How to survey runway roughness on the centreline and the main-gear wheel paths

A runway roughness survey is a longitudinal profile, measured along the centreline and along offset lines that approximate where the main-gear tyres run, at a fixed interval and with enough wavelength reach for the analysis that follows. The United States Federal Aviation Administration circular FAA AC 150/5380-9 (Guidelines and Procedures for Measuring Airfield Pavement Roughness, 30 September 2009) sets out the interval, the straightedge range and the survey lines for the Boeing Bump method, which tests a profile against straightedges of many lengths (FAA, 2009). This guide covers the lines, interval, device and evidence. It is engineering background, not operational guidance.

Key facts

• FAA AC 150/5380-9 (30 September 2009) sets a survey interval of 0.25 m (0.82 ft) and straightedges from 0.5 m to 120 m (394 ft) for the Boeing Bump method.
• With a 0.25 m interval and a 120 m longest straightedge, a point may have Ns = 120 / 0.25 − 1 = 479 straightedge lengths.
• FAA AC 150/5380-9 paragraph 2.2f recommends the centreline plus offsets of 3.05 m (10 ft) or 5.22 m (17.5 ft, as printed; 17.5 ft is 5.33 m), and says the gear location need not be matched exactly.
• Most laser profilometers developed for highway roughness capture wavelengths up to about 45 m, whereas an aircraft responds to wavelengths up to about 120 m (NLR-TP-2009-190).
• High-pass-filtered inertial profilers are not recommended for Boeing Bump use by FAA AC 150/5380-9 or Transport Canada AC 302-023.
• A synthetic 3,000 m runway at 0.25 m interval has 12,001 profile points per line, or 36,003 points on three lines.

Why is a runway profiled on several lines instead of one?

One line describes one strip of pavement, and an aircraft does not follow one strip. The nose gear runs near the centreline, while the main gear runs along two paths either side of it, so a centreline profile can miss a bump that only the main gear meets. FAA AC 150/5380-9 defines airfield roughness by its effect on aircraft: fatigue and wear, cockpit vibrations, excessive g-forces and reduced braking effectiveness. It does not define roughness by passenger ride quality, and it treats texture as a separate subject (paragraphs 1.2b to 1.2g) (FAA, 2009).

A longitudinal profile is the elevation of the surface plotted against distance along a line. Road practice profiles two wheelpaths, the strips that vehicle wheels follow, and the same idea applies here; see the guide to lane and wheelpath conventions across manuals. A runway has no painted lane for the gear, so the survey lines are chosen, not read off the pavement.

Which lines should be surveyed, and how do they relate to the gear?

The circular recommends the centreline plus a left and a right offset. An offset of 3.05 m (10 ft) can effectively address aircraft in Airplane Design Groups (ADG) II and III, and an offset of 5.22 m (17.5 ft, as printed in 2.2f; 17.5 ft is 5.33 m) can address ADG IV to VI. The same paragraph says the gear location need not be matched exactly, and advises avoiding longitudinal joints and light rows unless their effect is the subject of the survey (FAA, 2009). As an international comparator, Transport Canada AC 302-023 (Issue 02, 16 December 2016) normally uses 3.0 m offsets for moderately sized aircraft; the centreline (for the nose wheel) and offsets of 5 m to 6 m (nominally 5.25 m) "may also be desirable" (paragraph 5.3(1)) (Transport Canada, 2016).

Narrated film · 2:45

Surveying runway roughness on the wheel paths

Runway plan with the centreline and offset lines, the interval and the straightedge count.

