What Is Satellite Orbit Prediction and How Does It Enable Tracking and Pass Forecasting?
Updated
Satellite orbit prediction is the process of calculating where a satellite will be at a future time, using a mathematical model of its orbit plus a set of orbital parameters measured at a known reference time. It is what lets tracking software tell you, in advance, when a satellite will rise above your horizon, how high it will get, and which direction to point an antenna. Live tracking, by contrast, only tells you where the satellite is right now. Prediction is what turns a current position into a usable schedule.
Prediction vs. live tracking
These two are often confused because they appear in the same software window.
| Live tracking | Orbit prediction | |
|---|---|---|
| Question answered | Where is it now? | Where will it be at time T? |
| Input | Current orbital state | Orbital state + propagation model |
| Output | Present position, Doppler shift | Future pass times, max elevation, azimuth path |
| Typical use | Aiming a manually steered antenna | Scheduling contacts, automating rotator control |
Most tracking programs do both: they predict a pass, then use the same model in real time to keep the antenna pointed as the pass unfolds.
The inputs a prediction needs
Prediction is only as good as the orbit description you feed it.
Two-line elements (TLEs)
A TLE is a compact, standardized text format that encodes a satellite's orbital parameters at a specific moment. It contains the mean motion, inclination, eccentricity, argument of perigee, right ascension of the ascending node, mean anomaly, and a drag term. The moment the data refers to is the epoch, printed in the first line.
Keplerian elements
These are the classical six parameters that describe an ellipse in space: semi-major axis, eccentricity, inclination, right ascension of the ascending node, argument of perigee, and true (or mean) anomaly. A TLE is essentially a packaged, epoch-stamped version of these plus drag and mean-motion derivatives.
Epoch and its age
The epoch is the timestamp the elements are valid for. As time passes beyond the epoch, the prediction degrades, because the real orbit drifts from the model. This is why TLE age matters more than almost anything else for accuracy.
How predictions become pass forecasts
Once the software has the elements and a propagation model, it computes the satellite's position in an Earth-centered frame at each future time step. It then converts that to a topocentric frame relative to your station's latitude, longitude, and altitude.
From there it derives:
- Azimuth — the compass bearing to look toward
- Elevation — the angle above the horizon
- Range — distance to the satellite
- Doppler shift — the frequency offset caused by relative motion
A pass is simply the interval during which elevation stays above your minimum threshold (often 0° or a few degrees to avoid terrain). Within that interval the software reports acquisition of signal (AOS), the time of maximum elevation (TCA), and loss of signal (LOS). Those three numbers are the core of any pass forecast.
Why predictions go wrong
Errors come from a few well-understood sources:
- TLE age. The older the epoch relative to your prediction time, the larger the position error. Low-Earth-orbit satellites decay fastest in accuracy.
- Atmospheric drag. The drag term in a TLE is a fit, not a physical constant. It changes with solar activity, so predictions drift during high solar flux.
- Simplified models. SGP4/SDP4 are the standard propagators for TLEs and are accurate enough for most amateur and operational tracking, but they are not full numerical integrations.
- Station coordinates. A wrong latitude, longitude, or altitude shifts every predicted pass time.
- Maneuvers. If the satellite fires thrusters, the TLE no longer describes its orbit until a fresh one is published.
Practical rule: refresh TLEs before any critical pass, and treat predictions more than a few days out as approximate.
How prediction drives tracking software
In a typical workflow, prediction and tracking are two stages of one loop:
- Load a current TLE for the target satellite.
- Enter your station coordinates.
- Let the software compute upcoming passes and pick one.
- At AOS, the same propagator runs in real time, converting the predicted position into live azimuth/elevation commands.
- Those commands drive a rotator, or guide you manually, while Doppler correction tunes the radio.
This is the mechanism behind tools like Gpredict, which is described as free, real-time satellite tracking and orbit prediction software for Linux, Mac OS X, and Windows. Note that the site at oz9aec.dk currently shows only an "under construction" placeholder, so treat that description as the project's stated purpose rather than something you can verify from the page itself.
What to check before trusting a forecast
- Is the TLE epoch recent relative to your pass?
- Are your station coordinates correct to within a few kilometers?
- Is your minimum elevation threshold realistic for your horizon and obstructions?
- Has the satellite performed a maneuver since the last TLE?
If all four are satisfied, a pass forecast is usually good enough to schedule a contact and pre-point an antenna. If any are not, expect the satellite to arrive early, late, or lower than predicted.