An NDB meets the RNP for an NDB be it used to define the centerline of an airway or as an approach aid.
When constructing the approach procedure, allowance is made for the inaccuracies of the NDB and the obstacle clearance heights reflect the errors inherent in the system.
GPS can be more accurate than NDB and if an approach procedure was to be constructed, it's minima would reflect that accuracy.
However, when completing the NDB approach everyone constantly monitors the ident because there is no warning of failure. GPS does not have any ident or warning of failure that the CAA find as acceptable for instrument approaches. Thus there are not any GPS approaches yet.
The CAA is not overly worried that the Yanks can turn off the Navstar system at any time it chooses because the owner of the NDB at Shoreham is similarly entitled to act with their property.
The CAA is worried that errors and jamming can affect the Navstar system in unpredicatable and possibly unexpected ways possibly without any warning to the user. It is illegal for the owner of the NDB at Shoreham to mess with the signal from the NDB.
What are people using to assess how well the NDB is showing the approach centerline?
If it is the runway centerline then remember that the NDB is not on the runway and thus an angular difference exists between the runway centerline and the NDB approach. This lateral displacement gets further away from the runway centerline with distance. Remember that as long as it is within 30 degrees of the runway it counts as a straight in approach.
Also remember that any error is the result of the total system error i.e. the receiver error and the indicated error as well as pilot error.
Insted of worrying about coastal refraction which is not a significat factor on the 20 approach, pilots should remember the effects of quadrantal error especially with a strong westerly wind.
There is however one inherent system failure in the UK - NDBs are not reguluarly flight checked in the same way as other approach aids.
Regards,
DFC