S&P Global Inc. 0 Earnings Call

Hello everyone, welcome to today's webinar. My name is Sarah James, and I lead the Tech, Media & Telecoms News team within S&P Global Market Intelligence. It is my pleasure to moderate today's webinar, "Orbit as the Next Data Frontier: Capital Flows, Risks and Realities." Before I introduce my guests, a few housekeeping items. We recognize that the topic of today's webinar is of great interest. We want this to be an interactive session and encourage you to submit questions for discussion. At the bottom, these icons will allow you to interact with us throughout the session. I would like to point out the Q&A widget, which can be used to submit questions to the panelists as well as the survey widget. Please take time to fill out our short survey after the webinar. We really value your insight.

The webinar is being recorded. An on-demand version will be available shortly after we conclude. If you encounter technical issues during the program, please try refreshing your browser. If issues persist, please use the Q&A widget to contact us. A member from our technical team will assist you. It's my pleasure to introduce today's panel. I'll begin with my colleague, John Fletcher, a Senior Research Analyst with S&P Global Market Intelligence, Kagan. John leads the Americas research team for broadband, multichannel video, and mobile, with a focus on how the U.S. can close the broadband digital divide. Just as a hint, LEO satellites are expected to play a big role there. Next up will be Johan Vermaat, a senior research analyst at 451 Research from S&P Global Market Intelligence. Supporting the IoT edge and digital industries practice, Johan covers the digital transformation of the energy sector.

Also from S&P Global Market Intelligence's 451 Research is Ellie Brown, a research analyst covering quantum technologies. In terms of today's agenda, we'll begin with a quick scene setting from me and Ellie, followed by a 10-minute presentation from John about satellite broadband. Then a 15-minute presentation from Johan and Ellie about the infrastructure expansion that is currently underway in space. At that point, we'll begin answering questions, though we encourage you to use the Q&A widget to submit questions as they come to you during the presentations. Just to get us started, what are we talking about and why are we talking about it now? We would be remiss if we didn't at least talk about the SpaceX IPO earlier this year. That brought a lot of fresh attention to the space technology market for a lot of different reasons.

It was a huge market event, catalyzing the overall IPO market in the U.S. A total of $146 billion of common stock was offered in U.S. IPOs in the first six months of this year, dwarfing the $26.46 billion offered in the first half of 2025 and the $42.09 billion offered all of last year, according to S&P Global Market Intelligence data. Of the first half total this year, $86 billion was raised by the SpaceX IPO, accounting for more than half of the overall total. This webinar is by no means going to be all about SpaceX. We just can't ignore the impact that that company had on the space technology market and the interest around it. That's also true of the M&A, not just the IPO market. The space technology market has seen $338 billion in investment in the first half based on 193 transactions.

This includes both M&A and fundraising. That's up from just under $22 billion from 133 transactions in the first half of 2025. SpaceX is playing a major role here with its $250 billion combination with Elon Musk's xAI, or now SpaceXAI, and then some of its other follow-up transactions more recently. The space story goes so far beyond SpaceX's mega IPO and its mega M&A deal. It's also about how competitors are turning to M&A to compete with SpaceX, which is the market leader in both satellite broadband and rocket launch capacity. Amazon is racing to catch up with its LEO constellation, and in mid-April it bought the mobile satellite services provider Globalstar for over $11 billion.

The deal, which can largely be viewed as a spectrum play, will integrate Globalstar's satellite operations, infrastructure, and globally authorized mobile satellite service spectrum licenses into Amazon Satellite Network, Amazon Leo, and enable the deployment of direct-to-device services, D2D services, starting in 2028. In June, Rocket Lab agreed to buy Iridium for $8 billion, combining Rocket Lab's launch business with Iridium's mobile satellite network. We've got all of this deal activity that is really designed around these companies positioning themselves to better compete with SpaceX and getting their satellite networks up there, getting that launch capacity up, and seeing if they can play catch up.

Before we get too deep into these different constellations and why each was attractive, I want to hand it over to my colleague Ellie to walk us through some definitions and terminology, and then she'll hand it off to Johan and John to get us started. Ellie? Thank you so much.

To dive into the space economy, we figured we should probably talk about setting the scene for what we're actually looking at. Hopefully you can see on the slide, we've got a bit of a diagram showcasing some of the key terminology, and some of this might be familiar to our listeners. If not, great. If it is a review, take it as such. We want to talk about, in this webinar, mostly this focus on orbits. You might hear some terms such as low Earth orbit or LEO. That is on the left of your diagram, and you can see that usually that's consisting of satellites for communications, Earth observation, scientific research. As an orbital space, it's very full and very busy. You'll hear the term LEO quite a bit.

