How Satellites Orbit Earth: The Physics and Engineering Behind Eyes in the Sky

A satellite doesn't stay in orbit because it's beyond gravity's reach. At the International Space Station's altitude of 250 miles, gravity is still 89% as strong as on Earth's surface. The ISS is constantly falling — it falls about 4.9 meters every second. The reason it doesn't hit the earth is that it's also moving forward at 17,500 mph, fast enough that as it falls, the earth curves away beneath it at the same rate.

This is the fundamental insight of orbital mechanics, understood by Newton in 1687: orbit is perpetual falling, balanced by sideways speed. Every satellite in the sky, from the International Space Station to your GPS constellation to the Starlink network, is operating on this single principle. Understanding how engineers harness this physics reveals one of the most remarkable achievements in the history of technology.

Key Takeaways

• Satellites stay in orbit by falling perpetually while moving fast enough that Earth curves away beneath them
• Different orbits serve different purposes: low Earth orbit for imaging and communications; medium orbit for GPS; geostationary orbit for weather and TV
• The first satellite, Sputnik, was launched October 4, 1957; there are now over 8,000 active satellites
• GPS works by precisely timing signals from at least four satellites to calculate position to within a few meters
• SpaceX's Starlink constellation demonstrates the new era of commercial satellite deployment — over 6,000 satellites launched since 2019

The Physics of Orbit

Isaac Newton explained orbital mechanics with a thought experiment. Imagine firing a cannonball horizontally from the top of a mountain. At low speed, it curves to the ground quickly. At higher speed, it travels farther before landing. At exactly the right speed, it travels so far that as it curves toward the earth, the earth curves away at the same rate — the cannonball never lands. It's in orbit.

For Earth's surface, this orbital velocity is approximately 17,500 mph (28,000 km/h). At the International Space Station's altitude, it's about 17,150 mph — slightly less because gravity is slightly weaker farther from Earth's center, and thus less centripetal force is needed to maintain circular motion.

This velocity must be achieved horizontally, not vertically. A rocket going straight up simply falls back down. The trick of reaching orbit is achieving the horizontal velocity, which requires fighting gravity during the ascent and then accelerating horizontally once at altitude. This is why rocket launches look like a curve — they pitch over after clearing the lower atmosphere to spend most of their fuel on horizontal acceleration.

The orbital period — how long a trip around Earth takes — depends entirely on altitude. Higher orbits require less velocity (gravity is weaker) but longer paths, and they take longer. The ISS circles Earth every 92 minutes. A geostationary satellite at 22,236 miles altitude circles every 24 hours, matching Earth's rotation exactly and appearing to hover motionless over one spot.

The Orbital Zones and Their Uses

Low Earth Orbit (LEO): 100–1,200 miles altitude. Orbital periods of 90–110 minutes. This is where most human spaceflight happens (ISS, Space Shuttle), where most Earth observation satellites operate, and increasingly where commercial broadband constellations like Starlink live. The advantages: low signal latency (20–40 milliseconds for Starlink vs. 600ms for geostationary), high-resolution imaging possible with smaller telescopes. The disadvantages: each satellite covers a small area, so global coverage requires hundreds or thousands of satellites; atmospheric drag gradually decays orbits.

Medium Earth Orbit (MEO): 1,200–22,000 miles altitude. This is where GPS satellites live, at 12,550 miles. MEO offers a middle ground: each satellite covers a larger area than LEO, with reasonable signal delay. The GPS constellation's 31 satellites cover the entire planet simultaneously. Russia's GLONASS, Europe's Galileo, and China's BeiDou work similarly.

Geostationary Orbit (GEO): 22,236 miles altitude above the equator. At exactly this altitude, a satellite's orbital period equals Earth's rotation period, so the satellite appears stationary over one point. This makes it ideal for weather satellites (GOES satellites provide continuous images of the Americas), TV broadcast (a dish antenna can point at a fixed spot in the sky), and some communications. The main limitation: signal delay of ~600 milliseconds round trip, which makes real-time communications difficult.

