EE3306/ME3163 Group Report

Coral Reef Restoration

Poobalan Aatmika Lakshmi (A0238103R)
Koh Hui Yi, Phylicia (A0238871U)
Marcus Lim Sheng Jie (A0216307M)
Goh Yi Xuan (A0238652Y)

1
1 Abstract .........................................................................................................................................4
2 Introduction ...................................................................................................................................4
3 Current methods and their limitations ..........................................................................................4
3.1 Reef monitoring......................................................................................................................4
3.1.1 On-site surveys ................................................................................................................4
3.2 Reef restoration ......................................................................................................................5
3.2.1 Direct transplantation and Coral gardening .....................................................................5
3.2.2 Substrate Addition ...........................................................................................................5
3.2.3 Larval propagation ...........................................................................................................5
4 Solution .........................................................................................................................................6
5 Solution 1: Enhancing Coral Health and Reproduction Through IoT-Based Reef Monitoring for
Aquaculture Solutions ......................................................................................................................6
5.1 Overview ................................................................................................................................6
5.2 Motivations.............................................................................................................................6
5.3 IoT network ............................................................................................................................7
5.4 Data usage in aquaculture: ...................................................................................................10
5.6 Application of the data: Monitoring, forecasting .................................................................13
5.7 Enhanced proactive warning systems ..................................................................................14
5.8 Strategic acclimatisation for improved coral transplantation ...............................................14
6 Solution 2: Coral Restoration using 3D Printing Technology ....................................................14
7 Coral Model Design and Preparation Process ............................................................................15
7.1 Coral Designing – Photogrammetry .....................................................................................15
7.1.1 Structure from Motion ...................................................................................................15
7.1.2 Multi View Stereo .........................................................................................................15
7.2 Coral Modelling – Printing Design Method.........................................................................16
7.2.1 Gravity-Stimulated Printing Design Method .................................................................16
7.3 Design Customisation – Artificial Coral Substrates and Reef Structure .............................17
7.3.1 Larvae Settlement ..........................................................................................................17
7.4 Digital Twin – Design Optimisation ....................................................................................17
8 Coral Printing and Manufacturing ..............................................................................................18
8.1 Parallel Printing ....................................................................................................................20
8.2 Multi-axis Printing ...............................................................................................................20
8.3 Software Automation ...........................................................................................................20
2
 8.4 Remote Monitoring ..............................................................................................................20
9 Technical Feasibility and Business Viability ..............................................................................21
9.1 IoT feasibility and viability ..................................................................................................21
9.2 Additive manufacturing feasibility and viability .................................................................21
10 Conclusion ................................................................................................................................22
Bibliography ..................................................................................................................................23

3
1 Abstract
Coral reefs, critical for marine biodiversity and ecosystem services, face unprecedented threats,
with over half already lost or damaged. Conventional restoration methods suffer from limitations,
prompting the integration of Industry 4.0 technologies. This paper proposes a two-pronged
approach employing Internet of Things (IoT) and 3D Printing. The IoT system involves real-time
monitoring via wireless sensor networks, addressing challenges of cost and obtaining
comprehensive data. Additionally, advanced imaging and 3D Printing techniques, ensure
scalability, cost-effectiveness, and precision in manufacturing coral structures. The combination
of these approaches addresses the urgency of coral reef restoring by bringing transformative leap
in coral restoration practices.
2 Introduction
Coral reefs, comprise colossal living structures formed by countless tiny polyps and play a
vital role in marine biodiversity, coastal protection, fisheries, tourism, and scientific research [1].
These intricate ecosystems support a myriad of life forms and have played a pioneering role in
drug discovery [1].
Despite their significance, these complex ecosystems are under severe threat, with over
half of the world’s coral reefs already lost or damaged due to various stressors, including
overfishing, nutrient pollution, ocean acidification, and climate change impacts such as ocean
warming [2]. Without prompt intervention, it is estimated that coral reefs could entirely vanish by
2070 [3].
The urgent need for preservation and restoration requires addressing immediate challenges,
including sustainable management practices and leveraging technology like IoT for real-time
monitoring and swift response to changes. By combining sustainable practices with technological
advancements, we can strive to ensure the prolonged survival of these critical ecosystems.
3 Current methods and their limitations
Over the past few years, international efforts to monitor, assist recovery of damaged reefs
and allow for continued growth have been intensifying. However, there are multiple challenges
that need to be overcome.
3.1 Reef monitoring
3.1.1 On-site surveys
Traditional on-site surveys for monitoring coral reefs are expensive and rely on human
observers, introducing challenges such as limited objectivity, repeatability, and endurance [4].
Additionally, the vast diversity, size and remote locations of coral reefs make routine monitoring
resource-intensive, restricting the frequency of surveys. The reliance on scuba diving further adds
constraints, resulting in scattered data or limited coverage. While underwater digital photographs
aim to maximise data, their analysis demands hiring experienced biologists, causing delayed and
inefficiencies in information flow to conservation practitioners [5]. To address these issues, a more
cost-effective and efficient monitoring approach is essential.
4
3.2 Reef restoration
3.2.1 Direct transplantation and Coral gardening
Both direct transplantation and coral gardening involve relocating coral colonies or
fragments from damaged sites to new locations, with coral gardening including an intermediate
nursery phase. However, these strategies pose challenges. Transplanting coral fragments is costly,
labour-intensive, and relies on optimal environmental conditions at the new site for survival [6].
Cultured or transplanted corals may face difficulties adapting to their new surroundings,
encountering stressors like predation and water quality issues.
Additionally, coral gardening faces obstacles in promoting the growth of coral fragments
within nurseries. The materials used in these nurseries may not be environmentally friendly,
lacking resilience against damage over time. The health of corals within these nurseries is also at
risk due to concerns such as algae overgrowth and the rapid spread of diseases, particularly in
densely populated nurseries [6]. Addressing these challenges is crucial for the effectiveness of
coral restoration efforts.
3.2.2 Substrate Addition
Substrates, essential for coral recruitment, planting, and fish aggregation, represent
artificial structures designed to mimic natural processes and seamlessly integrate into reef
landscapes. Traditional methods employ concrete blocks and metal structures, lacking the ability
to replicate natural forms due to their standardised shapes. Moreover, these materials are non-
biocompatible, hindering optimal coral growth, as corals thrive on biocompatible materials like
calcium carbonate for skeleton formation [7]. Beyond their limitations in mimicking nature, these
structures are costly to design and deploy, restricting their use to small-scale projects [6].
Furthermore, their construction materials must adhere to eco-friendly standards and exhibit
resilience to degradation in oceanic environments [6].
3.2.3 Larval propagation
Larval propagation involves releasing coral larvae, either independently or as inoculated
substrates, at a restoration site following an intermediate collection and holding phase [6]. The
successful growth of coral reefs from larvae hinges on critical factors. Firstly, it necessitates
diverse sources of coral larvae to uphold genetic diversity within restored coral populations, a key
element for their resilience [6]. Secondly, the timing of the release is crucial, requiring optimal
ocean conditions for coral spawning. However, conducting larval propagation on a large scale is
costly, and there is limited evidence of long-term success [6].
Incorporating Industry 4.0 technology throughout each stage offers the prospect of
overcoming these limitations, markedly enhancing the efficiency and efficacy of coral restoration
endeavours. A blend of technologies, including Internet of Things (IoT) sensors, 3D Printing, and
Artificial Intelligence (AI), will be integrated into every phase, providing researchers with access
to more exhaustive data and insights. This facilitates quicker, more informed decision-making for
enhanced effectiveness.

