IRI vs rut depth vs texture depth: three records from one road profiler pass

IRI, rut depth and texture depth are three records from one pass of a laser profiler, each answering a different question. The International Roughness Index (IRI) summarises ride quality along a wheelpath, rut depth measures permanent wheelpath deformation across the lane, and macrotexture, reported as mean profile depth (MPD), measures surface coarseness; the South African road manual TMH 13 (Automated Road Condition Assessments, Committee Draft Final, May 2016) defines them in Parts C, D and E (COTO, 2016a) (COTO, 2016e) (COTO, 2016f).

Key facts

• South African road manual TMH 13 (Committee Draft Final, May 2016) defines the International Roughness Index (IRI), rut depth and mean profile depth in Parts C, D and E.
• The International Roughness Index (IRI) is calculated from the profile of one wheelpath and reported in m/km, whereas rut depth and mean profile depth (MPD) are reported in mm.
• TMH 13 Part D notes that rut depth in South Africa is commonly defined as the maximum permanent deformation under a two-metre straightedge placed across the rut.
• TMH 18 (Draft Version 4, Committee Draft 5.0) describes a combined instrument data (.cid) record holding IRI, rut depth and MPD with shared road, lane, km range and date.
• Texture depth reported as mean profile depth (MPD) describes macrotexture only, and TMH 13 Part E treats skid resistance as a separate wet-surface measurement.
• TMH 22 Part D (draft Road Asset Management Manual), section D.5.4, Table D-8, lists lane, wheelpath, test interval and stored interval for each instrument record separately.

How do IRI, rut depth and texture depth differ?

IRI, rut depth and texture depth differ in the question each answers, the profile it uses, its unit and the TMH 13 part that defines it. TMH (Technical Methods for Highways) is a series of South African road manuals now issued under the Committee of Transport Officials (COTO). TMH 13 is a committee draft that, under the document-version rules in each part, has no legal standing (COTO, 2016a). A wheelpath is the strip of lane that vehicle wheels follow.

The table draws intervals and wheelpaths from TMH 22 Part D (the draft Road Asset Management Manual, March 2013), which files roughness and texture as functional measures and rutting as structural distress, noting that permanent deformation is both (section D.4.1, Table D-2) (COTO, 2013).

Record and TMH 13 part (Committee Draft Final, May 2016)	What it measures	Unit	Interval and wheelpath set in	What it cannot tell you
IRI, Part C (Roughness)	Ride response to one wheelpath's longitudinal profile	m/km	TMH 22 Table D-8; TMH 13 Part A, A.4.3.5 and A.4.3.7	Where the roughness sits, its cause, or the treatment
Rut depth, Part D (Rutting)	Permanent wheelpath deformation, taken across the lane	mm	TMH 22 Table D-8; TMH 13 Part D, D.2.4 and D.4.2	Which layer deformed, or the ride quality
Texture depth as MPD, Part E (Skid Resistance and Texture)	Macrotexture: average profile depth over a short baseline	mm	TMH 22 Table D-8; TMH 13 Part E, E.4.3	Skid resistance, which needs a wet-friction measurement

TMH 18 (Road Asset Data Electronic Exchange Formats; Draft Version 4, October 2014, and Committee Draft 5.0, February 2016) carries all three in one file record.

What does a laser profiler measure on a road in one pass?

One pass yields one stream of measurements, and the three records are separate calculations on different slices of it. A laser profiler is a survey vehicle with height lasers, accelerometers and a distance measuring instrument that log road elevation against distance (TMH 13 Part C, section C.2.3.1) (COTO, 2016a). Modern profilers often measure the transverse profile on the same pass (section C.2.3.2), and Part D, section D.2.2.2, notes that one to three lasers are normally sampled at a higher frequency to capture texture (COTO, 2016e).

The longitudinal profile along a wheelpath gives IRI, the transverse profile across the lane gives rut depth, and the fast-sampled lasers give MPD. They also interact: coarse texture can add false roughness to laser-profiler IRI (Part C, section C.5.5.1), and rutting is one of the defects that produce roughness (Part A, section A.2.3.1) (COTO, 2016a) (COTO, 2016d). See the guide to TMH 13 and its seven parts.

Narrated film · 1:26

One pass, three records

A map splits one laser profiler pass into three records filed side by side in TMH 18, a compare scene sets what each records against what it cannot tell you, and a section strip shows synthetic Road S1.

