Baseline Habitat Setting for Future Evaluation of Environmental Status Quality of Jabal Ali Marine Sanctuary, Dubai, UAE

Abstract

Habitat mapping plays a crucial role in assessing marine protected areas (MPA) and implementing marine spatial management approaches. This study aims to present the spatial habitat distribution of the Jabal Ali Marine Sanctuary, considering the development projects implemented in its proximity. It serves as a reference for guiding conservation management efforts. The study focuses on in situ hyperspectral measurements of the optical properties of both the water column and the substrate. Additionally, a high density of geo-referenced spot checks were conducted, serving as sample points for ecological evaluation and ground-truthing. An “object-oriented” approach was adopted to generate the seabed map in two evaluated studies conducted in 2006 and 2017. While the 2017 survey identified 16 habitats, the 2006 study characterized only 10 habitats. These habitat maps serve as powerful tools for implementing mitigation measures and providing scientific support to mitigate the negative impact on the most crucial marine habitats within the context of a protected area management framework. Furthermore, monitoring the cover of the most important habitats provides an integrative indicator to maintain the good environmental status of the marine sanctuary. Based on this study, the information will be a reference for evaluating and synergizing the management approaches implemented by both the competent authority and the different stakeholders in the sanctuary.

1. Introduction

The continuous growth in population and the evolving development patterns of coastal cities have significantly altered the marine and coastal environment [1]. The Emirate of Dubai (UAE), boasting a 64 km shoreline length, stands out as one of the most prominent highly urbanized cities with a dynamic coastal area that has undergone remarkable changes in the past two decades [2]. From a humble fishing and trading post at the turn of the century to a sprawling metropolis, it is now among the most iconic engineered structures globally [3]. These engineered artificial habitats have become a prominent aspect of urbanization. However, besides being costly, they present challenges in ensuring the presence of biodiversity and effective management [4]. The interplay between natural and engineered habitats has sparked interest in studying the relationship, [5], and the mapping of these habitats has become a standard tool for conservation management and scientific research [6].

For mapping purposes, habitats are defined as spatially recognizable areas where physical, chemical, and biological environments can be distinguished from adjacent areas [7]. Habitat mapping technologies have rapidly improved in the last three decades and have been recognized as a fundamental tool for marine protected areas (MPA) evaluation and marine spatial management approaches [8,9]. In particular, benthic mapping is now acknowledged as foundational data for managing and conserving marine coastal areas, which are crucial for research and marine spatial planning [10,11,12]. Understanding benthic morphology can serve as a proxy through abiotic parameters [12,13,14] and may provide information about monitoring by reducing confounding habitat effects [15].

However, this management tool, intended to facilitate the transition to the sustainable use of marine resources, is limited to shallow depths and transparent water. This limitation could be addressed by multispectral remote sensing in combination with an in situ sampling of benthic habitats. Such data can provide valuable information for monitoring coastal water ecosystems [16,17,18,19].

Mapping the benthic habitats of Dubai is a challenging endeavor when considering short-term management due to the dynamic changes in its marine and coastal environments. Although Dubai is renowned for its massive development projects that have reshaped its coastline, it has one declared marine protected area (MPA). The marine protected (no-take) area of 21.65 km² (UNEP-WCMC (2022), known as the Jabal Ali Marine Sanctuary (JAMS) (Figure 1), is part of the wider terrestrial Jabal Ali Wildlife Sanctuary, which spans 76.69 km². Despite the presence of the Palm Jabal Ali and Waterfront complexes, along with other large coastal construction projects adjacent to the sanctuary, the waters still harbor unique biological diversity [20]. The Jabal Ali Marine Sanctuary boasts one of the highest assemblages of scleractinian corals in the UAE, with 34 coral species identified in the 1990s [21,22]. Additionally, most of the coral species observed in the UAE and listed in its species red list are present in the sanctuary. Furthermore, the Jabal Ali Marine Sanctuary has observed the presence of endangered species, such as the mottled eagle ray and Indo-Pacific humpback dolphin, while Hawksbill turtles nest on its sandy beaches and its seagrass beds are fishing nurseries and feeding grounds for endangered green turtles and vulnerable dugongs. A part of the coral reefs, mangroves, and seagrass meadows of the Jabal Ali Wetland Sanctuary has been designated as a wetland of international importance (Ramsar site) that covers 2.002 ha [23].

