Astronomy Impacts of Satellite Megaconstellations


In May of 2019, SpaceX launched a batch of 60 satellites into low Earth orbit (LEO) in the first of a series of launches designed to populate a “megaconstellation” of satellites called Starlink. A new study now examines how the presence of these satellites — and those of future megaconstellations — will impact optical astronomy.

A Prominent Population

large satellites in LEO

The number of objects (>100 kg) in lower LEO by year; click to enlarge. Before Starlink’s launch, there were fewer than 400 objects in this range. Now, Starlink (cyan) has begun to dominate this naked-eye-visible population. [Adapted from McDowell 2020]

With the goal of providing global internet access, Starlink and similar satellite megaconstellations sound like they should be a good thing. But this project, which is proposed to expand into a network of thousands of LEO satellites, has a drawback: these satellites are both large and orbit at low altitudes — which means they’re visible.

Right now, Starlink consists of 358 satellites orbiting in lower LEO (that’s below an altitude of 600 km); each of these satellites is 260 kg in mass and several meters across. If SpaceX launches the proposed number of satellites, there will eventually be more than 12,000 Starlink satellites — 9,000 of them in lower LEO — and they will dominate the naked-eye population by factors of 4 to 20.

Distribution of Starlink satellites

Simulated instantaneous distribution of Starlink satellites. [McDowell et al. 2020]

Looking Through Trails

How will reflections from these satellites impact optical astronomy observatories? Are we doomed to a future in which satellite trails ruin every exposure from ground-based telescopes?

In a new study, scientist Jonathan McDowell (Center for Astrophysics | Harvard & Smithsonian) models the proposed satellite population as a function of latitude, time of year, and time of night. He then explores the density of satellites that will be illuminated and visible to optical observatories around the world.

Evaluating Impact

number of Starlink satellites illuminated

The number of Starlink satellites illuminated above a given elevation for four cases: at the horizon, 5° above, 10° above, and 30° above. The satellites pose the greatest challenge for observations nearer to the horizon and near twilight. Note that this particular plot is for mid-summer at a low-latitude observatory (30°S): neither the best nor worst case. [Adapted from McDowell 2020]

McDowell finds that there will be a number density of 0.005–0.01 objects per square degree illuminated at elevations relevant to observing. Several factors influence which observations are most affected by these reflected-light sources, including:

  1. Latitude of the observatory and local time of year
    In local winter, observatories at low latitudes will have perhaps 6 hours of observing time each night without illuminated satellites visible. At the other extreme, in local summer, intermediate-latitude observatories (like those in much of Europe) will contend with hundreds of illuminated satellites visible above the horizon.
  2. Type of observing program
    Certain types of observing campaigns will be more affected. This includes those that rely on long exposures of large fields of view (like many trans-Neptunian object surveys, including those searching for the hypothesized Planet Nine) and those that rely on twilight observations (like near-Earth-asteroid detection programs).

An Uncertain Future

The upshot? Starlink probably doesn’t spell the end of optical astronomy, but McDowell’s work shows that it clearly will have an impact on astronomers’ abilities to conduct useful observations with ground-based telescopes. And with other countries working on their own megaconstellations, we can expect the challenges to astronomy to continue to grow.


Photograph of a batch of Starlink satellites shortly after launch. [Official SpaceX Photos – Starlink Mission]

In response to astronomers’ concerns, SpaceX recently launched a batch of Starlink satellites coated in a special material to reduce their reflectivity. The jury is still out on whether this approach is effective, but one thing is clear: we will certainly benefit from working together to find a solution that helps us to advance our technology while still being able to study and learn from the universe.


“The Low Earth Orbit Satellite Population and Impacts of the SpaceX Starlink Constellation,” Jonathan C. McDowell 2020 ApJL 892 L36. doi:10.3847/2041-8213/ab8016