Operational Landing Distance

Soort commissiepublicatie: 
Technische publicatie

Thomas Bos

Introduction

Data gathered by the Dutch Aerospace Laboratory NLR and the Flight Safety Foundation indicates that runway excursions are one of the most important threats to aviation safety today. According to the Flight Safety Foundation 94% of the casualties in 2008 occurred in runway excursion accidents. The current situation produces on average more than one runway excursion per week worldwide.

Lessons Learned

As a result of various runway overruns following high speed rejected takeoffs the National Transportation Safety Board (NTSB) published a Special Investigation Report in 1990 which has amongst others resulted in the review of certification requirements and the appearance of the Takeoff Safety Training Aid. As a result the common understanding of dry and wet runway friction characteristics dramatically improved. An important lesson from these efforts is the need for a thorough understanding of the underlying aircraft performance assumptions and their validity.

A similar momentum can be observed the last few years with the release of the FAA Safety Alert for Operators (SAFO) on landing performance assessments and upcoming rulemaking activity as a result of the 2005 Chicago Midway accident. This accident and several other incidents and accidents again underline the need for a thorough understanding of aircraft performance and the assumptions in performance data presented to and used by the flight crew.

Dispatch versus Inflight

A dispatch landing distance check must be performed which allows the airplane to land within 60% (1.67 factor) of the available runway when the runway is dry. When the runway is forecasted wet an additional 15% needs to be scheduled (1.92 factor) where it should be noted that this factor refers to the dry certified landing distance and hence does not give any clue with regard to an available margin as there are no certified wet runway landing distances. When the runway is forecasted contaminated European operators should schedule a margin of 15% with regard to approved contaminated runway data.

The only requirement with regard to inflight landing distance checks is the requirement for the commander to be satisfied that a safe landing can be made. Generally speaking this requires a comparison between a calculated required landing distance (LDR) and the landing distance available (LDA).

Now how does dispatch data relate to the inflight data? The dispatch requirement can be based either on the longest runway with zero wind and/or the most likely runway assigned considering amongst others the probable wind speed and direction. Depending upon the alternate aerodromes this requirement can be commercially interpreted and the exact assumptions of the dispatch check are generally not known to the flight crew identifying the need for an inflight assessment. Further stressing the need for an inflight check is the fact that with increasing uncertainties regarding the achievable braking action the lower the required margin is (dry → wet → slippery/contaminated).

Dispatch requirements are intended to provide an overall acceptable safety level whereas inflight landing distances are intended to guarantee an acceptable safety level of an individual flight.

Manufacturer Data

Dry runway dispatch landing distances are the result of dry runway certification flight testing representative of the maximum capability of the aircraft and always corrected back to a threshold crossing height of 50 ft regardless of the actual threshold crossing height during flight testing. This is one of the reasons for the comparatively high correction factors during dispatch as these landing distances are not at all representative of line operations.

Inflight landing distances are not subject to specific requirements and various formats are encountered based on various assumptions which as it turns out may be equally unsuitable for operational application.

For Boeing and Airbus the following applies at the time of writing:

  Boeing Airbus
Air Distance Dry/Wet Runways: 1000 ft (305 m) Dry/Wet Runways: ≈ 500 fpm sinkrate*
JAA/EASA Contaminated Runways:
7 seconds flare
JAA/EASA Contaminated Runways:
7 seconds flare
Operational Margin Dry Runways: 0 % 0 %
Other Runways: 0% / 15 % EASA/JAA
McDonnell-Douglas products: 0 %
Displacement / Impingement Drag Not included Included

* Slightly less than certification flight testing

It is the responsibility of the individual airline to adjust the data to reflect their operations and to provide a suitable operational margin. As aircraft manufacturers deal with many airlines the reference data is usually based on common assumptions in order to limit the amount of data furnished.  Furthermore the manufacturer data represents the airplane capability and does not address the specific training of an airline nor does it address the approach guidance at specific airports.

Critical Factors

It shall come as no surprise that the factors having the largest influence on the actual landing distance are the runway condition, the encountered wind, the threshold crossing height and touchdown point and the aircraft speed. A discussion on most of these factors follows.

Runway Condition

Limiting the discussion to Boeing and Airbus two different approaches are used.

Boeing uses both descriptive contaminants as well as equivalent aircraft braking actions as the basis of the performance information allowing both to be used for landing distance evaluations at the preference of the operator. The following table provides an insight in the Boeing performance levels:

RunwayState

Performance Level

Braking Action

Aircraft Mu*

Dry

Reference

Not applicable, flight test

Flight Test

Wet

Skid-Resistant**

≈ 70 – 80 % dry

Not applicable, assumes max manual

Speed Dependent

Wet

≈ 50 % dry

Speed Dependent /

averaged GOOD

Speed Dependent / averaged 0.2

Contaminated/Slippery

≈ 25 % dry

MEDIUM

0.10

Contaminated/Slippery

≈ 12.5 % dry

POOR

0.05

* Aircraft Mu should not be confused with Mu values reported by a friction measurement device as these can differ by more than a factor of two (see SNOWTAM table for GOOD/MEDIUM/POOR).