We also have some other orbits that we might be discussing. We've got our medium Earth orbit, which usually consists of some navigation satellites such as those found in GPS. We've also got geostationary orbits. Those would be communications, broadcasting, weather satellites, anything that you want good coverage within that orbital sphere. Some other terms that we want to bring up. We have cislunar space. That just means the region between Earth and lunar orbits. As the new space economy grows, this area is becoming more and more important. We're talking transportation, communications, logistics. How do we interact within space as a whole? Not just terrestrially, but as we go to the moon, as we build out some of our capabilities there, right? How do we focus on that cislunar space and expand that?

On your far right, we've got lunar orbits. There's a couple that we're going to talk about. We've got low lunar orbit, there's near-rectilinear halo orbits as well. As we develop the space economy and move towards the moon in a bigger way, lunar orbits are going to play just as big of a role as our Earth orbits play for us today, right? We're going to need satellites, we're going to need communication. All of those are important to keep in mind. We've also got the lunar surface. While we aren't going to discuss it as much in this webinar, it's a key part of the eventual new space economy. That would be lunar exploration, space mining, research stations, even tourism. Along the bottom of the slide, you can see some of the reasons why each of these regions matter.

We've got communications, navigation, Earth observation satellites, right? We talk about weather and monitoring. That's what a lot of the satellites that we see today have been working on. We're going to see an expansion of that. We've got space transportation for that cislunar space area. The lunar economy, which fits into this new space economy. Of course, driving a lot of this, in addition to economics, is science and exploration. We love exploring space. We're using it for science, for exploration, but also to feed back into the existing terrestrial economy. We've got this lovely give and take there. Anytime you hear some of these terms, this is what we mean. This is the scope of our new space economy. Not comprehensive, but hopefully a good setting of the scene.

I'm actually going to pass it back over to my colleague, Johan, who's going to talk about some of the differences between not just the new space economy, but how it compares to this older space economy that we might have seen previously. Johan? Thank you, Ellie, for setting the stage.

As you mentioned, the new space economy, that's what we're talking about. How is it different from the old space economy, the timeline. Looking at it, just a very brief historic overview. These are, I think in my life, my parents' life even, the events that got you up in the middle of the night, the things, the landing on the moon. That's why you bought a television and watched these. Obviously, this early space race was dominated by the U.S. and the USSR with national space programs, organizations like NASA, Roscosmos, in Europe later, the European Space Agency, with the big defense primes as the primary contractors that could build the space shuttle. They had very high launch costs, enormous, massive rockets.

The space shuttle or whichever launch vehicle bringing up huge satellites into geostationary orbits, primarily focused at weather information, communication satellites. In Europe, we could see American television shows, but also the navigation systems. Ellie just mentioned GPS, which is just the U.S. version. The type of satellite is GNSS, a global navigation system satellite. GPS is the U.S. version. GLONASS, the Russian version. Some of these Russian navigation satellites are already up to 50 years old and still operational, more or less. BeiDou is the Chinese version. In Europe, we have the Galileo GNSS satellites. It was primarily dominated by high launch cost and up to as much as 30,000 per kilogram. If we're talking about the new space economy, we see that emerge in the early 2000 as we begin to move towards more reusable rockets. The foundation of SpaceX. First private spaceflight missions.

Smaller satellites bring down the launch cost. It opens up to new commercial opportunities. Generally speaking about the new space economy, I'm seeing three primary drivers, which I highlighted in 2019, 2022, and this year. The IoT satellites, the Russia-Ukraine war, and this year, the orbital data center push. As cost goes down and the use cases increase, the number of satellites has increased significantly during the last decade. I'll hand it over to John to look at the numbers in the next section.

Thanks, Johan. My team is putting together a series of reports on satellite broadband next month. What I did is I went through those three reports that will be available to our subscribers and plucked out some of the more interesting tables I thought to talk about what Johan refers to as the new space economy. As we've mentioned, I'm going to be focused mostly on satellite broadband. We do have a report on D2D, another one on regulatory and threats in addition to the opportunities. The TLDR with regulations is that compared to terrestrial broadband operators, satellite operators have very little red tape to get up and running nowadays, which is interesting and gives them a little bit of a late mover advantage in a way. Ellie mentioned LEO and GEO.

I went on into the Kagan archives yesterday. I was wondering when satellite broadband started. Our first numbers were in 1998, 23,000 satellite broadband customers. At the time, those services used GEO, geostationary satellites, which are anchored to their position in the sky. For example, just envision where you're sitting right now as you watch this webinar, there's a satellite above your head. As the Earth spins, that satellite's still above your head morning and day. These satellites, they're 10,000 miles plus above the Earth. That impacts the speed you get with broadband and also the latency. The newer version of satellite broadband launched around 2020, including these LEO satellites. These things are only about 100 miles above the ground. It's very close to the Earth, and they're not anchored to the Earth. They're flying at 18,000 miles an hour above us over our heads.

