Diversity in the geomorphology of shallow-water...

Geomorphology xxx (2014) xxx–xxx

GEOMOR-04695; No of Pages 11

Contents lists available at ScienceDirect

Geomorphology

j ourna l homepage: www.e lsev ie r .com/ locate /geomorph

Diversity in the geomorphology of shallow-water carbonate depositional systems inthe Saudi Arabian Red Sea

Gwilym Rowlands a,⁎, Sam Purkis a, Andrew Bruckner b

a National Coral Reef Institute, Nova Southeastern University, Dania Beach, FL 33004, USAb Living Oceans Foundation, 8181 Professional Place, Suite 215, Landover, MD 20785, USA

⁎ Corresponding author.E-mail address: [email protected] (G. Rowlands).

http://dx.doi.org/10.1016/j.geomorph.2014.03.0140169-555X/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Rowlands, G., et aArabian Red Sea, Geomorphology (2014), ht

a b s t r a c t
a r t i c l e i n f o
Article history:Received 8 November 2012Received in revised form 18 September 2013Accepted 5 March 2014Available online xxxx

Keywords:Red SeaCarbonate systemCoral reefMorphologyGISRemote sensing

Coral reefs and their associated accumulations of carbonate sediment adopt particularly complex planformgeometries atop the coastal shelf of the Saudi Arabian Red Sea. By assembling 95,000 km2 of remote sensingdata into a GIS, this study aims to relate the morphology of these shallow-water depositional environments toprocesses that sculpt the coastal zone. A typology that sorts carbonate systems into end-members on the basisof their morphology and relationship to the coastline is developed. The resulting GIS was interrogated for spatialpatterns in the distribution and abundance of the end-members. While several depositional morphologies arepresent throughout the length of the Saudi Arabian Red Sea, the occurrence of others is restricted to narrow re-gions of latitude. Such differences in distribution can be explained in process-terms by the rift tectonics of the RedSea basin, spatial variability in the presence of sub-seafloor evaporites, and the input of siliciclastic detritus ontothe coastal shelf via wadis. This paper provides a foundation for understanding the morphological diversity ofshallow-water carbonate systems in both the modern ocean and rock records.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

The classification of coral reef morphology has a long history,startingwith Darwin (1842)who defined the three broad classes ‘fring-ing reef’, ‘barrier reef’, and ‘atoll’, terms still used to this day. Laterworkshave strived to capturemore subtlemorphological variations. For exam-ple, the typology of Hopley et al. (1989) splits the Great Barrier Reef(GBR) into nine classes while more than a thousand reef types havebeen defined by Andréfouët et al. (2006) at the global scale. The spanin number of classes presented by these studies is indicative of themorphological complexity of coral reef systems. Coral reefs are not byany means made of corals alone. Many other calcareous organisms,both animal and plant, contribute to the volume of a reef (Hart andKench, 2007). A “carbonate system”, as defined in this paper, thereforeconsiders the shallow-water (b30 m) accumulation of both reefs andassociated carbonate sediments. Carbonate systems can be partitionedon the basis of planform morphology of reefs and sediments intodistinct ‘morphological end-members’.

The classification of carbonate systems is relevant to a number ofdisciplines. Coral reef managers often lack knowledge on the diversityof carbonate systems over which they preside, their size, and howcarbonate systems are connected with one another (Andréfouët et al.,2006). A knowledge deficit is problematic because the morphologyof carbonate systems influences the biogeographic patterns of reef

l., Diversity in the geomorphtp://dx.doi.org/10.1016/j.geom

associated species (Bellwood and Hughes, 2001), and is an importantconsideration in the design of marine protected areas (McLeod et al.,2009). Modern carbonate systems also serve as analogs for understand-ing ancient carbonate systems (Montaggioni et al., 1986; Rankey, 2002;Purkis et al., 2007; Harris et al., 2011; Purkis et al., 2012a, 2012b; Harriset al., 2013). The definition of carbonate systems can also be used tostructure comparative analyses. For example, carbonate systems thatare similar in their appearance may be expected to exhibit similar car-bonate production values (Perry et al., 2008; Leon and Woodroffe,2013). Finally, carbonate system morphology provides much neededcontext for more detailed mapping of seabed habitats at fine spatialresolution (Andréfouët et al., 2006; Rowlands et al., 2012).

Diversity in themorphology of carbonate systems can sometimes beexplained in terms of underlying tectonics, the action of wind, wavesand currents, changes in the position of sea-level, antecedent topogra-phy, and climate (Darwin, 1842; Fairbridge, 1967; Maxwell, 1970;Purdy, 1974; Hopley, 1982; Bosence, 2005; Purkis et al., 2010; Harriset al., 2011; Purkis et al., 2012a; Leon and Woodroffe, 2013). The mor-phology of a carbonate system is not static through time. For instance,considering the carbonate systems of the Australian GBR, Hopley(1982) recognizes an evolutionary progression through differentmorphological end-members from juvenile (unmodified antecedentplatforms, submerged reefs, irregular patch reefs), tomature (crescenticreefs, lagoonal reefs) and finally to senile (planar reefs). Morphology istherefore a temporally dynamic property of a carbonate system.