5
4 Solution
We adopt a two-pronged approach to our solution, combining IoT and additive
manufacturing. IoT sensors will be strategically placed in reefs, providing crucial data for coral
acclimatisation in nurseries. This data will fuel Machine Learning models, generating valuable
insights and forecasts for enhanced understanding and proactive measures in coral restoration.
Simultaneously, in additive manufacturing, advanced imaging tools such as photogrammetry and
3D Printing will be employed to create more precise and resilient coral structures at potentially
reduced costs, thereby boosting the scalability of coral restoration efforts. Additionally, IoT
sensors will be integrated into 3D printers, simplifying maintenance and enabling rapid production.
5 Solution 1: Enhancing Coral Health and Reproduction Through IoT-Based Reef
Monitoring for Aquaculture Solutions
5.1 Overview
The advent of Internet of Things (IoT) technology has ushered in a new era of scientific
exploration and data acquisition. Within the realm of aquaculture, IoT enables the meticulous
monitoring of real-time water parameters, which is invaluable for the scientific understanding of
coral health and reproduction. Our solution introduces an IoT device that incorporates essential
water parameter sensors which will provide real-time data on the water conditions in existing reef
ecosystems. The sensor data will be leveraged on by replicating the exact conditions in a highly
controlled laboratory setting to cultivate more resilient coral species and determine the ideal water
conditions to stimulate controlled spawning events. Such aqua-cultured corals will eventually be
transplanted in the wild in the pursuit of advancing coral conservation.
5.2 Motivations

Figure 1: Current reef monitoring practices [13]

Motivations for IoT in aquaculture: Traditional data collection is prohibitively costly,
relying on human observers with limitations in objectivity, repeatability, and endurance [4].
Furthermore, the sheer diversity, size, and remote locations of coral reefs make routine monitoring
a resource-intensive task. IoT technology offers a cost-effective remedy by automating data
6
collection, elevating data quality, and lightening the financial burden associated with human
efforts, effectively addressing these cost-related challenges.
In a biological context, coral fragment transplantation is generally successful, with survival
rates ranging from 50% to 100% when fragments are transplanted into similar habitats [12]. Our
objective aligns with this success; we aim to increase the survival rates of lab-grown corals, even
when they have never encountered their natural habitats. We achieve this by replicating the
intended natural habitat within controlled tank environments, harnessing real-time IoT data on
water parameters from the original environment. This approach not only yields cost savings but
also enhances the survivability of lab-grown corals in more realistic conditions, thereby reducing
expenses when they are transplanted into the wild. Furthermore, augmenting survivability in lab
environments holds several potential benefits, including faster coral growth, the cultivation of
genetically resilient strains, and the opportunity to conduct controlled experiments for studying
coral health and adaptation to changing environmental conditions.
The high costs and the limited success in successfully reintroducing cultured corals to
natural reefs have bred uncertainty around adopting aquaculture methods for coral recovery.
Nevertheless, the pressing challenges presented by climate change, notably the widespread coral
bleaching and mortality, alongside persistent stressors, necessitate a proactive and intervention-
focused approach to coral reef conservation. This shift extends to reefs where restoration efforts
were previously deemed unfeasible [16]. To address these concerns, aquaculture techniques,
facilitated by IoT technology to replicate natural reef conditions in controlled tank environments,
promise to fortify coral resilience against climate-related stressors.
5.3 IoT network
In light of our deployment of sensors in remote areas across coral reefs, an efficient IoT
network capable of communication over vast distances becomes imperative. The data gathered by
these IoT sensors necessitates seamless transmission and storage for subsequent research. To
address this, we propose the integration of a Wireless Sensor Network (WSN) to facilitate the
communication and storage of the collected data effectively.

7
 Figure 2: Wireless Sensor Network (WSN)

A Wireless Sensor Network (WSN) comprises a base station (BS) situated above waters,
multiple sensor nodes (SN), and spatially distributed sensors, representing the underwater IoT
sensors mentioned earlier. Given the challenge of direct communication between underwater IoT
sensors and the BS, the implementation involves creating clusters. Within each cluster, an SN
functions as the cluster head, while individual sensors act as cluster members. The SN, acting as a
bridge, aggregates data from nearby sensors before uniformly transmitting the compiled data to
the BS. This hierarchical approach optimises transmission efficiency, leveraging the robust
transmission capacity and computation capability inherent in SNs [18].

Moreover, employing floating buoys positioned above the coral reefs can serve as housing
for the sensor nodes (SNs). By utilising reliable moorings, these buoys can withstand adverse
weather conditions and unforeseen situations [21]. This communication method, known as
Underwater Acoustic Communication (UAC), relies on the propagation of sound signals for data
transmission [19]. The incorporation of UAC aligns with the scalability requirement of our
solution, as WSN, with its extensive area coverage, can effectively span across the vast expanse
of coral reefs.

