Starlink Orbital Shells and Coverage Footprints: Why Altitude Changes the Network

Last updated: July 10, 2026

Introduction

Starlink coverage is easy to imagine as a set of circles on a map: each satellite sees a patch of Earth, and enough patches create global service. That picture is useful, but it hides the harder part of the network. A satellite footprint is not the same as usable capacity, and geometric visibility is not the same as a reliable internet connection.

Orbital altitude changes almost every part of the problem. A lower satellite has a shorter radio path and can support lower light-time delay, but it sees a smaller part of Earth’s surface. A higher satellite sees a wider area, but the path is longer, the signal spreads more, and the network may have different interference and debris tradeoffs. The altitude also affects drag, station keeping, deorbit behavior, and how often satellites may need to be replaced.

This article uses three simple Starlink-style reference altitudes: 340 km, 480 km, and 550 km. These are not meant to describe every active satellite at every moment. They are calculation points that help explain why shells matter. Recent FCC Starlink Gen2 authorizations include shells around the 340 km and 480 km classes, while the 550 km class is a useful comparison point from earlier Starlink authorizations and the familiar LEO broadband discussion.

The central lesson is that altitude does not merely change a coverage circle. It changes the whole network design: how many satellites are needed, how much overlap exists, how often handoffs occur, how spectrum can be reused, and how quickly failed spacecraft leave orbit.

Assumptions Used for the Estimates

The estimates use a spherical Earth with mean radius 6,371 km. A satellite is treated as if it is in a circular orbit at one fixed altitude. The footprint radius is the surface distance from the point directly below the satellite to the edge of visibility.

Two footprint concepts are used. The first is the geometric horizon footprint. This is the largest possible circle if the user could communicate with the satellite exactly at the horizon. It is a mathematical boundary, not a practical service boundary.

The second is a 25-degree minimum-elevation footprint. In this case, the user only counts the satellite as usable when it is at least 25 degrees above the local horizon. This is more realistic because low-elevation links pass through more atmosphere, face more obstruction, have worse geometry, and can create harder interference coordination problems. It is still only a model. Real systems use dynamic beam shapes, terminal rules, regulatory constraints, gateway availability, and capacity management.

The article separates official facts from estimates. The official facts are that Starlink is a non-geostationary satellite system in low Earth orbit, that FCC orders authorize specific SpaceX shells and operational constraints, and that NASA and ESA define LEO as a region close to Earth compared with higher orbits. The footprint radii are calculations from the reference assumptions, not official SpaceX service promises.

Physical Calculation

Horizon Footprint

For a satellite at altitude h, the horizon central angle can be estimated as:

theta = arccos(Re / (Re + h))

The surface footprint radius is then approximately:

radius = Re * theta

At 340 km altitude, the geometric horizon footprint radius is about 2,037 km. At 480 km, it is about 2,399 km. At 550 km, it is about 2,557 km.

Those numbers look enormous. A single satellite at 550 km altitude can geometrically see a disk thousands of kilometers across. But the edge of that disk is the horizon. A terminal near that edge would see the satellite extremely low in the sky, where buildings, terrain, trees, aircraft structures, ship motion, rain path length, and antenna constraints become much more important. For internet service, the horizon footprint is better understood as an upper bound than as a service area.

Minimum-Elevation Footprint

A more practical footprint uses a minimum elevation angle. With a 25-degree minimum elevation, the central angle estimate becomes:

psi = arccos((Re / (Re + h)) * cos(elevation)) – elevation

Using that assumption, the footprint radius is much smaller. At 340 km altitude, the practical 25-degree footprint radius is about 627 km. At 480 km, it is about 841 km. At 550 km, it is about 941 km.

This is the most important footprint result. Raising altitude from 340 km to 550 km increases the 25-degree footprint radius by roughly 314 km in this simple model. Because area roughly scales with radius squared, the difference in potential coverage area is larger than the radius difference suggests. But the wider footprint is not free. The signal travels farther, beam geometry changes, and the same satellite’s capacity may be spread across a larger region unless the system uses beamforming and scheduling to concentrate resources where demand exists.