It is doubtless that the Jabal Ali Marine Sanctuary is one of the most significant marine protected areas in Dubai and the Arabian Gulf. Despite considerable governmental efforts towards conservation, ongoing projects have the potential to impact the sanctuary’s ecological importance in the future. Currently, three major projects are being developed adjacent to the sanctuary in various phases: (1) the Hassyan Clean Coal Independent Power Plant, which despite its name, is powered by natural gas rather than coal; (2) the Hassyan SWRO Independent Water Project, involving a desalination plant; and (3) the Waterfront and the Palm Jabal Ali, a mixed-used development project.

Given the ecological significance of the sanctuary and the Dubai’s government’s vision for sustainable development by 2040 [24], the characterization of this study area provides valuable insights as a baseline for assessing Good Environmental Status (GES). This baseline will be crucial for future evaluations of human activities’ impact on the sanctuary, facilitating the proposal of mitigation measures if necessary to maintain the GES. Spatial habitat distribution and changes in the coverage of key habitats can serve as indicators for the maintenance of the GES. A reduction in the spatial coverage of important habitats, such as the seagrass meadows and the coral reefs, would indicate the need for mitigation measures. Conversely, maintaining or even increasing the coverage of important habitats would serve as an indicator of good management.

In this context, the present study aims to showcase the spatial habitat distribution of the Jabal Ali Marine Sanctuary and serve as a reference in its conservation management. The baseline habitat characterization, spatial coverage map, and calculated area of each habitat provided in this article hold significant potential as essential management tools and a reference document for scientists in the future. They allow for an objective comparison of human impact and natural events, such as climate change-derived variations in the sanctuary.

This project holds particular significance as the aforementioned developments may potentially impact the sanctuary in the future. Therefore, the primary objective of the present study is to establish an operational spatial habitat distribution methodology applicable to the shallow waters of the Arabian Gulf. Furthermore, this study aims to achieve the following objectives: (1) present an updated baseline habitat characterization and spatial habitat map of the sanctuary; (2) assess changes observed over a 10-year period between 2006 and 2017; and (3) quantify the area covered by each habitat, providing a benchmark for future studies.

2. Materials and Methods

The study area for mapping and ecological/environmental surveying is situated between the western edge of the Jabal Ali Palm breakwater and Ras Ghantoot, extending from the high tide water line to 2.7 km offshore (Figure 1).

2.1. Satellite Image Pre-Processing

The preparation of the WorldView-2 data involved radiometric, atmospheric, water column correction, and orthorectification [25]. Obtaining accurate quantitative information from WorldView-2 data requires the conversion of raw digital numbers to reflectance. Sensor characteristics, illumination geometry, and atmospheric conditions affect the signal received by a multispectral satellite sensor. For this study, an empirical line method was employed to calibrate the eight-band multispectral WorldView-2 image [25,26,27]. The correction for illumination geometry and atmospheric attenuation in the imagery was achieved by developing a non-linear relationship between the top of atmosphere spectral radiance and surface reflectance values measured from field targets. An accuracy assessment was undertaken by comparing image reflectance values against the surface reflectance values of these targets [27]. The overall agreement in reflectance, based on the root mean square error for the eight bands, ranged between 0.16 and 1.38%, with the greatest variance in the near-infrared bands. The orthorectification of the Worldview-2 imagery was undertaken using the sensor’s rational polynomial coefficients and ground control points collected using a differential GPS [28].