Read the film transcript

Roughness, rut depth and texture depth come from one pass of a laser profiler, so are they one measurement? Roughness is reported as the International Roughness Index, the IRI: how rough the ride is along a wheelpath, the strip of lane the wheels follow. Rut depth is the permanent deformation in the wheelpaths, measured across the lane. Texture depth measures macrotexture and is reported as mean profile depth, the MPD. One pass gives one stream of measurements, and each record is calculated from a different slice of it. The South African road manual TMH 13, Committee Draft Final of May 2016, defines them in Parts C, D and E. TMH 18 files them side by side. IRI cannot say where roughness sits or why, rut depth cannot say which layer deformed, and MPD is not skid resistance. Take synthetic Road S1, with every value invented, starting at window A. In window B, left IRI barely moves, yet left rut depth more than doubles. Window C doubles window A's IRI at almost the same rut depth, so rut depth does not explain it; MPD roughly halves. Never add the three into one condition. Name the TMH 18 version, Draft Version 4 or Committee Draft 5.0, because column names differ. One pass gives three records, each with its own question, so keep them apart. Next, read TMH 13 Part B on positioning, then TMH 22 Part D on storing the records.

What does IRI measure on a road?

The International Roughness Index (IRI) answers one question: how rough is the ride along a wheelpath? It is calculated from the profile of one wheelpath by simulating the response of a single wheel and suspension of a typical passenger car at 80 km/h, and is reported in m/km (TMH 13 Part A, section A.2.2.4; Part C glossary) (COTO, 2016d) (COTO, 2016a).

Because the calculation filters out wavelengths that do not bear on ride quality, Part A advises against interpreting IRI over very short lengths and recommends averaging over a longer section, as TMH 22 Part D, section D.4.3, also does (COTO, 2016d) (COTO, 2013). IRI cannot say where in an interval the roughness sits, what causes it, or what treatment follows. See the insight on IRI in South Africa.

What does rut depth measure?

Rut depth answers a different question from IRI: how deeply have the wheelpaths deformed? Rutting is the longitudinal permanent deformation in the wheelpaths of flexible pavements, and rut depth in South Africa is commonly defined as the maximum permanent deformation under a two-metre straightedge placed across the rut (TMH 13 Part D, glossary; Part A, section A.2.3) (COTO, 2016e) (COTO, 2016d).

A profiler measures the transverse profile with point sensors or scanning lasers and applies an algorithm instead of a straightedge. Part D, section D.2.3.1, describes straightedge, wire and pseudo-rut models, and sections D.2.2 and D.2.3.2 show that manufacturers differ and that sensor number and spacing change the result (COTO, 2016e). Part D, section D.4.2, lists average, maximum and 90th-percentile depth as typical columns, so the statistic is a specification choice.

What does texture depth (MPD) measure?

Mean profile depth (MPD) answers one question: how coarse is the surface at the scale of its aggregate? Mean texture depth (MTD) comes from the volumetric patch method, which spreads a known volume of material over the surface. MPD comes from a profiler as the average profile depth over a baseline, standardly 100 mm (TMH 13 Part E, section E.2.2 and glossary) (COTO, 2016f).

MPD is not skid resistance. Part E's glossary treats skid resistance as a wet-surface measure linked to microtexture, section E.2.4.1 explains that macrotexture mainly governs how quickly skid resistance falls as speed rises, and section E.2.4.2 states that no single skid-resistance value defines a safe or hazardous surface (COTO, 2016f). See the standards page on texture and skid resistance.

How are IRI, rut depth and MPD filed together?

The TMH 18 combined instrument data (.cid) record holds the three, and TMH 22 Part D lists each record's lane, wheelpath and interval (COTO, 2014) (COTO, 2013).

Which wheelpath and interval applies to each record?

TMH 22 Part D, Table D-8 (section D.5.4), lists each record's lane, wheelpath, test interval, stored interval, unit and direction, and notes that raw values should be stored at the lowest collected interval supplied (COTO, 2013). Texture shows the problem: Part E, section E.4.3, lists left and right wheelpaths, TMH 22 section D.4.4 allows left, right and/or centre, Table D-8 names a single wheelpath, and TMH 18 carries left, centre and right (COTO, 2016f) (COTO, 2013) (COTO, 2014). A specification should therefore name the test positions (Part A, section A.4.3.5) and the averaging procedure and reporting length (section A.4.3.7) (COTO, 2016d). See the guide to lane and wheelpath conventions.