Read the film transcript

An aircraft does not follow one strip of runway, so why would a survey measure only one line? A longitudinal profile is the height of the surface plotted against distance along a line. The nose gear runs near the centreline, while the main gear runs along two paths either side of it. So the survey takes the centreline plus a left and a right offset line. FAA AC one five zero, five three eight zero, dash nine, gives two offsets. Three point oh five metres suits the smaller airplane design groups, and five point two two metres suits the larger ones. Half a Boeing seven three seven main gear spacing is two point eight six metres, within point one nine of the first line. For a seven seven seven it is five point four nine metres, within point two seven of the second. The circular says the gear location need not be matched exactly, so these lines are chosen, not painted. The circular sets a survey interval of zero point two five metres. It tests straightedges, imaginary lines between two points, from zero point five to one hundred and twenty metres. Equation one counts the straightedge lengths at one point, and we build it term by term. Ns is the count. L max is the longest straightedge, one hundred and twenty metres. Delta x is the survey interval, zero point two five metres. Divide the first by the second to get four hundred and eighty, then subtract one. That leaves four hundred and seventy nine lengths. The device must keep the long wavelengths, the length of one repeat of an undulation. Most highway laser profilers capture wavelengths up to about forty five metres, while an aircraft responds to about one hundred and twenty. High-pass-filtered inertial profilers are not recommended for Boeing Bump use, because the filter removes the long wavelengths. Equation two counts the profile points on one line. N is the number of points. L is the surveyed length, here a synthetic three thousand metres. Delta x is the interval again, so divide, then add one, because both ends are measured. That gives twelve thousand and one points on a line, and thirty six thousand and three on three lines. Equation three gives the shortest wavelength the interval can represent. Lambda min is that shortest wavelength. Delta x is the interval, and the two is the sampling theorem: at least two samples per wave. Two times zero point two five metres is zero point five metres. Here synthetic waves are sampled every zero point two five metres. A two metre wave gets eight samples, and a half metre wave gets two, the limit. A three tenths of a metre wave gets one point two, and appears as a false wave one point five metres long. Run each line steadily, repeat one line, and check every file for gaps and spikes. Keep the offset, interval, device, direction, repeat runs and the raw profile file. This guide sets no limit, and decisions about operating a runway belong to the aerodrome operator.

The table sets those offsets beside published gear geometry.

Gear	Source geometry	Half-track	Nearest survey line	Difference
737-800 main gear	5.72 m spacing, 0.86 m between dual tyres (Boeing D6-58325-7 Rev C, 7.2.1)	2.86 m	3.05 m	0.19 m
777-200LR/-300ER/777F main gear	10.97 m between bogie centres, 12.90 m across outer tyre edges (D6-58329-2 Rev G, 7.2)	5.49 m	5.22 m	0.27 m

Both differences are small against the circular's wording that the gear need not be matched exactly (Boeing, 2025) (Boeing, 2024). On the 737-800, the 3.05 m line is 0.24 m from the nearer tyre centre of a dual set (calc/survey_lines.py). Paragraph 2.2f says the offset approximates the aircraft using the airport, and to take measurements at all locations if traffic contains all airplane groups.

What interval and straightedge range does the analysis need?

The circular sets a survey interval of 0.25 m (0.82 ft), calls it the FAA standard (paragraph 2.2c) and, in 2.2e, requires it for evaluating the Boeing Bump. It allows straightedges from twice the interval, 0.5 m, up to 120 m (394 ft) (paragraphs 2.2b and 2.2c) (FAA, 2009). A straightedge here is an imaginary line between two points on the profile (2.2a), and the Boeing Bump method tests many bar lengths at each point.

The formula, term by term: straightedges per point

Equation 1: Ns = L_max / Δx − 1

symbol	what it is in words	unit	where its value comes from	typical range
Ns	number of straightedge lengths possible at one profile point	count	the result	479 for this interval and range
L_max	longest straightedge	m (ft)	AC 150/5380-9 2.2b: 120 m (394 ft)	fixed by the circular
Δx	survey interval	m (ft)	AC 150/5380-9 2.2c, 2.2e: 0.25 m (0.82 ft)	fixed by the circular

The circular gives only the formula and the figure 479 (2.2d); the counting below is this guide's own reading. Straightedge lengths are whole multiples of the interval, kΔx, starting at k = 2 (the 0.5 m minimum) and ending at k = L_max / Δx. Counting k from 2 to that last value gives L_max / Δx − 1 lengths. Paragraph 2.2d (a) to (c) lets the point sit anywhere along each straightedge, so more than 479 positions are tested.