The footage was recorded using high-definition underwater cameras with overlays, capturing the prevailing current direction. A combination of underwater videos and photos was utilized to audit each seabed’s character from 45 sites arranged in a typical sampling grid. To adequately represent the ecological diversity of the seafloor in the study area, 16 habitat classifications were developed based on the collected field data. Detailed descriptions and photographs of each habitat class are provided in the Supplementary Material.

2.2. Satellite Imagery

Commercial earth observation satellite WorldView-2, operated by Digital Globe Corp, was utilized in this study. This satellite features five visible spectral bands, labeled “coastal”, “blue”, “green”, “yellow”, and “red” in increasing wavelength order, along with three infrared bands, labeled “red edge”, “near-IR 1”, and “near-IR 2.” In comparison to similar satellites like Quick Bird, which typically offer four multispectral bands, WorldView-2 strategically incorporates all eight bands, providing 1.8 m² resolution pixels and an unmatched capability to monitor shallow seabeds. Moreover, the WorldView-2 satellite offers quicker revisit periods than the QuickBird satellite [26]. Due to its higher orbit, the sensor can gather repeated data at look angles extremely close to the nadir.

2.3. Accuracy Test

To gather accurate assessment data and an inventory of benthic types within the study area, ground-truthing was conducted on the seafloor throughout the study area. Spot checks were carried out using a combination of snorkel and scuba, depending on water depth and turbidity. During each check, the seafloor was observed from an altitude of 1 m at the nadir, and a description of seabed morphology and species composition was recorded. Each spot was precisely located using DGPS for later retrieval of positioning concerning the imagery. The spot check data were subsequently categorized into broad substrate categories and compiled into a database of ground verification data for accuracy assessment purposes.

In line with the standard reporting conventions established in the literature analysis, classification accuracy was assessed using a standard error matrix. A matrix-based accuracy assessment was conducted to evaluate the reliability of the map product. This assessment included determining the overall accuracy of the classification, errors of omission, errors of commission, producer’s accuracy, and consumer’s accuracy. These measures quantitatively assessed how well the image classification was conducted and the level of accuracy achieved in the final map. A statistical accuracy assessment of the map product was performed against the 557 GPS-constrained ground observations obtained during the survey.

In the classification, both pixel-based classification for the 2006 survey and object-oriented classification for the 2017 survey were employed (Figure 2) [29].

2.4. Field Sampling for Habitat Classification Carried out in Jabal Ali Marine Sanctuary in 2006 Band 2017

2.4.1. Sampling Survey 2006

The data obtained from the 2006 survey (Figure 3,

Table 1. Breakdown of the area covered by each of the mapped assemblages.

Habitat Class	Description		Total Pixels
Dense Brown	Dense brown algal assemblage and sparse massive corals		514,860
Sparse Brown	Sparse brown algal assemblage and sparse massive corals		743,629
Sparse Porites	Sparse Porites assemblage with high algal cover		10,403
Sand with Algae	Sand with sparse macro-algal stands		89,947
Sand with corals	Sand and/or mud with coral cover		1046
HG Sparse Corals	Hardground with sparse Porites and faviid community		88,873
Dense Coral	Dense coral framework		3722
Sand	Bare sand		976,681
Mud	Uncolonised mud and silt		165,178
Seagrass	Mixed seagrass assemblage		34,109
Bare HG	Low-relief bare hardground		44,959

) were utilized to compile an ecological inventory of benthic habitats. Points were assigned color codes corresponding to their respective broad habitat categories. The survey was conducted from more than 570 spot observations of the seafloor across the study area. These observations aimed to gather accurate assessment data and establish an inventory of benthic diversity within the area. Spot checks were carried out using a combination of snorkeling and scuba diving. At each check spot, the seafloor was observed from an altitude of 1 m at the nadir, and a description of seabed morphology as well as its species composition was recorded. Digital GPS was used to precisely locate each spot for later reference. Photographic records were used to calibrate the estimated levels of substrate coverage, especially in areas with complex seafloor mosaics [30]. The spot check data were then categorized into eleven broad substrate categories and compiled into a database of ground verification data for accuracy assessment.

