The Contribution of Drones to the Monitoring of Rubble-Mound

Breakwaters

Maria Henriques

, Rui Capitão

, Conceição Fortes

, Rute Lemos

, Luís Gabriel Silva

Hugo Silva

and Rúben Gonçalves

National Laboratory for Civil Engineering (LNEC), Lisbon, Portugal

APS - Ports of Sines and the Algarve Authority, Sines, Portugal

Keywords: Drone, Breakwater, Monitoring, Sines.

Abstract: Breakwaters are built to promote sheltered areas, for people, ships, and harbour activities. In the design of

rubble-mound breakwaters, a common type of breakwater in many countries, including Portugal, it is assumed

that damage may occur in certain stretches of the structures, and therefore maintenance and repair works will

be quite certainly needed. To successfully carry out these interventions, in a timely and cost-effective manner,

the structures must be observed and monitored systematically. This enables one to follow their structural

behaviour and, through diagnosis analysis, to specify the most suitable timespan to undertake any necessary

intervention. The severity of the sea on the Portuguese coasts justified the establishment, by the National

Laboratory for Civil Engineering (LNEC), of a program of Systematic Observation of Maritime Works

(OSOM) which, in 2018, was improved with the introduction of drones to monitor the structural present

condition, evolution condition and risk condition of the structures, namely movements and falls of blocks in

the armour layers. This communication presents some results of the application of OSOM+ program on

breakwaters in Sines and Algarve (Faro-Olhão and Portimão) harbours, an activity that LNEC has developed

for the APS – Ports of Sines and the Algarve Authority.

1 INTRODUCTION

In 1986, the National Laboratory for Civil

Engineering (LNEC) started developing a

programme for Systematic Observation of Maritime

Works (OSOM) to monitor the behaviour of

rubble-mound breakwaters along the Portuguese west

and south coasts and recommend timely interventions

for their maintenance and/or repair. This maintenance

and/or repair works should be carried out at an early

stage following the acknowledgment of problems, at

the time these might likely affect a small area and

therefore repairs are simpler and less expensive.

The OSOM methodology is based on a series of

systematic visual observation campaigns that provide

the necessary information to feed the ANOSOM

database (Reis et al., 1995; Lemos et al., 2007),

which is meant to characterize the Present Condition,

https://orcid.org/0000-0001-8982-3967

https://orcid.org/0000-0003-3915-9951

https://orcid.org/0000-0002-5503-7527

https://orcid.org/0000-0003-0380-391X

the Evolution Condition, and the Risk Condition of

the observed maritime structures.

Since the 2010’s the OSOM programme has been

improved and, in 2017, when drone monitoring

started to be used in breakwaters, the programme

changed its name to OSOM (Capitão et al., 2018).

The use of drones improved the monitoring because

it provides detailed and more accurate information on

the condition of the structures, and made it possible

to quantify changes of settlements, volumes, etc..

This paper presents the expertise of LNEC in the

monitoring of rubble-mound breakwaters using

drones. Most of the situations presented here were

obtained in the breakwaters of Sines (Figure 1 and

Figure 2) and Algarve (Faro-Olhão inlet and

Portimão, Figure 3 and Figure 4), in an activity that

LNEC is providing for APS – Ports of Sines and the

Algarve Authority – since 2018 (Capitão et al., 2022).

160

Henriques, M., Capitão, R., Fortes, C., Lemos, R., Silva, L., Silva, H. and Gonçalves, R.

The Contribution of Drones to the Monitoring of Rubble-Mound Breakwaters.

DOI: 10.5220/0012691200003696

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 10th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2024), pages 160-167

ISBN: 978-989-758-694-1; ISSN: 2184-500X

Figure 1: Breakwaters of the port of Sines.

Figure 2: Breakwater of the leisure port of Sines.

Figure 3: Breakwaters of Faro-Olhão inlet.

Figure 4: Breakwaters of the port of Portimão.

2 THE BREAKWATERS

The three ports managed by APS – Sines, Faro and

Portimão – have several breakwaters, all

rubble-mound breakwaters (Figure 5 and Figure 6,

photos taken by one of LNEC's drones). Sines has a

total of six breakwaters (one of these, the less

exposed, is not monitored); Faro-Olhão inlet and

Portimão harbour have two breakwaters each. All the

breakwaters of Sines were constructed to protect

several infrastructures that exist in the area: five

terminals (liquid bulk, petrochemical, dry bulk,

liquified natural gas and containers), a logistic

activity zone, the fishing harbour, and the leisure port.