Worked numbers, from calc/survey_lines.out.txt: Ns = 120 m / 0.25 m − 1 = 479. Counting lengths from 0.5 m to 120 m in 0.25 m steps also gives 479.

It is a count, not a quality measure. The clause that applies the straightedge set is paragraph 2.4a of the circular, covered in the guide to FAA AC 150/5380-9.

How far must the profile reach in wavelength, and which device can deliver it?

Wavelength is the length of one repeat of a surface undulation. Most laser profilometers developed for highway roughness capture wavelengths up to about 45 m, whereas an aircraft responds to wavelengths up to about 120 m (NLR-TP-2009-190, paragraph 2.1) (NLR, 2009). A device that cannot see long wavelengths under-reports the slow undulations.

Two devices recur in the references:

• Inertial profiler. ASTM E950/E950M is device-independent and uses the inertial profiler as its example (ASTM, 2022). FAA AC 150/5380-9 (2.2g) and Transport Canada AC 302-023 (5.1(4)) do not recommend high-pass-filtered inertial profilers for Boeing Bump use, because the filter removes long wavelengths, and because acceleration, braking and cornering cause significant errors, so threshold-to-threshold profiles are hard to measure; the scope is Boeing Bump or simulated airplane accelerations (FAA, 2009) (Transport Canada, 2016). In report TC-23/09 an inertial profiler tracked a reference profile to about 67 m (220 ft) wavelength on the two profiles shown (App E, Fig E-2, p. E-4); do not treat that as a general figure (FAA, 2023).
• Rolling inclinometer. A hand-operated device on a rigid beam up to 0.305 m (1 ft), used at up to 4 km/h (TC AC 302-023 5.1(3)), described by ASTM E2133 (ASTM, 2025).

"Profilograph" is the United States term many readers search for. In report TC-23/09, simulated profilograph windows are research parameters, not FAA limits (FAA, 2023).

How many points will the survey produce?

Fix the arithmetic before mobilising.

The formula, term by term: profile points per line

Equation 2: N = L / Δx + 1

symbol	what it is in words	unit	where its value comes from	typical range
N	profile points on one line	count	the result	12,001 for the example
L	surveyed length	m	the runway plan	3,000 m, synthetic
Δx	survey interval	m	AC 150/5380-9 2.2c	0.25 m

This is derived by counting: L / Δx intervals have one more point than intervals, because both ends are measured. Worked numbers for a synthetic 3,000 m runway: N = 3,000 m / 0.25 m + 1 = 12,001 points per line; three lines give 36,003 points and five lines give 60,005. It assumes uniform spacing along a continuous line.

The formula, term by term: shortest resolvable wavelength

Equation 3: λ_min = 2 Δx

symbol	what it is in words	unit	where its value comes from	typical range
λ_min	shortest wavelength a sampled profile can represent	m	the result	0.50 m for the example
Δx	survey interval	m	AC 150/5380-9 2.2c	0.25 m

This is the sampling theorem: a wave needs at least two samples per wavelength. The two are numerically equal to the circular's minimum straightedge, twice the interval; its own reason is that three profile points are needed to obtain a deviation (2.2c). Worked numbers: λ_min = 2 × 0.25 m = 0.50 m, so the wavelength span from 0.50 m to 120 m is 240 to 1. In practice the usable shortest wavelength is longer than this bound.

How is the survey run, step by step?