The breakdown of the mapped assemblages, along with the pixels counted in the process, is presented in Table 1.

2.4.2. Sampling Survey 2017

The 2017 study analyzed 45 locations arranged in a standard sampling grid (Figure 4). Each sampling location and its observations are briefly described in

Table 2. Geographic locations of the 45 stations from which seabed character was validated in the field. Habitat classes were attributed to each ground control point following the inspection of field videos and digital underwater photographs. Images of each habitat class can be found in Supplementary Materials Annex 1.

Stations	Longitude	Latitude	Habitat
1	54.897° E	24.9129° N	Sparse brown algal assemblage and sparse massive corals
2	54.883° E	24.9234° N	Bare sand
3	54.8719° E	24.933° N	Bare sand
4	54.901° E	24.924° N	Hardground with sparse Porites and faviid community
5	54.896° E	24.930° N	Sparse brown algal assemblage and sparse massive corals
6	54.88° E	24.937° N	Sparse brown algal assemblage and sparse massive corals
7	54.878° E	24.945° N	Mud and silt
8	54.903° E	24.936° N	Construction submerged with corals
9	54.893° E	24.943° N	Construction submerged with corals
10	54.891° E	24.946° N	Construction submerged with corals
11	54.913° E	24.929° N	Mixed seagrass assemblage
12	54.915° E	24.936° N	Mud and silt
13	54.919° E	24.943° N	Caulerpa meadows
14	54.912° E	24.954° N	Dredged channels
15	54.929° E	24.942° N	Dense coral framework
16	54.922° E	24.949° N	Bivalves on sand
17	54.933° E	24.948° N	Mixed seagrass assemblage
18	54.925° E	24.957° N	Mixed seagrass assemblage
19	54.919° E	24.963° N	Caulerpa meadows
20	54.908° E	24.968° N	Dredged channels
21	54.939° E	24.950° N	Mixed seagrass assemblage
22	54.936° E	24.956° N	Dense brown algal assemblage and sparse massive corals
23	54.944° E	24.959° N	Mixed seagrass assemblage
24	54.934° E	24.965° N	Mixed seagrass assemblage
25	54.945° E	24.967° N	Mixed seagrass assemblage
26	54.92° E	24.974° N	Bivalves on sand
27	54.957° E	24.964° N	Hardground with sparse Porites and faviid Community
28	54.945° E	24.973° N	Bivalves on sand
29	54.936° E	24.981° N	Bare sand
30	54.964° E	24.970° N	Dense coral framework
31	54.953° E	24.980° N	Construction submerged with corals
32	54.980° E	24.972° N	Caulerpa meadows
33	54.971° E	24.976° N	Construction submerged with corals
34	54.965° E	24.979° N	Construction submerged with corals
35	54.954° E	24.994° N	Construction submerged with corals
37	54.974° E	24.995° N	Mud and silt
38	54.967° E	24.994° N	Mud and silt
39	55.00° E	24.988° N	Sand with sparse macro algal stands
40	55.007° E	25.007° N	Caulerpa meadows
41	55.011° E	25.020° N	Caulerpa meadows
42	55.0244° E	24.999° N	Construction submerged with corals
43	55.0188° E	25.0254° N	Construction submerged with corals
44	55.031° E	24.997° N	Dense coral framework
45	55.039° E	25.005° N	Caulerpa meadows

, while

Table 3. Habitat classifications were constructed after examining the field data collected from 45 locations.