Orbital Speed and Handoff Timing

Altitude also changes orbital motion. The reference calculations give about 7.71 km/s and a 91.2-minute period at 340 km, about 7.63 km/s and a 94.1-minute period at 480 km, and about 7.59 km/s and a 95.5-minute period at 550 km.

The speed differences are modest, but the footprint differences are meaningful. A lower satellite moves slightly faster and has a smaller useful footprint. That combination tends to increase the importance of constellation density and handoff management. The user does not care which satellite is overhead; the user cares whether the network can keep a stable route as satellites enter and leave useful view.

Operational Reality

Coverage Is Not Capacity

A map can show that a satellite is visible. It cannot show that the satellite has enough capacity for everyone in the footprint. Internet service depends on beams, spectrum, terminal density, gateway or laser backhaul, routing, and traffic demand. A rural area with few users and a clear sky can be easy to serve even with modest capacity. A city, event venue, busy coastline, flight corridor, or shipping lane can be visible to satellites and still be capacity constrained.

This is why Starlink cannot be evaluated only by asking whether a satellite can see a place. The better questions are: how many satellites are visible at useful elevation angles, how many beams can serve the area, how much spectrum can be reused without harmful interference, where is the nearest internet handoff, and how much traffic is competing for the same resources?

Overlap matters because it gives the network choices. If multiple satellites can serve a user, the system can balance load, avoid an obstructed path, prepare for handoff, or choose a better route to a gateway. If only one satellite is available, coverage may exist but resilience is weaker. Dense overlap is especially important for mobility, high-latitude coverage, and areas with uneven demand.

Why Lower Altitude Can Help

Lower altitude shortens the radio path. That can reduce ideal light-time and improve link geometry for users near the satellite’s center of view. It can also help orbital debris mitigation because objects at lower altitude generally reenter sooner after losing propulsion than objects at higher altitude. FCC discussions of SpaceX modifications have repeatedly treated lower altitude as relevant to user experience, interference geometry, and orbital environment considerations.

But lower altitude is not automatically superior. A smaller footprint means more satellites are needed to provide continuous service across the same region at the same minimum elevation. More satellites can increase capacity if spectrum reuse and beams are designed well, but they also require manufacturing, launches, tracking, collision avoidance, and disposal management. The network becomes less like a few large umbrellas and more like a fast-moving woven fabric.

Why Higher Altitude Can Help

Higher altitude provides a wider footprint. That can reduce the number of satellites needed for basic geometric coverage and can give each satellite a longer useful pass over a region. A wider view can help continuity, especially in sparse areas where capacity demand is low.

The tradeoff is distance. At 550 km, the ideal vertical one-way light-time is about 1.83 ms, compared with about 1.13 ms at 340 km. That difference is small compared with terrestrial routing delays, but it is still part of the latency budget. More importantly, a higher shell changes slant range across the footprint and changes the radio link budget. The practical network design must balance coverage width against signal quality, spectrum reuse, handoff behavior, and orbital sustainability.

Drag and Replacement Reality

Low Earth orbit is not empty in the engineering sense. Atmospheric density is thin, but it is not zero. At lower altitudes, drag becomes more important. Satellites need propulsion for orbit raising, collision avoidance, station keeping, and end-of-life disposal. If a spacecraft fails, lower altitude can help it decay faster, which is good for long-term debris risk. But the same environment can make routine operations more demanding.

This creates a replacement reality. A low-altitude broadband constellation is not a monument placed in orbit once. It is maintained. Satellites age, fail, deorbit, get replaced, and are upgraded. Lower shells can make debris mitigation more favorable, but they also require the operator to sustain a pipeline of spacecraft, launches, ground operations, and software updates. A Starlink shell is therefore both a coverage layer and a logistics commitment.

The FCC’s Starlink orders reflect this operational framing. They discuss orbital shells, frequencies, deployment flexibility, collision risk, propulsion, disposal reporting, and conditions on authorization. Those details are not side notes. They are the regulatory version of the same physical truth: a satellite internet constellation is a dynamic system, not a static map.