Furthermore, in the event of a data transmission failure from a SN to the BS, SNs, armed
with knowledge of their spatial locations (dimensions x, y, and z), can be reconfigured for data
transmission to other SNs through multi-hop forwarding, a form of communication in radio
networks [20]. This capability proves advantageous in minimising information flow delays
compared to on-site surveys for coral reef monitoring, as it mitigates the risk of data loss.

Nonetheless, several challenges must be addressed. Transmitting data underwater

consumes significant energy, with greater distances between SN and sensors resulting in increased
energy loss along the transmission path. Furthermore, acoustic signals generated by shadows and

8
marine creatures can interfere with data transmission between SN and sensors. Lastly, the necessity
to replace batteries for ocean deployed SNs can be time-consuming and environmentally wasteful.

Thus, to enhance the solution and address the challenges mentioned, the Energy Balanced
Reliable and Effective Clustering (EBREC) method [22] can be employed. EBREC is an algorithm
designed to prolong the lifespan of SNs by optimising energy consumption. It achieves this through
a consistent sleep and wake cycle, minimising energy usage within the network. As the network's
nodes are organised into clusters, the sleep and wake cycle will be automatically triggered at
suitable intervals, ensuring efficient energy management.

Figure 3: Data transmission for our IoT solution

Following data collection, the information from the sensors will be temporarily stored on
SNs before transmission to gateway nodes (BS). Subsequently, the data will be transferred from
the BS to high-level nodes, including cloud servers, using IP addresses. The data will then be
accessible to researchers and conservationists through cloud service providers, facilitating its
utilisation in the monitoring and restoration of coral reefs [23].

Figure 4: Sensors used to monitor real-time water parameters

9
 Sensor Purpose
Monitoring water temperature is crucial for understanding the thermal
Temperature Sensor conditions of the reef, as coral health is sensitive to temperature
changes [9].
Monitoring pH levels helps assess ocean acidification, which can
pH Sensor
impact the growth and health of coral reefs.
Measuring the salinity (salt concentration) of the water is essential for
Salinity Sensor
understanding the water quality and its influence on reef ecosystems.
Dissolved Oxygen These sensors measure the amount of oxygen in the water, which is
Sensor vital for the survival of reef organisms.
Photosynthetically Help assess the availability of light for photosynthetic processes in
Active Radiation corals and other reef organisms.
(PAR) Sensor
Monitoring nutrient levels, such as nitrate and phosphate, is important
Nutrient Sensor to understand nutrient cycling and potential nutrient pollution in reef
environments.
Cameras can provide visual data for reef assessment, helping to
Underwater Cameras document the condition of the coral, the presence of marine life, and
potential stressors.
These sensors can be used to monitor the soundscape of the reef,
Acoustic Sensors including the vocalisations of marine life and the impact of human
activities.
Table 1: Sensors and their purposes
Several sensors, as cited, are presently integrated into NexSens operations. The company
utilises Coral Reef Early Warning System (CREWS) buoys to effectively monitor changes in
ocean acidification, a critical threat to coral health, yielding successful outcomes [15]. It's
noteworthy, though, that this solution primarily centres on monitoring water parameters for
research purposes and pre-emptively tracking changes in ocean acidification or other variables
posing serious threats to corals. However, it does not actively address the restoration of damaged
coral reefs.
5.4 Data usage in aquaculture:
Consequently, our solution strives to emulate the real-time sensor conditions within a
controlled laboratory environment. This feasibility is exemplified by Australia's National Sea
Simulator, SeaSim, which exercises precise control over parameters such as pH, temperature, and
lighting in grow-out tanks [11]. Similarly, we propose a spacious open-plan area comprising
controlled environment rooms, enabling the fine-tuning of environmental factors like temperature,
lighting, and moonlight cycles to manipulate the natural reproductive cycles of corals.
Furthermore, the facilities can cater to the growth and nurturing of larvae and early life stages of
these organisms.
In fact, exposing corals to real-time environmental data enhances their acclimation to
specific conditions, boosting their survivability and growth in the wild. Furthermore, a nuanced
10
understanding of variations in measured water parameters allows aqua culturists to tailor
laboratory tanks, subjecting corals to controlled stress. Research findings support this approach,
with analyses of various time series indicating that corals exposed to moderate temperature
fluctuations demonstrate heightened resilience to bleaching compared to those experiencing minor
temperature variations [9].

Figure 5: Coral growth monitored in laboratory environments [10]

One significant advantage lies in comprehending coral growth within laboratory settings,
a feat challenging to achieve in the wild due to the costly and inaccurate nature of relying on human
observers for measuring and monitoring morphological aspects. In contrast, lab environments
provide a clear and replicable means to track coral growth (Figure 5), ensuring accurate
representation of their development in the natural environment, as these conditions are faithfully
replicated in tanks using the IoT sensor data collected.

11
 Figure 6: Coral aquaculture process [10]
In fact, the real-time data from IoT sensors, allow researchers to capture subtle variations
in water conditions. This precision is crucial for recognising environmental triggers that influence
coral health and spawning events. Coral reproduction, notably mass spawning events, is a critical
component of coral population maintenance. IoT technology enhances our ability to manage and
sustain coral populations by brings a scientific lens to the understanding and optimisation of these
events:
a) Predicting spawning times: IoT sensors are adept at detecting subtle environmental cues
that precipitate spawning events. Through the collection of extensive temporal data,
researchers can make predictions regarding the temporal occurrence of these events,
streamlining the process of gamete collection for controlled coral propagation.
b) Controlled spawning: Controlled environmental conditions, manipulated with the aid of
IoT technology, can induce spawning events in controlled laboratory settings. This
precision facilitates comprehensive scientific research on coral reproduction and enhances
the success of propagation endeavours.
c) Adaptive cultivation: IoT technology enables meticulous control of coral aquaculture
systems, facilitating the manipulation of environmental conditions to replicate those of
warmer or more challenging habitats. This enables the acclimation of corals to higher
temperatures over time, a scientific avenue that promises to enhance coral survival under
adverse conditions and even cultivate more heat resistant corals.
5.5 Further Improvements
Morphological methods are the mainstay for monitoring coral reef health [20], focusing
largely on bleaching and lack intervention options. As such, our solution stands to benefit from
the integration of additional (bio)sensors for on-site measurements, extending the device’s utility
beyond aquaculture to actively protect reefs. These sensors directly measure physiological
compounds and environmental factors, addressing the 20% of global corals threatened by harmful

