Space-based data centers powered by solar panels sound like a climate solution for AI's energy crisis, but a critical accounting gap is hiding massive environmental costs that most companies aren't measuring. Two peer-reviewed studies published in 2025 reached opposite conclusions about whether orbital computing is cleaner than terrestrial data centers, not because of scientific disagreement, but because they counted different environmental impacts. One study found orbital facilities could be carbon-neutral within years; the other concluded they're an order of magnitude worse than any power grid on Earth, even under optimistic assumptions. The stakes are enormous. Data center energy demand is projected to exceed 1,000 terawatt-hours (TWh) by 2026, roughly equivalent to Japan's entire annual electricity consumption. Several U.S. states, including New York and Oklahoma, have already imposed emergency moratoriums on new data center connections because the electrical grid cannot handle the load. This constraint has triggered a race to launch computing infrastructure into orbit, with Google's Project Suncatcher, NVIDIA-backed Starcloud, Axiom Space, SpaceX, and the Chinese government all announcing plans for space-based data centers. The problem is not the solar panels in orbit. It's everything else. What's the Real Carbon Cost of Launching Data Centers Into Space? The industry's headline claim is straightforward: orbital data centers powered by solar produce a fraction of the carbon of terrestrial equivalents. Starcloud claims they're ten times cleaner; Google describes Project Suncatcher as a route to carbon-neutral AI infrastructure. Both statements are technically defensible, but only if you count a single metric: the energy consumed once the hardware is already in orbit. Saarland University researchers built a comprehensive lifecycle model called ESpaS that includes launch emissions, hardware manufacturing, and reentry. Their findings were sobering: orbital compute carries an effective carbon intensity of 800 to 1,500 grams of CO2 equivalent per kilowatt-hour (gCO2e/kWh). That's worse than every country's electrical grid on Earth. The industry figure of 134 to 165 gCO2e/kWh counts only the energy dimension once the satellite is operational, a valid metric that happens to be the only framing under which orbital compute looks competitive with renewable-powered terrestrial data centers. The disagreement matters because of a structural asymmetry: NTU Singapore's paper, which found orbital data centers could become carbon-neutral, was a perspective piece funded in part by a research lab with commercial interests in space-grade semiconductors. Saarland's study was an empirical model with open-source code that any researcher can reproduce and verify. How Do Rocket Launches and Satellite Reentry Create Hidden Climate Damage? The most carbon-intensive operational rocket currently flying is SpaceX's Starship, which emits 5,490 tonnes of CO2 per launch. The European Space Agency established a precise threshold for orbital computation to match the emissions of renewable-powered terrestrial data centers: launch vehicles would need to emit approximately 1.9 kilograms of CO2 per kilogram of payload. Every rocket currently operational sits between 10 and 25 times above that line. There's a counterintuitive twist: reusable rockets don't automatically reduce emissions intensity. Falcon 9 in reusable configuration actually produces more CO2 per kilogram of payload than in expendable mode, because retaining propellant for landing recovery reduces payload capacity while keeping total launch emissions roughly constant. Reusability reduces cost and launch frequency, but the math on emissions per unit of cargo doesn't improve. The reentry problem is even more troubling. When satellites reach the end of their operational life, they burn up in Earth's atmosphere. Until recently, this was treated as a clean, cost-free disposal mechanism. New atmospheric science tells a different story. Currently, approximately 887 tonnes of material are injected into Earth's atmosphere annually from human-made satellite reentry. NASA's 2024 Technical Memorandum projects that figure will exceed 30,000 tonnes per year by 2040, driven entirely by megaconstellation maintenance cycles. The first wave of Starlink satellites launched in 2020 is now reaching the end of its five-year design lifetime, meaning the reentry surge is not a future event but is already accelerating. The chemistry of this burning is where the real climate risk emerges. Aluminum structures vaporize into fine nanoparticles that drift up and accumulate in the stratosphere, the atmospheric layer that includes the ozone layer. Scientists at NOAA have already detected spacecraft-origin metals, including exotic elements with no natural atmospheric sources. They estimate that roughly one in ten of the relevant particles already carry traces of spacecraft material, a share that could reach 50 percent as constellations scale. Recent modeling suggests the consequences include measurable changes in upper-atmosphere temperatures and wind patterns, and ongoing disruption to the chemistry that sustains the ozone layer. The ozone hole, which was the defining environmental crisis of a previous generation, was caused by a far smaller class of atmospheric interference. Steps to Evaluate the True Climate Impact of Space-Based Computing - Demand Full Lifecycle Accounting: When evaluating claims about orbital data centers, require companies to disclose carbon costs across the entire lifecycle, including launch emissions, hardware manufacturing, operational energy, and end-of-life reentry, not just the energy consumed in orbit. - Check the Rocket Emissions Baseline: Compare the specific launch vehicle's CO2 emissions per kilogram of payload to the ESA's threshold of 1.9 kg CO2 per kg payload. Current rockets exceed this by 10 to 25 times, making orbital compute significantly dirtier than renewable-powered terrestrial alternatives. - Assess Atmospheric Chemistry Impacts: Look beyond carbon dioxide to understand how satellite reentry injects catalytically active metals and rocket soot into the stratosphere, which can alter ozone chemistry and trap heat far more effectively than ground-level emissions. Rocket exhaust adds a second, unrelated problem. Soot from rocket fuel, when released high in the stratosphere, traps heat far more effectively than the same soot at ground level. Research published in Earth's Future puts the difference at around 500 times greater warming potential per unit mass. Neither of these mechanisms, the metal contamination from reentry or the soot from launch, appears in any published lifecycle emissions model for orbital data centers, including the most rigorous one available. Every estimate of how green these facilities might be is incomplete. The trajectory of space-based infrastructure is accelerating rapidly. The number of objects placed in orbit annually grew from under 500 in 2018 to nearly 3,000 by 2023. The forecast to 2032 shows an acceleration to around 10,000 objects per year, which makes sustainability questions urgent while infrastructure decisions are still being made. For now, space data centers are not a meaningful contributor to global emissions. All orbital launches in 2024 combined emitted roughly 0.4 million tonnes of CO2, approximately 0.04 percent of aviation's annual output. The problems arise when looking at projections. If space-based computing scales as companies are planning, the hidden costs of launch and reentry could dwarf the energy savings from solar-powered orbital infrastructure. The accounting gap needs to close before the infrastructure does.
