Authorisation Required Approaches: What the Curved Path Gains, and What It Costs

An airliner follows a curved approach path over a mountain valley at dusk, banking to maintain the arc as it aligns with the distant runway.

At some valley-floor airfields, the extended runway centerline runs into a mountain a few miles from the threshold: a direct approach does not exist, and what replaces it – a mandatory visual maneuver or an offset final approach course with raised minima – requires exactly the visibility that is lacking on the day the valley is closed. A constant-radius arc solves the geometry differently: it guides a descent where the terrain leaves a corridor. What this curve gains is counted in hundreds of feet and days of operation. What it costs never stops, and it is this permanence, far more than the difficulty of flight, that explains why these approaches remain rare and local.

The problem that the curve solves, and that no altitude solves

A direct approach requires a cleared volume: a corridor aligned with the runway, wide enough to contain navigation imprecision, and high enough to overfly what exceeds. On a confined airfield, this corridor exists – it is the valley – but it does not point toward the runway. The valley alignment and the runway axis form an angle, and neither will be moved.

Conventional solutions all work on the same variable: altitude. The minimum is raised until the protected area clears the ridges. The final approach course is offset, and the offset is paid for in required visibility. The end of the flight is entrusted to the crew in the form of a mandatory visual maneuver, with a minimum altitude that has little to do with that of a precision approach. The similarity of these solutions is evident when they are lined up side by side: they trade guidance for a vertical margin at the precise moment when guidance is most valuable.

The curve, on the other hand, shifts the problem from one axis to another. Instead of aligning the trajectory with the runway and raising the protected area above the terrain, the protected area follows the corridor that the terrain leaves free, then the runway centerline is joined by an arc a few miles from the threshold. The obstacle is no longer cleared vertically: it is avoided laterally. The whole authorization logic flows from this shift, because avoiding laterally requires knowing, at all times and without margin, where you are laterally.

Why the direct axis does not exist on a valley-bottom platformTop view. The extended runway axis, in dotted lines, crosses a massif. The trajectory with required authorization, in solid lines, arrives from the bottom of the valley, follows the terrain, then joins the axis by a constant-radius arc a few miles from the threshold.terraindirect trackhits the high groundconstant-radius arconly cleared corridorrunway
The clear corridor exists, but it does not point toward the runway. Compare the two trajectories: the dotted line is short and impractical, the solid line is longer and protected along its entire length. The connection is made only a few miles from the threshold, leaving little time between exiting the turn and making a decision.

The curve avoids obstacles by turning around them instead of flying over them. This allows lower minimums and more reliable operations in bad weather, but it requires precise navigation and constant monitoring.

At Courchevel Airport (LFJT), an Airbus A320 with RNP AR capability can use a constant-radius arc approach in poor visibility. The procedure starts 6.5 NM from the threshold, with a 1.5 NM radius turn at 150 knots, joining the runway axis at 1,000 feet. Without this approach, the airport would require 2,500 feet of visibility, reducing operations by 30 days per year.

Anatomy of a constant-radius arc

A constant-radius segment is not a turn: it is a ground track, defined by a center, a radius, and two reference points, coded as such in the navigation database, entered and exited tangentially. The computer does not have to decide where the maneuver begins; it must maintain a circle against a wind whose effect changes with each degree traveled.

The in-flight consequence is mechanical. On an imposed radius, the required bank angle increases with the square of the ground speed: a tailwind encountered in the turn does not only cause drift, it consumes the available bank margin, and this margin is not large. Hence the speed limitations on the chart; hence, also, the expectation that these segments are flown coupled to the autopilot or, at minimum, the flight director. The trajectory to be maintained is not a trajectory that can be recovered by eye.

The crew loses three freedoms they usually do not think about. The turn is neither shortened nor extended. No radar vectors are given on this segment, and the protected area is that of the coded trajectory, not that of a neighboring trajectory that would have the same effect on the screen. Finally, the displayed lateral deviation is scaled to the applicable RNP value: the same offset symbol no longer represents the same distance as in a standard approach. Same image, tolerance several times tighter.