12
substances [17]. These encompass heavy metals, pharmaceutical residues, natural toxins,
microplastics, hydrocarbons, toxic ions, etc., primarily due to their potential impact on coral reef
physiology and ecology.
Device Parameter Measured Pros Cons
Optical sensor • Viruses - Simple and cheap - High specialised
• Toxins - Portable technique
• Drugs - Customisable - Stability of the
• Antibodies - Label-free technique components
• Tumour biomarkers
• Tumour cells
• Ions
Electrochemic • Pathogens (cells) - High precision - Highly
al biosensor • Herbicides (sensitive and selective) specialised
• Cancer and disease - User friendly technique
biomarkers - Not time-consuming - Stability of the
- Economical components
Microbial • Petroleum - High precision - No direct
biosensor hydrocarbon (sensitive and selective) measurement
• Heavy metals - User friendly - Stability of the
- Economical components
- Specific design
Table 2: Additional sensors and their pros and cons
5.6 Application of the data: Monitoring, forecasting
The above sensors collect diverse data types, including sound, images, and numerical data.
Processing and understanding the intricate relationships among these factors can be challenging
for researchers. In addressing this complexity, the collaboration of chemometrics and artificial
intelligence (AI) becomes invaluable for gaining insights into the interactions among these
multifaceted factors [25]. AI-based analyses, tailored to analytical requirements and experimental
setups, have the potential to reveal valuable information from the data, occasionally unveiling
hidden insights that conventional approaches may overlook [25]. Moreover, this analysis can be
completed in a fraction of the time required for conventional methods, expediting the flow of
information from monitoring programs to conservation practitioners.
Machine Learning (ML) is a subset of AI that leverages data and algorithms to mimic
human learning processes and, over time, enhance its accuracy [24]. Among the diverse range of
machine learning models available for forecasting, we have opted for the Long Short-Term
Memory (LSTM) model in our solution [26]. The LSTM (Long Short-Term Memory) model
improves upon Recurrent Neural Networks (RNNs), designed for sequential data modelling. In
RNNs, raw data is processed, retaining knowledge from past inputs for accurate predictions.
LSTM extends memory duration, learning from historical inputs, preventing small gradient issues,
and boosting future value predictions [27].

13
 In our solution, the variability of numerous water parameters over time underscores the
need for continuous monitoring to uphold laboratory consistency. This is particularly critical given
the dynamic nature of these parameters. To address this challenge, forecasting emerges as a
valuable tool, specifically in the realm of water quality management.
Utilising forecasted data pertaining to real-time environmental parameters collected from
our IoT sensors, like water temperature, pH levels, and dissolved oxygen becomes instrumental in
anticipating potential deviations in water quality. This proactive approach enables effective
management to uphold optimal conditions for aquatic organisms within the laboratory
environment. Therefore, the deployment of a machine learning model with long-term memory,
such as the LSTM model, becomes imperative for the accurate prediction of future values, ensuring
the sustained quality and integrity of the laboratory setting.

5.7 Enhanced proactive warning systems

Moreover, forecasting serves as an advanced early warning system, empowering scientists
to replicate extreme water parameter variations in controlled laboratory environments using
forecasted data. This capability empowers aqua culturists to implement preventive measures
effectively, such as adjusting water parameters or administering treatments, to mitigate the impact
of unexpected and extreme water conditions. Thus, the controlled laboratory setting facilitates the
exploration of real-time environmental changes and allows the development of response strategies.

5.8 Strategic acclimatisation for improved coral transplantation

By using LSTM to anticipate future water parameters in the environment where corals are
presently being aqua-cultured, we can simulate these conditions within the laboratory setting. This
proactive approach ensures that as the corals mature and become ready for propagation, they are
already acclimatised to the anticipated environmental conditions. This strategic acclimatization
process has demonstrated a notable increase in the success rate of coral transplantation [17].

The infusion of IoT technology into coral aquaculture exemplifies the transformative
potential inherent in this scientific paradigm. Through the sustained real-time monitoring of water
parameters, researchers gain profound insights into the intricacies governing coral health and
reproductive processes. This wealth of knowledge not only cultivates the development of heat-
resistant coral strains but also heralds a new era in marine ecology research and conservation
endeavours. In essence, IoT technology stands poised to propel marine science and coral
conservation to unprecedented heights.

6 Solution 2: Coral Restoration using 3D Printing Technology

The significance of structural complexity in ecology, closely tied to biodiversity and habitat
carrying capacity [28], underscores the need for innovative solutions in coral restoration. Additive
manufacturing, specifically 3D Printing (3DP) technology, emerges as a promising avenue for

14
replicating the intricate morphology of natural coral formations. Leveraging advanced techniques
such as Photogrammetry and Gravity-Stimulated Printing Design, we can intricately design
artificial corals that address the challenges associated with morphological complexity crucial for
larval settlement and polyp transplantation.
In our proposed solution, we advocate for the use of environmentally friendly Calcium
Carbonate Precipitation (CCP) ink, complemented by Fused Deposition Modelling (FDM) for
coral fabrication. To further enhance efficiency, we recommend the incorporation of parallel
printing and multi-axis printing technologies, supported by advanced software automation and
remote monitoring. This holistic approach not only ensures scalability, cost-effectiveness, and
precision in 3D printing coral structures but also positions artificial corals for optimal deployment
in large-scale reef restoration initiatives.

