// UK Pilot Resource — CAP 1404 Ed. 6 (June 2023) · CAP 670 (2019)
The radar used to allege your infringement has a certified positional error of up to 926 metres — and the airspace boundary drawn on the controller's screen can be a further 450 metres from its true position. If the alleged infringement is smaller than those tolerances, the evidence may not be capable of proving it occurred. You are entitled to understand these limits and to challenge evidence that falls within them.
This guide provides factual information only. It is not legal advice. For formal proceedings or licence action consult a qualified aviation solicitor or contact AOPA UK or the LAA.
The NATS CAIT system that auto-triggers infringement reports applies no transponder tolerance whatsoever. Any Mode C return that appears to breach the airspace floor — even by a single 100 ft quantisation step — automatically generates an MOR and the 5 NM/3,000 ft safety bubble, without any allowance for known instrument inaccuracy. An expert engineer analysis cited in Pilot magazine found combined altitude errors could reach 500 ft under normal conditions. Source: Pilot magazine, Nov 2020.
Act immediately. Radar data and ATC recordings are held for limited periods. Send your written evidence request the moment you receive notification of an alleged infringement. Do not wait.
An alleged infringement typically begins with contact from NATS (or the relevant ANSP), followed by a CAA review under the CAP 1404 process. Nothing has been decided at this stage. The CAA's Infringement Co-ordination Group (ICG) must first confirm that an infringement actually occurred. That determination rests on radar data which has well-documented, regulated accuracy limits set out in CAP 670.
CAP 1404 Edition 6, p.11 lists as the first evaluation question: "Can the ICG confirm an infringement actually occurred?" If the margin of alleged penetration falls within the documented accuracy tolerance of the radar, the answer should be "not confirmed" — and the case should not proceed to remedial action.
All UK ATC radar must meet performance standards set in CAP 670. A fully compliant en-route radar can place your aircraft up to 926 m (0.5 NM) from its true position on any single plot — and still be operating within its certified specification. CAP 670 SUR02.49
The airspace boundary on the controller's display is a video map overlay, not a GPS line. CAP 670 SUR11.131 permits that boundary to be up to 450 m (0.25 NM) from its true surveyed position. Sensor error and map error can act in the same direction — combined, that is potentially 1.4 km of permitted error.
SSR altitude is derived from your altimeter via Gillham code encoding. ICAO Annex 10 permits encoding tolerances of ±100 ft (up to ±200 ft at code transitions), plus display quantisation of 100 ft. A combined error of 200–300 ft is normal operation within specification.
CAP 670 mandates documented alignment checks, Remote Field Monitor serviceability logs, and performance monitoring records. If these cannot be produced for the period around the incident, the reliability of the data as evidence is unverified.
If the allegation is that your aircraft penetrated the lateral boundary of controlled airspace — a CTR, CTA, TMA, ATZ or other zone — the evidence comes from PSR or SSR positional data. The regulatory accuracy requirements for that data are set out in CAP 670, with specific numerical thresholds in the EUROCONTROL ATM Surveillance System Performance Specification.
CAP 670 SUR02.47–49 references the EUROCONTROL ATM Surveillance System Performance Specification for the specific performance thresholds. Those figures at the 95th percentile are:
These are 95th percentile thresholds — a fully compliant radar may exceed them on 1 in every 20 plot updates. This is not a fault; it is normal operation within the certified specification. An alleged infringement of 50–100 m is between 5 and 18 times smaller than the en-route tolerance alone. [CAP 670, SUR02.45–49]
| Error Source | Max Permitted Error | Reference | Notes |
|---|---|---|---|
| PSR/SSR positional accuracy — 5 NM en-route | ≤ 0.5 NM (926 m) | CAP 670 SUR02.45–49 | 95th percentile; 5% of plots may exceed this |
| PSR/SSR positional accuracy — 3 NM terminal | ≤ 0.25 NM (463 m) | EUROCONTROL ATM Surveillance Spec | 95th percentile; the reference value for terminal radar |
| Display map feature accuracy (en-route / approach) | ≤ 450 m (0.25 NM) | CAP 670 SUR11.131 | Permitted error of displayed boundary vs surveyed position |
| Display map acceptance threshold at commissioning | ≤ 900 m (0.5 NM) | CAP 670 SUR12.85 | New-system flight trial acceptance criterion |
| PSR north alignment (recommended tolerance) | ≤ 0.1° | CAP 670 SUR04.9 | Recommendation only — not an absolute requirement |
| Combined worst case (sensor + map error, same direction) | ~0.75 NM | SUR02 + SUR11 combined | Both errors at maximum, acting in the same direction — within full specification |
If the CAA alleges you were 50–200 m inside controlled airspace, the maximum permitted positional error of a fully compliant terminal radar (463 m at the 95th percentile) is up to nine times larger than the alleged infringement. The permitted display map error adds a further 450 m. The smaller the alleged infringement, the more powerful this argument becomes.