What does the TMH 18 combined record carry?

Section 5.6 of TMH 18, in both Draft Version 4 and Committee Draft 5.0, describes the .cid file (COTO, 2014) (COTO, 2016i). For a single-file submission the three must have been measured at the same time and place, because they share one road identifier, lane code, km range and date. Each row holds an average and a standard deviation for IRI (left and right wheelpath), a half-car roughness index (HRI), rut depth (left and right) and MPD (left, centre and right).

Name the version in every example: the left-wheelpath IRI column is AVG_IRI_L in Draft Version 4 and IRI_L_AVG in Committee Draft 5.0. In either version the file has no combined index, no column for the rut-depth algorithm or the MPD baseline, and neither the maximum nor the 90th-percentile rut depth. See the guide on how a profiler run becomes a register value.

What does a worked example of one profiler pass show?

Synthetic example. Every value below is invented. Synthetic Road S1, lane P1, one profiler pass, three consecutive 100 m windows (an assumed length), each a Draft Version 4 .cid row.

• Window A, chainage (distance along the road) 3,200 m to 3,300 m: IRI 1.7 m/km left and 1.9 m/km right; rut depth 2.4 mm left and 3.1 mm right; MPD 0.62, 0.55 and 0.60 mm (left, centre, right).
• Window B, 3,300 m to 3,400 m: IRI 1.8 and 2.0 m/km; rut depth 5.6 and 6.3 mm; MPD 0.58, 0.54 and 0.57 mm.
• Window C, 3,400 m to 3,500 m: IRI 3.4 and 3.7 m/km; rut depth 2.7 and 2.9 mm; MPD 0.31, 0.28 and 0.30 mm.

Between A and B the left IRI barely moves (1.7 to 1.8 m/km) while the left rut depth more than doubles (2.4 to 5.6 mm), so a roughness-only report would miss the deformation. Between A and C the left IRI doubles (1.7 to 3.4 m/km) while rut depth hardly changes, so rut depth does not explain C; the profile, video and a visual assessment must. MPD in C is about half that in A (0.31 against 0.62 mm on the left), a statement about macrotexture only (COTO, 2016f).

Check value: if window A stored ten left rut values, nine of 2.0 mm and one of 6.0 mm, they average (9 × 2.0 + 6.0) ÷ 10 = 2.4 mm, and the row cannot show the 6.0 mm spot; its standard deviation shows spread, not the maximum.

What does TMH 13 leave undecided, and what is often misread?

TMH 13 leaves the rut statistic, test positions and averaging to the survey specification (Part A, sections A.4.3.5 and A.4.3.7); Part D takes the two-metre straightedge as the South African reference, but the device's algorithm must still be named (COTO, 2016e). Interpretation ranges appear in Part A (sections A.2.2.4 and A.2.3.3) and Part E (section E.2.4); this guide reprints none, and the road authority's terms of reference decide which apply (COTO, 2016d) (COTO, 2016f).

Common misreadings:

• One condition. The three records do not add up to one condition: their units and questions differ, and the .cid row carries no combined value.
• A rough ride means deep ruts. Window C doubles window A's left IRI at almost the same rut depth, and window B shows the reverse: IRI responds to ride-relevant wavelengths along a wheelpath, not to rut depth across the lane.

What should be kept with the three records?

For each survey and .cid row, keep:

• the .cid file as delivered and a note of its TMH 18 version, which the file does not state (no header, section 1.1) (COTO, 2014);
• the shared road identifier, lane code, km range and date, with the run's direction and measurement speed;
• the position each column describes and the intervals behind each average;
• the device description: laser count, sampling rate, rut-depth algorithm and MPD baseline;
• the dated profiler validation record (Part C, section C.4; Part D, section D.3; Part E, section E.3) (COTO, 2016a) (COTO, 2016e) (COTO, 2016f);
• operator flags and the raw values at the lowest collected interval;
• the specification clause that named the interval, wheelpaths and statistics.

Part D, section D.4.4.1, and Part C, section C.5.5.1, note that lateral placement varies between runs and affects rut depth and roughness (COTO, 2016e) (COTO, 2016a); see the insight on repeatability and uncertainty.

Frequently asked

What is the difference between IRI, rut depth and texture depth?

