Artificial Fish Reef Site Evaluation Based on Multi-Source High-Resolution Acoustic Images

Abstract

Marine geophysical and geological investigations are crucial for evaluating the construction suitability of artificial fish reefs (AFRs). Key factors such as seabed topography, geomorphology, sub-bottom structure, and sediment type significantly influence AFR design and site selection. Challenges such as material sinking, sediment instability, and scouring effects should be critically considered and addressed in the construction of AFR, particularly in areas with soft mud or dynamic environments. In this study, detailed investigations were conducted approximately seven months after the deployment of reef materials in the AFR experimental zones around Xiaoguan Island, located in the western South Yellow Sea, China. Based on morphological factors, using data from multibeam echosounders and side-scan sonar, the study area was divided into three geomorphic zones, namely, the tidal flat (TF), underwater erosion-accumulation slope (UEABS), and inclined erosion-accumulation shelf plain (IEASP) zones. The focus of this study was on the UEABS and IEASP experimental zones, where reef materials (concrete or stone blocks) were deployed seven months earlier. The comprehensive interpretation results of multi-source high-resolution acoustic images showed that the average settlement of individual reefs in the UEABS experimental zone was 0.49 m, and their surrounding seabed experienced little to no scouring. This suggested the formation of an effective range and height, making the zone suitable for AFR construction. However, in the IEASP experimental zone, the seabed sediment consisted of soft mud, causing the reef materials to sink into the seabed after deployment, preventing the formation of an effective range and height, and rendering the area unsuitable for AFR construction. These findings provided valuable scientific guidance for AFR construction in the study area and other similar coastal regions.

1. Introduction

The offshore area, where the ocean meets land, is a crucial part of the Earth’s source-to-sink surface system. This region has been extensively studied and developed, with frequent activities taking place due to its critical importance. As a land–sea interaction zone, offshore waters are affected by biological, runoff, coastal currents, tides, waves, wind, and solar radiation, as well as by human activities [1,2,3,4,5]. The seabed topography and geomorphology of continental margins are shaped by a combination of geological tectonics, sea-level changes, hydrodynamics, hydrochemistry, climate, biology, and human influence [6,7]. Understanding the patterns and distribution characteristics of offshore topography and geomorphology, identifying the factors that control their formation and evolution, and uncovering their underlying mechanisms would provide background data for the development of the marine economy, disaster prevention and reduction, scientific research, environmental protection, and digital ocean initiatives. These insights also offer a critical scientific foundation for the exploration and development of marine resources, coastal engineering construction, maritime transportation, and national defense and security [8].

Aquaculture development is one of the most promising areas of economic activity in coastal sea zones. However, balancing the environmental impacts of marine aquaculture with its economic benefits is a globally recognized challenge for the sustainable development of this industry [9,10]. The development of artificial benthic habitats has been a long-standing global venture to enhance coastal fishery resources [11]. Zeng et al. [12] give insights into how bioacoustics have been applied to fisheries management in artificial fish reef (AFR) reserves. The first issue to be addressed in the design and construction of AFR is scientific site selection, which is a key factor related to success or failure [9,10]. Several factors must be considered when selecting a site for AFR construction, including marine functional zoning, fishery resource distribution, hydrodynamic conditions, underwater topography, geomorphology, and seabed sediment characteristics [4,9,10,11,12,13,14,15]. Many studies have mainly focused on issues such as reef materials, biological enrichment effects, and species identification and abundance investigation [3,4,12,14,16,17,18]. The state effect of reef materials after deployment is mainly studied by numerical simulation prediction [10,15], and there are few studies evaluating the effectiveness of reef construction through actual measurements [11]. Ali et al. [11] proposed the method of side scan sonar (SSS) coupled with scuba diving observation for the enhanced monitoring of the pre-deployed AFR. However, no attention was paid to the effects of seabed microgeomorphology, sediment characteristics, and shallow geology features on the reef.

