or--

WW: Yeah. So actually, that's a good question. Our primary science data product is what we call a delay Doppler map. And it's really a 2D image. It's got the delay on one axis, and the Doppler shift of the signal on the other.

And so what we're talking about here is you can imagine the GPS constellation broadcasting its signals down towards the earth. And the CYGNSS satellites receive the signal directly from the GPS satellites. But they also receive the signal that bounces off the ocean surface.

And depending on exactly how that signal bounces off the surface, the geometry is a little bit different, and so the time that it takes for the signal to reach our satellite is a little bit different. So we can compare how soon did we get the direct signal, and how soon did we get the reflected signal. And that's the delay component of it.

And then also, the Doppler shift component is, how did the frequency of this signal change a little bit due to the relative motion of our satellite and the transmitter on the GPS satellite? So you put that together, and you also look at the received signal strength.

And so you can color code this image and say, show red areas where the signal is the strongest. But you show this signal intensity spread across a map of the delay and the Doppler shift. And from that, you can actually create a picture. And it kind of has a little bit of a crescent or a horseshoe type shape on it.

And really, that kind of dictates how fast the wind is that-- or what the roughness is. And then we can back out what the wind speed is there. So that's our primary data product. And that's what we downlink 24/7.

There's a bit of processing that goes on on the satellite to generate that data, though. And we can also collect the very raw form of those GPS signals that we were receiving. Then we can downlink that onto the ground. And we can process it in any number of ways. And it provides a lot of flexibility to do other things.

LP: So these satellites were launched in late 2016?

WW: That's right.

LP: OK. And so when you started, was there some uncertainty about the outcome, whether this data collection method would really work?

WW: No. We were quite confident that it was going to work. To our advantage, this science instrument is developed by Surrey Satellite Technology in the UK, actually. And it's really kind of a commercial GPS receiver. It's pretty fancy, but you can buy it.

And they flew a similar instrument on a spacecraft called TechDemoSat. And as the name would suggest, that was a technology demonstration mission. So they had a number of different technologies, one of which was this GNSSR receiver.

And so we were able to get some of the raw data actually, like we're just talking about. They didn't have any onboard processing of the data like we were just talking, but they downlinked some of that raw data. And that's actually what we used to develop the processing techniques that run onboard the spacecraft.

And so after we developed that in our software, we could run the real TechDemoSat data through our algorithms, and prove that it was going to produce the science data products that we were interested in. So we were pretty confident that this measurement technique was going to work, certainly for the ocean-related wind speed retrievals.

Some of the land science stuff is-- we had ideas that it might be interesting. But we weren't so sure that it was going to be as cool as it is. And so that was a bit of a surprise after we launched it, how good some of the land science data was.

LP: So as you mentioned earlier, you've been able to find other uses for this data?

WW: Yeah.

LP: That's cool. When tropical weather flares up, we hear about those planes flying into hurricane winds and assessing the wind speeds. And those planes give us those updates every few hours, and tell us whether or not this storm is scary, how intense it is.

So how is the CYGNSS system different from other hurricane data collection methods, like planes?

WW: That's a good question. Those planes actually have instrumentation on them that's very similar to what CYGNSS does. The difference is instead of using the GPS signal, they actually carry a transmitter on the aircraft itself. And so they do a similar type of measurement.

They also do several other types of measurements. And the planes can fly through different altitudes in the storm. They can collect pressure information, humidity, other things that CYGNSS doesn't necessarily do. But they certainly don't have the global reach that CYGNSS has.

Like I said, we're 24/7, 365, all around the world. And those planes typically don't get launched until the storms are really threatening the United States. And they're really complementary data sets.

In fact, we, especially in the 2017 hurricane season, we did a lot of coordinated underflights-- underflight, overflight, depending on what perspective you want to look at. But we provided NOAA, who operates those planes, with the track of our satellites as we were going to be going over the storms.

And we tried our best to have them fly the airplane right underneath the spacecraft along the same path at the same time, so we could make the measurements of the same area in the storm. And so that way, we can compare the CYGNSS data to the data that the hurricane hunter aircrafts are producing.

And it's really one of our best ways to validate our high wind speed measurements for CYGNSS. So they're very complementary techniques.

LP: Do you think CYGNSS could replace those planes at some point?

WW: I don't think necessarily replace. I think that, like I said, they're very complementary. And I think that if CYGNSS got to the point of being a more operational mission, where we would downlink our data within only a couple hours after flying over the storm, then we would kind of be getting toward the same type of real time information gathering that those planes produce. But I think there's definitely room for both of them to exist well into the future.
Barry Wind Data Courtesy of University of Miami

The CYGNSS wind speed data collected around the world on July 13, 2019, as Hurricane Barry made landfall on the U.S. Gulf Coast.

