How to rate gravel road condition in South Africa under TMH 9 Part E

TMH 9 Part E (Committee Draft Final, May 2016) is the unpaved-roads part of the South African visual assessment manual for road pavements, and it sets out how to rate a gravel road one fixed segment (the length of road with one recorded rating (COTO, 2016b)) at a time (COTO, 2016k). Part E chooses no treatment and calculates no index. TMH (Technical Methods for Highways) is a series of road manuals issued under the Committee of Transport Officials (COTO); TRH (Technical Recommendations for Highways) is its companion series.

Key facts

• TMH 9 Part E (Committee Draft Final, May 2016), the unpaved-roads part of the South African visual assessment manual, states that it replaces the earlier Draft TMH 12.
• TMH 9 Part E has three assessments: material properties, surface distress rated by degree and extent, and functional items such as roughness, each given one degree for the whole segment.
• TMH 9 Part E prefers assessment in the dry season, with at least one stop on each segment for a closer assessment of material quality, layer thickness and general performance.
• The overall condition rating on the TMH 9 Part E form is not used in data processing; the rating provides checks on the other ratings.
• Draft TMH 22 (March 2013), the road asset management manual, names the unpaved condition index CIUNPAVED but leaves the comment and weighting columns of Table E-8 empty.
• A poor Part E rating does not by itself select a treatment; TRH 20 (Draft Version 1.5, April 2009) notes some low-volume roads rated poor can be safe and accessible.

Why does a gravel road need its own rating method?

The paved-road distress list does not carry over to a gravel road. TMH 9 Part B, for flexible pavements, lists surfacing and structural items such as cracking, bleeding, pumping and patching, and none is a distress row on the gravel form (COTO, 2016j). Potholes, rutting, roughness and drainage appear in both parts with different descriptions. Part A holds what all parts share: the definitions of degree (severity) and extent (how widespread), segment rules and procedure (COTO, 2016b).

A gravel road is a formed unpaved road with an imported gravel wearing course (COTO, 2016k), the layer that protects the subgrade (the in-situ material beneath) from wear by vehicles (COTO, 2016b). Part E names three unpaved road types (tracks, earth roads and gravel roads) and covers the last two (COTO, 2016k). TRH 20 (Draft Version 1.5, April 2009) calls these roads unsealed (Department of Transport, 2009).

What does TMH 9 Part E ask you to rate on a gravel road?

Part E asks for three assessments, material properties, surface distress and functional performance, all rated on one segment (COTO, 2016k).

• Material properties (section E.3): gravel quality and the factors behind it, maximum stone size, grading, plasticity, layer thickness, exposed subgrade and subgrade quality.
• Surface distress (E.4): potholes, corrugations (ridges across the road), rutting, loose material, stoniness and erosion, each rated by degree and extent; stoniness and erosion are each split in two, giving eight form rows.
• Functional performance (E.5): roughness, trafficability (whether a saloon car can pass), safety, drainage on the road and drainage from the road, each given one degree for the segment, with a problem row for the cause.

Narrated film · 1:25

Rating a gravel road, step by step

A comparison of paved and gravel distress rows, a step rail for rating one unpaved segment, a chainage line of Synthetic Road G1 with illustrative marks and no rating values, and the records to keep.

Read the film transcript

How do you rate a gravel road under TMH 9 Part E, the South African visual assessment manual? The paved-road distress list does not carry over: cracking, bleeding, pumping and patching have no gravel row. Part E rates each segment, the length of road with one rating, on material properties, surface distress and functional items. Confirm the edition, form and accreditation, then fix the segments from the register. Calibrate the team, assess in the dry season, and stop on each segment to inspect the gravel closely. Rate the gravel's material properties, then each surface distress, degree first and extent second. Give each functional item one degree, record the overall condition, then photograph, check and file. On this synthetic segment, twenty-four pothole stretches of about twenty metres cover seven point five per cent of its length. That percentage is only a guide, so note how widely the stretches spread, and where corrugations lie. We keep the form and its edition, the chainage, the photographs and the calibration notes, so ratings can be checked. Part E chooses no treatment and calculates no index; Draft TMH 22 names the unpaved condition index. Part E states that it replaces the earlier Draft TMH 12, a different manual from TRH 12. Rate each segment on Part E's own items, and keep the record behind every rating. Next, read TMH 22 for condition indices, and TRH 20 for unsealed road maintenance.

How do you rate one gravel segment, step by step?