Speculative Lens

Under a modest speculative lens, Starlink is not a planet covered by invisible domes. It is more like a set of moving lanes wrapped around Earth. At 340 km, the lanes are close, fast, and narrow. At 480 km, they are still close but cover more ground. At 550 km, the lanes widen further, giving the network more geometric reach while asking it to manage longer paths and different orbital behavior.

From the ground, the experience can feel simple: a terminal finds the sky and the internet works. From the network’s point of view, the shell is constantly rewriting itself. A satellite that matters now will soon move on. Another will take its place. The footprint is not a permanent circle; it is a temporary promise renewed by the next spacecraft.

That is where the analogy becomes useful without turning into hype. Starlink suggests an internet that is less tied to fixed towers and cables, but not free from infrastructure. The infrastructure is simply moving. The cost of that movement is paid in orbital mechanics, launch cadence, spectrum coordination, and replacement cycles. The result feels futuristic mainly because the maintenance work is happening above the atmosphere instead of beside a road.

What This Does Not Prove

These footprint estimates do not prove where Starlink service is available, how fast it will be, or how many users a satellite can support. Real availability depends on regulatory approval, terminal rules, obstructions, local demand, gateway access, satellite density, network policy, and service plan limits.

The estimates also do not prove that one altitude is the universal best choice. A 340 km shell may be attractive for latency and disposal behavior, but it has a smaller practical footprint. A 550 km shell may provide wider geometric reach, but it has longer paths and different operational tradeoffs. A 480 km shell can be understood as one possible compromise, not as a magic altitude.

The calculations do not include Earth oblateness, terrain, atmospheric refraction, antenna gain patterns, beam steering limits, phased-array scan losses, rain fade, frequency-specific path loss, or interference coordination. They are first-order geometry. That is enough to explain why altitude matters, but not enough to engineer a real constellation.

Finally, coverage footprints should not be confused with business coverage. A satellite may physically see a country where the service is not licensed. A terminal may have sky visibility but no active service plan. A ship or aircraft may be inside a beam but subject to mobility rules. Physics enables service; operations and regulation decide whether service is delivered.

Conclusion

Starlink orbital shells matter because altitude changes both physics and operations. At 340 km, the practical 25-degree footprint radius is about 627 km in the reference model. At 480 km, it grows to about 841 km. At 550 km, it reaches about 941 km. The horizon footprints are much larger, but horizon visibility is not a good proxy for reliable service.

Lower altitude can reduce path length and improve debris decay behavior, but it demands more density and careful handoffs. Higher altitude can widen coverage and extend useful visibility, but it increases distance and changes link and interference conditions. The best network is not simply the lowest or highest shell. It is the shell design that balances latency, capacity, overlap, launch logistics, orbital safety, and regulatory limits.

The realistic interpretation is that Starlink coverage is not a set of static circles. It is a managed moving system. Altitude defines the geometry, but capacity, routing, handoffs, spectrum, gateways, and replacement cadence define the user experience.

Sources

FCC DA 26-36, SpaceX Gen2 Starlink upgrade applications partial grant: https://docs.fcc.gov/public/attachments/DA-26-36A1.pdf
FCC FCC 21-48, SpaceX third modification order: https://docs.fcc.gov/public/attachments/fcc-21-48a1.pdf
FCC DA 21-34, SpaceX polar shell partial grant and earlier shell background: https://docs.fcc.gov/public/attachments/DA-21-34A1.pdf
NASA Low Earth Orbit FAQs: https://www.nasa.gov/humans-in-space/leo-economy-frequently-asked-questions/
ESA Low Earth Orbit explainer: https://www.esa.int/ESA_Multimedia/Images/2020/03/Low_Earth_orbit
ESA Types of Orbits: https://www.esa.int/Enabling_Support/Space_Transportation/Types_of_orbits
SpaceX Starlink Technology: https://starlink.com/technology
Source access date: July 10, 2026.