Anatomy of a constant-radius segmentThe solid trajectory enters with a straight segment, follows a quarter-circle defined by a coded center and radius, and exits tangentially. The thin dotted lines delimit the protected corridor on each side. A second dotted curve shows a trajectory whose drift is not compensated: it exits the corridor outside the turn.straight segmentarc centercoded radiusconstant-radius arccorridor: twice the RNPtangential exituncompensated driftthe flight path leaves the corridor
The trajectory is a ground circle, not a turn left to judgment. Center, radius, and reference points are coded; entry and exit are tangential. Drift is continuously compensated, otherwise the trajectory escapes outside the turn, where there is no secondary area.

What the curve actually buys

The first gain is a minimum. When the protected area follows the valley, the obstacle that dictated the minimum altitude is no longer under the trajectory, and the decision height drops to a value that no visual maneuver can approach. The margin is generally measured in hundreds of feet – a magnitude, never a rule, since each procedure is adapted to its own obstacle survey.

A pilot views a winding road through a mountain valley from the cockpit of a helicopter at sunset.
Complex terrain demands precise navigation and planning.

The second gain is the only one that really interests an operator: regularity. The question is not whether you can land on a given day, but how many days a year the destination remains usable without diversion, without extra fuel, without cancellation decided the day before. It is this line that finances the rest. It makes the investment obvious for those who serve the valley all season, and indefensible for those who go there twice a month.

Finally, there is what the ground no longer has to provide. Where no antenna can be sited, installed, aligned, or protected from ground reflections, a trajectory calculated from satellites is indifferent to topography: it requires a reliable obstacle survey and precise coding, nothing more. It is precisely for this reason that these procedures appear first in places where no ground equipment could serve them: their interest is not that they are more modern, it is that they are possible.

Three decision levels for the same airfieldSide view. A massif occupies the left, the runway the right. Three dotted levels overlap: the minimum altitude of a mandatory visual maneuver, that of a classic offset approach, then the decision height of an RNP approach with required authorization, the lowest. The guided trajectory passes through the three levels, getting closer and closer to the threshold.the margin between these levels is measured in hundreds of feetguided trajectory toward the thresholdmandatory visual maneuverconventional offset approachdecision heightRNP approach, authorisation requiredthe marginterrainrunway
The lower the minimum, the closer the decision point is to the threshold. The three levels describe the same airfield treated in three ways. The gain is not seen on a single flight but in the annual count of days when the destination remains usable.
What the curved trajectory brings, and the permanent counterpart it requires
Domain Provided gain Corresponding permanent load
Minima Guidance maintained to a decision height, where the visual maneuver imposes a much higher minimum altitude Protected corridor without secondary area: performance must be maintained, not just declared
Regularity Days of operation gained in seasons of poor visibility, diversions, and extra fuel avoided Unavailability of equipment or crew removes access without notice
Ground installations No antenna to install, align, protect from reflections, or calibrate in the valley Obstacle survey, procedure validation, and maintenance on the designer’s side
Trajectory Shorter route, chosen overflown areas, stabilized descent path toward the threshold Speed limits, constrained bank angle, no shortcuts or extensions possible
Crew A guided approach rather than a visual maneuver in poor visibility in mountainous terrain Individual qualification, skill maintenance, familiarization with the airfield if necessary

The price on the aircraft side, the price on the crew side

The aircraft’s eligibility is not an evaluation: it is written, in the flight manual or its supplements, and it applies to a specific configuration. Two aircraft of the same series may differ on a version of the computer or a display standard. What is expected, from one reference to another, is summarized in the same few lines: dual satellite-updated navigation, redundant computers and displays, guidance capable of maintaining an arc, a deviation scale linked to the RNP value, a performance loss alert, a terrain-based ground proximity warning system. The lower the required value, the more demanding the demonstration; and when the go-around also requires a finer value than usual, it is generally the subject of a separate demonstration, most often assuming an engine failure.