Geophysical techniques for shallow waters are mainly used in the planning stage to assist investigations. In addition to surveying water depth and seabed landforms, it is also an important application to determine the geological conditions below the seabed, using reflected acoustical signals from sonar, which assist in determining material changes qualitatively and evaluating the geologic structure with minimum drilling data [19]. Underwater acoustic techniques have proven to be valuable tools for studying seabed geomorphology and evolution, offering the ability to map these features across a wide range of spatial scales [8,20,21,22]. Meanwhile, acoustic technology has also demonstrated its ability in the assessment of fish abundance, species identification, and spatial distribution characteristics of benthic organisms in AFR areas [4,14,17]. A large number of successful cases demonstrate that the quantitative geophysical methods have high potential in nearshore and offshore site characterization [19], and in identifying fish species and assessing detailed habitats around AFRs [17]. However, different types of acoustic devices have their own application scenarios due to operating principles and acoustic frequency. Multibeam echosounder (MBE) is suitable for bathymetric topographic surveys and can achieve a seabed spatial resolution of 20 cm in shallow water by adjusting the beam angle and beam number [23]. SSS is suitable for fine measurement of detailed features on the seabed, with a seabed spatial resolution of 2 cm [24]. A sub-bottom profiler (SBP) is suitable for measuring shallow geological features below the seabed, and the vertical resolution of the parametric equipment can reach 12 cm (layer thickness) [25]. Therefore, multi-source acoustic images are of great significance in the measurement of underwater targets. High-resolution seabed imagery derived from multisource acoustic data has enabled the delineation of submarine geomorphology in remarkable detail [11,26].

In recent years, with strong support from national and local governments for the construction of marine ranches, multiple AFR experimental zones have been established in the offshore waters of Shandong Province. The waters surrounding Xiaoguan Island in the western South Yellow Sea have become a key experimental area for AFR construction due to their favorable geographic location and healthy ecological environment. Xiaoguan Island is situated in Laoshan Bay on the southeast coast of the Shandong Peninsula and on the western side of the South Yellow Sea (Figure 1). This area is known for its natural scenery, good geographical location, and excellent ecological environment. Influenced by a variety of natural forces and human activities, the sea area near Xiaoguan Island features a tide-controlled insular-bay sedimentary environment. The seabed in this region exhibits a distinctive offshore topography, geomorphology, and submarine microgeomorphology.

In this study, we proposed a combined method of using multi-source high-resolution acoustic images (MBE, SSS, and SBP) for investigating seabed microgeomorphology and evaluating the suitability of an AFR site through a typical case, providing a reference for similar research and engineering construction. Using multi-source acoustic survey data obtained in October 2020, we conducted a comprehensive analysis of the sedimentary environment, seabed topography, geomorphology, and Holocene geological characteristics of the surrounding waters of Xiaoguan Island. The primary focus was on identifying and analyzing the characteristics of seabed microgeomorphology types with scales ranging from centimeters to meters, similar to that of AFR materials such as stone blocks or concrete modules (from 0.5 m to 2.0 m). Based on these findings, we further evaluated the suitability of AFR construction in this area by assessing the state of the seabed after the deployment of AFR materials. Figure 2 illustrates the research framework, including the data acquisition, data processing, interpretation, and AFR construction suitability analysis.

2. Materials and Methods

2.1. Geological Setting

The surrounding waters of Xiaoguan Island are located in the transitional zone of the Jiaolai depression, Jiaonan uplift, and Qianliyan uplift. Previous studies on the region’s tectonic structure, stratigraphy, and magmatic rocks indicated that during the pre-Sinian period, the area transitioned from geosynclinal sedimentation to the formation of a crystalline basement. Afterward, this region underwent erosion and denudation following folding uplift [27]. In the early stage of the Yanshan movement, a large depression occurred in the northern part of the region, resulting in the deposition of the Late Jurassic strata. In the later stage of the Yanshan movement, frequent magma activity led to large-scale acidic granite intrusions, creating a unique and striking landscape [27,28]. The area around Xiaoguan Island is not currently experiencing active tectonic movements, with no reactivation of faults or evidence of Cenozoic volcanic activity.

The terrain around Xiaoguan Island is characterized by undulating landscapes, with coastal areas marked by low mountains and hills that extend toward the sea, forming capes or peninsulas. These landforms control the scale of Xiaodao Bay to the west and influence the seabed geomorphology where the underwater slope transitions to shallow sea plains. Alluvial deposits resulting from the long-term weathering and erosion of Xiaoguan Island and its nearshore rock mass contribute significantly to the seabed sediments in this region [28]. Driven by winter winds, these suspended substances, carried by strong nearshore currents, have served as the primary material source for nearshore and coastal sediments in the South Yellow Sea for nearly 7000 years [29].