Positional error is not uniform across radar coverage. PSR azimuth resolution is determined by antenna beamwidth — typically 1–2° at the 3 dB point. At 30 NM, 1° of azimuth error produces approximately 555 m of lateral positional uncertainty. At 40 NM, it exceeds 700 m. Request the documented positional accuracy of the specific installation at the range and bearing of the alleged infringement, not just the headline specification figure. [CAP 670 SUR02.18–21]
CAP 670 requires PSR systems to use Permanent Echoes (PEs) for geographic alignment checking, and all SSR systems to have a mandatory Remote Field Monitor (RFM) that continuously monitors target bearing, range, peak power, side lobe suppression, and pulse spacing. Controllers or maintenance engineers must check the range/bearing error against established tolerances at regular intervals, and those records must be retained. [CAP 670 SUR04.8–16; SUR05.32–47]
Request the PE check records and RFM logs for the 90 days before the incident. What were the measured bearing and range errors? Were they within the defined tolerances? If checks were overdue or results are unavailable, the calibration state of the radar at the time cannot be established.
The airspace boundary on the controller's screen is a video map overlay referenced to a geodetic coordinate system (WGS-84 is recommended). The map must undergo formal verification and validation before entry into service, and the accuracy of features must be documented. [CAP 670 SUR11.119, SUR11.121–126, SUR11.131, SUR11.134]
Request the map validation records for the specific boundary in question. If the displayed boundary was itself 300 m displaced from the true boundary, an aircraft shown 50 m inside the displayed boundary could actually have been 250 m outside the real boundary.
CAP 670 requires all failures to meet required performance to be recorded and logged, and mandates that the CAA Regional Inspector is informed if performance is not met. Request the fault log, BITE records, and performance deficiency log for the 30-day period around the incident. Any logged performance failures — interference, equipment degradation, transmitter power issues — are directly relevant to the reliability of the data being used as evidence. [CAP 670 SUR12.98–101]
A single radar plot showing marginal penetration is far weaker evidence than a continuous track well inside the boundary over multiple scans. A single anomalous plot is consistent with a tracking error, a false target, or a momentary position error. CAP 670 requires tracking performance assessment to determine track deviations, which must not exceed the maximum horizontal position error. [CAP 670 SUR12.53–57]
Request the complete plot-by-plot track history. How many scans showed the aircraft inside the boundary? By how much on each? An abrupt position jump between adjacent scans is consistent with a tracking artefact, not aircraft movement.
If your aircraft had a GPS device or EFB application (SkyDemon, Garmin Pilot, ForeFlight, or similar), the track log is likely your most powerful evidence. GPS horizontal accuracy is typically ±5–15 m at the 95th percentile under normal conditions — up to 90 times more precise than a compliant terminal radar. Preserve the full track log immediately, export it (GPX, KML), and overlay it on a current chart showing the airspace boundary.
Note that CAP 1404 Edition 6 (p.11) explicitly asks whether you used a VFR moving map. A "yes" answer supported by a GPS track showing no infringement directly addresses the ICG's evaluation criteria. [CAP 1404 Ed. 6, p.11]
If the allegation is that you entered airspace vertically — above a ceiling or below a floor — the evidence comes from SSR Mode C altitude data. This has a well-documented error chain governed by CAP 670 and ICAO Annex 10, Volume IV.
