Lunar reference station
What accuracy could a reference station near the lunar south pole give, and where?
A differential reference station at a surveyed site computes corrections for the lunar navigation satellites it sees; a user nearby applies them, and the errors the two share cancel. This view maps the error that is left over a south-polar grid and a day of epochs: the corrected percentile error per point, how far from the station it stays within a threshold, availability and protection levels. Every figure comes from the Kshana engine's lunar-station-coverage report. What this shows.
MODELLED. Under the stated public assumptions, not a statement about the performance of any real system. Not affiliated with or endorsed by any agency.
The constellation, its signal-in-space error, the station's survey error, the correction latency and the link quantization are inputs stated in the scenario file, each with its source or marked illustrative. No figure here has been checked against a real station or flight data.
Starting the engine…
Corrected error over the south-polar region
South-polar stereographic: the pole in the centre, the grid's edge on the rim, 0° longitude up, 90° east to the right. Each cell is one grid point; the heavy line is where the corrected error crosses the threshold, and the dashed line is the coverage radius per azimuth.
Region figures
Over the whole grid: means weighted by area, and the worst point.
Coverage radius per azimuth
Along each bearing from the station, the distance at which coverage first drops below 95 %.
As a table
| Azimuth, ° | Radius, km | At the grid edge | Points along it |
|---|
Availability
An epoch is unavailable at a point when the user sees fewer than four satellites, or the station does not see every satellite the user uses; no error is computed for it.
Grid points, as a table The same points as the map, nearest the station first.
| Latitude, ° | Longitude, ° | From the station, km | Corrected error | Uncorrected error | Availability | Coverage | HPL median |
|---|
The engine's statement, inputs and link budget
Inputs, as the engine read them
Error model
Correction-link budget at the station and at the worst point
What this shows
- The corrected error at a grid point is the stated percentile (95 % unless the file says otherwise) of the user's three-dimensional position error after applying the station's corrections, over every available epoch and Monte Carlo trial. The uncorrected error is the same percentile without corrections, from the same random draws; their ratio is the improvement factor.
- Coverage at a point is the share of epochs whose percentile error is within the accuracy threshold. The coverage radius along a bearing is the distance from the station at which coverage first drops below 95 %.
- Protection levels (HPL, VPL) bound the horizontal and vertical error at a stated integrity risk; the alert limit they are compared with is illustrative, as no public value exists.
- What it leaves out. The errors the station and the user share cancel; what is left grows with the distance between them, plus the station's survey error, the age of the corrections and the link's rounding. Terrain masking beyond the elevation mask, multipath and the receiver are not modelled. Every figure is MODELLED under the stated public assumptions, and says nothing about how any real system performs.
- The engine runs here as WebAssembly. Nothing is fetched from another site and nothing you set is sent anywhere.