2.2. MBE Survey

MBEs have proven effective for acquiring high-resolution seafloor bathymetric data and acoustic imagery in various water environments [8,16,29]. For the shallow water AFR experimental zone, bathymetric data were collected using a Reson SeaBat T50-P MBE system (Teledyne Reson A/S, Slangerup, Denmark), operating at a frequency of 190–420 kHz. T50-P is a portable, shallow-water type device with relatively high and wide-band frequency that enables complex objects and more detailed images of the seafloor. The system was operated from a transducer mounted on a ship hull, which emitted 512 beams, spaced 0.5° by 0.5° over an angular sector of 150°. Its vertical resolution exceeded 0.6 cm, excluding potential errors from position, attitude, and refraction uncertainties. In addition, a Trimble SPS 351 differential global positioning system (Trimble Navigation Co., Ltd., Long Beach, CA, USA.), Applanix PosMV 320 (Trimble Applanix Co., Ltd., Richmond Hill, ON, Canada.) motion sensor, and AML MicroX sound velocity profiler (AML Oceanographic Co., Ltd., Victoria, BC, Canada.) were used for bathymetric measurement.

Almost full-coverage MBE of the entire study area was conducted in this work. The typical survey line spacing of MBE was 50 m and 100 m, and the line spacing was adjusted according to the water depth in shallow water. It was not fully covered in areas with a water depth of less than 5 m, and previous data (2014) was used to supplement the coverage. Teledyne PDS software (Version 2.1) was used to carry out data acquisition and processing of the MBE data. Before data collection, parameter settings and system calibration were set in strict accordance with the operation specifications. After the MBE system was installed, a typical characteristic terrain area was selected in the study area for calibration. The calibration module of PDS software (Version 2.1) was used to calculate the accurate correction values of roll, pitch, and yaw, and input them into the acquisition software. The MBE data process included data checking, sound velocity correction, tide correction, and error point elimination. Finally, the water depth grid data were obtained and a seabed terrain rendering map was output (Figure 3).

2.3. SSS Survey

Given its reliable performance in locating and characterizing AFR structures, SSS is considered the most preferable method to monitor both artificial benthic habitats and natural benthic habitats [11]. The SSS (side scan sonar) system offered significant advantages over the MBE in measuring the seabed microgeomorphology. The acoustic wave is emitted diagonally downward at a small angle toward the seabed with fan-shaped signals covering more than a single direction. Since the angles between the emitted waves and the seabed surface are relatively small, the travel distance can be converted to a horizontal distance [19]. One key benefit of this approach was the ability to provide high-resolution complete spatial coverage of the study area [16,29,30]. An Edgetech 4200 MP dual-frequency SSS systemEdgetech Co., Ltd., West Wareham, MA, USA was used to acquire the microgeomorphic features, operating at frequencies of 100 and 400 kHz, with horizontal beam angles of 0.64° at 100 kHz and 0.3° at 400 kHz, and a range resolutions of 8 cm at 100 kHz and 2 cm at 400 kHz. The sonar towfish was towed 2 m laterally from the ship’s hull using a horizontal brace and a pulley to minimize wake interference. The built-in heading, pitching, and rolling sensors corrected the towfish’s attitude in real time. The SSS survey for the study area used a 75-m or 100-m line spacing grid pattern, achieving approximately 100% overlap.

SSS data processing was carried out using the SonarWiz software (Version 6.04). After importing the data, we applied filtering, seabed tracking, and time-varying gain to obtain the best waterfall images [8,16]. Subsequently, after sound velocity, beam, and slant range correction, we completed the recognition, digitization, and mosaic output, as shown in Figure 4, which allowed for consistent segmentation and interpretation with respect to sediment type, feature delineation, target recognition, and so on [16].

2.4. SBP Survey

SBP is mainly used to survey the sub-seabed geological structure, sediment thickness, or buried materials below the seabed. Because of this, lower frequency waves, which have larger penetrating depths, are used, usually ranging between 500 Hz and 40 kHz [19]. To meet this frequency range, a more diverse range of vibration sources is used, depending on the required depth and resolution. To minimize sediment reverberation, especially for the upper layer of the seabed surface, a system with highly directional beams is required. Parametric SBP systems meet this requirement by generating low-frequency narrow beams without side lobes [29,31].