1. Confirm the edition, the form and your accreditation. Agree with the client which edition and form apply, and check that your accreditation covers unpaved roads, since Part A (section A.5.3) accredits assessors per pavement type (COTO, 2016b). Keep: edition, form version and accreditation.
2. Fix the segments before driving out. Take start and end chainage (distance along the road) from the register. Part A (section A.3) sets the standard unpaved rural length and the short-length rule at either end, and recommends that the authority predefines segments (COTO, 2016b). Keep: identifiers, chainage, direction.
3. Calibrate the team. Hold the calibration session that Part A (section A.5.4) places at the start of internal quality control (COTO, 2016b). A shared trial length rated against Part E is an addition Part A does not prescribe. Keep: calibration notes and trial ratings.
4. Assess in dry conditions. Prefer the dry season, as Part E does (COTO, 2016k). Part E's season and speed ceiling differ from Part A's (section A.5.1) (COTO, 2016b), so record which you followed. Keep: moisture condition (wet, moist or dry).
5. Stop on each segment and rate the trafficked portions. Stop at least once per segment for a closer look at the gravel. Rate the trafficked portions, not the shoulders or the windrows (ridges of loose gravel) a grader leaves (COTO, 2016k). Keep: stop positions.
6. Fill in the header and rate the gravel. Complete the header on the form (section E.8), then rate each material property in section E.3, with thickness measured in wheel-track holes where more accuracy is needed (COTO, 2016k). Keep: thickness measurements.
7. Rate each surface distress by degree and extent. Judge the degree against the Part E description, then the extent over the segment, with Part A's percentages as a guide (COTO, 2016b). Keep: the ratings and any localised degree-5 defect.
8. Mark the form and scrape with a geological pick. Mark a dot per observation while driving and settle the average at the end (COTO, 2016b). Scrape corrugations and loose material with a geological pick (COTO, 2016k).
9. Rate the functional items and mark the cause. Give each functional item one degree and mark causes of poor ratings on the problem row. Rate roughness from the comfortable and safe driving speed, setting aside geometry and road width (COTO, 2016k). Keep: cause marks.
10. Record the overall condition and other problems. Enter the overall rating last: Part E says it is not used in data processing but checks the other ratings, and asks for other maintenance items in the comments (COTO, 2016k). Keep: the comments.
11. Photograph, check and file the record. Take at least one photograph per segment, placed by GPS (global positioning system). Part E does not ask for one; it lets a reviewer check. Read back every entry and file the form with evidence. Keep: form, photographs, check date.

How do degree and extent work on a gravel segment?

Degree rates how severe a distress is and extent how widespread it is along the segment (COTO, 2016b). Part E applies Part A's definitions (sections A.2.2 and A.2.3) to every surface-distress item, illustrates degrees 1, 3 and 5, and uses 2 and 4 where necessary (COTO, 2016k). The insight on keeping degree, extent and evidence together explains why the two stay separate.

Where two combinations occur on one item, Part A's generic examples record the highest product of degree and extent and treat an isolated degree-5 defect as localised distress needing routine attention. Part A warns that even experienced assessors tend to overestimate extent, more so as severity rises, and gives percentages of segment length as a guide only (COTO, 2016b).

What does a rated gravel segment look like?

Synthetic example: every value is invented. Synthetic Road G1 starts at a municipal boundary at km 3.60, so its first segment runs to km 10.00 (6,400 m), because Part A section A.3 absorbs a short length at the start of a road into the next segment (COTO, 2016b). Trip-meter notes show potholes in 24 stretches of about 20 m: 24 × 20 m = 480 m, which is 480 / 6,400 = 7.5 % of the segment. Check value: 480 m is well under 6,400 m, so a low extent is plausible, though Part A's percentages are only a guide.

Item	Synthetic observation	Synthetic rating
Potholes	24 stretches of about 20 m in parts of the segment; two deep potholes	degree 3, extent 2; deep pair in comments
Corrugations	900 m on the two steepest descents	degree 3, extent 3
Roughness	Comfortable speed drops over the corrugations, grade aside	degree 3

Real ratings come from Part E Tables E.2 to E.20 and Part A Tables A.1 and A.2, not reprinted here. Potholing is recorded as degree 3, extent 2 rather than degree 5, extent 1 for the deep pair, because Part A uses the higher product. Roughness should agree with the rows above: Part E lists corrugations and potholes among the factors influencing roughness (COTO, 2016k), and Draft TMH 22 (the road asset management manual, March 2013, section D.5.11) describes checking ratings for conflicts (COTO, 2013); a roughness of 1 here would need rechecking.

How do Part E ratings reach TMH 22 and the TMH 18 file?

Part A calls Parts A to E companion manuals to TMH 22 (COTO, 2016b). Draft TMH 22 names the unpaved condition index CIUNPAVED in its section E.2.2 and lists in Table E-8 the items it draws on, which largely match the Part E form (COTO, 2013). It says the deduct point method, which scores each degree and extent pair, calculates the index, but leaves Table E-8's comment and weighting columns empty and works only surfaced-road examples in Appendix J-5, so the receiving system's documented method decides the index.