There's actually a pretty cool website called satellitemap.space. You can find where you live, zoom in, and they track the Starlink satellites. You can see how fast 18,000 miles an hour is. It was pretty shocking to see that. 100 miles above your head, you can get a lot better speeds, a lot lower latency. It changed things for satellite broadband. Johan mentioned Sputnik. Since Sputnik until Y2K, we went from one to 700 satellites. As the costs have come down and the new space economy has emerged, since 2000, we've added over, I think that number, is this from orbitalradar.com? It's more like 17,000 satellites in space right now. It's ramped up. You talk about a hockey stick of growth. This is quantifying how fast and how many more satellites there are now. About two-thirds of those satellites belong to Starlink alone.

Digging in a little bit to the components of these LEO operators. Their current constellation as of March, across the big ones, there's about 10,000 here, and their total plan that they've got approval from to eventually put up in space. That hockey stick's not stopping anytime soon, basically, is the takeaway here. This also in the bar chart underscores the scale at which Starlink and Amazon Leo are going to have when we just look at these LEO satellites orbiting currently and in the future. Shifting gears a little bit and looking at back on Earth, we look at the broadband market in the U.S. in four different buckets. Cable broadband, telco, which includes fiber and DSL, satellite, which is what we're here talking about, fixed wireless.

Fixed wireless has been the major growth story in the United States since 2020. I think in the last couple of years, satellite is now becoming another growth story. Again, back in 1998, it started modest. It never really cracked 2 million subscribers up until about 2020. Now we have, just in the last year, if you look at that market share table down there, satellite has gone from 2% share to over 3% share in the United States just based on Starlink's growth alone. The next slide digs in a little bit more to the components in the U.S. for satellite broadband. You've got the two legacy operators, then, of course, Starlink ramping up. On the left is total subscribers, on the right is net additions by quarter. Starlink really ramped up recently compared to their competitors.

Looking at the economics a little bit on this line chart on the left here, this is average revenue per user per month for broadband services. That blue line that starts out in 2023 is the highest. That's just Starlink's ARPU. You can follow that line as it falls to the right and down through 2026. You envision, okay, they're competing for market share by pulling down their price. The other four lines are the other four broadband technologies, which have basically tracked inflation for the most part. I think things started to soften up last year as the combination of Starlink and fixed wireless led to what I call the broadband price wars last year. This is now no longer a low-hanging fruit market. It's pretty saturated. Leading up to the IPO, Starlink rates in the U.S. were as low as $30 a month, which is an incredible discount compared to what the competitors were offering.

I think they still have some $30 a month service offerings, but on the table on the right is looking at the economics of the SpaceX connectivity division, which is basically Starlink. They're getting a good profit margin already in terms of EBITDA and adjusted EBITDA in the 60% range in 2025. On the next slide, it underscores these are the top 10 countries in the world by satellite broadband penetration. There's two takeaways here. Number one is it's not a primary connection method for broadband anywhere. The biggest adoption rate is in Qatar at about 6%. So it's not the leading broadband method anywhere.

The other takeaway here is I think eight of these top 10, if you look at the countries involved, U.S., Australia, Russia, Canada, these are geographically large countries with the population spread out, which really hits home to what satellite broadband's sweet spot was. It's really good for rural area and rural connectivity, the reason is because running wires, whether they be fiber or DSL or cable, to remotely populated areas, the return on investment just isn't there for a lot of companies. I've heard some companies say it's as high as 100 years until they get that money back from digging a trench and running a line and connecting the home. You put satellites up and you connect with those homes automatically, as long as they have a line of sight to the sky, there's no trees or mountains in the way.

I did because I covered wireless for 15 years. I was curious about what the spectrum map looks like for the big three broadband operators in the U.S. Starlink, Viasat, and Hughes, they have far more megahertz worth of spectrum than the terrestrial wireless operators. I think, if you look at Starlink, 15,000 total megahertz is about 15 times what one of the big three wireless operators has in terms of low and mid-band spectrum. They have about 1,000 megahertz to run your smartphones and also this fixed wireless technology. The biggest difference is these guys all share Ka-band and Ku-band spectrum. The terrestrial wireless operators, they buy at auction and they pay a nice premium to have access to spectrum chunks that's just theirs. They don't share it at all.