Polar Orbits: Satellites that pass over the poles have orbits inclined roughly 90° to the equator. As Earth rotates beneath them, they gradually see the entire planet's surface. Earth observation satellites, spy satellites, and weather satellites use polar or near-polar orbits to map the entire globe.

GPS: The Greatest Practical Application of Orbital Mechanics

The Global Positioning System is arguably the single most useful satellite constellation ever built. Originally deployed by the US military for precision navigation and weapons guidance, its full capabilities were made available to civilian users in 2000 when the Clinton administration eliminated Selective Availability — the deliberate degradation that had limited civilian GPS to 100-meter accuracy.

GPS works through precise timing. Each GPS satellite continuously broadcasts a signal stamped with the exact time it was sent, using atomic clocks accurate to nanoseconds. A GPS receiver computes how long each signal took to arrive and multiplies by the speed of light to get distance. With distances from three satellites, the receiver can compute its position in two dimensions (trilateration). A fourth satellite resolves the third dimension and corrects the receiver's clock error.

The accuracy depends on how precisely the receiver can measure signal timing. Consumer GPS achieves 3–5 meter accuracy. Military differential GPS and modern civilian augmentation systems achieve centimeter accuracy. This level of precision enables not just navigation but precision agriculture, autonomous vehicle positioning, surveying, and timing synchronization for financial markets and power grids.

What makes GPS a triumph of engineering is reliability. The constellation must maintain continuous global coverage. Satellites fail — there are typically 31 operational satellites where 24 provide full coverage, with the extras providing redundancy. Each satellite is replaced before it fails, in a process managed by the US Space Force that has maintained uninterrupted GPS service since 1995.

The Satellite Industry in the SpaceX Era

For most of the Space Age, launching a satellite cost roughly $10,000–$20,000 per kilogram. Building a communications satellite cost hundreds of millions of dollars. Access to space was limited to governments and the largest telecommunications companies.

SpaceX changed this. The Falcon 9's reusable first stage, which flies back to land on a drone ship after launch, cut launch costs to approximately $2,700 per kilogram — roughly a 5–7× reduction. The Starship, which aims to make both stages fully reusable, targets costs below $100 per kilogram.

The Starlink constellation demonstrates what lower launch costs make possible. SpaceX has launched over 6,000 Starlink satellites since 2019 — more than double all other satellites ever launched by all other organizations combined. Starlink provides broadband internet to over 4 million customers in rural areas, ships, and aircraft where ground-based internet doesn't reach.

This democratization of space access is bringing new actors into orbit. Companies are deploying Earth observation constellations that can photograph every point on Earth multiple times per day. Climate monitoring satellites are tracking deforestation, methane emissions, and ocean temperatures in near-real time. The era of space as a specialized domain for a few powerful nations is ending.

Source: SpaceX (2025) — Starlink deployment updates and constellation status reports.

Perturbations and station-keeping

While the concept of a stable orbit seems straightforward — gravity provides centripetal force — real orbits are never perfectly stable. Multiple forces perturb satellite paths: the non-spherical shape of Earth (an oblate spheroid, wider at the equator), gravitational pull from the Moon and Sun, solar radiation pressure, and residual atmospheric drag in low Earth orbit.

Earth's oblateness (described mathematically by the term J₂, the second dynamic form factor) causes the orbit's orientation to change gradually — a phenomenon called precession. Sun-synchronous orbits exploit this precession: they're inclined so that the orbital plane rotates at the same rate Earth orbits the Sun, keeping the same local solar time every day. This is ideal for Earth observation because lighting conditions remain consistent.

Geostationary satellites must fight several perturbations. Gravitational anomalies on Earth's surface create regions of stronger and weaker gravity. Solar radiation pressure pushes on the spacecraft (a faint but constant pressure from photons). The Moon and Sun tug on the satellite. Over weeks and months, these effects accumulate, nudging a geostationary satellite east or west. To maintain station, geostationary satellites carry fuel and perform station-keeping maneuvers — small thruster burns that correct the orbit. This is a major operational cost and ultimately limits satellite lifetime; when fuel runs low, the satellite is retired.