LP: Hurricane Barry made landfall in Louisiana as a Category 1 hurricane last month. Did CYGNSS contribute any insight on that storm in particular?

WW: We did take some special data over Barry. We had several spacecraft that had-- several of the eight had nice tracks over the center of the storm. So we did do some of our special data collections.

You know, I guess this hurricane season is off to a little bit of a slow start. We haven't seen, in the Atlantic at least, too many strong systems developing yet. But yeah, we do our best to collect special data overall of the active storms. And that data has been collected and added to the archive.

LP: Let's talk about Hurricane Harvey. That was a major storm. It made landfall along the Texas coast near Port Aransas as a Category 4. That was back in August 2017. And as it moved inland, it caused devastating flash flooding.

So in this case, CYGNSS did provide useful data in the aftermath of Hurricane Harvey. What did it reveal?

WW: That's an interesting one. That was-- certainly, it was a very strong storm. So we did lots of special collections over Harvey for wind speed as it approached. But it was around the time, it was in 2017, so the first hurricane season. And we were really first figuring out that we could do this new, cool land science, where we can see standing water.

And so the flooding was just so widespread from Harvey, that we were able to make some time-lapse movies. They're not necessarily a lot of frames, but time-lapsed movies showing how all the floodwaters inundated the whole Houston and Southeast Texas area. And the flooding increased, increased, increased more. And then over the next few days, the flood started to recede and flow back out into the ocean.

And so yeah, this inland water detection that CYGNSS is able to do, that was right around the time when we were first really figuring that out. And so we were able to produce some cool movies showing how the flooding evolved and then receded from Harvey.

LP: So long term, images like that could help people on the ground probably understand better where the worst flooding is, and hopefully, help people on the ground.

WW: Yeah, definitely. Situations of flooding like that from hurricanes. Also, we've been doing similar stuff over the last couple of years looking at rivers when they flood. It may not be a storm right there where the flooding is happening. It may be upstream. But large rivers can flood. And so we're contributing to information about that.

And there's also a lot of interest in places like the Florida Everglades or the Amazon Basin, where it's really hard to get people on the ground there, because it's remote and in the forest. It's hard to get there. But CYGNSS is able to kind of keep tabs on the standing water in those locations, and provide a history of what's going on there over years. So that's a great data set. A lot of people are interested in that.

LP: So have you ever had to be on location for a hurricane, or in the aftermath of a hurricane for CYGNSS?

WW: I personally have not. I think the closest we came was there was a storm that was headed toward San Antonio, I think in 2017. I don't remember the name of it. But it definitely fizzled out.

But I don't know if you remember, they shut down lots of grocery stores and other places right around that time. And we thought we were going to get hit. And we thought it was kind of ironic. Because CYGNSS had just launched, and then a hurricane is coming straight for us in San Antonio. But it turned out to be not so bad.

LP: Yeah. So as a-- in my past life as a news reporter, I did cover tropical weather. And it was Hurricane Dolly along the Texas coast in 2008. And as I mentioned before, the hurricane's one thing, but the aftermath is another.

And just to see power out for days, and the flooding, especially in some of those smaller neighborhoods and areas. And people just trying to live for at least the next couple of weeks after that. It's rough. So I think anything we can do to better predict the storm's intensity and path and help people, I think it's great.

WW: Oh, yeah. Absolutely.

LP: So what are your hopes for the CYGNSS mission long term?

WW: Long term for me, as much as I'm interested in the hurricane science, I'm not a scientist. I'm an engineer. So I take a lot of pride on being part of the team that developed these satellites, and did it for such a low cost, relatively.

So I'm looking forward to the CYGNSS constellation continuing to do good science for years to come. And so I want to see the spacecraft survive as long as they can. I mean, they'll eventually return to Earth, burn up in the atmosphere.

And just kind of like your cell phone and your laptop, the battery in the spacecraft is only good for so many charge/recharge cycles. And other systems-- we have moving parts, reaction wheels in the spacecraft for attitude control.

Space is a harsh environment. So eventually, these systems will start to degrade, and the satellites will go offline and re-enter. But I'd like to see them last for a long time and continue to do good science. And I'm looking forward to applying all the lessons that we learned here at Southwest Research to building the next mission, the next set of satellites.

LP: So they're just going to, what did you say? Blow up in the atmosphere?

WW: Definitely don't tell NASA they're going to blow up.

LP: I didn't say blow up. What was the correct term?

WW: I said burn up.

LP: Burn up. There we go. They're going to burn up. And that seems like such a sad ending for such a great mission.

WW: Yeah. Yeah.

LP: I guess that's the reality, though.