Habitat Class	Description	Ground-Truth Image
1 Dense brown algal assemblage and sparse massive corals	A low-relief hardground (cap rock) terrace colonized by a dense and diverse assemblage of turfing and fleshy algae.
2 Sparse brown algal assemblage and sparse massive corals	A low-relief hardground (cap rock) terrace colonized by a sparse assemblage of turfing and fleshy algae. Occasional live massive corals and sponges are also present.
3 Sand with sparse macro-algal stands	An expanse of macroalgae in which thalli are interspersed by unconsolidated sediment with predominantly fine-grained trapped carbonate sediments.
4 Hardground with sparse Porites and faviid community	Areas of sparse coral, typically an average of about 10%, were defined as less than 25% the cover of substratum with living colonies. Corals can occur in two distinct assemblages
5 Dense coral frameworks	Singled out as the most important habitat, the category “dense corals” is made up within the Jabal Ali Marine Sanctuary primarily by densely packed (30–70%) coral cover
6 Bare sand	Expansive sheets of mixed-grain skeletal and carbonate sand with variably characterized bedforms and sparse patches of macroalgae
7 Mud and silt	Accumulations of fine-grained mud-to-silt in tranquil areas. The sediment surface is variably covered by blue-green algal mats. Infauna is limited because of local anoxia developing within a few centimeters of the sediment surface. Typically highly turbid
8 Mixed seagrass assemblage	The mixed assemblage of seagrass in a dense meadow is usually dominated by the species Halodule uninervis, Halophila ovalis, and Halophila stipulacea. Leaf densities vary and in some areas are >100 leaves per 10 cm−2
9 Sand and/or mud with corals	Expansive sand sheets are variably colonized by macroalgal stands. Scattered live corals, coral skeletons, and coral rubble locally accumulate and provide a habitat for a broader assemblage of turf algae. Live corals typically consist of faviids, siderastreids, and poritids. Occasional pockets of seagrass are also present but at low density.
10 Sparse faviid communities with high algal cover	A low-relief hardground interspersed locally by sand pockets. The hardground hosts a mixed assemblage of flesh macroalgae, Padina and Sargassum, in particular, interspersed by a sparse assemblage of faviid corals.
11 Construction submerged with corals	Corals have variably settled on all artificial marine structures in the study area. The most abundant settlers are faviid corals, primarily of the genus Favia and to a lesser extent, Platygyra and Cyphastrea. Porites and Siderastrea recruits are also relatively common.
12 Construction submerged—bare	Like the previous class but with a limited settlement of macroalgae and corals
13 Bivalves on sand	Expansive sheets of coarse-grained skeletal sand colonized by an assemblage of dense bivalves, Chama reflexa, Spondylus marisrubris, and the pearl oyster, Pinctada radiata, primarily. Accumulations of bivalve shell detritus provide a stabilized habitat for fleshy macroalgae, Padina and Sargassum, primarily, and the black sea-squirt, Phallusia nigra.
14 Caulerpa meadows	Caulerpa racemosa is a non-native fronded seaweed in the Arabian Gulf. Stands of Caulerpa persist in restricted areas, within the Palm particularly, where the alga forms monospecific meadows of variable density in sandy to muddy habitats.
15 Dredged channels	Areas of the artificially deepened seabed, typically at depths too deep for the benthos to be reliably resolved from the satellite. The depth is >8 m, and the seabed is not regularly agitated by swell waves, which allows for the accumulation of mud-to-silt-sized sediment, often inhabited by thick mats of blue-green algae.
16 Construction emergent	Artificial land created in the offshore zone. While the top of these features is barren or urbanized, the littoral fringe (the habitat immediately adjacent to the waterline) is characterized by a distinctive dark-colored habitat that is formed by blue-green algae preferentially inhabiting the spray zone.

provides descriptions of habitat classes along with visual material. Following the acquisition of field and remote sensing data, an “object-oriented” approach was adopted to generate the seabed map. This approach, which leverages the high fidelity of the WorldView-2 data more effectively than the strategy used in 2005–2007, involved segmenting image data into objects at multiple scale levels [24]. Objects were assigned class rules based on spectral signatures, shape and textural and contextual relationships. This method surpasses pixel-based classification by utilizing spectral, spatial, textual, and contextual information to describe each class during classification processing and also performs image segmentation into groups of pixels [25,31].

Upon reviewing the gathered field data, 16 habitat classifications were identified (Table 2 and Table 3), each accompanied by comprehensive descriptions and images (Supplementary Materials Annex 1). This comprehensive documentation aims to accurately depict the ecological condition of the seabed within the research area.