This study considers the morphological diversity of carbonate depo-sitional systems in the Saudi Arabian Red Sea (Fig. 1), which Bosence

ology of shallow-water carbonate depositional systems in the Saudiorph.2014.03.014

http://dx.doi.org/10.1016/j.geomorph.2014.03.014

mailto:[email protected]


http://www.sciencedirect.com/science/journal/0169555X


Fig. 1. a) The Red Sea basin. Note rotated north axis. Letters and boxes indicate position of image subsets shown below; b) subsets of QuickBird satellite imagery that illustrates the sixteenmorphological end-members of the Saudi Arabian Red Sea typology (Fig. 2). In each image, the horizontal scale bar represents 1 km. First row (A–E) are carbonate systems with an‘Attached Form’, second row (F–I) are carbonate systems in the ‘Tower Group’, third row (J–M) are carbonate systems in the ‘Platform Group with an Aggraded Edge’ and forth row(N–P) are carbonate systems in the ‘Platform Group with a Non-aggraded Edge’. Names of Carbonate Systems in key are abbreviated; see Fig. 2 for details.

2 G. Rowlands et al. / Geomorphology xxx (2014) xxx–xxx

(2005) broadly explains in terms of the input of siliciclastics ascontrolled by the hyper-arid Arabian climate, combined with the actionof rift-related tectonics. Despite the relatively small size of the Red Sea, awide variety of carbonate systems have been documented (Sheppardet al., 1992; Dullo andMontaggioni, 1998; Bosence, 2005). Nonetheless,there has not been an effort to collate thismorphological diversity into a

Please cite this article as: Rowlands, G., et al., Diversity in the geomorphArabian Red Sea, Geomorphology (2014), http://dx.doi.org/10.1016/j.geom

logical scheme of classification, nor to analyze the spatial distribution ofthis morphological diversity, as, for example, has been undertaken inthe GBR (Hopley et al., 1989) and Torres Strait (Leon and Woodroffe,2013), Australia.

This study will firstly propose a robust typology for identifyingmorphological end-members of the carbonate systems of the Saudi



3G. Rowlands et al. / Geomorphology xxx (2014) xxx–xxx

Arabian Red Sea. Second, through the development of a geographic in-formation system (GIS), the spatial distribution of the morphologicalend-members will be quantitatively analyzed for trends and patterns.Thirdly, the spatial distribution of the coastal shelf, sub-seafloor evapo-rite thickness, and terrestrial drainage basins feeding into the SaudiArabian Red Sea are mapped into the GIS. Lastly, this study aims to ex-plain the spatial distribution of the carbonate system morphologicalend-members in process-terms.

2. Material and methods

2.1. Development of a typology for Saudi Arabian Red Sea carbonatesystems

To quantify the diversity of carbonate systems, a typology that sortsSaudi Arabian Red Sea carbonate systems into morphological end-members is developed (Figs. 1 and 2). This contrasts typologies gearedto different goals, such as Andréfouët et al.'s (2001) hydrodynamic clas-sification of Pacific atolls. The typology of this paper is built upon thevisual interpretation of satellite and aerial imagery (Fig. 1). Remotesensing data have been shown to be a viable and efficientmeans to clas-sify carbonate systems in other regions (Hopley et al., 1989; Purkis andPasterkamp, 2004; Andréfouët et al., 2006; Harris et al., 2011; Purkiset al., 2012a; Leon and Woodroffe, 2013). Because the waters of the

Fig. 2. Typology of Saudi Arabian Red Sea Carbonate Systems. Boxed text represents different tboxes represent the sixteen morphological end-members of the decision tree (Fig. 1).


Red Sea are so clear, seabed morphology can be appraised down todepths of at least 30 m using satellite imagery. This study draws on a95,000 km2 collation of QuickBird (2.5 m pixel; DigitalGlobe) andLandsat (30 m pixel; USGS) satellite imagery acquired under clear andcalm sea conditions (Bruckner et al., 2011; Rowlands et al., 2012). Thisvast archive of remote sensing imagery encompasses all shallow watercarbonate systems of the Saudi Arabian Red Sea. As detailed byBruckner et al. (2011), in the rare cases where the quality of these im-ages did not allow the shallow seabed to be appraised, the clarity ofthe imagery was improved through the processing of atmospheric andwater surface effects. The satellite data were augmented with thepublic-domain imagery available as an ESRI ArcGIS base-layer andthrough Google Earth, as well as georeferenced British and SaudiArabian nautical charts.

The typology of Saudi Arabian Red Sea carbonate systems takes theform of a decision tree (Fig. 2). The criteria for branch-points in thetree are objective and reproducible. Criteria include differences in thespectral content and texture of satellite image pixels, measurementsof the size and shape of carbonate systems as measured in the GIS,andwater depth assessed fromnautical charts. At the top of the decisiontree are two ‘Forms’, which split to four ‘Groups’, under which nest thesixteen morphological end-members (Fig. 1b).