In this study, a high-resolution SBP survey was conducted on the east side of Xiaoguan Island, with five survey lines (about 32 km), as shown in Figure 4. The layout of survey lines was based on the principle of passing through typical geomorphic units, with a main survey line spacing of 1.5 km and a connecting survey line spacing of 2.5 km. A parametric SBP system GeoScope-100 (Ocean Physics Co., Ltd., Shanghai, China.)was used to collect the SBP data, with an operational first frequency of 90–120 kHz and an optional difference frequency of 5–20 kHz. The pulse length ranged from 0.05 to 1 ms, with a vertical resolution of up to 4 cm (theoretically). The SBP transducer was installed in the center of the ship’s hull. To avoid acoustic interference, the SBP was not operated simultaneously with other acoustic instruments, such as MBE or SSS. During the survey, a motion sensor was used for real-time heave correction.

The Triton Perspective software (Version 7.7) was used to process the SBP data. After importing the data, we applied high- and low-pass filtering, average gain, and time-varying gain to optimize the images. Subsequently, we performed seabed tracking and digitized strong reflection interfaces and any abnormal reflection events. Finally, an intuitive three-dimensional SBP display map was generated (Figure 5).

2.5. Sediment Survey

A grab sampler(Haiyan Electronics Co., Ltd., Shandong, China) was used to collect sediment from the seabed in the study area. The layout of sampling stations was based on the principle of uniform distribution within the study area and representation of typical geomorphic units, with intervals ranging from 0.8 km to 1.5 km. After sampling, we classified and described the samples on deck according to their engineering geological characteristics. A total of 48 sediment samples were classified into three types: mud (Si), which accounted for the majority with 40 samples; sand (S), which accounted for 3 samples; and mixed mud and sand (Ms), which comprised 5 samples. The locations of these samples are shown in Figure 4.

3. Results

3.1. Geomorphology Types

Submarine geomorphology refers to the various morphological features of the seabed, which encompass characteristics, genetic types, distribution patterns, and evolution processes [29]. Full-coverage bathymetric data are essential for studying submarine topography and geomorphology [8,26]. Combined with previous studies, the geomorphology in this study was divided into three types and zones, namely, the TF, UEABS, and IEASP. Their distribution characteristics are shown in Figure 3.

TF geomorphology was located in the northwest region of the study area, forming a shallow beach on the west bank of Xiaodao Bay. This beach had a relatively gentle slope of about 0.09°, featuring a prominent headland in the southeast, with its current form the result of long-term erosion of the bedrock coast and nearby sedimentation [32]. This slightly inclined beach was formed by the accumulation of mud and sand on the shore by turbulent waves. At the upper part of the beach, wave action extended above the sea surface, with the lower part of the beach extending to the area where waves broke beneath the surface. This beach was located in a high-energy environment between high and low tide levels. As waves lost energy while propagating inland, the energy sediment accumulated, forming strip-like, rhythmic coastal features that ran roughly parallel to the shore, known as submarine sand embankments. The formation of these coastal features was driven by self-organization mechanisms, where the interaction of hydrodynamics and seabed relief promoted the growth of small perturbations on the bottom [33].

The UEABS geomorphology was found in the western region of Xiaoguan Island and Rabbit Island. It was partially adjacent to the TF geomorphology in Xiaodao Bay, with the remainder extending toward the continental bedrock coastline. The seabed slope was generally steep, at about 0.15°. This sediment was highly active under the action of waves and tidal currents, with strong erosion and accumulation, resulting in landforms such as sandbars and tidal current sand ridges.

IEASP geomorphology was found in the eastern sea area of Xiaoguan Island and Rabbit Island. Generally, we did not observe any obvious abrupt uplifts or depressions on the seabed. The slopes in this region were gentle, ranging from 0.01° to 0.05°. We identified a rich source of terrestrial materials, with strong modern sedimentation, which mainly consisted of mud or sandy mud.

3.2. Microgeomorphology Features

Seabed microgeomorphology refers to the microscopic morphology and features of the seafloor. Due to their extremely high resolution and working efficiency, SSS systems offer significant advantages for surveying seabed microgeomorphology and underwater target detection. In this study, SSS was used for the identification and delineation of seabed micro characteristics. The six submarine microgeomorphology types observed in this study consisted of disturbed zones, sand waves, tidal sand ridges, sand dunes, natural reefs, and AFR (as shown in Figure 4 and Figure 6).

The disturbed zones were distributed mainly in the eastern part of Xiaoguan Island and appeared as a small range of branches in the SSS image (Figure 6a). Small-scale physical disturbance was readily detected using the SSS system, with tracks characterized by high acoustic backscatter, while the seabed between the furrows exhibited low backscatter [2,34]. During the SSS survey, a large sand suction vessel was operated in the study area, and we speculated that these disturbed zones were caused by the vessel’s operation on the seabed, which should be interpreted as the scars of chain anchors rasping the seabed, as a result of the vessel drifting [2]. The fine and shallow nature of these disturbances made them undetectable in the MBE images.