**Grooved or Porous Friction Course runways meeting stringent maintenance requirements.

 

Airbus uses only the descriptive contaminant as the basis of their performance information and includes displacement/impingement drag as a function of contaminant depth in the landing distance calculation.

For runways equipped with a grooved or Porous Friction Course overlay and meeting specific maintenance criteria either generic landing performance based on roughly 70 – 80 % dry can be used or airport specific data following flight testing. However if this data is not in the operators AFM the wet runway data should be used whenever moisture is present, i.e. not dry.

What may come as a surprise is the fact that the well known SNOWTAM correlation between measured friction coefficient and braking action is only considered valid for compacted snow and ice covered runways and even that conclusion may not be valid under all circumstances. For other winter contaminants notably slush and wet snow friction measurements should be considered unreliable. This does not mean however that no equivalent braking action has to be reported as practiced by certain states as this would leave the crew without any information.

Measured or Calculated
Coefficient

Estimated Surface Friction

MOTNE Code

0.40 and above

GOOD

5

0.39 to 0.36

MEDIUM/GOOD

4

0.35 to 0.30

MEDIUM

3

0.29 to 0.26

MEDIUM/POOR

2

0.25 and below

POOR

1

9 - unreliable

UNRELIABLE

9

The big question is how accurate is all of this? An interesting summing up of assumed accuracies of friction measurements was presented by the Norwegian Accident Investigation Board (AIBN) during the ISASI 2007 meeting in Singapore

Year

Organization

Accuracy -
Uncertainty

Remarks

1962

ICAO

± 0.01

Reported by
a state

1974

ICAO

± 0.20, ±
0.15

Wet
surfaces

1974

ICAO

± 0.15, ±
0.10

Compacted
snow and ice surfaces

1990

NASA

± 0.10

Aircraft in
the loop

2005

ASTM

± 0.20
→ ± 0.05

Use of ASTM
standard E2100-04

In certain situations friction readings on winter contaminated/slippery runways can produce results that would translate to braking action GOOD whereas in reality the aircraft would experience POOR.

Landing performance assessments should therefore not rely solely on friction measurements; needless to say that optimization or calculations with presumed two digit accuracy should be avoided altogether.

Wet Runways

The quality of a runway when it is raining generally deteriorates relatively slowly for instance by the accumulation of rubber. As such the quality of the runway can be assessed periodically without the need for actual friction measurements during rain. This is done by so called maintenance friction measurements where a measuring device is used on a dry runway with an artificial wetting system to spray water ahead of the measuring wheel. ICAO defines three different maintenance levels with different threshold values as a function of the type of friction measurement device. These levels are known as the Design Objective Level (DOL) for newly constructed runways, the Maintenance Planning Level (MPL) when the runway or parts thereof have deteriorated up to a point where maintenance needs to be scheduled and the Minimum Friction Level (MFL) below which the airport should issue a NOTAM “slippery when wet”.

Unfortunately the Minimum Friction Level does not correlate with aircraft performance although an attempt is made in the ICAO Airport Services Manual. Ideally the absence of a NOTAM “slippery when wet” should allow the use of standard wet performance (≈ braking action GOOD) whereas the presence of the NOTAM would require the use of more conservative data such as braking action MEDIUM or worse.

Another limiting factor for wet runways is the actual water discharge capacity in combination with rain intensity. With increasing rain intensity at a certain point a limit will be reached where the water level will exceed the texture depth. Obviously the rain intensity required for this phenomenon will be higher for grooved or Porous Friction Course runways than for regular runways. In both cases at that point the positive effect of texture has reached its limit and a further increase in rain intensity will start to flood portions of the runway with possible hydroplaning and considerably reduced braking action.

Attempts to predict this actual limit as a function of the runway geometry by laboratory testing and analytical analyses are considered inaccurate as a result of various factors such as crosswind, measurement of rain intensity, variations in runway quality, rubber and paint markings.

Wind

In dispatch performance calculations it is only allowed to take account of 50% of the reported headwind where 150% of the tailwind should be taken into account. According to ICAO Document 7401/AIR/OPS/612 Final Report of the Standing Committee on Performance (1953) this rule is intended to cover variations of the wind around the forecasted wind. Current ICAO Annex 3 requirements specify a wind reporting accuracy of 2 knots and 10 degrees whereas gusts should be reported when equal to or exceeding 10 knots or the variation in wind direction exceeds 60 degrees. Although the 50% / 150% factors may generally cover variations in wind speed and direction, certain variations are not adequately covered by these factors, e.g. a shifting crosswind or headwind condition to a tailwind condition. Flight crews have no insight in the actual accuracy of wind reporting as this will depend largely on the airport geography (e.g. terrain roughness, buildings, runway layout, etc.), the quality of the meteorological services and the amount and position of measuring poles.