The minimum equipment list then becomes an access document as much as a departure document. A single inoperative item can remove the RNP approach capability while leaving the aircraft perfectly fit for the rest of the day. It is a preparation decision – destination choice, diversion choice, fuel load – and not a descent decision.

On the crew side, the requirement is individual, and that is where rostering gets stuck. Specific training and associated checks, flying constant-radius segments, degraded modes and loss of performance, discontinuing the approach from any point on the path, temperature window and local altimeter setting: this is acquired, then maintained. On many of these platforms, there is also the particular familiarization due to the airfield’s classification, which targets the captain and their recent experience at the location. An RNP AR approach is therefore not performed with an eligible aircraft and an available crew: it is performed with two qualified pilots, whom the rostering system must identify well before the day of the flight.

Nothing in the cockpit will prevent you from selecting it. The procedure appears in the navigation database of all aircraft receiving the same update, eligible or not. The computer displays, sequences, and flies it without asking questions. The deviation screen does not help either: its scale following the applicable RNP value, it presents the same image as in a standard approach while the distance represented by the same symbol offset is much shorter – nothing in what is displayed indicates that the corridor has narrowed. The only lock is administrative – aircraft configuration, operator certification, qualification of the two pilots – and it is verified on the ground, not on final.

Performance monitoring, the line that no one budgets

The lateral containment of an RNP AR approach has no secondary area: the evaluated space is the one the aircraft must hold, with nothing added. Such a regime cannot rest on a postulated performance. It rests on a measured performance, which creates for the operator a permanent monitoring obligation, distinct from everything else they do.

The total error is broken down into three contributions that do not have the same owner or the same remedy. The procedure definition first: what is in the navigation database versus what is published on the chart. The navigation system error next: the distance between the calculated position and the real position, a question of sensors, antenna installation, and signal availability. Finally, the flight technical error: the distance between the calculated position and the trajectory to be followed, i.e., what the autopilot and the crew hold. The third is the one that flight recordings show the best, and it is often the only one that is looked at.

Decomposition of total error and confinement limitsHorizontal bar starting from the trajectory axis. Three contributions add up: procedure definition, navigation system, then trajectory maintenance, the latter highlighted. Two vertical dotted markers indicate once the RNP value, to be maintained almost all the time, and twice this value, the confinement limit beyond which there is nothing.total system errordefinitionsensor and receivertrajectory maintenanceonce the RNPmaintained almost all the timetwice the RNPconfinementtrajectory axislateral deviation
Three contributions, three owners, one budget. The procedure coding and the sensor performance escape the crew; holding the path is entirely their responsibility. It is also the contribution that flight recordings show the best, so it is the first to monitor.

Monitoring requires defined events, extracted, read by someone who knows what they mean, then reported: a navigation anomaly is reported to the service provider and the authority, as it concerns everyone who flies the same procedure. And the loop returns to the aircraft’s eligibility: a drift that sets in on a given aircraft stops being a maintenance problem to become a certification problem.

A burden that is not a file but a rhythm

What costs, in this regime, is not the initial request, however voluminous it may be. It is the three clocks that start ticking once certification is obtained, and none of them stops between rotations.

On every flight: the recorded deviations, the crew reports, the verification of the loaded trajectory against the day’s chart. Every update cycle, i.e., every 28 days: the coded procedure must match the published procedure, and an unnoticed modification does not announce itself – it presents itself as a normal approach, with one less reference point or a different radius. In the longer term: obstacle survey to be reviewed on the designer’s side, training to be renewed, certification to be re-examined. Three rhythms, three responsible parties, and one common consequence if one of them falls behind.

The three clocks of an RNP approach with required authorizationThree superimposed timelines starting from the delivery of certification. The first has a marker for each flight, the second a marker every 28 days, the third a few spaced markers for amendments and periodic reviews. None of them stops between rotations.certification deliveredon every flightevery 28 dayslong termrecorded deviations, reports, trajectory verifieddatabase compared to the published procedureobstacle survey, training, certification reviewthe load is not a file, it is a rhythm
The real cost is measured in repetitions, not in pages. The initial request is made once; the three clocks, on the other hand, continue to tick as long as the procedure is in the network. An operator who no longer feeds the fastest of the three loses the ability to demonstrate its performance, even before actually losing it.