The submarine sand waves were distributed around Xiaoguan Island, Rabbit Island, and Benchmark Rock, which were presented as corrugated shapes in the SSS images. Figure 6b shows the sand waves on the east side of Rabbit Island. These micro sand waves consisted of regular and constantly moving sand deposits formed on the seabed as a result of local hydrodynamic changes caused by the reefs [26]. Due to their small scale, they were not visible in the MBE images.

Tidal sand ridges were distributed near the bedrock coast in the west of the study area and in the southern headland and followed a northwest–southeast trend. These ridges appeared as distinct, strip-like strong reflection zones on the SSS images (Figure 6c). They were formed by tidal currents, with large amounts of sand accumulating in parallel ridges aligned with the direction of the current. The tidal sand ridges were relatively large and could be clearly identified in the MBE images.

The submarine sand dunes (sand embankments) were found to be distributed in Xiaodao Bay and along the west side of the mouth between Xiaoguan Island and Rabbit Island, creating long trip-like strong reflection zones on the SSS mosaic image. Affected by the topography, the area near the SSS towfish exhibited strong reflection, while the area far away from the towfish presented obvious shadows (Figure 6d) [20,35]. The sand embankment in Xiaodao Bay consisted of a sandy sedimentary body formed by the sediment carried by periodic ebb and flow currents in the breakwater zone. The sand dune on the west side of the entrance between Xiaoguan Island and Rabbit Island was produced by a confined, strong tidal current, which became unconfined after passing through the mouth gate, leading to turbulence and vorticity, with a decline in velocity leading to deposition of the transported suspended sediment load [7,36,37]. Due to their large scale, these sand dunes were clearly visible in the MBE images (Figure 3).

Natural reefs were widely distributed in the shallow water on the west side of the study area, around Xiaoguan Island, Rabbit Island, Benchmark Rock, and the western headland and bedrock coast. Figure 6e shows an underwater natural reef near Benchmark Rock, which was partially exposed at the water surface at low tide. These natural reefs, as substrates exposed to the seabed, were easily identifiable in the MBE [7] and SSS images. However, for smaller independent reefs, the boundaries were more clearly delineated in the SSS images.

The western and southern waters of Xiaoguan Island are designed as the China Qingdao Laoshan Bay National Marine Aquaculture Demonstration Zone, where several AFR zones have been deployed, as shown in Figure 4. Figure 6f presents a relatively scattered AFR experimental zone on the sandy seabed. The individual reef module formed a cube with a side length of 2 m, with octagonal openings on each surface (Figure 6g). The AFRs could be identified in the MBE and SSS images. Scattered reefs were clearly delineated and shaped in the SSS images; however, their boundaries could not be clearly identified in the MBE images.

3.3. Holocene Strata

The results of the high-resolution parametric SBP survey indicated that the seabed geological structure on the east side of Xiaoguan Island was simple with uniform thickness, and the sedimentary thickness was affected mainly by the top interface of the underlying bedrock. In areas far away from the island, the sedimentary thickness increased, and the SBP survey could not penetrate the bedrock. According to the acoustic sequence of the SBP data, the seabed strata on the east side of Xiaoguan Island could be divided into three sequences, as shown in Figure 7.

Sequence I was a well-stratified Holocene marine sequence, characterized by parallel or subparallel internal reflections, with its bottom boundary forming an unconformity interface. The thickness of the Holocene was primarily controlled by the underlying bedrock interface, and, at the far end of the outlying island, it remained stable, generally ranging from 9.0 to 12.0 m. According to the characteristics of internal reflection, we divided sequence I₁ into two subsequences: I₁ and I₂. Subsequence I₁ was a shallow marine sequence formed during the modern period of high sea level, with weak hydrodynamics during sedimentation. Its interior was mainly dominated by blank reflections, supplemented by parallel reflections. According to the sediment survey results, we speculated that the lithology of this layer consisted of organic-rich mud with locally mixed sand. Subsequence I₂ represented a shallow marine sequence formed during the transgression period. The internal reflections were mainly parallel, where a typical progradational sequence formed near the island. We speculated that the lithology of this layer was muddy clay or muddy silty clay.