7 Coral Model Design and Preparation Process

7.1 Coral Designing – Photogrammetry
Photogrammetry encompasses methods of image measurement and interpretation to derive
the shape and location of an object from one or more photographs of the object [29]. It involves
capturing a series of two-dimensional (2D) images of an object or a scene from multiple angles
and then using specialised software to process these images into a detailed, three-dimensional (3D)
digital model. Unlike 3D scanning, where lighting is critical for the accuracy of data output,
photogrammetry technology does not require additional equipment for active light sources.
Photogrammetry technology is advantageous as it only requires simple equipment, such as digital
cameras, to capture photographs of the object and gather the data to create a 3D digital coral model.
There are two main techniques used for 3D photogrammetry, Structure from Motion (SfM) and
Multi View Stereo (MVS).
7.1.1 Structure from Motion
SfM uses a series of 2D images from different angles and analyses the spatial relationships
between them to create 3D models. These images can be taken using standard cameras without the
need of expensive equipment. SfM uses Point Correspondence Identification algorithm to identify
and match common distinct feature points in the images. The software will compute the position
and orientation of the camera when each image was captured, establishing the spatial relationships
between the series of photographs.
7.1.2 Multi View Stereo
MVS has a similar imaging approach as SfM, using several images to reconstruct the
artificial 3D models. MVS differs from SfM as they use stereo matching and triangulation
techniques to estimate the 3D coordinates of distinct features in multiple images instead of
estimating the photo location, its orientation, the camera parameters. This allow MVS to obtain
more dense 3D data: depth maps, point cloud, signed distance function or a mesh [30], creating a
more accurate model of 3D coral structures.

15
 The convergence of both methods will create a more accurate and high-resolution imaging
used to construct a 3D point cloud. The combination of identifying the relative spatial positions
and coordinates of distinct features across the series of photographs will eliminate the estimation
error arising from both technologies individually and obtain a denser 3D data collectively. From
the data analysed, the software will build a 3D mesh representing the surface of the subject and
textures are then applied to the 3D base model create a more intrinsic and realistic appearance for
the corals.
7.2 Coral Modelling – Printing Design Method
Morphological complexity and diversity are key factors in creating artificial coral reefs and
reef structures to aid coral restoration. Given that the structural complexity of coral reefs is the
most conspicuous feature in a coral reef ecosystem, intricate and realistic details must be replicated
in the 3D model of artificial corals to achieve effective restoration. Hence, modelling corals from
Photogrammetry directly may not be enough to produce the most effective shape for growth,
needing an additional design method to refine the coral details. As most of the currently available
slicers for 3DP on the market are mainly used for thermoplastic 3DP such as Fused Deposition
Modelling (FDM) and some low-viscosity Paste-Based Extrusion (PBE) which have less
morphological properties, research has been done to establish an alternative method for improving
structural complexities. Hence, to achieve a more advanced morphological result, we will leverage
the Gravity-Stimulated Printing Design method (GSPD).
7.2.1 Gravity-Stimulated Printing Design Method
GSPD uses a comprehensive parametric algorithm that provides direct control of the
printing toolpath using gravity as an active forming tool instead of the traditional 3DP method of
creating a volumetric geometry shape of final 3D-printed product [31]. This control allows for the
creation of more detailed and realistic artificial coral and reef structures that mimic the complex
structures of natural coral reefs.

Figure 7: Final Shape using Printing Toolpath

16
 In addition to the complex design optimisation, GSPD method produces more stable and
durable artificial coral structures which are better able to withstand harsher conditions, maintaining
their integrity over a longer period compared to currently existing 3D-printed coral structures. The
structures produced via GSPD can offer improved habitat features for marine life as well. Since
this technique can closely replicate the natural coral formations, the printed reefs can better support
the settlement and growth of corals and coral larvae.

7.3 Design Customisation – Artificial Coral Substrates and Reef Structure

Designing coral skeletal structures that serve as a substrate for coral growth involves
creating artificial frameworks that models the structural elements of natural coral skeletons.
Customised and targeted design is important to support and promote the settlement, attachment
and growth of coral larvae and coral polyps.
7.3.1 Larvae Settlement
Coral larvae has many sensory cells and exhibits a complex searching behaviour to find
the most appropriate place to settle; and some abiotic factors that affect their choice of settlement
include light, currents and presence of hard substrate and biotic factors such as composition of the
biofilm, availability of certain crustose coralline algae and the presence of competitors [32]. They
are highly sensitive to their environmental conditions such as the water temperature, ocean
warming, and acidification and these factors affect the larval survival, swimming and settlement
behaviours [33]. This will result in potentially shortening the competency time of larvae and limit
dispersal distance, hindering the rate of growth of corals. Hence, it is critical that the 3D printed
artificial structures imitate exactly the irregularities and microhabitats found on natural reefs. The
creation of microscale features such as crevices and irregular surfaces, identical to the natural
corals, will also provide a more ideal settlement site for the larvae.
Batches of coral larvae will be tested for settlement in laboratories and nurseries to observe
the rate of growth and their survivability. This iterative will continue to facilitate the selection of
coral designs that achieves the optimal conditions for coral larvae settlement. Ultimately, the
design that promotes the highest rate of restoration will be selected for production. Similarly, an
iterative process will be performed for 3D printed reef structures that host coral polyps for coral
transplantation. The ability to create minute details such as crevices that allow the coral polyps to
attach themselves securely onto the design is also a crucial factor that improves corals survival on
artificial reefs. With the use of our imaging and 3D design methods – SfM and MVS with GSPD,
artificial corals and reef structures can then achieve the morphological complexities that existing
3D printing methods lack of, solving various issues revolving around structural complexities, coral
transplantation and larvae propagation.
7.4 Digital Twin – Design Optimisation
Through the concept of digital twinning, a virtual replica or simulation of the physical 3D-
printed coral structures can be created, allowing for precise monitoring, testing, and validation of
designs before actual production. This process involves using advanced computer modelling and
simulation software to generate an exact copy of the 3D-printed coral structures, which can be
17
extracted from the Photogrammetry and GSPD data. This digital twin will accurately replicate the
dimensions, design, and features of the physical structures of the coral reef, as well as each specific
coral species.
Once the digital twin is created, it offers a platform for real-time monitoring and analysis
of the structures and can simulate the behaviour and responses of the coral structures under various
environmental conditions, such as variations in water currents or exposure to environmental
stressors. Through this, it can identify flaws and weakness in the designs prototyped and gather
data related to stress points, structural integrity, and possible weaknesses for future design,
allowing continuous analysis and refinement of the 3D-printed coral structures before processing
for 3D printing and manufacturing.
8 Coral Printing and Manufacturing
3D printing corals stems from the urgent need for innovative solutions to address the rapid
degradation of natural coral ecosystems worldwide. These artificial corals serve as tools for reef
restoration, providing a sustainable and scalable approach to counteract the detrimental effects of
climate change, coral bleaching, and habitat destruction. The technology allows for the creation of
customisable structures that can support marine life and promote ecological diversity while
mitigating the challenges faced by traditional coral restoration methods. By replicating the
morphological complexity of natural corals, 3D printing offers a promising avenue to contribute
to the conservation and preservation of vital marine ecosystems.
The process of 3D printing has several advantages over traditional fabrication methods due
to the convenience for rapid prototyping, ease of customisation, and potential for scalability. Coral
species vary in size, shape, and structure requiring the design and fabrication process to be flexible
so that it can be easily applied to a wide range of complex geometries. Hence, the ability to
maintain the repeatability and accuracy of 3D printed objects is a key element for scalability of
coral fabrication. One significant challenge in scaling up coral reef restoration efforts is the need
for faster growth methods. 3D printing emerges as a solution by creating predetermined calcium
carbonate structures, accelerating coral growth. However, current 3D printing solutions face
limitations in design complexity, speed, accuracy, and customisation [34]. To overcome design
complexities, the proposed methods of SfM and MVS for photogrammetry and gravity-stimulated
printing design methods will be utilised, eliminating the need for post-printing adjustments such
as cutting trimming and sanding, enhancing the efficiency of creating 3D coral models. The
following solutions will discuss the solutions to improve speed, accuracy and customisation.