3. Results

3.1. Habitat Map Distribution

The 2006 survey revealed a notable predominance of carbonate sand bares (Figure 5). A nearly equal split between hard- and soft-dominated substrates was observed. Bare sand constituted the highest proportion of pixels, representing 36% of all classified pixels, while dense corals accounted for only 0.1% of the pixels. However, when considering all areas with corals combined, the coral community in the reserve accounted for 3.5% of the investigated area.

In addition to an evident dominance of bare carbonate sand in the 2006 survey, there was a nearly equal split between hard- and soft-dominated substrates, as illustrated in Figure 5. Live and dense coral cover at the time of the survey comprised 3% of total areal coverage, including potential coral habitat which is largely sediment-free and thus conducive for coral settlement that covered half of the studied area. The areal coverage map depicted a nearly equal co-dominance between soft sediment-covered habitats supporting an extensive stand of seagrass and macro-algae (41.2%) and areas of hardground characterized by algal cover (50.9%). In the majority of cases, these hardground areas supported at least sparse coral coverage.

The 2017 habitat map (Figure 6) illustrates artificial islands surrounded by a sporadic halo of mud, variably overlapping offshore sand sheets. Beyond the Palm and Waterfront structures, the seabed primarily consists of unconsolidated sand sheets interspersed with patches of sparse seagrass and linear outcrops of hardground. Ground control data confirms that these hardground patches are densely populated by fleshy seaweed, with varying densities of live coral colonies. Sparse seagrass patches are often found sheltered behind the hardground patches, where their slight topographic relief allows for the accumulation of stable pockets of soft sediment conducive for seagrass growth.

The small-scale gradients along the shore, which are influenced by constructed structures (Figure 7), exhibit a narrow halo of the construction submerged with coral rims encircling the Palm Jabal Ali and the Waterfront. The dredged channels appear filled with mud (Figure 7A,C,D). As one moves away from the construction areas, the seabed maintains its typical composition of offshore sand sheets, colonized by sparse seagrass, Caulerpa, and bivalves, interspersed with patches of coral gal-covered hardground. Habitat diversity peaks in the nearshore zone (Figure 7B) where dense seagrass meadows thrive within sand pockets amidst a matrix of hardground featuring mixed Porites assemblages. This zone serves as a crucial habitat where stands of live coral have flourished.

Among the 16 habitats characterized during the 2017 survey, bare sand emerged as the largest habitat covering 21.07 km² (25.3%), while construction-submerged areas with coral presence exhibited the smallest area coverage at 0.32 km², n (0.4%) (Figure 8). The majority of coral formulations within the reserve are integrated with the other habitats, making accurate coral coverage estimation challenging. Focusing on identified dense coral habitats, minimum coral coverage extends to 0.58 km² (0.7%) within the reserve. In contrast, mixed seagrass meadows occupied a significantly larger area, totaling 4.79 km² (5.7%), which is eight times the coverage of the dense coral habitat. Human-induced impacts are evident in the prevalence of mud and silt (12.04 km², 14.5%), primarily associated with the construction of the Palm Jabal Ali and the Waterfront islands. Additionally, construction-emergent habitats (12.04 km², 14.4%) and the dredged channels (3.68 km², 4.46%) collectively account for 33.4% of the protected area, indicating a substantial coverage of anthropogenically modified habitats.

Conversely, key habitats crucial for biodiversity conservation, such as mixed seagrass assemblages (4.79 km²), dense coral framework (0.58 km²), dense brown algae assemblage, and sparse massive corals (4.79 km²), collectively reach a coverage of 10.16 km².

3.2. Comparison of Relative Habitat Contribution of the 2017 and 2006 Habitat Map

In contrast to the 16 habitats identified in the 2017 survey, only 10 habitats were characterized in the 2006 study. Figure 9a–c summarizes the comparable habitat data between both surveys and their change in coverage over the 10-year time period. The total coverage of the comparable habitats between both habitat maps was 41.96 km² in 2006 and 58.20 km² in 2017.