We're always looking for simple ratios to make at Kagan, one of the things I looked at here was, well, how many subscribers per satellite in orbit are these big three operators dealing with? That's the graphic on the right there. I went through for Starlink and tried to count the satellites in orbit above the lower 48 and Hawaii and Alaska, it was about 200. I saw yesterday someone had estimated it was more like 500 satellites above those countries that could be in range of your home connection. Anyways, using my more conservative 182 satellites, you divide that by 3 million subscribers, it's only 17,000 subscribers per satellite for Starlink. That really underscores a big difference between LEO and GEO, just in terms of the volume of satellites we're talking about here with LEO compared to the legacy satellite broadband services.

As Sarah mentioned earlier, there's been a couple big satellite deals. Thinking of a ratio to compare these guys to looking at LEO and GEO, it's a mixed bag right now in terms of total deal value divided by satellites acquired. We do this a lot for towers for wireless communication companies. If there's a big tower deal, we give an average of a value per tower, it's more like $200,000-$1 million per tower. Compared to here we have it's in the millions per satellite acquired, the two deals from this year so far are definitely ranking up at the top of this table, just underscoring just how popular this. It's a good story right now, it's a hot space, the deals are pretty expensive. My final slide, this one is a little bit of a head scratcher for me.

Let's start with the bar chart on the right. That is the cost to build a terrestrial communications tower for smartphone connection and for fixed wireless. It's about half a million dollars to build a tower in the United States. The red tape, the costs involved, the time involved to get approval with local, state, county, et cetera, to build your tower, it can take many years to get approval and build these things. The accountants say, "Okay, we're going to amortize this asset at a 30-year cadence." I think that's a lot of us have 30-year mortgages. It's a real estate business. That makes sense. I think if you maintain your tower properly, that tower can last forever. The Eiffel Tower is over 120 years old. They maintain it. They make sure it's not rusting. That thing will last forever.

Shifting gears now and looking at satellite. I was curious about for those satellites launched, on average, what does it cost? I'm looking at V3, which Starlink just launched last week, some test launches for this. Looking forward a little bit, for all those LEOs up there that Starlink put, it's about $1.4 million to build and to launch and get these things in orbit per satellite. The thing that confuses me a little bit is that the accountants say the amortization schedule is just five years for these things. After five years, the lifespan of that LEO satellite is we can bring it back through the atmosphere and let it burn up. We don't need that satellite anymore. The economics of that is a little bit confusing to me.

I think what Johan mentioned earlier about the cost, this $1.4 million per satellite is not set in stone. That might make this make a little bit more sense. It is kind of interesting to think of, none of these wireless operators are burning down a tower after 30 years of operation. They can use them a lot longer than that. There's the economic question here, there's also a little bit, there's definitely an environmental concern with how many hundreds of satellites are going to start falling through the atmosphere and burning up and hitting birds and adding to pollution and things like that. On that note, I'm going to switch gears and pass the microphone back to Johan.

Thank you, John, for that introduction. You primarily talked about communication satellites. In the introduction, I mentioned the IoT satellites as a push for the first wave of expansion. Roughly since 2019, we've seen dozens of new startups focusing on small satellites in low Earth orbit to serve those IoT use cases based on low bandwidth communication protocols such as LoRaWAN, NB-IoT, and some proprietary protocols. What we put on the slide is that we've been monitoring the adoption of some of these use cases in our 451 Research Voice of the Enterprise: The OT Perspective survey. We have monitored the adoption of satellite technology by enterprises, and especially in the oil and gas sector. Already 67% of respondents indicate they leverage satellite connectivity, and of these, two-thirds monitoring remote assets is by far the most popular use case, followed by several environmental use cases.

Asset tracking is mentioned by 30% of oil and gas companies, primarily focusing on locating very expensive equipment that tends to get lost and someone can't find it. If we talk about these IoT satellites, we've seen many startups come into play like 2019, 2021. The difference is, compared to the old navigation satellites and communication satellites in geostationary orbit, these operate in low Earth orbit, they're a lot smaller and therefore also a lot cheaper. Compared to the GLONASS satellite operating for 50 years, John just mentioned the amortization, five years life expectancy. In low Earth orbit, the CubeSat has become the standard. CubeSat refers to cubes of 10 by 10 by 10 centimeters in different configurations. You can have a 2U, 4U, or a 6U satellite configuration totaling the number of 10 by 10 by 10 cubes.

Obviously those are a lot lighter than the old mega satellites that operate in geostationary orbit. We've been tracking those companies, those IoT startups for several years, we saw a myriad of startups tailoring to different use cases, each launching with an idea, a specific use case that gained some traction. Monitoring fishing boats in Indonesia, or herd cattles on the Argentinian plains, or pipelines in Siberia. Many of these startups, they launched their own proprietary constellation, or at least they filed their plans with the FCC, spectrum applications. Earlier, John mentioned roughly 15,000 active satellites. Looking back at the forecast we did in 2023, we saw that the total application for spectrum amounted over to over 100,000 satellites by 2033. Since we're only three years into that decade forecast, the numbers are falling short of the FCC filings.