Sequence II consisted of an unstratified Late Pleistocene continental sequence, with weak chaotic reflections. The bottom interface of this sequence in the near island area comprised the bedrock surface, while, in the far island area, its bottom interface could not be tracked. We determined that the lithology of this layer was sandy clay. Sequence III comprised a layer of native bedrock.

4. Discussions

4.1. Data Accuracy and Limitation Analysis

4.1.1. Data Accuracy

The data for the underwater terrain rendering map were derived from MBE measurement and historical data collection. Due to the insufficient coverage range of MBE, historical depth data (2014) was used to supplement the TF area with a depth of 5 m or less. Except for the TF area, the rest area adopted the MBE measurement data. The error in MBE measurement came from multiple aspects: sea conditions, sound speed in the water, water clarity and salinity, and unreasonable equipment parameter settings [23]. In addition to taking measures to reduce errors during the measurement process, it was also necessary to perform accuracy checks on the final data. The accuracy of water depth measurement was verified using overlapping point difference data. The result showed that only 0.08% of the overlapping points have a difference of more than 0.4 m, with a root mean square error of 0.05 m.

There are many factors that affect the SSS detection accuracy, such as sea conditions, ship speed, towfish posture and height, seabed undulations, sound speed, and non-uniformity [24]. At present, there is no uniform evaluation method for the actual measurement resolution of SSS (this is an important issue worth exploring in future work). In practical work, it is usually quantitatively described based on the operating environment and the nominal parameters (theoretical values) of the sonar. The nominal resolution given by the manual of Edgetech 4200 MP is 2 cm for high frequency (400 kHz) and 8 cm for low frequency (100 kHz). It can be seen that high-frequency images can discover more detailed features of the seabed. According to the comparison of adjacent survey lines, the effective coverage rate of SSS achieved 100% except for the aquaculture area, with a 50–75% overlap in areas shallower than 5 m, and a 100% overlap in areas deeper than 5 m.

The SBP vertical resolution was verified by the depth of the stratified interfaces at the overlapping points of intersecting survey lines. Five survey lines produced six intersection points, resulting in a total of eighteen overlapping points (three interfaces, Figure 7). The comparison results showed that the depth differences of stratified interfaces at the overlapping points were 0.11–0.21 m, indicating that the vertical accuracy of SBP was better than 0.21 m.

4.1.2. Data Limitation

The site selection of AFR involves the intersection and integration of multiple disciplines, such as ecology, geomatics, geomorphology, geology, materials science, and hydrology [9,13]. It must be acknowledged that this article has certain limitations in terms of data.

(1) The lack of biological data from AFR experimental areas makes it impossible to evaluate their bioaccumulation effects. In the later work, we will consider using underwater cameras and acoustic imaging methods to investigate the fish species and quantity in the reef experimental area [4,11,17].

(2) This study only provides single-period data after the deployment of AFR materials, without data before the deployment, making it impossible to make detailed comparisons.

(3) The lack of hydrological data such as waves and currents limits the depth of scouring analysis around AFRs. The basic hydrological data can be used to numerically calculate the scouring depth and volume around the reefs.

4.2. Suitability Evaluation of AFR Construction

According to the European Artificial Reef Research Network, an artificial reef can be defined as a submerged structure intentionally placed on the seabed to mimic certain characteristics of a natural reef [13]. AFRs can be designed to enhance or restore marine habitats [38]. A key aspect of successful artificial reef development is the thorough geophysical and geological evaluation of candidate reef sites [9]. The AFR experimental zones in the study area were nearshore aquaculture reefs, typically located in waters with a depth of from 2 to 30 m, with flat, hard seabed to ensure stability and resistance to sedimentation. We assessed the suitability of constructing AFRs in the study area based on seabed characteristics, such as seabed topography, microgeomorphology, and seabed sediment.