Take the four gates in the order in which certification placed them. The aircraft as it is this morning, minimum equipment list in hand; the crew as it is composed, the two seats; the procedure as it is coded in the current cycle, verified against the chart, reference point by reference point; then the day’s conditions, temperature, local altimeter setting, wind in the turn. And always keep a way out that requires no special authorization: a diversion that can be joined with a standard approach.

td>Crew
Who produces what, at what rate, and what happens in the absence of the object
Actor Produced object Rate Consequence of absence
Procedure designer Obstacle survey, trajectory, validation At creation, then at each amendment and periodic reviews The procedure is withdrawn, or its minima are raised
Navigation database provider Coding of reference points, arcs, and constraints At each update cycle The approach disappears from the screens – or, worse, remains with an obsolete trajectory
Operator Certification, procedures, training, deviation monitoring Permanent Access is lost for the entire fleet, not just for one aircraft
Maintenance Eligible configuration, elements required by the minimum equipment list At each departure The aircraft remains fit for flight, the approach does not
Verification of the loaded trajectory, conduct, report On every approach The deviation is not seen: monitoring loses its input

Why they remain rare, and why they remain local

A procedure of this type exists only where four conditions are met: a decisive gain, i.e., an airfield that would otherwise be lost; a designer and a state ready to survey, validate, and maintain the data; part of the fleet eligible; and enough qualified crews to serve it without planning contortions. Remove one and the procedure is not designed – or, worse, it is published and little used, which produces a skill that nobody really keeps current.

The asymmetry explains the rest. The gain is local: an airfield, often only one runway orientation, sometimes only one wind case. The load, on the other hand, is global: it affects the fleet, the training system, deviation monitoring, and documentation monitoring. Two rotations per week pay for the same permanent structure as twenty. That is why these approaches are concentrated on a handful of platforms and, on each, on a handful of operators – most often those for whom the destination is structural.

There is even a second-order effect, rarely mentioned. Because they are rare, the crews who perform them perform them little; because they perform them little, skill maintenance goes through the simulator rather than the line; and the cost goes up a notch. Rarity sustains itself. It is not an anomaly to be corrected, it is a datum to be integrated into the system’s sizing, on a par with the number of qualified captains on the platform.

What to check before counting on it

In your own documentation, four points deserve careful reading, and are verified faster than one might think:

  • the declared capability for your exact configuration, including the serial number and standard, and not for the type in general;
  • the RNP values covered by the certification, in approach and go-around: coverage up to a certain value does not open procedures that require a finer one, even if the computer accepts them;
  • what the minimum equipment list removes from this capability, element by element, and at what time of the day this reading must be done;
  • the published temperature window and the requirement for a local altimeter setting, which close the procedure without any equipment failing.

There remain gray areas that manuals rarely address head-on: the fate of a procedure amendment discovered mid-cycle, wet leasing and the shared responsibility it implies, and the case of the airfield where the RNP AR approach is the only practicable entry. On this last point, the right answer is not an approach answer but a preparation answer: diversion and fuel, decided at the moment you choose to go.

For those preparing a qualification or a fleet change, finally, this capability is a line that weighs on a schedule – but it is a line that expires. The useful question is not “Are you qualified?” but “When did your last recurrent training cover a constant-radius segment and a discontinued approach from the turn?” This is worked on upstream, in a training path as in the field reports exchanged on the professional forums.

This article describes a mechanism, it has no authority. The applicable text, the operator’s approved documentation, and the aircraft’s flight manual prevail over everything that precedes; for a given procedure, it is the chart in force in the day’s cycle that is authoritative.

Commentaires

Leave a Reply

Découvrez l’écosystème FL330

FL330, c’est toute l’aviation professionnelle — explorez nos autres espaces.