The altitude number displayed to the controller is the output of four sequential steps, each introducing potential error:
| Error Source | Magnitude | Reference |
|---|---|---|
| Gillham encoding tolerance | ±100 ft (±200 ft worst case) | ICAO Annex 10, Vol IV |
| Mode C display quantisation (100 ft steps) | Up to ±100 ft | SSR system design |
| Altimeter instrument error (certified aircraft) | ±30–75 ft typical | EASA Part 21 / airworthiness standards |
| QNH setting / atmospheric variation | Variable | METAR / AIP meteorological data |
| ADS-B ground domain altitude resolution (minimum) | Must not degrade below 100 ft | CAP 670 SUR07.38 |
| Combined plausible error | 200–400 ft | All sources combined |
If you are alleged to have been 100 ft above a ceiling or below a floor, the Gillham encoding tolerance alone (±100–200 ft) may equal or exceed the alleged infringement. A displayed value of FL055 (5,500 ft) could represent any actual altitude between approximately 5,200 ft and 5,800 ft under normal, compliant operation. Additionally, CAP 670 SUR07.31 states that only barometric altitude shall be displayed to controllers for separation purposes — if GPS-derived geometric altitude has been used in the allegation, this is procedurally incorrect. [CAP 670 SUR07.29–31]
Obtain the METAR for the relevant aerodrome at the time of the incident and compare the QNH with the value recorded in the ATC system. CAP 670 SUR11.18–20 requires QNH display capability and mandates double-entry validation for any manual QNH change. Any discrepancy shifts the displayed altitude for every aircraft in the system. [CAP 670 SUR11.18–21]
Your transponder and altitude encoder are required to be calibrated within airworthiness requirements. Even a correctly calibrated system operates within a permitted tolerance. If you have your aircraft's most recent transponder calibration and altimetry check records, produce them. They establish that your equipment was operating correctly and quantify the known error tolerance of your specific installation.
Mode C altitude is reported at each radar scan (typically every 4–12 seconds). A single anomalous readout is consistent with a transient Gillham code decode error. Request the complete sequence of altitude readouts over the relevant period. If only one or two scans show an altitude outside the limit while adjacent scans are within it, this pattern is characteristic of a momentary decoding error, not a sustained altitude exceedance. [CAP 670 SUR12.47–48]
Send your evidence request by email (with read receipt) and recorded post simultaneously — to the CAA SARG and the relevant ANSP (typically NATS, or the aerodrome operator). Under the UK GDPR and Data Protection Act 2018, you are entitled to your personal data, which includes radar track data associated with your registration or callsign. Submit an SRG1605 form in parallel for the occurrence report.
A real case illustrating the systemic problems described in this guide. A pilot flying to Guernsey in February 2025 was accused of infringing Southampton's controlled airspace by "a few metres." What followed exposed deep flaws in the CAA's infringement process.
The pilot's own GPS track data showed the aircraft remained 30 metres outside controlled airspace at its closest point. Despite this, the CAA pursued the infringement allegation based on radar data from a station 120 km away — well beyond optimal range. The closer Southampton radar, which would have been more accurate at that location, showed no infringement. The CAA chose to rely on the distant radar that supported the allegation and disregard the closer one that didn't.
When the pilot responded to the Mandatory Occurrence Report, the assigned CAA official conducted what the pilot characterises as a biased investigation:
NATS initially refused to provide the radar trace data. The pilot was forced to escalate through a Freedom of Information appeal to the Information Commissioner. When the data was finally released, it turned out to be simple JPG images — not complex operational data requiring special handling. The refusal appeared designed to prevent scrutiny of the evidence, not to protect operational security.
Analysis of the released radar traces revealed a software bug in NATS' evidence recording tool that shifts aircraft positions between different display views. In some views the aircraft appeared outside controlled airspace; in others the same data showed it inside. This means the "evidence" of infringement may be an artefact of a display bug — not a reflection of where the aircraft actually was.
Full discussion of this case: EuroGA forum thread — "CAA Infringements Process: A scandal"