In Algarve, on the Faro-Olhão area, the construction

of the breakwaters in the 1930’s created a channel that

allowed the development of several ports, while in

Portimão the breakwaters were built at the mouth of

a river to ensure protection at its entrance.

These rubble-mound breakwaters consist of a core

of finer material covered by large blocks forming the

so-called armour layer. The blocks are of rock being

that in areas more exposed the armour layer blocks

are in made concrete in several shapes (tetrapod,

Antifer, or parallelepiped). The superstructures of the

breakwaters are also made in concrete. Some

breakwaters have public access, while others are in

(very) restricted non-public access areas.

Figure 5: The head of Sines west breakwater.

Figure 6: The head of Faro-Olhão inlet west breakwater.

The Contribution of Drones to the Monitoring of Rubble-Mound Breakwaters

161

3 OSOM PROGRAMME

The OSOM – Systematic Observation of Maritime

Works – programme has been developed by LNEC to

help the owners of breakwaters and other maritime

structures.

The objective of this programme is to monitor the

behaviour of these structures and recommend timely

interventions for their maintenance and/or repair. The

OSOM methodology is based on a series of systematic

visual observation campaigns complemented with data

from drone photogrammetric surveys. All the data

provide the necessary information to feed the

ANOSOM database (Maia et al., 2017), which is

meant to characterize the Present Condition, the

Evolution Condition, and the Risk Condition of the

structures. Based on this information, it is then

possible to establish when, where and under what

circumstances maintenance or repair works should be

carried out.

More recently, ANOSOM-WEB interface app

was developed, a web mapping platform accessible

by any device (smartphone, tablet, or PC) equipped

with web connection. During visual observation

campaigns, this app allows the observer to carry out,

on a mobile device, various operations such as, as

examples, consultation of information from previous

campaigns (Figure 7) and accessing, in situ, the

current, evolution and risk conditions of the structure,

and whether the structure needs immediate repair or

maintenance works.

Figure 7: Interface ANOSOM-WEB – Breakwater Sines

west.

4 THE DRONES AND THE

FLIGHTS

The first surveys with the drone (2017), used a DJI

Inspire V1 and a camera Zenmuse X3; since June

2022, it has been used a DJI Matrice 300 RTK and a

camera Zenmuse H20 (

Figure 8

and

Figure 9

). Due to

the use of this drone with RTK, the security of the

landing was improved, especially with mild to strong

winds (up to 54 km/h). Both cameras have sensors

CMOS, the first one with 12.4 MP, the second with

20 MP. At an altitude of 100 m, a pixel in a

photograph taken by the X3 covers about 5 cm

, with

the H20 3.4 cm

. H20 camera has zoom capabilities.

Before 2020, the breakwaters were fully surveyed,

e.g., the flights covered an area comprising from the

head to root of the breakwaters, always avoiding

flight over pipes of gas or of liquid petrochemical

products. After 2020, due to the European legislation,

in some breakwaters the covered area was reduced to

keep flights more than 150 m away from buildings.

This rule led to some limitations on the image

acquisitions, leading that the seaside armour layer of

one of the breakwaters has been only covered with

oblique photos, with the drone flying over the sea.

Figure 10 presents the two flight plans made over this

breakwater, where the area shaded in blue is the

photographed area, and the green line represents the

flight path. The left image in the Figure 10 shows the

flight path over the head of the breakwater (nadir

photos), while the right image shows the flight path

over the sea (oblique photos).

Until now the RTK was used only with navigation

proposes; in a near future the data collected will be

integrated in the processing, to reduce the number of

ground control points (GCP) needed. Concerning the

information needed for processing the aerial images,

it is mandatory to have GCP (Figure 11) marked on

the surface of the breakwaters, and clearly visible in

photographs. The points have been coordinated with

GNSS (Henriques et al., 2014).

For the necessary flight permissions, relevant

national entities were contacted beforehand. These

include the National Aeronautical Authority and

Local Port Authorities (these are compulsory),

relevant Aerodrome and Heliport Authorities, and the

Institute for Nature and Forest Conservation,

depending on the breakwater locations.