Particularly, SpaceX aimed at 42,000 by 2030. They're still a little under 10,000. They're falling short on their own ambitions. In many cases, the business case proved hard. Some of the startups pivoted, abandoned their own constellation plans, started piggybacking on other constellations, or just refocused to the specific use case data collection. Since the starting of the war in Ukraine, we've seen a next drive emerge. Spy satellites, of course, have been the exclusive domain of national governments, departments of defense. During the war in Ukraine, that shifted. Military on both sides started using commercial observation satellites. Whether it's hyperspectral imaging, synthetic aperture radar, or just high-resolution optical satellites to assess troop movements, even identify targets Starlink became essential for drone navigation.

Other than just the military use of commercial satellites, another trend emerged as private citizen started using Maxar, Planet, ICEYE satellites for open source intelligence to support Ukrainian Army in identifying targets or simply to confirm battlefield claims or investigate rocket damage on civilian targets. Currently, we're in 2026, and we're seeing a major push in FCC filings for orbital data centers. The primary push, I think, is the debate that's going on by the proliferation of AI and the power consumption of AI data centers and the availability of energy. The immediate reflex is like, "Okay, we got space in space and it's cold, so we don't have cooling issues. We don't have CO2 emissions. We can directly capture energy from the sun, so we solve the energy problem." It's a little bit more complicated than that, and Ellie will go into that later.

The FCC filings this year, they've been accumulating to pretty much of a perfect galactic storm with obviously Starlink, SpaceX being most ambitious. Again, announcing plans for 1 million orbital data centers. Blue Origin, 51,000. There's a couple in the bottom that are still in development, don't have finite numbers on their constellation plans. There's a very small red dot on the left-hand side, Kepler Communications, which should be even smaller. It's so small you wouldn't see it. It's a bit larger than comparative, but Kepler Communications is an example of a startup that has an orbital data center in space. To put that into perspective, its first orbital data center has 44 GPUs, and compare that to a hyperscale campus. To put it in context like the IoT satellite ambitions, there is a difference between FCC spectrum applications and what will actually be built.

Are we chasing stellar mirages or dreams? We're seeing definite use cases for space edge computing. Some of the drivers that we're seeing, the energy constraints, cooling, CO2 emissions, might look to be an answer, but Ellie will tell you that's more complicated. On the inhibitor side, we see the same things emerge as inhibitors. The energy constraints, cooling, radiation shielding, orbital BECA hole add to that. If we're looking at the trends and challenges, we see the vertical integration that Sarah mentioned, Amazon, Iridium, Rocket Lab. It's not just satellite companies buying competitors operating in the same space, but it's space companies buying capabilities in different orbital planes.

Starlink, for instance, operating in low Earth orbit, buying larger communication satellites in geostationary orbit, or it's the vertical integration that concentrates the entire technology stacks and not just the satellite capabilities, but also the launch capabilities, the rocket technology, or robotic engineering companies for autonomous operations. For instance, for in-orbit servicing, as these data centers could have a lifespan of five years, they would need servicing. One of the trends that we primarily see emerging, the space edge compute. As all these Earth observation satellites capture more and more images, high-resolution images of Earth, it makes sense to do the first processing in space. Space edge compute before sending down the results to Earth to avoid downlink constraints, rather than sending up AI inferencing workloads from Earth up to space. Space communication. As Ellie mentioned, if we're moving on towards lunar space operations, we need to establish that space communication infrastructure to enable asteroid and lunar mining.

The business case remains a hard nut to crack. There's sustainability and sovereignty, which we'll touch upon later. As the example from the Ukraine-Russia war already also showed is that we have a governance issue to solve if military uses commercial satellites. How does that happen? How does that work? Who's responsible? There's a whole lot of other challenges that we will dive into.

All right. Perfect. I'm going to take over the screen a little bit and talk about some of those challenges that Johan had mentioned. Obviously, this slide says sustainability, but I would love it if we framed it more as practicality. Right? When we're talking about expanding the space economy, moving into these new levels of satellite launches, stuff in space, we need to think practically about what that actually means in terms of how do we make this happen, and then also the impact if we make it happen. We have a great opportunity here as we're scaling into space to move into this new frontier and to build it sustainably from the ground up. That's not something that you always get the opportunity to do, right?