4.2.1. UEABS Experimental Zone

The Xiaodao Bay AFR experimental zone was located on the west side of Xiaoguan Island, with a water depth of from 5.6 to 7.8 m, corresponding to the characteristics of the UEABS geomorphology zone. The microgeomorphology in this area was characterized by scattered artificial reef individuals or deposits, as shown in Figure 3 and Figure 4. The reef materials in this area were deployed approximately seven months before the survey. Limited by the MBE resolution, only the rough outlines of the reef deposits, micro pits, and pockmarks formed by scattered reef individuals could be identified from the MBE image (Figure 8a). The MBE results revealed minimal erosion around the reefs in the Xiaodao Bay experimental zone, with reef outlines and pockmarks visible on the seabed. The SSS mosaic provided a clearer delineation of individual reef structures, which appeared as square-shaped areas with strong reflections and weak reflections in the center (Figure 8b). Sediment analysis confirmed the presence of mixed mud and sand (Ms), with gravel inclusions that likely contributed to the seabed’s stability (Figure 8c). High-resolution SSS images of individual AFRs (Figure 8d) further demonstrated their intact structure, clear acoustic shadows, and minimal scouring, confirming that the UEABS zone supports effective reef deployment.

Acoustic shadows are crucial features in SSS image recording, resulting from the low acoustic energy reflected back to the towfish from the local seafloor, with these shadows considered valuable for underwater target identification and height estimation [4]. As shown in Figure 9, we considered the propagation path of sound waves emitted by sonar towfish as a straight line. In this case, the towfish, the seabed, and the outermost tip of the target (AFR) shadow approximately formed a right triangle. According to the principle of similar triangles in geometry, the height of AFR is estimated, as shown in Equation (1):

$$H_a={\displaystyle \frac{L_s\times H_t}{L_{es}}}$$

(1)

where H_a is the height of the underwater target (AFR), L_s is the length of the shadow, H_t denotes the height of the towfish to the seabed, and L_es is the distance from the towfish to the outermost tip of the acoustic shadow.

Thus, the settlement of AFR could be calculated using Equation (2):

$$∆H_a=d-H_a$$

(2)

where ∆H_a is the settlement amount of AFR, and d is the actual side length of a cubic reef module (=2 m).

To assess the settlement of AFRs in the experimental area after deployment, we selected individual reefs with clear, identifiable outlines and normal postures from the SSS images (Figure 8d). By interpreting the corresponding parameters from the images, we calculated the settlement amount of individual reefs using Equations (1) and (2), as shown in

Table 1. Settlement of AFRs in the UEABS experimental zone.

No.	Ht (m)	Ls (m)	Les (m)	Ha (m)	∆Ha (m)
A	4.9	6.65	19.94	1.63	0.37
B	5.0	16.66	56.34	1.48	0.52
C	4.8	24.76	97.57	1.22	0.78
D	4.7	5.88	16.29	1.70	0.30
E	4.7	31.11	83.02	1.76	0.24
F	4.9	5.78	19.07	1.49	0.51
G	4.9	11.66	38.82	1.47	0.53
H	4.9	24.56	90.43	1.33	0.67
I	4.6	10.86	40.08	1.25	0.75
J	4.7	17.34	56.15	1.45	0.55
K	4.7	4.82	12.24	1.85	0.15

. The results showed that in the sandy seabed of the experimental area, after approximately seven months of deployment in the sandy seabed, the settlement was minimal, ranging from 0.15 to 0.78 m, with an average amount of 0.49 m. The low-energy reflection area on the side of the reef closest to the towfish suggested weak scouring around the reef. However, the shape of the reef was complete, the outline was clear, and the reef block showed only slight subsidence into the seabed. Therefore, according to the perspective of seabed morphology and sediment characteristics, this experimental area was suitable for AFR construction.

4.2.2. IEASP Experimental Zone

The IEASP AFR experimental zone was located in the southeast region of the study area, with depths ranging from 11.0 to 12.3 m. The seabed was flat, with a sloping erosion-accumulation continental shelf plain landform. Except for the disturbed zones and AFRs, the seabed in the eastern area of Xiaoguan Island was smooth. In the SSS records, these smooth areas were characterized by constant low backscatter [30]. The microgeomorphic feature of the AFR zone consisted of a strip-like artificial reef accumulation body, as shown in Figure 4. The AFRs in this area were deployed about seven months before this survey, using crushed and block stones as the main materials. In the MBE image (Figure 10a), the outline of the AFR was not visible, and no obvious topographic anomalies were detected, only a vague high point on the seabed. The SSS image (Figure 10b) revealed a clear outline of the artificial reef, however, no obvious shadow area was observed. The sediment type in this area consisted of soft mud, gray–black, and fluid plastic (Figure 10c).