Five days before the survey, national weather

forecast sources were checked (Henriques et al.,

2022). Rain, wind velocity and wind gust velocities

were verified, although wind gust information was

scarcely found.

GISTAM 2024 - 10th International Conference on Geographical Information Systems Theory, Applications and Management

162

Figure 8: DJI Matrice 300 RTK drone while taking off

Sines east breakwater.

Figure 9: Preparing the flight on Sines west breakwater.

Shown on top is the RTK GNSS antenna base.

Figure 10: Flight plans over the fishing harbour breakwater

at Sines. Left: head – vertical photos; Right: sea face –

oblique photos.

Figure 11: Location of GCP in the leisure port of Sines

breakwater.

The flights have the following parameters: altitude

of the flight: or 30 m or 40 m (depending on the length

of the breakwater or local conditions); overlap 80%

(both longitudinal and transversal). The flights are

always autonomous (i.e. the missions are planned and

upload to the drone before the flight) and made during

low tides to maximize the area exposed. In Table 1 it is

presented some data related to the flights in three

breakwaters in Sines (the shortest, and the two

longest). The pixel size of the orthomosaics produced

is, in the three cases, 1.4 cm.

Table 1: Data concerning three drone surveys at Sines.

Length

Width

N.º

fli

hts

Duration

fli

hts

N.º

hotos

Leisure

250 m

30 m

1 12 min 178

East North

Sec.

1000 m

55 m

2 56 min 982

West

1500 m

65 m

3 74 min 1092

5 DATA PROCESSING

Detection of changes in a rubble-mound breakwater

armour layer is normally done by comparing the

digital surface models (DSM) from two surveys

performed at different dates.

To obtain it, following processing steps are used,

in this order, and for each breakwater: point cloud

computation, DSM creation (Figure 12). After the

orthomosaic is generated (Figure 13).

It has been used Agisoft Metashape Pro software.

Processing parameters used: i) Photo alignment

quality: high; ii) Dense cloud quality: high; iii) Depth

filtering intensity: mild.

10 m

0 m

Figure 12: DSM (matrix form) of the head of Sines east

breakwater.

The Contribution of Drones to the Monitoring of Rubble-Mound Breakwaters

163

Figure 13: Head of Ericeira breakwater. Two orthomosaic

extracts with five years of difference are shown.

For presentation purposes, in reports, DSM are

sometimes generated in the form of a mesh (Figure 14

and Figure 15).

Figure 14: Mesh of the head of Sines east breakwater.

Figure 15: Mesh of the head of Sines west breakwater.

It was determined the planimetric and altimetric

RMSE (Root Mean Squared Error, see Table 2) for

some breakwaters using the equations presented by

FGDC (1998). Here it is presented the values obtained

in Sines east north section breakwater, in 2020. The

data used is the horizontal distance between a point

coordinated and its image in the orthomosaic (

Figure

), and the vertical distance between this point and

the point cloud. In the table are included the results

obtained for the check points (points materialized like

the GCP, used to assess the accuracy of procedures).

Figure 16: Point coordinated (in yellow), its image in the

orthomosaic and vector between both (its length is the

planimetric distance used).

Table 2: RMSE Sines east north section breakwater.

Planimetric Altimetric N.ºPoints

Check Points 2.3 cm 2.5 cm 36

Ground Control Points 1.9 cm 2.0 cm 20

6 ANALYSES OF DATA

Since 2018, only few movements of blocks were

detected and even those were non-significant. Some

cases are presented here.

As stated before, geometric changes in the armour

layer were detected through the analysis of the DSM:

by computing the difference between two DSM

outputs one can notice if there are changes. QGIS

software is used for this task. In reports, the

presentation of significant changes is accompanied by

extracts from orthomosaics to present the evolution.

This is the case of Figure 17 to

Figure 20

, that show

extracts of orthomosaics and DSM that illustrate

some of the detected changes. Differences between

DSM are illustrated in colour to clearly identify

relevant evolution.

Figure 17: A block became more eroded in a period of two

years.

Figure 18: A block that broke.

GISTAM 2024 - 10th International Conference on Geographical Information Systems Theory, Applications and Management

164

Figure 19: A tetrapod that rotated.

Figure 20: Displacements of blocks.