Bare sand and mud/silt habitat types dominated in both years, covering 15.25 km² and 21.01 km², respectively, in 2006 and making up 36.98% and 36.02% of the covered area in 2017 (Figure 9a,b). This dominant and ubiquitous habitat only decreased by 0.7% and did not change in the 10-year time period. The highest decreases (10–15%) were found for dense brown algal assemblage and sparse massive corals (DBA) as well as for sparse brown algal assemblage and sparse massive corals (SBA) (Figure 9c). Thus, brown algae and corals seem to have suffered the most due to human modifications in the study area. In contrast, the highest increases (14.5% and 6.9%) were found for mud and silt (MS) as well as for mixed seagrass assemblages (SG) (Figure 9c).

4. Discussion

Remote sensing and habitat maps are widely used in coral reef conservation efforts [32], the localization and selection of new MPAs [33], and the design of climate-resilient MPA networks [34]. Once established, MPAs can use habitat maps as reference benchmarks for effective management and to assess the impact of human activities, such as offshore nutrient spills, brine discharges, and climate change-driven warming and acidification on benthic fauna [35,36,37]. Although in the past 10 years, there have been very few anthropogenic activities in the JAMS, the results from the habitat map in this study highlight the habitats that require management interventions. The Waterfront and the Palm Jabal Ali have been halted, while the power plant is still in its construction stage and the desalination plant is still in its earliest land preparation stage. The impact of operationalization of these projects will have a potential effect on the coral reefs and other important habitats in the sanctuary.

The comparison of habitat maps from 2006 and 2017 in the study area spanning Ras Ghantoot to Jebel Ali reveals the prevalence of corals as a prominent characteristic. The abundance of hard substratum is encouraging, as it suggests that coral settlement dynamics are not limited by substrate availability, potentially supporting a significant coral resource. Coral density is influenced by sand movement across the hardground, with areas of higher topography typically experiencing less sand accumulation and consequently facilitating better coral growth.

While the increase in the contribution of the dense coral framework (DCF) from 2006 to 2017 is relatively small at 0.85%, this expansion from 0.06 km² in 2006 to 0.58 km² in 2017 signifies an almost tenfold increase in the area covered by this crucial habitat for diversity conservation. Throughout the area, corals tend to co-exist closely with algae, particularly with the Padina–dominated and the Sargassum/Hormophysa–dominated assemblages.

Competition between coral and algal communities arises due to their similar ecological requirements [38,39]. However, despite the slow growth of corals, the constant presence of corals prevails over the seasonal and inter-annual variability of algae. Algal communities exhibit frequent shifts, readily abandoning or colonizing areas swiftly. These ecological dynamics were observed commonly in the study area. There is no discernible pattern to the spatial distribution of these patches, and it has to be assumed that it is largely chance settlement that determines their presence or absence.

On the other hand, due to their density and spatial extent, macroalgal meadows are one of the most important eco-elements in the study area. Algal meadows usually harbor similar fauna to seagrass meadows. The area between Ras Ghantoot and the Palm Jabal Ali where seagrass meadows are relatively rare and sparse, makes the macroalgae meadows the primary habitat for all organisms requiring plants as their shelter (many of the bivalves, foraminifera, worms, crustaceans, etc.). Due to their capability of rapidly taking up and metabolizing nutrients and removing them from the water column, macroalgae are significant ecological indicators [40]. However, it is imperative to pay attention to brown algae and coral habitats in the future, as the persistence of this trend could result in the disappearance of these crucial habitats from the protected area.