In the South African road manual TMH 13 (Committee Draft Final, May 2016), the International Roughness Index (IRI) in Part C summarises ride quality along a wheelpath in m/km, rut depth in Part D describes permanent wheelpath deformation across the lane in mm, and texture depth reported as mean profile depth (MPD) in Part E describes macrotexture in mm. One laser profiler pass yields all three records.

Is texture depth the same as skid resistance?

No. In TMH 13 Part E (South African Technical Methods for Highways, Committee Draft Final, May 2016), texture depth reported as mean profile depth (MPD) describes macrotexture, the coarse unevenness of the surface, in millimetres. Skid resistance is a wet-surface friction measurement made with a friction-measuring device, not a texture laser. Macrotexture influences how friction falls with speed, but an MPD value cannot replace a skid-resistance measurement.

What does MPD mean on a road survey?

Mean profile depth (MPD) is the average profile depth over a baseline, standardly 100 mm, reported in millimetres as a measure of macrotexture. A laser profiler reports it under TMH 13 Part E, the South African road manual. Mean texture depth (MTD) is a different measurement, made by spreading a known volume of material on the surface, so the two should not share a column without a stated conversion.

Do IRI, rut depth and MPD share one TMH 18 file record?

Yes, in the combined instrument data (.cid) record. TMH 18, the South African road-data exchange formats manual (a COTO draft), describes it in section 5.6 of both Draft Version 4 and Committee Draft 5.0. The International Roughness Index (IRI), rut depth and mean profile depth (MPD) share the road identifier, lane code, start km, end km and measurement date, so they must come from the same time and place. Column names differ between versions, so name the version.

Which wheelpath do IRI, rut depth and MPD each use?

TMH 22 Part D (the draft Road Asset Management Manual, March 2013), section D.5.4, Table D-8, lists the lane and wheelpath for the International Roughness Index (IRI), rut depth and mean profile depth (MPD) records, and they are not set out alike. TMH 18 carries left and right IRI and rut depth, and left, centre and right MPD. A survey specification should state which positions it needs and how each average is formed.

Can IRI, rut depth and MPD be combined into one condition score?

Not by adding or averaging the International Roughness Index (IRI), rut depth and mean profile depth (MPD), because their units and questions differ. An index that draws on some of them, such as the TMH 22 Part E pavement index CIPAVE (also printed CIFPAVE), states its own inputs and weights. The TMH 18 combined instrument data record holds the three without any combined value, so any index sits outside the file and should be documented separately.

Where this connects

TMH 13 Part B (Positioning) describes the GPS, inertial and distance measuring equipment that fix where each record starts and ends; read it before comparing two surveys of one road (COTO, 2016c). TMH 22 Part D describes how the records are stored, reported and quality-controlled in an asset register (COTO, 2013). Last reviewed on 21 October 2026 against the TMH 13, TMH 18 and TMH 22 editions named above.

References

1. COTO (2016a). TMH13: Automated Road Condition Assessments. Part C: Roughness. Committee Draft Final, May 2016. Original source.Sections C.2–C.5; Table C.5 is for calibration-section selection
2. COTO (2016c). TMH13: Automated Road Condition Assessments. Part B: Positioning. Committee Draft Final, May 2016. Original source.Positioning and distance measurement guidance
3. COTO (2016d). TMH 13: Automated Road Condition Assessments. Part A: General. Committee Draft Final, May 2016. Original source.Sections, tables and clauses are cited inline in each guide
4. COTO (2016e). TMH 13: Automated Road Condition Assessments. Part D: Rutting. Committee Draft Final, May 2016. Original source.Sections, tables and clauses are cited inline in each guide
5. COTO (2016f). TMH 13: Automated Road Condition Assessments. Part E: Skid Resistance and Texture. Committee Draft Final, May 2016. Original source.Sections, tables and clauses are cited inline in each guide
6. COTO (2014). TMH 18: Road Asset Data Electronic Exchange Formats. Draft Version 4, October 2014. Original source.Sections, tables and clauses are cited inline in each guide
7. COTO (2016i). TMH 18: Road Asset Data Electronic Exchange Formats. Committee Draft 5.0, February 2016. Original source.Sections, tables and clauses are cited inline in each guide
8. COTO (2013). Draft TMH 22: Road Asset Management Manual. Draft, March 2013. Original source.Sections, tables and clauses are cited inline in each guide

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