Atmospheric drag is particularly important for low Earth orbit satellites, where a thin (but measurable) atmosphere exists. Above 2,000 km, atmospheric density is negligible. But at the ISS altitude of 250 miles, the residual atmosphere gradually drags the station down. The ISS is periodically boosted back to higher altitude by cargo vehicles that fire their engines — a process called reboost. Without this, ISS would decay and re-enter Earth's atmosphere within several years.

The study of these perturbations is essential for mission planning. Engineers use sophisticated propagation software (tools like STK — Satellite Toolkit) to predict satellite positions years in advance, accounting for all perturbations.

Orbital debris and the space junk crisis

One of the most pressing challenges facing satellite operations is orbital debris — "space junk." There are currently over 34,000 catalogued debris objects larger than 10 centimeters in Earth orbit. The actual number is far higher: estimates suggest over 900,000 pieces between 1-10 cm (a paint fleck traveling at orbital velocity can puncture a spacecraft).

This debris comes from multiple sources: spent rocket stages, defunct satellites, fragments from collisions and explosions, and material shed from operational spacecraft. The 2007 Chinese anti-satellite test destroyed one satellite, creating an estimated 2,000-3,500 trackable debris pieces. The 2009 Iridium-Cosmos collision (an operational communications satellite and a defunct Russian military satellite) generated another debris cloud.

Kessler Syndrome is a scenario proposed by NASA scientist Donald Kessler in 1978: if debris density becomes too high, collisions would create more debris, causing more collisions in an escalating cascade. Certain orbital regions (particularly LEO at 800-1,000 km altitude, where many Earth observation and communications satellites operate) could become hazardous for generations. We're already seeing increased collision avoidance maneuvers — every year, hundreds of satellite operators need to move their spacecraft to avoid projected impacts.

Mitigation strategies now include designing satellites to deorbit at end-of-life (burning up in the atmosphere), developing active debris removal systems, and creating deorbiting strategies for constellation operators. The 2025 European ARES mission tested harpoon-based debris capture. Companies like Axiom Space and RemoveDebris are developing commercial technologies to remove large debris. This is increasingly urgent as mega-constellations like Starlink grow.

Source: ESA Space Debris Office (2024) — ESA's Space Debris Mitigation Guidelines and Discos Database.

The future of satellite technology: mega-constellations and beyond

The satellite industry is experiencing a historic transition. Historically, satellites were expensive, specialized instruments operated by governments or large corporations. A single geostationary satellite cost $200-500 million and served a handful of purposes.

The new model is mega-constellations — thousands of identical, mass-produced satellites launched at scale. Starlink (6,000+), Amazon's Project Kuiper (3,236 planned), and European constellation plans represent a shift toward satellite broadband as a primary connectivity option. These constellations are designed to launch, operate, and deorbit on decade timescales, with scheduled replacements.

Smaller satellites are also democratizing Earth observation. Companies like Planet Labs operate constellations of small imaging satellites that can photograph the Earth daily at moderate resolution (3-5 meters per pixel). This is good enough to track agriculture, forestry, water resources, and climate indicators. Planet's constellation of nearly 500 satellites generates terabytes of imagery daily — far more data than can be manually analyzed, driving adoption of AI-based image analysis.

The technological trajectory is clear: lower cost, higher cadence deployment, and integration with terrestrial networks. Satellite internet is no longer a niche technology but infrastructure. As launch costs continue falling and satellite design becomes more modular, we'll see continued expansion in satellite applications — from climate monitoring to telecommunications to IoT connectivity.

Emerging Launch Providers and Competitive Dynamics

SpaceX revolutionized launch costs, but they're no longer alone.

Chinese providers (China Great Wall Industry Corporation, China National Space Administration) have achieved similar cost reductions and now launch roughly 40% of global launch volume. Chinese rockets (Long March series) recover boosters less frequently than Falcon 9 but achieve comparable per-kilogram costs through different engineering approaches.

Other US providers: Blue Origin (working toward New Glenn), Axiom Space, and Relativity Space (3D-printed rockets) are developing launch vehicles aiming for comparable costs. By 2026, the launch market is far less monopolistic than it was in 2015.