18
 Figure 8: Conceptual model showing 3D CoraPrint, a scalable multistep fabrication approach
for coral restoration [41]

Developed by the 3D CoraPrint team, Calcium Carbonated Photo-initiated (CCP) ink

mirrors the natural composition of coral skeletons, creating a 3D artificial model of corals that
closely resembles the mineral structures. Unlike conventional materials such as ceramic and
cement, which introduce foreign elements into marine ecosystems, using CCP ink is a more stable
and nontoxic alternative. CCP ink solidifies upon exposure to UV laser light. This property ensures
a rapid and efficient printing process, allowing for the printing of intricate coral structures with
minimal time investment. The swift solidification with UV exposure also contributes extensively
to the scalability of 3D coral manufacturing as post-processing time for solidification is reduced.

Common methods used for 3D printing include Fused Deposition Modelling (FDM) and
Stereolithography (SLA). In our attempt to improve the 3D printing techniques for coral structure,
we will leverage on improving FDM and its processes for coral printing. While SLA has the
advantage of producing prints with higher resolution and smooth surface finishing, it is a specific
3D printing method that uses epoxy resin which can be toxic to marine life and ecosystems.
Moreover, resins are expensive and are specifically produced by SLA printer's manufacturer [35],
further amplifying the issue of high cost in 3D printing. On the other hand, FDM allows the use of
a diverse range of materials, including biodegradable and environmentally friendly filaments
which provides the flexibility to use CCP ink instead of traditional materials. To further improve
this printing technique, we will utilise parallel printing and multi-axis printing to address the
scalability issue and propose an automated and monitored 3D printing process.

19
8.1 Parallel Printing
Compared to using a single extruder to build the 3D object, multi-head 3D printers can be
used to implement parallel printing. Multi-head 3D printers can deposit different materials
simultaneously, facilitating more efficient production of different coral parts, mitigating the need
for equipment changes. This also saves administrative time in moving materials around is cost-
effective as by reducing the need for multiple changes and equipment switches [36]. Furthermore,
multiple extrusion heads enable more complex design to be created as they can work concurrently
from various angles.

8.2 Multi-axis Printing

Current 3D printers still require the use of support structures, and the stair-step effect will
be quite evident if the printers are used to print complex and intricate structures such as corals.
Thus, we can use of multi-axis 3D Printers to resolve these problems. A multi-axis 3D Printer
works by either rotating the print-bed or the extruding head [37], which allows the printed structure
to easily be re-oriented at any given point in time during printing. As such, this is useful in
overcoming the issue of gravity-induced distortion and need for support structures [38], thereby
reducing the cost because support structures normally require additional materials. In addition, the
multi-axis 3D Printer allows for the manufacturing of intricate structures as the re-orientation of
structures allow for material to easily be deposited along the curved layers of the structure [39].

8.3 Software Automation

Utilise advanced slicing software like Orcaslicer which converts 3D models into layers and
generates Geometry code that guides the printer [40] for enhanced accuracy and resolution.
Checkpoints in this code can also be used to program connecting points to automate removal of
finished parts, the start of new print jobs and feeding of materials. Furthermore, optimisation of
process can be implemented by using advanced scheduling algorithms to maximise the utilisation
of each printer.

8.4 Remote Monitoring

IoT sensors on key components of the 3D printer, such as motors, extruders, and
temperature controls can be used to collect real-time data on the performance and status of the
printer. This data can be used for predictive maintenance, minimising downtime and enhancing
production cycles.

20
9 Technical Feasibility and Business Viability
In this section, we examine the technical feasibility and business potential of the previously
discussed solutions.
9.1 IoT feasibility and viability
The implementation of IoT sensors is technically viable, given the diverse range of sensors
already developed for our specific requirements [23]. Applications like water quality and
temperature monitoring offer readily available sensor options. The adaptability of sensor sizes and
designs also allows for seamless integration into the environment, ensuring easier installation.
Moreover, modern IoT sensors are designed to be waterproof and exceptionally durable, making
them well-suited for marine environments [42]. Furthermore, advancements in wireless
connectivity and satellite communications enhances real-time information flow to researchers in
remote marine locations.
The business viability of IoT sensors is equally promising. Developing and selling these
sensors for marine settings holds profit potential, with anticipated demand set to rise as coral reef
restoration efforts intensify in the coming years [43], driven by the sustainability goals of
governments and conservation organisations. Beyond sensor development, businesses can offer
valuable services, including data analytics and predictive modelling, utilising insights derived from
IoT sensor data.
9.2 Additive manufacturing feasibility and viability
3D printing corals has become increasingly feasible due to significant progress in printing
synthetic corals over the last decade [44]. In our coral design solutions, easily accessible
algorithms for photogrammetry and the extensive use of Digital Twin provide a solid foundation.
Thus, only minor adjustments are necessary to seamlessly integrate data from coral designing into
Digital Twin for in-depth analysis. Moreover, the availability of parallel printing and multi-axis
printing offers versatility, requiring simple printer customisations to reproduce various coral
structures.
When considering the business viability of 3D printing, critical factors include the type of
3D printer, software, and materials utilised in coral structure development. While the initial
acquisition cost of 3D printers, especially multi-axis 3D printers, can be substantial, our solution
incorporates GSPD, remote monitoring, and software automation to enhance the quality and
accuracy of printed coral structures, minimising material wastage. Additionally, using ceramic or
cement fillings as infill for 3D models, instead of CCP ink, not only improves structural integrity
but also reduces material costs. Ultimately, the successful restoration of corals through artificial
structures holds the potential to support fisheries and tourism, contributing to overall business
viability [44].