A lot of times you start with a project and you're already halfway through it, and you look back and you say, "Oh, I would've done that differently." Right? As we're moving into these new frontiers, we have the chance to really start smart and build things in a way that will last, in a way that is actually good long term and works well. That's kind of what we're going to talk about in the next couple of slides here. As we talk about sustainability within the space economy, the long-term growth of space economy depends on managing resources responsibly. Resources come in lots of different shapes and sizes. Obviously, you think environmental resources. As we talked about, we have orbital resource constraints as well. We'll dive in a little bit into that in a bit more detail.

We like to look in sustainability across an entire lifespan of an operation. As we discuss the space economy, obviously we've talked about satellites, we've talked a little bit about data centers in space, launch systems, right? All of that, plus the supply chain goes into these considerations, along with end of life considerations and de-orbiting procedures. Because as John had mentioned, we can't just leave stuff up there forever, right? It's not like building a tower. It's not going to work for 30, 100, 150 years. There are other considerations that we need to keep in mind as we expand into space, because the environment there is, frankly, very different than what we're used to on Earth. Each stage of this space life cycle includes different sustainability considerations that are going to be magnified as we try to hit these proposed expansions into the space ecosystem.

First, looking at supply chain, right? When we're talking data centers in space, when we're talking satellites, compute, communication, rocket launch technology, that stuff doesn't just magically materialize in space. It is based ultimately on Earth, right? We need physical materials to make all of this happen. Some of the same constraints and supply chains that we're used to dealing with on Earth for compute construction, for communication, that all comes into play in the same way as we enter the space economy. Carbon fiber, aluminum, titanium, rare earth minerals, all of these have some sustainability implications from extraction and processing. They're well documented on Earth. It kind of doesn't matter whether you're building a data center on Earth or in space, you need those materials to do so.

We can get into asteroid mining later, you need to get the material from somewhere, and that takes work, right? That has add-on consequences. As we move along kind of this space story, we need to get stuff into space, right? That also takes work. Traditionally, we have rocket propellants, all of those from solid rocket fuels to refined kerosene. Those release emissions into the atmosphere. You can talk about emissions from terrestrial data centers, as we move into space, it's not emission-free, it's just you change kind of where those emissions are released. A lot of that is from the launch technology. With traditional rocket propellants, most of the time, they've included large amounts of ozone-depleting chemicals, including black carbon, and we have nitrogen oxides.

Part of the issue with rocket launches that isn't really factored in when we're talking terrestrial data is that you're actually launching into the atmosphere. Where those emissions are released really changes the impact of those emissions. They can be a lot more, I don't know, destructive and harmful if they're released into the atmosphere as opposed to closer to Earth's surface. As rocket launches continue to scale, we talked about the cost of rocket launches going down, we also need to think about the environmental cost of rocket launches. Yes, you can make it cheaper and cheaper to get into space, if you're still being massively pollutant while you're doing it, we should consider that as well. Fossil fuels burned by the space industry right now only make up about 1% compared to those burned by conventional aviation.

The numbers are really, really small right now. If we look at all of these FCC filings, as we look at how things have really scaled up in terms of what's going where, moving into space, the projections for where we want to go, that ratio is going to change. We're lucky now because we have the opportunity to really target that and think about that from the ground up, no pun intended. Pollution launches is definitely something that we need to keep in mind. Not only are we talking stratosphere and mesosphere and the atmosphere impact of all of these emissions, we also need to keep in mind launch site ecology. We've talked a lot, there's been a lot of studies done around rocket launches, the impact on local plants, animals, the environment. We talk about well-executed rocket launches.

You can do a pretty good job planning and saying, "Okay, if this launches successfully, this will be the impact on the local ecology." Of course, anyone who has tracked anything with the space economy knows that not all rocket launches work successfully. You have catastrophic failures, you have learning events, I'm sure you're learning quite a bit, when your rocket blows up on the pad, what does that mean for the local landscape, for vegetation, and for the animals? Plenty of space agencies, including NASA, monitor launch sites and try to minimize launch impacts. That's not the case across the board. There are several launch sites around the world that have historically just kind of been used for rocket launches without much concern for the local environment, creating large zones of pollution caused by years of leaking and toxic rocket fuel seeping into the soil.

We can really cause a lot of problems if we aren't thoughtful about how we're launching stuff into space. Now, when we move on to the third section of this space life cycle, which is the actual in-orbit section, I'll talk about it a little bit more in the next slide. We can get something called orbital congestion. Even though space is big and we have relatively few numbers of satellites now compared to what we want to put into space, we have limited orbital bandwidth for what can go where. On the next slide, I'll talk a little bit more about the congestion story. We also have this new kind of type of satellite that is being designed to specifically serve as orbital mirrors, beaming sunlight on demand to different places around the world.

