The geostationary belt is some of the most valuable real estate in space. Communications, weather and government satellites sit about 35,786 km above the equator, alongside retired spacecraft, rocket bodies and fragments. Keeping track of all of it is a core job of space domain awareness, and most of that tracking is still done by ground-based optical telescopes with cooled scientific cameras.
GEO asks different things of a camera than low Earth orbit does. This post walks through the choices that size a camera for GEO surveys, using Finger Lakes Instrumentation (FLI) Kepler cameras as the worked example. For fleet-level planning, see our post on SDA sensor networks, and for daylight LEO work, our post on daytime satellite tracking.
Why GEO is its own problem
A LEO satellite crosses the sky in minutes. A geostationary object barely moves relative to the ground, so it stays in roughly the same patch of sky all night. That sounds easier, but GEO objects are tens of thousands of kilometers away, so they are faint, and the smallest debris is fainter still. The belt also stretches in a band across the whole sky visible from a site, so a survey has to cover a lot of area every night.
So a GEO camera has two jobs at once: reach faint objects, and cover a wide band of sky quickly. Field of view, pixel scale and readout behavior all trade against each other.
Staring or sidereal tracking
GEO surveys usually run in one of two modes. In staring mode, the telescope mount holds still relative to the ground. GEO objects stay as points, and stars trail across the frame at the sidereal rate, about 15 arcseconds per second of time near the celestial equator. In sidereal mode, the mount follows the stars, so stars are points and GEO objects trail instead.
Staring mode concentrates a GEO object's light in a few pixels, which helps detection. Sidereal mode gives clean star images for astrometric reference. Many programs use staring mode for discovery and switch modes or fit star trails for precise positions. Either way, the camera must handle a frame that contains both points and streaks.
Pixel scale and streak length
Pixel scale ties the camera to the telescope. In arcseconds per pixel, it is 206.265 times the pixel size in microns, divided by the focal length in millimeters.
Take a 500 mm focal length as an example. The 9 µm pixels of the KL4040FI give about 3.7 arcseconds per pixel. In staring mode, a star near the equator then moves about 4 pixels every second, so a 2 s exposure turns stars into streaks about 8 pixels long while GEO objects stay compact. Longer exposures lengthen the streaks, which spreads star light over more pixels and makes star-crowded frames easier to separate from the point-like objects you are hunting.
Too coarse a scale merges faint objects into the sky background. Too fine a scale spreads light over many pixels and shrinks the field. Our post on pixel size and sampling covers the matching math in more detail.
Field of view and survey rate
Survey rate is roughly field area times the number of fields you can visit per hour. At the same 500 mm focal length, the KL4040's 4096 × 4096 sensor covers about 4.2° on a side, roughly 18 square degrees per pointing. The KL6060FI, with 6144 × 6144 pixels at 10 µm, covers about 7.0° on a side, about 50 square degrees, nearly three times the area per frame.
Large formats need optics that cover them. The KL6060's 86.9 mm diagonal calls for a telescope or lens with a big, well-corrected image circle, and the filter and shutter path must clear that format too. Our post on back focus, flanges and filter wheels helps plan the optical path.
Going fainter: exposure, stacking and synthetic tracking
Kepler cameras are designed for fast, continuous, short-exposure imaging. FLI does not recommend single Kepler exposures longer than about 5 minutes. For GEO, exposures are usually far shorter than that, because the sky moves and the survey has ground to cover.
To reach fainter debris, many programs stack frames. Synthetic tracking, described by Shao and colleagues in 2014, shifts and adds many short frames along trial motion vectors, so a faint moving object builds up signal while noise averages down. It works best with low read noise and short, continuous exposures. Kepler's 16-bit images combine two 12-bit readouts at different gains, which helps keep bright stars and faint objects in the same frame.
Kepler comes in front-illuminated and back-illuminated versions, including the KL4040BI and KL6060BI. Compare their data sheets with your sky brightness, targets and exposure plan before choosing.
Timing and astrometry
A GEO observation is only useful if its position and time are right. Kepler sensors use a rolling shutter with a global-reset mode, so rows are read at slightly different times. Ask how frame start times are defined across rows, and build that into your astrometry pipeline. A GPS timestamp option is listed for Kepler and Aurora cameras; contact FLI for current status on your model. If several cameras share a site, our post on synchronizing multiple cooled cameras covers shared triggers.
Data volume
Continuous surveys produce a lot of data. One 16-bit 4096 × 4096 frame is about 34 MB, and one 6144 × 6144 frame is about 75 MB. Over a long night, that adds up quickly, so size the host storage, processing and network link along with the camera.
Running unattended
GEO survey sites are often remote and run all night without staff. User-replaceable shutter, fans, and digital board make Kepler ideal for remote installations. The shutter isn't needed for exposure timing; it makes calibration darks convenient, and a blocked filter-wheel position works too. Anti-dew is recommended at cold sites, and the KL4040 is rated to operate from 0 to 40 °C. Our post on remote and polar observatories covers site planning.
The Kepler SDK runs on Windows and Linux, and FLI provides an ASCOM driver for Kepler cameras. Drivers and manuals are on our support page.
Choosing the camera
Kepler CMOS cameras are our SDA line, built for continuous 24-hour use. The KL4040 suits telescopes with moderate image circles and programs that want a balance of field and data rate. The KL6060 suits survey telescopes that can fill its larger format and want the most sky per frame.
Questions to settle before you choose
- Will you survey in staring mode, sidereal mode or both?
- What focal length, aperture and image circle does your telescope offer?
- How faint must you go, and will you stack or use synthetic tracking?
- What timing accuracy does your astrometry need?
- How much data can your host and pipeline handle each night?
- What temperatures and humidity will the site see?
Data-sheet values are typical. A production test summary exists for each camera and can be provided by serial number, which helps when you model survey performance. Our post on acceptance testing explains what to check on arrival.
Talk to our sales team
FLI cameras are Made in USA, and our sensor chambers are assembled in our 1,000+ sq ft ISO Class 7 clean room. See how on our in-house manufacturing page.
Send our sales team your telescope, survey plan and site details, and we'll help you choose a camera for the GEO belt. Request a quote or email sales@flicamera.com.