1. Fix the runway ends, the chainage origin and the direction of the first run.
2. Choose the lines: centreline, plus left and right offsets approximating the aircraft using the runway; take all locations if traffic contains all groups (2.2f).
3. Mark each offset from the centreline and avoid longitudinal joints and light rows.
4. Select a device that retains the long wavelengths, and record its settings.
5. Set the interval to 0.25 m (0.82 ft).
6. Verify the device against a reference such as a rod-and-level survey (TC AC 302-023 5.1(2), usually the standard for evaluating other profiling equipment) or a rolling inclinometer (ASTM E2133), on a short section.
7. Run each line at a steady speed, in a recorded direction.
8. Repeat at least one line and compare the two runs.
9. Check each file for gaps, spikes and consistent start and end positions.

Run lines separately: as reviewed in the 2023 report, ProFAA uses one survey line (App E pp. E-3/E-4) (FAA, 2023). See the insight on measurement repeatability and uncertainty and the guide to checking a roughness survey on a validation section.

What should be kept with the survey?

Evidence this stage produces, for each line:

• offset from the centreline, with the marking method;
• interval, device model and serial number, and filter settings;
• run direction, start and end references and speed;
• repeat runs and their comparison;
• the chainage reference shared with any texture or friction runs on the same lines, as described in the guide on macrotexture and microtexture;
• the raw profile file as delivered.

See the insight on chainage, direction and wheelpath.

What does this guide leave undecided?

It does not set a roughness limit, compute a Boeing Bump Index, run a spectrum or estimate a load. The Boeing Bump Index comparison covers the index and the bump guide the loads. Decisions to operate, restrict or close a runway belong to the aerodrome operator under its State's rules.

Common misreadings: the offset lines are not painted lanes; the circular's 3.05 m and 5.22 m values are not exact gear tracks; and a profile that looks smooth to a highway laser may still hide long wavelengths.

The procedure, step by step

1. Fix the runway ends and chainage

   Record the runway ends, the chainage origin and the direction of the first run, so every later file shares one reference.

2. Choose the survey lines

   Take the centreline plus a left and a right offset approximating the aircraft using the runway, for example 3.05 m (10 ft) or 5.22 m (17.5 ft, as printed in 2.2f; 17.5 ft is 5.33 m); measure all locations if traffic contains all airplane groups.

3. Mark the offsets

   Mark each offset from the centreline and avoid longitudinal joints and light rows unless their effect is being evaluated.

4. Select a suitable device

   Choose a profiler that retains the long wavelengths; high-pass-filtered inertial profilers are not recommended for Boeing Bump use, nor where acceleration, braking and cornering cause errors.

5. Set the survey interval

   Set the interval to 0.25 m (0.82 ft), so straightedges from 0.5 m to 120 m give Ns = 479 lengths per point.

6. Verify the device on a short section

   Check the profiler on a short section against a reference such as a rod-and-level survey (TC AC 302-023 5.1(2)) or a rolling inclinometer (ASTM E2133) before the main runs.

7. Run each line

   Run each line at a steady speed in a recorded direction, keeping the lateral position on the marked offset.

8. Repeat at least one line

   Repeat a line and compare the two runs to separate real features from measurement artefacts.

9. Check each file

   Check every file for gaps, spikes and consistent start and end positions before leaving the runway.

Frequently asked

Where should the survey lines be placed on a runway?

FAA AC 150/5380-9 (2009), paragraph 2.2f, recommends the centreline plus a left and right offset. An offset of 3.05 m (10 ft) can effectively address Airplane Design Groups II and III, and 5.22 m (17.5 ft, as printed; 17.5 ft is 5.33 m) groups IV to VI. The offset approximates the aircraft using the airport, so the gear need not be matched exactly.

What survey interval does the Boeing Bump method use?

FAA AC 150/5380-9 (2009) calls a survey interval of 0.25 m (0.82 ft) the FAA standard in paragraph 2.2c, and paragraph 2.2e requires it for evaluating the Boeing Bump. Straightedges run from twice the interval, 0.5 m, to 120 m (394 ft), so a profile point can be tested against up to 479 straightedge lengths.

Why are high-pass-filtered inertial profilers not recommended for runway bump analysis?