Additionally, the deposition of fine sediments in proximity to man-made structures is a common occurrence and can be attributed to two primary reasons. First, these structures create sheltered environments that are conducive to the accumulation of mud in areas that would otherwise be cleaned by swell waves and bottom currents. Second, the dredging and filling process suspends sediment particles of all sizes in the water column, facilitating the drifting and settling of mud. These outcrops elongate the coastline, typically aligning parallel to the easterly longshore current prevalent in this Gulf region. [41,42]. Such observations are the dynamics that are occurring in the study area. The artificially engineered structures have transformed into a unique habitat [5], as classified in the 2017 survey, which has influenced the dynamics of the sediments in the study area and increased the areal coverage of the mud and silt compared with the 2006 study. A notable trend emerged wherein brown algae and coral mixed assemblages were substituted by mud silt and seagrass assemblages. This shift can be attributed to construction activities, which impede coastal currents, impacting corals and leading to the accumulation of fine particles like mud. Conversely, these altered conditions may facilitate the growth of seagrass assemblages. While muddy habitats may not be of immediate conservation interest, the proliferation of seagrass communities holds the potential for fostering high biodiversity, including epifauna and iconic marine mammals such as dugongs [43]. The increase could be an opportunity for habitat managers to characterize infaunal community-habitat relationships [44]. This study provides a higher resolution of the habitat dynamics in the sanctuary and is an essential management tool amidst the anthropogenic pressures in the study area.

The 2017 habitat map, along with the observed habitat coverage changes from 2006 to 2017, establishes a baseline for assessing future conservation efforts and potential degradation in the study area. This is particularly significant considering the higher resolution of the 2017 habitat map and its completion prior to the implementation of planned human activities, such as #1 the Hassyan Clean Coal Independent Power Plant, #2 the Hassyan SWRO Independent Water Project, and #3 the Waterfront and the Palm Jabal Ali. The impact assessments of these projects have highlighted changes in the ecological dynamics in the area including water quality, water temperatures, current dynamics, habitat, and species conditions, among others. The habitat map can provide a reference for prioritizing conservation measures. The complexity of the projects can be arduous. However, they are essential and strategic, particularly in their contributions to the attainment of the Dubai 2040 vision. Therefore, the provided habitat map serves as a vital reference benchmark for evaluating the impact of these aforementioned projects on habitat preservation and biodiversity conservation within the protected area [45].

Overall, data indicates significant changes in the distribution of benthic habitats and communities between 2006 and 2017, with certain habitat categories witnessing an increase in coverage while others exhibited a decline.

Monitoring the habitat cover of these crucial habitats serves as an integrated indicator for maintaining the good environmental state of the marine sanctuary. In the event of a significant reduction in an important habitat, a prompt investigation into the cause is essential, followed by the implementation of further mitigation measures and conservation approaches.

5. Conclusions

The presented habitat maps provide a robust tool for implementing mitigation measures and offering scientific support to identify successful conservation strategies while preventing negative impacts on vital marine habitats such as the following:

(i)

Vast meadows of seagrass beds are known to be crucial as fish nurseries and feeding grounds for dugongs and green turtles. The latter species has sporadically utilized seagrass regions in this vicinity for feeding in recent years. Hawksbill turtles primarily lay their eggs on the sandy beaches between Ghantoot and Palm Jabal Ali.

(ii)

Small yet dense clusters of hard coral communities (>40% cover), featuring sensitive table corals (Acropora spp.).

(iii)

Mixed ecosystems with a variety of seasonal brown macroalgae and sporadic but significant concentrations of poritid and faviid hard coral colonies.

(iv)

Abundant fish communities connected to high-relief hard bottom habitats found in breakwaters with coral colonization and natural coral reefs.

(v)

From the biodiversity conservation standpoint, the loss of brown algae and sparse corals from muddy habitats is concerning. However, the extension of seagrass assemblages and the slight increase in the areal coverage of dense coral frameworks are positive signs for the conservation effort.

It is noteworthy that only 12.1% of the protected area comprises habitats essential for biodiversity conservation. Given this limited coverage, preserving the ecological integrity of seagrass and coral communities becomes paramount for their protection and conservation.

In summary, this study demonstrated evident changes between the 2006 and 2017 surveys. The latest 2017 habitat map serves as baseline information for assessing the future impacts of human activities within the protected area and evaluating its management strategies.