WW: That's the reality. Lots of spacecraft do that.

LP: But then you'll learn from what you-- well, you'll learn from them. And you can rebuild probably more satellites, bigger and better.

WW: Yeah, definitely. Although, bigger isn't necessarily always better, talking about--

LP: Same size and better.

WW: Yeah. Or even smaller. Because--

LP: Smaller and better.

WW: --the reality is that the cost of getting this hardware into space is high. So the smaller and lighter the spacecraft can be while still doing the good science, the more of them you can launch for the same price.

So there's been some discussion about whether there's going to be a CYGNSS follow-on mission. And I know the University of Michigan is working on some next-generation type instrument hardware that could potentially launch.

So we may combine-- or someone may combine, if we find the right opportunity to do the mission, some other science instruments as well. Maybe some instruments that measure precipitation and cloud imaging, along with the CYGNSS payload, next-generation type CYGNSS payload. And that way, having all these-- a few different instruments on the same spacecraft would-- you just enable even more science.

LP: How did you become involved in CYGNSS? Did you, at the start of your career, envision supporting a NASA hurricane mission?

WW: I definitely had no idea that I would be doing anything related to hurricanes. I came to Southwest Research and worked on a few different projects before CYGNSS. CYGNSS was definitely cool, because it was the first spacecraft that we built here at Southwest Research. We have a long history of building science instruments of different types for NASA missions, but we had never actually built the spacecraft here in San Antonio.

And so even when I was in school and pursuing an engineering degree, I always thought it would be great to design-build a spacecraft. That would be awesome.

So when I found out that we were going to propose to build the actual spacecraft here at San Antonio, I was definitely very excited, and wanted to get on the team doing that. And was able to kind of get my foot in the door with the people that were working on the proposal.

So I helped a little bit with the proposal, and then pretty much have been contributing to the CYGNSS mission all the way through operations now.

LP: Got to be experience of a lifetime.

WW: Yeah, absolutely. I mean, it's been great for me. It's been tremendously rewarding. And it's really set the stage for, I think-- it's really a pathfinder for the company here, building small satellites. And also, kind of just setting an example in the larger space community of the cool things you can do with relatively cheap, small spacecraft.

LP: What your team has learned from CYGNSS has led to other projects for SwRI specifically NASA's-- OK, bear with me as I pronounce this. It's NASA's Polarimeter to Unify the Corona and Heliosphere, or PUNCH mission. OK, so that mission will take images beyond the sun's outer corona. So connect the dots for us. How did CYGNSS lead to PUNCH, which is awesome?

WW: So, yeah. PUNCH is awesome. And I've been working on that as well. And we just recently found out a few weeks ago that NASA had selected this mission for implementation. Totally different science. Nothing to do with hurricanes.

As you just mentioned, it's studying the kind of a gap in our understanding of the solar wind and the sun's corona, and really, what's going on as the mass of solar particles get ejected off the sun. And how does that become the solar wind, just trying to kind of fill in some gaps in our understanding there. Tremendously oversimplified that, because the science is awesome.

LP: That's fine. That works for me.

WW: But what's-- to tie it back to CYGNSS, it's different science, but the small spacecraft are very similar. And the PUNCH spacecraft are a little bit bigger than CYGNSS, but kind of the same size. And again, we are designing something to fit in a relatively small rocket, so PUNCH will be four spacecraft.

But almost-- there's so many things that we learned for CYGNSS are directly applicable to building the spacecraft for PUNCH. Mistakes that we made, things we realized that worked really well, we can apply all of that to building PUNCH. So we don't have to figure everything out for the first time now. We proved that we can do it on CYGNSS. And so now, we're just ready to do it even better on PUNCH.

LP: Yeah. So you're just going further and further.

WW: Yeah.

LP: Great. OK. Well, congratulations with all your success on CYGNSS, and now this new development, PUNCH. So we'll be watching to see where that takes your team. That's great.

WW: Thanks.

LP: So great conversation about what it takes to understand and track hurricanes today, and ultimately, the goal to use that information to keep people safe. So thank you for the work you do, Will. It's important.

WW: Thank you, Lisa.

LP: And that wraps up this episode of Technology Today.

Subscribe to the Technology Today Podcast to hear in-depth conversations with people like Will Wells, changing our world and beyond through science, engineering, research, and technology.

Connect with Southwest Research Institute on Facebook, Instagram, Twitter, and LinkedIn. Check out the Technology Today Magazine at And now is a great time to become an SwRI problem solver. Visit our career page at Thanks for listening.


SwRI’s Space Science and Engineering Division develops technology that helps clients better understand weather and climate patterns on Earth as well as atmospheres on distant planets and moons.