ICAO Doc 9157 Part 4: Aerodrome Design Manual, Visual Aids

Doc 9157 Part 4 does not allow or forbid drones to check PAPI angles, because it is guidance. It does describe the method. Paragraphs 8.3.43 to 8.3.46 say an unmanned aircraft system “may be used as an alternative method for measurement of PAPI settings”, with a validated camera and possibly special authorization from the State.

Doc 9157 is ICAO’s Aerodrome Design Manual, published in six parts, and Part 4 is the one on visual aids. The fifth edition dates from 2021 and has 19 chapters and 6 appendices. Readers usually come to it for one of three things. They want the PAPI verification methods in Chapter 8, the lighting maintenance guidance in Chapter 18, or the intensity measurement methods in Chapter 19. This page says what those chapters contain and what status they carry. It does not repeat the six-part overview on the Aerodrome Design Manual glossary page.

Guidance that sits under Annex 14

The foreword says the material is closely associated with the specifications in Annex 14, Volume I, and that the manual’s main purpose is to help States implement them. Section 1.1, the introduction to Chapter 1, says the information in that chapter is for illustrative purposes and is not necessarily meant to imply ICAO approval or endorsement of the operational practices described. The foreword itself calls the contents guidance. The manual is neither a Standard nor a Recommended Practice.

The Standards and Recommended Practices are in the Annex . Its foreword says Notes give factual information or references and do not form part of the Standards and Recommended Practices. Most of the Annex’s pointers to Part 4 sit in Notes. The Note after 5.3.1.8 covers measuring the temperature of inset lights. The Note after 5.3.5.32 gives additional guidance on PAPI light unit characteristics. Note 1 after 5.3.5.46 points to Part 4 for guidance on what to do when an existing object extends above the obstacle protection surface of the system. The Note after 10.5.1 covers maintenance criteria, the site standard and a higher main beam average intensity. A value that appears only in Part 4 is a design or maintenance suggestion.

One pointer is not a Note. Footnotes to the obstacle light tables in Annex 14 Chapter 6 say that for flashing lights the effective intensity is determined in accordance with Part 4, and CS-ADR-DSN repeats that wording in its obstacle light tables. Tables form part of the Standard or Recommended Practice they accompany, so on that one point the Annex’s own tables refer to the manual’s method.

The wording is not uniform. Most paragraphs say “should” or “may”. A few say “must” or “shall”, for example 8.3.56 on installation tolerances and 8.3.60 on action when an object penetrates the obstacle protection surface. Check those against Annex 14 Chapter 5 and the EU specification before quoting them as requirements. The ICAO Annex 14 page and the EASA CS-ADR-DSN page set out those two texts.

What Part 4 covers

The chapters group into ground aids and markings, light characteristics, the visual slope guidance systems, runway and taxiway systems, obstacles and frangibility, and maintenance and measurement. The titles below follow the table of contents of the copy read.

This page reads Chapters 8, 18 and 19 in detail, because they deal with checking, maintaining and measuring lights already installed.

Checking a PAPI: Chapter 8

Section 8.3 describes the PAPI as a four-unit wing bar, with the unit nearest the runway set highest and a normal difference of 20 minutes of arc between the setting angles (8.3.5). Table 8-1 gives the settings for a typical 3 degree approach: 3°30′ for unit D, 3°10′ for C, 2°50′ for B and 2°30′ for A. The manual says a satisfactory transition from white to red is not more than 3 minutes of arc deep at azimuths up to 8 degrees either side of the beam centre, and not more than 5 minutes at 15 degrees (8.3.10).

The foreword lists new fifth edition material on PAPI testing criteria, single and doublet lens designs and the inspection and verification methods that follow. It names the clinometer, the ground check, the image analysis method and automatic monitoring. The table sets out each method as the manual describes it.

The manual gives the PAPI a maintenance routine as well. A daily check confirms that all lamps are lit and illuminated evenly, that there is no damage, that the change from red to white is coincident across the elements of a unit and that the lenses are clean (8.3.48). Monthly checks look for vehicle damage, rodents, water and insects, and realignment may be done then (8.3.49). Quarterly checks clean the reflector, lenses and filters (8.3.50). Paragraph 8.3.47 says this is a recommended routine that may be altered to suit local conditions. The PAPI inspection service page explains what a measured transition angle looks like in a protocol.

Unmanned aircraft in 8.3.43 to 8.3.46

The four paragraphs sit under the heading “Use of an unmanned aircraft system (UAS)”. Paragraph 8.3.43 says a UAS “may be used as an alternative method for measurement of PAPI settings”. It lists the parts of a typical system as the aircraft, a control or remote pilot station, a data link, possibly launch and recovery equipment, and a ground processing unit. It says an RTK base station is necessary to obtain high dimensional precision, and that the data can be viewed at the site and recorded.

Paragraph 8.3.44 describes a typical operation. The aircraft is at least 300 m from the PAPI. Vertical scanning gives the heights h1 and h2, the upper and lower limits of the red to white transition zone. The unit setting angle is then the arctangent of (h1 plus h2) divided by 2d, where d is the horizontal distance of the aircraft from the PAPI. Figure 8-24 draws the geometry. Paragraph 8.3.45 uses horizontal scanning to find the lateral limits of the system by noting where the signals from units D and A disappear.