Figure 21

presents the comparison of DSM of the

head of

Faro-Olhão west breakwater (2023-2018) and

Figure 22 presents the two meshes produced by Metashape

where one can easily see

the changes in the position of

the blocks and verify that the concrete structure of the

crest is less protected.

Figure 21: Block movements in the Faro-Olhão west

breakwater between 2018 and 2023 detected by DSM

comparison.

2018

2023

Figure 22: Meshes of Faro-Olhão west breakwater show the

movements of the blocks.

Figure 23 shows the aerial photo and the orthomosaic

at the Faro-Olhão inlet west breakwater. There, it can

be seen the quite apparent (and, lately, quantified)

erosion of sand in the northern area, on the land side,

leaving the foundation of the breakwater exposed.

←

north south

→

Figure 23: Faro-Olhão inlet west breakwater (aerial photo

and orthomosaic). Indication of the area analysed situated

on the north side. The red arrow shows the beginning and

direction of the X axis of the graph of Figure 25.

Figure 24 shows DSM outputs from different dates.

Figure 25

shows the profile drawn at the deepest

location. The direction of the X axis is presented in

Figure 23

The Contribution of Drones to the Monitoring of Rubble-Mound Breakwaters

165

0m 3m

Figure 24: DSM of the area analysed in three years and

colour scale.

Examination of the historical images available on

Google Earth revealed that, in 2006, this problem

already existed, although in a much smaller extent. As

only planimetric information exists, the only values

that can be obtained are distances. It was found that

in 2006 the largest depression had an opening of 6 m.

This area increased until it was clear, from the 2013

image, that, the depression stabilized. In 2023, the

opening was 22 m, according to the DSM, as can be

seen in

Figure 25

Figure 25: Profiles, starting from de breakwater, in the

direction of land calculated in the lowest area.

The analyses presented in this paper were based

on products fully generated by LNEC. There is other

information, collected in previous years, which exists

in the archives of APS or in the National Geographic

Information System (orthophotos and DSM) whose

analysis has been included in reports present to APS.

7 CONCLUSIONS

The paper presents the contribution of the aerial

photogrammetric surveys made with the help of

drones to the monitoring of rubble-mound

breakwaters. The information generated from the

photographs, especially DSM and orthomosaics,

contribute to better quantify evolution of observed

structures under OSOM - Systematic Observation

of Maritime Works programme.

OSOM , which initially was primarily based on

qualitative analyses, can now also include additional

quantitative analysis, and detect very small changes,

in the order of a few centimetres, in the structures,

considering the currently attainable quality of the

surveys and, consequently, of the generated products.

This paper focused on the use of surveys to

monitor already existing maritime structures, but the

photo surveys can also have other uses. For example,

by evaluating whether the geometry of the structure

matches its design (Henriques, 2016 and Henriques et

al., 2017), which is especially useful during

construction or repairs of the structure.

Between 2018 and 2023, the changes in the

breakwaters of APS were small. The need to quantify

these changes in the form of volumes, something

simple to do when there are point clouds, was not

considered important. For this reason, this data has

not been included in the reports prepared for the APS

although it was computed to be included in papers

(Henriques, 2016 and Henriques et al., 2016).

In the processing phase, to include drone

coordinates obtained by RTK will reduce the number

of GCP needed. This is an important step to be taken

very soon. Regarding the improvement in the quality

of results that would result from using more complex

flight plans, such as those in the form of grids, or

complementary flights with a camera in an oblique

direction, it is necessary to point out that the time

available for carrying out flights is reduced because

these must be done during low tides. The

improvement in quality of the results is not so

relevant as to justify these procedures.

Concerning the detection of the movements of the

blocks, there are interesting approaches. More complex

analysis of the data produced (like the ones presented

by Soares et al., 2017 and 2022, or Arza-García et al.,

2024) are still academic advances which should evolve

into procedures to be applied in production work (such

as the one presented in this paper). As pointed out,

there were no damages in APS breakwaters during the

period 2018-2023 so there was not the need to more

complete analysis of the breakwaters, as the one

presented by Florio et all. (2024). There is an area in

which research is really needed which is the detection

of the movements of submerged blocks. This research

may need to include the development of equipment.

For instance, Sakamoto et al. (2024) present the

results and limitations of the use of a green laser

scanner in shallow waters.