According to the high-resolution parametric SBP image (Figure 10d,e), the seabed reflection lines in the area across the AFR were smooth, with no obvious terrain undulations, which also confirmed the conclusion that the MBE data did not effectively reflect the morphology of the reef. However, acoustic shielding zones spanning 36 and 132 m were observed in the two images. The energy at the microsequence interfaces below the seabed and inside the sequences was significantly weaker than that in the adjacent unshielded zones, and the seabed of the shielded zone formed more obvious double and triple reflections. These characteristics indicated that although the reef materials were submerged below the seafloor, they still partially shielded sound waves and remained present on the seabed surface.

The measurement results showed that the reef materials sank into the mud layer (Sequence I₁) after deployment, preventing the formation of an effective reef structure. Therefore, from the perspective of seabed morphology and sediment characteristics, the IEASP experimental zone was unsuitable for AFR construction. SBP data showed that the thickness of the surficial layer in the IEASP zone was uniform, ranging from 2.1 to 4.8 m. The surface sediments were mainly soft mud (26 samples, 86.7%), with only 3 samples being sand (10%) and 1 sample being mud mixed with sand (3.3%). It was the latest soft fine-grained sediment deposited during the modern high water level period, lacking consolidation and compaction, resulting in that reef materials such as stone blocks or concrete modules easily sinking into the seabed after being deployed. It was difficult to form obvious undulations on the seabed, so it cannot be recognized on the MBE images. However, the change in the surface sediment caused by the deployment of reef materials led to the enhancement of the scattering intensity of SSS so that the outline of the AFR materials can be roughly recognized in the acoustic image. However, due to insufficient vertical relief of the AFR materials, no significant acoustic shadow was observed.

5. Conclusions

High-resolution acoustic images improve the observation of the AFR distribution characteristics, which facilitates calculating the number of modules and their size. In addition, digital information based on images enables the potential establishment of an AFR national database to be used to conduct comparative studies of post-deployment performances. Therefore, the precise mapping of the AFR site via multi-source acoustic images can aid in proper deployment planning in the future [11]. In this study, we conducted a comprehensive investigation using MBE, SSS, and SBP to explore the submarine topography, microgeomorphology, and Holocene strata of the AFR experimental zone around Xiaoguan Island in the western South Yellow Sea, China. Based on this investigation, we assessed the suitability of AFR construction. The high-resolution acoustic images and their interpretation in this study demonstrated their effectiveness in revealing underwater detailed features and supporting site evaluation for AFR development.

Three types of submarine geomorphology were identified in the surrounding waters of Xiaoguan Island: TF, UEABS, and IEASP. TF was located on the western bank of Xiaodao Bay, featuring a relatively gentle slope, which extended to a strip-like submarine sand embankment where waves broke below the sea surface. UEABS was found in the western waters of Xiaoguan Island and Rabbit Island, with significant changes in the seabed slope, along with strong erosion and obvious accumulation. IEASP was located in the eastern sea area of Xiaoguan Island and Rabbit Island, where the seabed was generally flat with gentle slopes and no significant uplifts or depressions.

We evaluated the suitability for AFR construction by examining the seabed characteristics, such as the topography, microgeomorphology, and sediment types in the experimental zones. The UEABS zone, represented by the Xiaodao Bay AFR experimental zone, featured gentle terrain slopes, stable microgeomorphology, and hard seabed sediments. After approximately seven months of deployment, the settlement was minimal, ranging from 0.15 to 0.78 m, with an average amount of 0.49 m. Following reef material deployment, the surrounding area showed weak or no erosion, with the reef structure forming an effective range and height, making it suitable for AFR construction. By contrast, the IEASP area in the eastern region of Xiaoguan Island had gentle terrain slopes and a stable microgeomorphology. However, the SBP data showed that the thickness of the surficial layer (I₁) was the latest soft fine-grained mud deposited during the modern high-water-level period, lacking consolidation and compaction, resulting in the reef materials easily sinking into the seabed after being deployed, preventing the formation of an effective reef structure. Therefore, for materials such as stone blocks or concrete modules, the IEASP area was unsuitable for AFR construction. Perhaps future multi-functional integrated reef modules with lightweight materials or new foundations will be suitable for this type of construction environment [18].

Finally, it should be pointed out that this article has certain limitations in analyzing the suitability of AFR construction solely based on the topography, geomorphology, and shallow geological features of the seabed. The lack of biological data makes it impossible for us to evaluate the bioaccumulation effect. Single-period data make it impossible for us to conduct long-term monitoring and verification. The lack of hydrological data makes it impossible to predict the overall scouring volume around reefs from numerical calculations. These limitations provide direction and guidance for our future research.