Even non-orbital mirror satellites can reflect quite a bit of sunlight toward Earth. When you put satellites into orbit, they actually can be really detrimental, especially to research in astronomy and physics, as you're trying to look past our Earth orbit and into the greater cosmos. We need to keep in mind this for us here on Earth, also as we try to continue our exploration further out, we need to make sure that we aren't blocking ourselves off intentionally or unintentionally. Finally, at the end of a satellite's life, we want to keep in mind some of the sustainability considerations there as well. That would be disposal without causing massive problems. One of these solutions to help prevent overcrowding orbits is this idea of de-orbiting satellites. There's a couple of ways to do that.

You can safely degrade an orbital satellite and let it burn up in the atmosphere on purpose. This is controlled. This is meant to clear up overcrowded orbits instead of having a bunch of dead satellites orbiting around causing problems. De-orbiting satellites is pretty standard practice anymore. In low Earth orbit, satellites are usually finished with their mission, and they're mandated by the FCC and the European Space Agency to de-orbit as soon as possible, and it's actually baked in that they aren't supposed to orbit any more than five years past their mission completion. There is a plan already in place to try and help keep those orbital lanes as clear as possible. The problem, of course, with burning up satellites as they reenter the atmosphere is just like burning junk here on Earth, they release additional pollution into the atmosphere.

That adds to the emission problem. We aren't just emitting as we launch these satellites and those systems, they're also further emitting as they reenter the atmosphere. Some satellites, we don't de-orbit at all. We actually move them into a graveyard or a junk orbit, which happens further up. Instead of burning up inside of Earth's atmosphere, these satellites are moved into these kind of long-term storage orbits that are outside of our low Earth orbits and our geostationary orbits and are meant to kind of be long-term holding cells, graveyards, for these dead satellites. It does help clear operational orbits, but we have this nice, beautiful, unpolluted, so to speak, space that we are now just dumping junk into. It's worth keeping in mind that as we have graveyard orbits, they're going to stay there, these satellites, for thousands or even millions of years.

They don't break down in the same way that you have stuff weather here on Earth. There's not air resistance, there's not weathering. If you put something up there and you don't deal with it can stay up there for much longer than any of us are going to be around. It's worth keeping in mind our long-term impact as well. Those are some of the things to look at from a sustainability perspective. Just a little bit more on this congestion sort of story. We've gone through a lot of numbers already on this webinar, talking about what is up in space, right? It's interesting to think about because most of the time you look up at the night sky or even the sky in the day, and you don't see much, right? You've got your clear blue sky. You've got lots of stars.

Sometimes you'll see the fun SpaceX constellations going around in their little trains. There's actually a lot of space debris up in space, most of which we can't see with the naked eye. With these numbers, we've got from the European Space Agency, and we've had since 1957 more than 7,000 rocket launches total, placing more than 26,000 satellites into orbit. These aren't satellites that are operational, right? These are just any satellite that has ever been placed into orbit since 1957. The total mass of all of these space objects in orbit is more than 16,000 metric tons, okay? We talk about satellites as these nice little packages, but they don't always stay that way. I mentioned that things don't weather in space. You don't get degradation. That happens as long as they don't run into other things.

If we have collisions in space, things are moving very quickly up there, and that can cause debris fields that can really amplify the amount of stuff that we've got in orbit at any given time. On the right-hand side, we've got some estimated numbers of objects in orbit. There's no way to really count. All of these are based off of models, again, in this case, from the European Space Agency. As far as objects go, those that are greater in size than 10 centimeters, we've got about 54,000 objects, right? That could be active satellites. That could be pieces of satellites. Anything that's 10 centimeters or bigger, we've got 54,000. As we move into these smaller and smaller scales, we've got millions and millions of very small pieces of space junk that are just orbiting around.

You can think, well, a small thing isn't a problem. Yeah, think about a bullet, right? A small thing moving very quickly can become very problematic very quickly. One sort of term you might read about or come into contact is this idea of the Kessler effect, which is that these small pieces of space junk can create these catastrophic chain effects where all of these interactions and collisions can lead to full orbital bands that are so full of space junk that you can't launch satellites, that they interfere with communications, that we can't use an orbital band because it is so full of junk and it's just not safe. While that hasn't happened yet, it is something that really needs to be considered as a we must avoid this scenario at all costs sort of thing.

Having a proper plan in place to make sure we're de-orbiting satellites in a safe, sustainable way, that we're dealing with this space junk problem, right? We already have a lot of space junk that's up there. If we could minimize the amount that is in our orbits that we want to use, that would be very important as well, especially as we're trying to expand and really regulate what's going on up in space. With that being the case, right, we really need to focus on international standards, space traffic management, actively removing the debris that's there. Then, of course, as we've already started, we've got this reusable launch system already in place. We've got work being done for more efficient fuels. The work towards building a sustainable space economy is definitely actively happening right now.