2020

2022

2023

GISTAM 2024 - 10th International Conference on Geographical Information Systems Theory, Applications and Management

166

ACKNOWLEDGEMENTS

The work has received funding through LNEC

research projects “Digital images for support of civil

engineering studies – DICE” and “Risk and Safety in

Ports and Maritime Structures”.

REFERENCES

Arza-García, M., Gonçalves, J.A., Ferreira Pinto, V.,

Bastos, G. (2024). On-Site Stability Assessment of

Rubble Mound Breakwaters Using Unmanned Aerial

Vehicle-Based Photogrammetry and Random Sample

Consensus. Remote Sensing 2024, 16, 331.

https://doi.org/10.3390/rs16020331

Capitão, R., Lemos, R., Henriques, M.J., Fortes, C.J.E.M.,

Neves, M.G., Silva, L.G. & Reis, M.T. (2018).

Systematic observation of maritime works. The new

OSOM+. In Proc. 5as Jornadas de Engenharia

Hidrográfica, Instituto Hidrográfico.

Capitão, R., Fortes, C.J.E.M., Lemos, R., Silva, L.G.,

Neves, M.G., Henriques, M.J., Martins, T. (2022).

OSOM+: Aplicação às estruturas marítimo-portuárias

do porto de Sines. In 10

Jornadas de Engenharia

Costeira e Portuária, PIANC.

FGFC (1998). Geospatial Positioning Accuracy Standards.

Part 3: National Standard for Spatial Data Accuracy.

Federal Geographic Data Committee,

FGDC−STD−007.3−1998.

Florio, A., Di Luccio, D., Robustelli, U., Cutugno, M.,

Pugliano, G., Benassai, G. (2024). UAV close-range

photogrammetry for breakwater damage assessment in

a Tyrrhenian coastal site. Coastal Engineering Journal.

https://doi.org/10.1080/21664250.2024.2308388

Henriques, M.J., Fonseca, A., Roque, D., Lima, J.N. &

Marnoto, J. (2014). Assessing the quality of a UAV-

based orthomosaic and surface model of a breakwater.

In FIG Congress 2014, International Federation of

Surveyors.

Henriques, M.J. (2016). Surveillance par drone des digues

à talus. Revue XYZ n. 148, Association francophone de

topographie.

Henriques, M.J., Roque, D., Santos, A.V. (2016).

Monitorização de quebra-mares com veículos aéreos

não tripulados. In I Seminário Internacional UAV,

Universidade Nova de Lisboa.

Henriques, M.J., Lemos, R., Capitão, R., Fortes, C.J.E.M.

(2017). The monitoring of rubble-mound breakwaters.

An assessment of UAV technology. In INGEO2017, the

International Conference on Engineering Surveying,

Slovak University of Technology & LNEC.

Henriques, M. J., Roque, D. (2022). Planning UAV

surveys: can we rely on wind forecasts? In 5

Joint

International Symposium on Deformation Monitoring

(JISDM 2022), Universitat Politècnica de València.

Lemos, R., Santos, J.A. (2007). ANOSOM. Análise da

observação sistemática de obras marítimas. In 5

Jornadas Portuguesas de Engenharia Costeira e

Portuária, PIANC.

Maia, A., Rodrigues, A, Lemos, R., Capitão, R., Fortes,

C.J.E.M. (2017). A Web platform for the systematic

monitoring of coastal structures. In GISTAM 2017,

SCITEPRESS.

Reis, M.T., Silva, L.G (1995). Systematic Observation of

Maritime Works. ANOSOM Database: User’s Manual.

Report NPP, LNEC.

Sakamoto, N. and Nishiyama, S. (2024). 3D monitoring of

coastal erosion control structures using UAV. The

International Archives of the Photogrammetry, Remote

Sensing and Spatial Information Sciences, Vol.

XLVIII-4/W9-2024, https://doi.org/10.5194/isprs-

archives-XLVIII-4-W9-2024-313-2024

Soares, F., Barbon, V. (2023). Image segmentation of

breakwater blocks by edge-base Hough transformation.

Journal of Applied Geodesy, 17(2), 131-137.

https://doi.org/10.1515/jag-2022-0044

Soares, F., Henriques, M.J., Rocha, C. (2017). Concrete

Block Tracking in Breakwater Models. In FIG Working

Week 2017.

The Contribution of Drones to the Monitoring of Rubble-Mound Breakwaters

167