The ratings also travel in TMH 18 road-data exchange files. TMH 18 (Draft Version 4, October 2014) section 5.5 lists an unsurfaced visual condition summary with the extension .vcu and refers back to Part E (COTO, 2014); TMH 18 (Committee Draft 5.0, February 2016) lists it under the extension vgi (COTO, 2016i). Draft Version 4 codes surface-distress degree and extent 1 to 5, whereas Committee Draft 5.0 codes 0 to 5, matching degree 0 (no distress visible) in Part A, and adds a sand road type. Agree the version and the coding of no distress with the receiver.

What records should you keep from each gravel segment?

Keep, for each segment:

• the completed form, with the manual edition and form version;
• segment identifier, chainage, direction, date, assessor, moisture condition, stop positions and thickness measurements;
• GPS-placed photographs named by road, segment and chainage;
• notes of localised degree-5 defects and anything that limited the view;
• the calibration notes and quality-check result (COTO, 2016b), and the TMH 18 version of any exchange file.

What does Part E leave undecided, and where do readers go wrong?

On treatment, TRH 20 notes that some low-volume roads described as poor by the visual assessment method can still be safe and accessible, and asks for traffic, geometry, climate and purpose to be weighed (Department of Transport, 2009). The purposes also differ: Part A says a TMH 9 assessment does not aim to find uniform sections of distress (COTO, 2016b), while TRH 20's road and condition survey does (Department of Transport, 2009), so a Part E segment is not a treatment section. Part E rates thickness in categories, not volumes; the insight on unpaved road upgrading evidence treats volumes separately. Part E's front matter says committee drafts have no legal standing, so the client's terms of reference decide the edition and form.

Common misreadings:

• Treating TMH 12 as current. Part E states that it replaces the earlier Draft TMH 12, which is effectively withdrawn (COTO, 2016k). TRH 20 (2009) still points to that draft, dated 2000 (Department of Transport, 2009). TMH 12 is a different manual from TRH 12, on flexible pavement rehabilitation (Department of Transport, 1997).
• Treating ReCAP as TMH 9. The ReCAP (Research for Community Access Partnership) raters' guideline Part E (Final, September 2020), UK Aid-funded and written for low-volume rural roads, draws on the COTO draft TMH 9 (2016) but is not a COTO document; it has its own procedure and weighting table for a visual condition index (ReCAP, 2020b).
• Reading a gravel quantity of none as a failed road. Part E rates gravel quantity as none even where the exposed subgrade appears to perform adequately (COTO, 2016k), so the rating records missing gravel, not failure.

The procedure, step by step

1. Confirm the edition, the form and your accreditation

   Agree with the client which edition and form apply, and check that your accreditation covers unpaved roads, since Part A (section A.5.3) accredits assessors per pavement type. Keep: edition, form version and accreditation.

2. Fix the segments before driving out

   Take start and end chainage (distance along the road) from the register. Part A (section A.3) sets the standard unpaved rural length and the short-length rule at either end, and recommends that the authority predefines segments. Keep: identifiers, chainage, direction.

3. Calibrate the team

   Hold the calibration session that Part A (section A.5.4) places at the start of internal quality control. Rating a shared trial length against Part E is a useful addition Part A does not prescribe. Keep: calibration notes and trial ratings.

4. Assess in dry conditions

   Prefer the dry season, as Part E does. Part E's season and speed ceiling differ from Part A's (section A.5.1), so record which you followed. Keep: moisture condition (wet, moist or dry).

5. Stop on each segment and rate the trafficked portions

   Stop at least once per segment for a closer look at the gravel. Rate the trafficked portions, not the shoulders or the windrows (ridges of loose gravel) a grader leaves. Keep: stop positions.

6. Fill in the header and rate the gravel

   Complete the header on the form (section E.8), then rate each material property in section E.3, with thickness measured in wheel-track holes where more accuracy is needed. Keep: thickness measurements.

7. Rate each surface distress by degree and extent

   Judge the degree against the Part E description, then the extent as how widespread it is over the segment, with Part A's percentages as a guide. Keep: the ratings and any localised degree-5 defect.

8. Mark the form and scrape with a geological pick

   Mark a dot per observation while driving and settle the average at the end. Scrape corrugations and loose material with a geological pick.

9. Rate the functional items and mark the cause

   Give one degree each to roughness, trafficability, safety and both drainage items, and mark causes of poor ratings on the problem row. Rate roughness from the comfortable and safe driving speed, setting aside geometry and road width. Keep: cause marks.

10. Record the overall condition and other problems

    Enter the overall rating last: Part E says it is not used in data processing but checks the other ratings, and asks for other maintenance items in the comments. Keep: the comments.

11. Photograph, check and file the record

    Take at least one photograph per segment, placed by GPS (global positioning system). Part E does not ask for one; it lets a reviewer check. Read back every entry and file the form with evidence. Keep: form, photographs, check date.