FAA AC 150/5380-9 paragraph 2.2g and Transport Canada AC 302-023 paragraph 5.1(4) do not recommend them for Boeing Bump use. The filter removes long wavelengths, and acceleration, braking and cornering cause significant errors. An aircraft responds to wavelengths up to about 120 m, while most highway laser profilometers capture up to about 45 m (NLR-TP-2009-190).

Is a survey line the same as the aircraft wheel path?

No. A survey line is a chosen reference that approximates where the main-gear tyres run for a group of aircraft. FAA AC 150/5380-9 says the gear need not be matched exactly. Half the Boeing spacings (5.72 m and 10.97 m), computed here, give 2.86 m and 5.49 m, which differ from the circular's 3.05 m and 5.22 m by under 0.3 m.

How many profile points does a runway survey produce?

Points per line equal length divided by interval, plus one. For a synthetic 3,000 m runway at a 0.25 m interval, that is 12,001 points per line, 36,003 points on three lines and 60,005 points on five lines. Plan storage and repeat runs from this count before mobilising.

Where this connects

The circular is covered in FAA AC 150/5380-9 explained, and the frequency view of the same profile is in running a PSD at take-off speeds. The hub lists the governing documents. Last reviewed on 3 November 2026 against FAA AC 150/5380-9 (2009), Transport Canada AC 302-023 (Issue 02) and the reports named above.

References

1. FAA (2009). AC 150/5380-9, Guidelines and Procedures for Measuring Airfield Pavement Roughness. 30 Sep 2009, no revision letter, Active. Original source.Foreword; para 1.2b-g; para 2; 2.1; 2.2a-g; 2.3a-b; 2.4a-d; 2.5a-b; Fig 2-3; Fig 2-4; Ch 3 (3.1, 3.2)
2. Transport Canada (2016). AC 302-023, Measurement and Evaluation of Runway Roughness. Issue 02, effective 16 Dec 2016 (current per TC page, checked 3 Nov 2026). Original source.4.3(3); 5.1-5.3; 6.1 with Fig 1 and Table 1; 6.2 (IRI, RMSVA, RCI)
3. NLR (2009). Evaluation Methods for longitudinal evenness of runway pavements: An overview of recent Dutch experiences, NLR-TP-2009-190 (P.A. van Gelder, M.J.A. Stet). June 2009. Original source.Sec 1 pp. 5-6; Sec 2.1 p. 7 and Table 1; 2.2; Elaborate Survey pp. 9-10; 3.1 p. 11; 3.3 p. 13; 3.5 p. 16; 4 p. 18
4. FAA (2023). Method to Identify and Quantify Multiple-Event Roughness, DOT/FAA/TC-23/09 (T. Gerardi, M. Gerardi, APR Consultants). Final Report, January 2023 (does not constitute FAA policy). Original source.Exec summary pp. ix-x; Sec 2 p. 2; Sec 2.1-2.3 pp. 2-3; Sec 3 p. 4; Sec 3.1 p. 9; Table 1 p. 7; App E pp. E-3/E-4, E-8/E-9
5. Boeing (2025). 737 Airplane Characteristics for Airport Planning, D6-58325-7. Rev C, October 2025. Original source.Section 7.2.1, p. 7-5
6. Boeing (2024). 777-200LR/-300ER/-Freighter Airplane Characteristics for Airport Planning, D6-58329-2. Rev G, December 2024. Original source.Section 7.2, p. 7-5
7. ASTM (2022). Standard Test Method for Measuring the Longitudinal Profile of Traveled Surfaces (ASTM E950/E950M). E950/E950M-22. Original source.Scope 1.1
8. ASTM (2025). Standard Test Method for Using a Rolling Inclinometer to Measure Longitudinal and Transverse Profiles of a Traveled Surface (ASTM E2133). E2133-25. Original source.Scope 1.1

Use the manual edition and project requirements that apply to your work. Worked examples use synthetic values and are not site-specific design advice.