There's certainly more to do and more to think about. It's never just as easy as, well, let's just send the data center up to space and we'll be done with it there. All right. One quick note that I wanted to make on sovereignty. I know we're getting close on time. We had a really great look-forward journal article published by a bunch of 451 Research analysts come out recently. There's a lovely QR code in the bottom right corner of your screen if you want to read the full report discussing this idea of compute sovereignty and kind of what that means terrestrially. I did want to flag just really quickly this idea of sovereignty as a driving factor within the space ecosystem as well.

Governments across the world are looking at sovereign cloud environments, domestic AI, high-performance computing, quantum computing skill and capacity, just to make sure that we've got secure data storage transmission, that things are built in a safe, sovereign way. As we look at space and space infrastructure, sovereignty as well is going to continue to come up as a key driver here because the same questions and the same discussions we're having here on Earth are going to apply to space. That includes with data sovereignty, application sovereignty, compute sovereignty as we look at data centers in space a little bit more. That's going to play a driving role in that. What's next? We've got all sorts of stuff coming up, space mining, infrastructure build-out, the new space race regulations.

There's a lot that's going on, and we're excited to keep tracking what's going on within that. I know we're close on time. I'm going to pass it back over to Sarah, and she can manage any Q&A.

Thanks, Ellie. We had a couple of questions come in, and I'm going to race through them so that we can cover as much ground as possible. We had a question come in about satellite lifespans. Are they considered technically not usable anymore after the five-year timeframe we mentioned? That's a really interesting question, and honestly, it depends on the constellation we're talking about. Right now we're talking, that five-year timeline has mostly to do with the Starlink LEO satellites that we've spent a lot of the webinar talking about. Some of the constellations that have different constructions, like Telesat their construction, although LEO, their satellites are going to be a little bit higher up. They're going to be a little bit bigger.

They envision launching far fewer of them, and those satellites are expected to last, I think, closer to 15 to 20 years if I remember correctly, as opposed to that much shorter life cycle for the lower Earth orbit LEO satellites that Starlink, but then also Amazon's LEO envisions using. That is in terms of these different weighing these different constellations, that's definitely something to keep in mind. Ellie, we had a question come in about space radiation damage data centers and what are you hearing on that?

Yeah, for sure. Johan and I have actually been on some great briefings around that. Yes, long story short, radiation can cause problems in space. Obviously, when you're on Earth, you have atmospheric protection from a lot of space radiation. That's not the case when you're up in orbit. There are a couple of ways to manage that. You can either help secure any of the chips behind different types of materials to block against radiation. There are radiation-hardened chips that are available. I'm not sure, Johan, if you want to talk a little bit more about some of the radiation solutions that we've learned about in some of our briefings.

Well, looking at the traditional old space economy, you use ruggedized chips, specially manufactured to withstand radiation, tested $200,000 a piece, lifetime 30 years plus, still operational in the International Space Station, that's not keeping pace with commercial off-the-shelf capabilities. Yeah, we see companies working on shielding technology, whether it's the chip itself. In the semiconductor industry, we're seeing fully depleted silicon on insulator process technology just to harden the chips or a specialist working on more of the casing in various materials to withstand the radiation.

John, we had a quick question come in about your slide 21. How many users per tower versus satellite? You brought up cost, just wanted to get a clarification on that one.

Yeah. Is this the number of users- per terrestrial tower?

Per tower versus per satellite. Yep. Including if you're a tier 1 operator, you have 100 million subscribers, maybe most of those are humans.

Some of those are double counting. There's some tablets, there's some smart watches, there's some cars included in there. It's just a perfect human ratio. Assuming 50,000 pole towers, the number's about 26,000 people per tower. If we add in small cells and DAS networks, that number shrinks down to about 1,200 people served per tower. There's a ton of caveats there. I think, generally speaking, the terrestrial networks serve a lot less people per tower than the satellite networks do. Good question. Well, I know we are at time, and that you can continue to use the Q&A widget to submit your questions and we will try and follow up with you.

I want to thank John, Ellie, and Johan for their insightful presentations. We covered a lot today, so if you have any follow-up questions, please use the contact widget and we will be glad to assist. For those who want to review anything we covered, this session is recorded and you'll receive a copy shortly, so you can access it on demand at your own convenience. When we close out of the webinar, you'll be routed to our webinar survey form. We hope to see you at future events. Thank you all so much for your great time and questions.