ESA and Europe: The European Space Agency is developing Ariane 6, which targets lower costs than Ariane 5 but likely won't match SpaceX's economics. Europe emphasizes independent launch access rather than cost competitiveness.

India and Japan: Both countries have demonstrated successful satellite launch. India's ISRO offers competitive pricing; Japan's Space Exploration Technologies aims to compete with SpaceX on cost and reusability.

Competitive landscape: Multiple providers keep costs competitive and prevent monopolistic pricing. Lower launch costs enable new entrants (Planet Labs, Axiom Space, smaller nations) to access orbit.

Spectrum, Regulation, and International Coordination

Satellites don't just occupy orbital altitude — they occupy radio spectrum, which is globally finite and regulated by the International Telecommunication Union (ITU).

ITU coordination: Different satellite constellations operate at different frequencies. Starlink operates at Ku-band and Ka-band. Other constellations (OneWeb, Telesat, Amazon's Kuiper) operate at different frequencies to avoid interference. The ITU coordinates these allocations and prevents international conflicts.

Regulatory challenges: The rapid growth of mega-constellations (thousands of satellites) has outpaced regulatory frameworks. Questions include:

• Orbital altitude regulations: Some nations are pushing for deorbiting requirements to prevent debris accumulation.
• Environmental impact: Megaconstellations affect astronomy (image contamination) and atmospheric chemistry (satellite reentry).
• Equitable access: Should developing nations have preferential access to spectrum? How do we prevent developed-nation dominance of space resources?

ITU meetings (held every few years) attempt to harmonize rules, but consensus is difficult with 193 member nations with competing interests.

Earth Observation and Climate Monitoring

One of the most valuable satellite applications is Earth observation — imaging Earth's surface to monitor climate, agriculture, deforestation, and disasters.

Temporal resolution: Traditional satellite imaging offered 1–2 images per location per month. Planet Labs' constellation of 500+ small satellites now provides 5–10 images daily of most Earth locations. At 3–5 meter resolution, this is good enough to track agriculture, deforestation, urban sprawl, and water resources.

Climate applications: Satellites monitor:

• Sea surface temperature (indicators of El Niño, ocean warming)
• Sea ice extent (Arctic/Antarctic)
• Glacier extent and volume changes
• Atmospheric CO₂ concentrations
• Methane emissions from oil fields, landfills
• Forest biomass and carbon stock

These measurements are essential for tracking progress on climate goals and understanding climate change dynamics.

Real-time disaster monitoring: Floods, wildfires, earthquakes, and hurricanes are tracked in near-real-time via satellite. In 2023, satellites detected the Morocco earthquake minutes after occurrence, enabling rapid emergency response. This represents a genuine improvement in disaster management.

Challenge: The data volume is enormous — terabytes per day. Processing requires AI and cloud computing to extract actionable information.

Source: NASA Earth Observatory (2024) — Climate and Earth Observation satellite data products.

Starlink's Terrestrial Impact and Rural Connectivity

Starlink has achieved coverage in remote areas where traditional broadband is economically infeasible.

Coverage gaps: About 25% of Earth's population (mostly in developing nations, rural areas, and extreme terrain) lacks reliable broadband access. Ground-based infrastructure is expensive to deploy in these regions.

Starlink economics: By 2026, Starlink reportedly provided service to 4+ million customers globally. Latency (~40–50 ms) is good enough for video conferencing, gaming, and streaming. Speed (50–200 Mbps) meets most needs. Cost ($100–150/month) is cheaper than satellite internet was in 2015 but pricier than terrestrial broadband in developed areas.

Limitations remain: Weather affects latency and throughput. In heavy rain, Starlink can drop. This makes it less reliable than fiber but more reliable than previous satellite internet.

Competition: Amazon's Kuiper (2024 deployment), OneWeb, and smaller constellations are entering the market, likely reducing costs and improving service.

Impact on inequality: If satellite internet reaches rural and developing-nation populations, it could dramatically improve educational and economic opportunities. This is potentially transformative but requires affordability and regulatory coordination.