21
10 Conclusion
In harnessing the capabilities of Industry 4.0 technologies for the restoration of coral reefs,
it's crucial to recognise their role as supportive tools rather than ultimate solutions to the underlying
problem of reef degradation. While these technologies contribute to accelerating restoration
efforts, they do not directly address the root causes of the issue. To truly address and mitigate reef
decline in the long term, a comprehensive approach is necessary, confronting the primary drivers:
climate change, pollution, and destructive fishing practices.
It is imperative to avoid overemphasis on artificial structures and sensors, recognising that
these interventions should complement, not replace, established conservation techniques.
Excessive allocation of resources to high-tech solutions may inadvertently divert funds from
proven methods such as effective policing within marine protected areas [45]. Balancing
technological innovation with a commitment to tackling the fundamental challenges is essential
for achieving sustainable and enduring coral reef conservation.

22
Bibliography
[1]
National Oceanic and Atmospheric Administration, “How Does Climate Change Affect Coral reefs?,”
[Link], 2019. [Link]
[2]
National Ocean Service, “What Are Corals? - Corals: NOAA’s National Ocean Service Education,”
[Link], 2019. [Link]
[3]
The World Counts , “The World Counts,” [Link].
[Link]
[4]
D. O. Obura et al., “Coral Reef Monitoring, Reef Assessment Technologies, and Ecosystem-Based
Management,” Frontiers in Marine Science, vol. 6, Sep. 2019, doi:
[Link]
[5]
M. González-Rivero et al., “Monitoring of Coral Reefs Using Artificial Intelligence: A Feasible and Cost-
Effective Approach,” Remote Sensing, vol. 12, no. 3, p. 489, Jan. 2020, doi:
[Link]
[6]
M. Mer, “AS A STRATEGY TO IMPROVE ECOSYSTEM SERVICES CORAL REEF RESTORATION
A guide to coral restoration methods,” 2020. Accessed: Nov. 10, 2023. [Online]. Available:
[Link]
[7]
L. Boström-Einarsson et al., “Coral restoration – A systematic review of current methods, successes,
failures and future directions,” PLOS ONE, vol. 15, no. 1, p. e0226631, Jan. 2020, doi:
[Link]
[8]
L. Gastoldi and S. Cinti, “(Bio)sensors applied to coral reefs’ health monitoring: a critical overview,” Green
Analytical Chemistry, p. 100049, Jan. 2023, doi: [Link]
[9]
I. M. Soto, F. E. Muller Karger, P. Hallock, and C. Hu, “Sea Surface Temperature Variability in the Florida
Keys and Its Relationship to Coral Cover,” Journal of Marine Biology, vol. 2011, p. e981723, Aug. 2011,
doi: [Link]
[10]
K. Damjanovic, L. L. Blackall, P. Menéndez, and M. J. H. van Oppen, “Bacterial and algal symbiont
dynamics in early recruits exposed to two adult coral species,” Coral Reefs, vol. 39, no. 1, pp. 189–202,
Nov. 2019, doi: [Link]
[11]
A. D. Ridlon et al., “Culturing for conservation: the need for timely investments in reef aquaculture,”
Frontiers in Marine Science, vol. 10, Apr. 2023, doi: [Link]
[12]
M. Mahmoud et al., “Survivorship and growth rates for some transplanted coral reef building species and
their potential for coral reef rehabilitation in the Red Sea,” vol. 23, no. 2, pp. 183–193, 2019, Accessed:
Nov. 11, 2023. [Online]. Available:
[Link]
[13]

23
Aacl and Bioflux, “Analysis of transplanted coral growth using the rock pile method in Karimunjawa
National Park, Central Java, Indonesia,” vol. 16, no. 1, 2023, Available:
[Link]
[14]
V. Schoepf, H. Sanderson, and E. Larcombe, “Coral heat tolerance under variable temperatures: Effects of
different variability regimes and past environmental history vs. current exposure,” Limnology and
Oceanography, vol. 67, no. 2, pp. 404–418, Dec. 2021, doi: [Link]
[15]
K. Kubasky, “The Coral Reef Early Warning System (CREWS),” NexSens, Oct. 31, 2019.
[Link]
[Link] (accessed Nov. 11, 2023).
[16]
“National Sea Simulator Capabilities,” AUSTRALIA INSTITUTE OF MARINE SCIENCE.
[Link] (accessed Nov. 11, 2023).
[17]
J. Patterson, K. I. Jeyasanta, N. Sathish, J. K. P. Edward, and A. M. Booth, “Microplastic and heavy metal
distributions in an Indian coral reef ecosystem,” Science of The Total Environment, vol. 744, p. 140706,
Nov. 2020, doi: [Link]
[18]
J. Zhang, X. Wang, B. Wang, W. Sun, H. Du, and Y. Zhao, “Energy-Efficient Data Transmission for
Underwater Wireless Sensor Networks: A Novel Hierarchical Underwater Wireless Sensor Transmission
Framework,” Sensors, vol. 23, no. 12, p. 5759, Jan. 2023, doi: [Link]
[19]
H. Chen, Y. Zhu, W. Zhang, K. Wu, and F. Yuan, “Underwater Acoustic Micromodem for Underwater
Internet of Things,” Wireless Communications and Mobile Computing, vol. 2022, pp. 1–20, Sep. 2022, doi:
[Link]
[20]
“Eigendecomposition-Based Partial FFT Demodulation for Differential OFDM in Underwater Acoustic
Communications | IEEE Journals & Magazine | IEEE Xplore,” [Link].
[Link] (accessed Nov. 11, 2023).
[21]
C. Albaladejo, F. Soto, R. Torres, P. Sánchez, and J. A. López, “A Low-Cost Sensor Buoy System for
Monitoring Shallow Marine Environments,” Sensors, vol. 12, no. 7, pp. 9613–9634, Jul. 2012, doi:
[Link]
[22]
S. Kaveripakam and R. Chinthaginjala, “Energy balanced reliable and effective clustering for underwater
wireless sensor networks,” Alexandria Engineering Journal, vol. 77, pp. 41–62, Aug. 2023, doi:
[Link]
[23]
M. Palaniswami, A. Rao, and S. Bainbridge, “REAL-TIME MONITORING OF THE GREAT BARRIER
REEF USING INTERNET OF THINGS WITH BIG DATA ANALYTICS.” Accessed: Nov. 11, 2023.
[Online]. Available: [Link]
[24]
IBM, “What is Machine Learning?,” [Link], 2023. [Link]
[25]