Frequently asked

Is TMH 12 still used to rate gravel roads in South Africa?

Not as the current method. TMH 9 Part E (Committee Draft Final, May 2016), the unpaved-roads part of the South African Technical Methods for Highways visual assessment manual, states that it replaces the earlier Draft TMH 12, which is effectively withdrawn. TRH 20 (Technical Recommendations for Highways, Draft Version 1.5, April 2009) still points to the draft because it predates Part E. TMH 12 is a different manual from TRH 12, which covers flexible pavement rehabilitation.

Can the paved-road distress list be used to rate a gravel road?

No. TMH 9 Part B, the flexible-pavements part of the South African visual assessment manual for road pavements, lists surfacing and structural items such as cracking, bleeding, pumping and patching, and none of them is a distress row on the gravel form. Potholes, rutting, roughness and drainage appear in both parts, but each part describes its degrees differently. Use Part E for earth and gravel roads and Part A for the shared definitions of degree and extent.

What is the difference between degree and extent in a Part E rating?

In TMH 9, the South African visual assessment manual for road pavements, degree measures how severe a distress is, and extent measures how widespread it is along the segment. Part A (sections A.2.2 and A.2.3) defines both, and Part E applies them to potholes, corrugations, rutting, loose material, stoniness and erosion. Functional items such as roughness, trafficability, safety and drainage carry one degree for the whole segment, with a problem row for the cause.

Which condition index uses the ratings from a gravel road?

Draft TMH 22 (March 2013), the South African road asset management manual, names the unpaved condition index CIUNPAVED and says the deduct point method, which scores each degree and extent pair, calculates it. TMH 9 Part E contains no index calculation. The comment and weighting columns of Table E-8 are empty in that draft, so the receiving system's documented method decides the index. Keep the ratings as observed, with the manual edition and form used.

Who may rate a gravel road under TMH 9?

TMH 9 Part A (section A.5.3), in the South African visual assessment manual for road pavements, describes accreditation of visual assessors for each pavement type and says only assessors with a valid accreditation are allowed to carry out visual assessments. An assessor accredited for another pavement type should confirm that the accreditation covers unpaved roads. The client's terms of reference decide how this applies to a particular survey.

Does a poor rating mean a gravel road must be regravelled?

No. A TMH 9 Part E rating describes condition on a segment and does not select a treatment. TRH 20 (Draft Version 1.5, April 2009), the South African Technical Recommendations for Highways manual on unsealed roads, notes that some low-volume roads described as poor can still be safe and accessible, and asks for traffic, geometry, climate, the purpose of the road and the distance travelled to be weighed before a measure is chosen.

Where this connects

TMH 22 turns ratings into indices, valuation and plans; read its Part D (usage and condition data) and Part E (indices) (COTO, 2013). TRH 20 describes the design, construction and maintenance of unsealed roads; its chapter 7 sets out how an investigation leads to maintenance measures (Department of Transport, 2009). For the paved counterpart, see the walkthrough for rating a flexible road pavement; for consistency checks, the map of survey quality control routes; for district RRAMS (rural roads asset management system) data, the journey from register to first asset management plan.

Last reviewed 21 October 2026 against the editions named above.

References

1. COTO (2016k). TMH 9: Manual for Visual Assessment of Road Pavements. Part E: Unpaved Roads. Committee Draft Final, May 2016. Original source.Sections, tables and clauses are cited inline in each guide
2. Department of Transport (2009). Draft TRH 20: Unsealed Roads: Design, Construction and Maintenance. Draft Version 1.5, April 2009. Original source.Sections, tables and clauses are cited inline in each guide
3. COTO (2016j). TMH 9: Manual for Visual Assessment of Road Pavements. Part B: Flexible Pavements. Committee Draft Final, May 2016. Original source.Sections, tables and clauses are cited inline in each guide
4. COTO (2016b). TMH9: Manual for Visual Assessment of Road Pavements. Part A: General. Committee Draft Final, May 2016. Original source.Section A.2: degree and extent
5. COTO (2013). Draft TMH 22: Road Asset Management Manual. Draft, March 2013. 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. Department of Transport (1997). TRH12: Flexible Pavement Rehabilitation Investigation and Design. Draft TRH12, Pretoria: Department of Transport for COLTO. Original source.Chapter 1: rehabilitation design process; investigation, evaluation, analysis and economic appraisal
9. ReCAP (2020b). Raters' Guideline for Visual Assessment of Low Volume Rural Road Pavements. Part E: Unpaved Roads. Final, September 2020 (GEN 2132A); M Pinard and R Geddes, Civil Design Solutions; London: ReCAP for DFID (UK Aid). Original source.Preface and acknowledgements (relationship to the COTO draft TMH 9); section E.2.1 (assessment procedure); section E.7 and Table E.7-1 (Visual Condition Index weightings)

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