Future Challenges: Debris, Spectrum Saturation, and Sustainability

As mega-constellations proliferate, several challenges loom:

Orbital debris escalation: Every mega-constellation adds thousands of satellites. If even 10% are not properly deorbited, orbital debris will accumulate catastrophically over decades. Kessler Syndrome becomes a genuine threat.

Spectrum saturation: There are only so many radio frequencies. Allocating spectrum for multiple mega-constellations is increasingly contentious. Future constellations may face spectrum scarcity.

Sustainability concerns: Rocket launches emit CO₂ and other pollutants. Scaling to thousands of launches/year raises environmental questions. Satellite reentry burns up atmosphere, adding trace elements. Long-term environmental impact is still uncertain.

Astronomical impact: Bright satellites interfere with ground-based astronomy. Professional observatories lose observation time due to satellite contamination. Some proposed solutions (darkening satellites, careful constellation geometry) add cost.

These challenges suggest that the satellite era, while enabling unprecedented connectivity and observation, comes with externalities society is still learning to manage.

The takeaway

Satellites remain in orbit through the elegant principle of perpetual falling — gravity constantly pulls them toward Earth, but their forward velocity causes Earth to curve away beneath them. Different orbits serve different purposes: GPS constellations in medium Earth orbit provide global navigation; geostationary satellites provide continuous weather monitoring and broadcasting; mega-constellations in low Earth orbit offer broadband connectivity. The past two decades have seen a revolution in launch costs and satellite economics, enabling new actors to access space and deploy satellites at unprecedented scale. Yet this growth creates new challenges — orbital debris, spectrum congestion, and the need for responsible long-term planning. The physics of orbit that Newton understood in the 17th century now powers a critical 21st-century infrastructure, connecting billions of people and observing a planet in rapid change.

Frequently Asked Questions

Q: Why don't satellites fall down?
A: They are falling — constantly. But they're moving sideways so fast that as they fall, Earth curves away beneath them at the same rate. They're in continuous free-fall around the planet. This is orbital mechanics: the balance between falling due to gravity and moving forward horizontally.

Q: What is the Kessler Syndrome?
A: A scenario proposed by NASA scientist Donald Kessler in 1978: if the density of debris in low Earth orbit becomes too high, collisions would generate more debris, causing more collisions in a cascade that could make certain orbits unusable for generations. The rapid growth of satellite constellations makes managing orbital debris increasingly urgent.

Q: How long do satellites last?
A: Varies by type and orbit. Geostationary communications satellites typically operate for 15 years; they carry fuel to station-keep against gravitational perturbations and solar pressure. Low Earth orbit satellites have shorter lifespans — some as little as 5 years — partly due to degradation from radiation and the low residual atmosphere that gradually drags them down.

Q: How do satellites handle the temperature extremes of space?
A: In Earth orbit, a satellite in direct sunlight can reach +120°C; in Earth's shadow it can drop to -170°C. Satellites use combinations of reflective surface coatings, heaters, heat pipes, and radiators to maintain electronics within operating temperatures. This thermal management is one of the more challenging satellite engineering problems.

Q: Can satellites see individual people?
A: High-resolution commercial imaging satellites can resolve objects about 25–30 centimeters (10–12 inches) across — enough to distinguish a car but not reliably identify a person. Military reconnaissance satellites are assumed to have significantly better resolution. The limiting factor is the atmosphere, which introduces distortion that adaptive optics can partially correct.

Q: What happens when a satellite's mission is over?
A: Modern satellites are increasingly designed to deorbit safely, where they re-enter the atmosphere and burn up. Older satellites were sometimes left in "graveyard orbits" — higher than geostationary, out of the way. Some active satellites are pushed into higher disposal orbits; others (particularly in LEO) are deorbited within a few years of mission end. International guidelines now require deorbit within 25 years of end-of-life.

---

Read next:

• How GPS Works — The Physics of Positioning
• The Internet Actually Works — Packets and Protocols Explained
• How the World's First Computer Worked — The Colossus and ENIAC