A common misconception by flight crews is the assumption that the maximum wind intensity will coincide with the most veered wind direction reported (northern hemisphere). In reality the suggested Coriolis effects are not predominant in the earth’s boundary layer and the maximum wind intensity may coincide with a backed wind direction. It should also be noted that wind shifts may occur when initiating a missed approach is not feasible anymore.

As such there is a need for more detailed insight in wind behavior near the surface and enhancement of wind measurement data which can improve reliability and validity of promulgated wind reports.

Threshold Crossing Height and Touchdown Dispersion

As discussed previously manufacturer data is normally based on a threshold crossing height of 50 ft and a touchdown point that may not be achievable in line operations. The actual crossing height over the threshold will depend on a lot of factors such as the positioning of the aiming point, the accuracy of the guidance used (precision versus non-precision), the glidepath angle, PAPI / ILS alignment, calibration reference for visual slope indications, aircraft type and last but not least piloting accuracy and missed approach criteria. See the following table for threshold crossing heights for an on-glideslope airplane with a glideslope intercept of 1000 feet and 1500 feet from the threshold.

Airplane

Gear Height (ft)

1000 ft Runway/GS intercept

Gear Height (ft)

1500 ft Runway/GS intercept

737-3/4/6/7/8/9

~ 33-36

~ 64-67

757-2/300

~ 27-28

~ 58-59

767-2/3/400

~ 22-25

~ 53- 56

747-400

~ 12

~ 43

777-2/2LR/3/3ER

~ 19-22

~ 50-53

 

For a 3 degree glidepath the effect of a higher threshold crossing height is best illustrated by the following figure:

landing_dist

 

Another complicating factor is the actual position of the aiming point. ICAO Annex 14 contains the requirements for the runway markings:

rwy_markings

 

For runways longer than 2400 meters the touchdown zone has a total length of 900 meters and the aiming point markings are located 400 meters beyond the threshold.  This means that when touching down at the aiming point the landing distance is increased by 95 meters as compared to a 1000 ft (305 m) reference. When a touchdown would occur at the end of the touchdown zone the landing distance would increase by 595 meters. Runways with a length between 1500 m and 2400 m should have a touchdown zone with a length of 600 m and aiming point markers 300 meters beyond the threshold.

An important thing to consider here is the fact that ICAO Annex 14 prescribes the minimum distance from the threshold and an IFALPA survey indicates that in reality things may be quite different. As an example it was found that for Charles de Gaulle airport all aiming point distances exceed 400 m and the aiming points for runways 27L and 27R actually exceed 500 m.

FAA research has confirmed that actual air distances may exceed the current assumptions used in the manufacturer advisory landing distance tables for dry and wet runways*. The FAA research shows the effect of pilot motivation and glideslope/runway intercept on the actual air distance achieved in service. The figures below present this air distance for two airports: Washington National (6869 ft / 2094 m runway) and London Heathrow (12743 ft / 3884 m runway) with glideslope/runway intercept points of 1000 ft / 305 m and 1157 ft / 353 m respectively. When the data is reviewed it appears that when the runway is shorter and has approach guidance that results in a shorter aiming point the flight crew will put the airplane on the runway sooner (in less distance).

* EASA CS-25 AMC 25.1591 requires an air distance based on a 7 second flare time for contaminated surfaces which results in an allowance of 1300 to 2000 ft.

Interesting in this context are the autoland certification criteria where the autoland landing distance should be validated by a set of flight tests and needs to be published when exceeding the landing distance for a manual landing. A Monte Carlo analysis should guarantee that actual autolands beyond the touchdown zone lighting (specified as 2700 ft / 823 m) are improbable (equivalent to three standard deviations). In reality a three standard deviation analysis will generally result in an air distance of 2100 ft or less.

Conclusions

Stabilized approach criteria and general guidelines such as “land in the touchdown zone” do not match the assumptions in the currently provided data by the manufacturers. Although there is industry activity at the moment and the reference values will likely change in the near future operators should still adapt the performance data for their specific operations.

Operators should adjust performance data to reflect their Standard Operating Procedures (SOPs) and should include suitable air distances (either a conservative value or based on Flight Data Analyses* and if required airport specific).

Operators should include an operational margin and in line with the FAA SAFO (Safety Alert for Operators) on landing performance assessments 15% is recommended.

Shortcomings in runway state reporting require a conservative approach reflecting the scale of the inaccuracies.

Flight crew should take uncertainties in wind reporting into account and use conservative wind values.

Insight in the actual aircraft landing capability is useful but should only be used as a reference for emergency situations at the discretion of the captain.

Operators should use a common format for operational landing distances minimizing the possibility of errors when flight crews transition to other aircraft types.

* Proper interpretation of Flight Data Analyses may require support from the manufacturer.