Paragraph 8.3.46 sets the conditions. The method needs a visible spectrum camera or sensor that is “operationally validated against the perceived total image quality”, and it “may require special authorization from the competent State’s authority that has jurisdiction”.

In the text read, unmanned aircraft appear nowhere else. A search of the extracted text finds the terms only in the abbreviations list and in these four paragraphs, so the manual describes no unmanned method for runway, approach or taxiway lights. In the copy read, the pages that carry these paragraphs are marked 12/7/21, No. 1, which is the date and number of Amendment No. 1. A copy of the 2021 text without that amendment may not contain them.

Two other ICAO papers show how the paragraphs are read. A vendor paper to the 2023 SAM Implementation Group cites paragraph 8.3.43 for the point that drones are an efficient and less expensive option, and presents the company’s own drone method. A Kazakhstan information paper to the 2025 Assembly says existing ICAO material was developed with manned operations in mind and lacks procedures, performance criteria and safety parameters specific to unmanned systems. It asks ICAO to consider whether Annex 14, Doc 9137 Part 8 and Doc 9157 Part 4 need updating. Both are one party’s view, not ICAO text. The regulatory acceptance guide covers how authorities have handled drone methods.

Maintenance and measurement: Chapters 18 and 19

Chapter 18 gives guidance on Annex 14 paragraph 10.5, the preventive maintenance system for approach and runway lighting that supports Category II and III operations (18.1.2). It repeats the Annex Recommendations on in-field measurement in 18.3.8 and 18.3.9, then adds reasoning and practice.

• Why output matters. In low visibility a 50 per cent drop in light output reduces the range of a lighting aid by approximately 10 per cent, and in-pavement fittings often lose far more than 50 per cent without a good maintenance regime (18.3.3).
• What inspection cannot see. Visual inspection and maintenance schedules alone at best find lamp failures, gross misalignments and structural damage (18.4.2).
• How often. The measurement interval is set for each location. It may be weekly at some aerodromes, and monthly or longer at many once the system meets the standard (18.4.3). The Annex floor of twice a year for in-pavement lights and once a year for other lights on a CAT II or III runway is Recommendation 10.5.6, and the EU AMC repeats it.
• What the equipment records. Test equipment should measure and record the isocandela diagram, alignment and colour of each light with the lighting at 100 per cent (18.4.5). A vehicle speed above 50 km/h is generally found acceptable (18.4.4).
• Inset and elevated lights fail differently. Inset output is generally deficient because of contamination and optical faults, while elevated lighting is primarily affected by misalignment of the whole fitting (18.4.13).
• Approach lights and PAPI. Approach lighting outputs are more difficult to monitor, and an airborne inspection method using cameras and image processing offers one potential means (18.4.15). The paragraph names no platform. Paragraph 18.5.6 says regular visual flight inspections should be carried out where photometric measurement of approach lighting is not practicable, until a practicable means is found.

Chapter 19 covers how light intensity is measured. Paragraph 19.2.18 explains the site standard. All specified isocandela values are minimums, so units that exceed them can create an imbalance between systems, and an operator should know the output of the units first procured and keep later purchases at that level. The photometric measurement guide covers the field methods, and the serviceability limits guide covers the Annex 14 and EU percentages.

How EASA and the Czech L 14 cite it

The EU texts cite Part 4 mostly in guidance material. The guidance material to ADR.OPS.C.015, points (b) and (f), says further guidance on how to conduct a flight check is in Chapter 8 of Part 4 and in Chapter 5 of Part 8 of Doc 9137 (EAR for Aerodromes, pages 488 and 489). The AMC to the same points asks for flight checks of the lighting systems at least yearly, and the CS-ADR-DSN page sets out the AMC. GM1 ADR-DSN.M.655 points to Part 4 for the calculation for siting a PAPI or APAPI on a runway with ILS or MLS (page 708), and a note to Figure M-4 does the same for harmonisation of PAPI, ILS and MLS signals (page 704).

Chapter numbers need care. GM1 ADR-DSN.N.775 points to Part 4, Chapter 11 for signs (page 760). In the fifth edition Chapter 11 is on autonomous runway incursion warning and signs are Chapter 12. Confirm a chapter against the edition in hand.

The Czech L 14, effective 25 December 2025, lists all six parts of Doc 9157 among the manuals related to its specifications. Note 4 to the summary notes of its Appendix 2 sends the reader to Part 4 for the maintenance criteria for aeronautical ground lights and the site standard. Annex 14 carries the same pointer in its Note after 10.5.1 and in note 4 to its Appendix 2.

Where a drone survey fits

TarmacView’s PAPI inspection reads each unit’s transition angle from RTK-positioned drone frames of the colour change. That is the kind of measurement 8.3.43 to 8.3.46 describe. Two of their conditions are not something a provider can settle alone. The camera has to be operationally validated, and the State may require special authorization. Whether the method serves as an aerodrome’s periodic check is decided in its maintenance programme with its authority. The manual does not approve a system or set the tolerance a measurement must meet. The comparison with flight inspection sets out what each method checks, and the wider airfield lighting inspection service covers the other systems the manual treats.

What this page is based on

The text was read from a copy of the fifth edition with Amendment No. 1 and Corrigendum No. 1 hosted by the Swiss Federal Office of Civil Aviation. Short phrases are quoted and the rest is paraphrased, because ICAO holds the copyright. Amendment No. 2 of 15 September 2025 was not read, so any change it makes to Chapter 8 is not reflected here.