24
A. Mayfield and C. Lin, “Leveraging the power of artificial intelligence to identify resilient reef coral
genotypes for cryopreservation,” Platax, vol. 19, pp. 43–57, 2022, doi:
[Link]
[26]
S. saxena, “LSTM | Introduction to LSTM | Long Short Term Memor,” Analytics Vidhya, Mar. 16, 2021.
[Link]
[27]
N. Donges, “Recurrent neural networks 101: Understanding the basics of RNNs and LSTM,” Built In, 2019.
[Link]
[28]
R. Yanovski, P. A. Nelson, and A. Abelson, “Structural complexity in coral reefs: Examination of a novel
evaluation tool on different spatial scales,” Frontiers in Ecology and Evolution, vol. 5, no. APR, p.
undefined–undefined, 2017, doi: [Link]
[29]
T. Luhmann, S. Robson, S. Kyle, and J. Boehm, Close-Range Photogrammetry and 3D Imaging. Walter de
Gruyter GmbH & Co KG, 2023. Accessed: Nov. 11, 2023. [Online]. Available:
[Link]
D+photogrammetry&ots=7dW9VUgJOp&sig=1V0XnW8t-
ijrDZQ_TwdZ9fiR7g4&redir_esc=y#v=onepage&q=what%20is%203D%20photogrammetry&f=false
[30]
T. Rouch, “PatchMatch Multi-View Stereo,” Medium, Apr. 10, 2023.
[Link]
fc46e5dfe912#:~:text=Multi%2DView%20Stereo%20(MVS) (accessed Nov. 11, 2023).
[31]
O. Berman et al., “Design and application of a novel 3D printing method for bio-inspired artificial reefs,”
Ecological Engineering, vol. 188, p. 106892, Mar. 2023, doi:
[Link]
[32]
“SECORE International | Larval settlement,” [Link].
[Link]
[33]
J. Mallon et al., “The influence of deoxygenation on Caribbean coral larval settlement and early survival,”
Frontiers in Marine Science, vol. 10, Oct. 2023, doi: [Link]
[34]
A. Humanes, Alasdair Edwards, J. Guest, and J. Bythell, “3D printing coral reefs can create new habitat –
but it doesn’t tackle human destruction,” The Conversation, Oct. 08, 2018. [Link]
printing-coral-reefs-can-create-new-habitat-but-it-doesnt-tackle-human-destruction-103927
[35]
“Fused Deposition Modelling vs Stereolithography | Prototal UK,” Oct. 19, 2022.
[Link]
[36]
J. K. Placone and A. J. Engler, “Recent Advances in Extrusion-Based 3D Printing for Biomedical
Applications,” Advanced Healthcare Materials, vol. 7, no. 8, p. 1701161, Dec. 2017, doi:
[Link]
[37]

25
F. Hong, B. Lampret, C. Myant, S. Hodges, and D. Boyle, “5-axis multi-material 3D printing of curved
electrical traces,” Additive Manufacturing, vol. 70, p. 103546, May 2023, doi:
[Link]
[38]
A. Eckhoff, “Removing The Boundaries of 3D Printing with Multi-Axis Additive,” [Link].
[Link]
axis-additive/ (accessed Nov. 11, 2023).
[39]
G. Schuh, G. Bergweiler, G. Lukas, S. Hohenstein, and J. Schenk, “Feature-based Print Method for Multi-
Axis Material Extrusion in Additive Manufacturing,” Procedia CIRP, vol. 93, pp. 85–89, 2020, doi:
[Link]
[40]
“3D Printing Software for Designing, Slicing, Printing, and Monitoring | Obico Knowledge Base,”
[Link], Oct. 09, 2023. [Link] (accessed Nov. 11, 2023).
[41]
H. I. Albalawi et al., “Sustainable and Eco-Friendly Coral Restoration through 3D Printing and
Fabrication,” ACS Sustainable Chemistry & Engineering, vol. 9, no. 37, pp. 12634–12645, Sep. 2021, doi:
[Link]
[42]
J. Trevathan, S. Schmidtke, W. Read, T. Sharp, and A. Sattar, “An IoT General-Purpose Sensor Board for
Enabling Remote Aquatic Environmental Monitoring,” Internet of Things, vol. 16, p. 100429, Dec. 2021,
doi: [Link]
[43]
“Can coral restoration reverse long-term declines in coral reef growth? | U.S. Geological Survey,”
[Link]. [Link]
declines-coral-reef-growth
[44]
“Coral Relief: How 3D Printing Is Reviving the Ocean’s Key Ecosystems,” [Link].
[Link]
[45]
“Tourism,” Coral Reef Alliance. [Link]
[46]
“Why 3D printing coral reefs won’t save our oceans,” The Independent, Oct. 17, 2018.
[Link]
[Link]

26