The logic of the decision tree proceeds as follows (Fig. 2). The ‘edge’of a carbonate system is defined by its boundary with either land or

ypes of Carbonate System, while unboxed text is a decision point in the typology. Shaded



Fig. 3. GIS grid cells: a) Grid extends across all Saudi Arabian Red Sea Carbonate Systemsand is used to assess their distribution and abundance; b) Grid overlain on a Landsatsatellite image of a Planar Platform System. GIS grid cells measure 1 × 1 km.


deepwater. The system's ‘interior’ lies inboard of the edge. ‘Attached’ or‘Detached’ represents a logical place to start the typological hierarchysince it distinguishes systems where the main geological foundation –

coastal shallows – is apparent, from offshore systems that require anantecedent platform or mechanism of geologic uplift to explain theirformation. The distinction also separates attached systems which re-ceive terrestrial sediments from detached systems which do not. A car-bonate system is defined ashaving a ‘Detached Form’ if separated fromalandmass by water deeper than 50 m, the most appropriate contouravailable in the nautical chart data. Landmass in this context includesislands N3 km2, a threshold which discriminates rock islands fromsmaller sand cays. The ‘reef terrace’ refers to the area between the sea-ward edge of the system and the coastline. The reef terrace may there-fore encompass a number of recognizable features such as reef crests,back-reefs, shallow lagoons, patch reefs, small islands, and sand cays.The term ‘fringing’ is used in naming both narrow system morphologi-cal end-members (Fringing, Fringing with embayment), which have asimple morphology consistent with the common use of the term(Smithers, 2011). Various thresholds were tested and 2 km was foundto be the most logical to differentiate these simple fringing morphol-ogies in the GIS from more complex forms. Though both ‘DendriticSystems’ and ‘Unorganized Systems’ exhibit shallow water extendingoutboard from the coastline, in the latter case, reef growth is lesspronounced and often punctuated by channels running parallel to thecoastline.

Within the typology (Fig. 2), carbonate systems with a DetachedForm fall into two groups. Detached Form carbonate systems that arecircular in plan-view fall into the ‘Tower Group’. To facilitate rapid as-sessment of systems in the GIS, circular was defined as having roundededges and a ratio of length towidth of b3. The remaining carbonate sys-tems with a Detached Form are encompassed by the ‘Platform Group’.Carbonate systems in both the Tower and the Platform Groups fit intothe typology based on the characteristics of their edge and interior.The approach to classifying morphological end-members outlined inthis paper therefore follows that of Hopley (1982) and Harris andVlaswinkel (2008) from whom many terms in this work are taken.For example, the term “aggraded” relates to the situation where reefand/or sediment has filled the available accommodation space andbuilt to, or above, sea-level. Features of a carbonate system that areaggraded include reef crests, shallow reef flats, back reefs, small islandsand sand cays. Because of the differential absorption of light through thewater column, aggraded coral reef frameworks and sediments can beidentified by a strong spectral response in the red band of the satelliteimagery. Aggraded coral frameworks appear honey-brown to green incolor in an image composed of the red, green and blue satellite bands.Breaking waves may indicate aggraded structure, but since our imageswere acquired under clear calm sea conditions, they are not alwayspresent and therefore not a necessary criterion. Shallow (b1 m) waterdepths from nautical charts provide further evidence for the occurrenceof aggraded coral frameworks and sediments.

Morphological end-members with a Detached Form are identifiedbased on how much and where (edge vs. interior) the system that hasaggraded. The term ‘fully-aggraded’ describes cases where N80% of thecarbonate system's edge or interior has built to sea-level. Occurrencesof deepwater (greater than ~15m), such as narrow channels and reen-trants, are permitted within areas of seabed classified as fully-aggradedprovided deep waters do not occupy N20% of the total system area.‘Partially-aggraded’ describes cases where between 30% and 80% ofthe carbonate system's edge or interior has built to sea-level. The occur-rence of aggraded coral framework and sediment is patchy and often in-terspersed with deep water channels, which may attain widths ofseveral hundred meters. ‘Non-aggraded’ describes cases where b30%of the carbonate system's edge or interior has built to sea-level. Use ofthe term ‘ribbon’ to describe ‘Ribbon Platform Systems’, is consistentwith the “linear, long, winding” aggraded central axis (Andréfouët andCabioch, 2011).


To illustrate the use of the typology, consider how the followingcarbonate system would be handled in the decision tree (Fig. 2). Thesystem is separated from land by a structural low of several hundredmeters in depth and is therefore deemed in the typology to have a De-tached Form. The system is not circular and therefore it falls into thePlatform Group. Because all the edges of this carbonate system arefully-aggraded but the interior is only partially-aggraded, its morpho-logical end-member classification is a ‘Platform System with LagoonalPatches’ (Fig. 1b-L).

2.2. Spatial distribution of carbonate systems

On the basis of the typology, carbonate systems throughout theSaudi Arabian Red Seaweremapped from satellite imagery and nauticalcharts using a 1 km× 1 kmgrid (Fig. 3). This grid is hereafter referred toas being composed of ‘GIS grid cells’. Only GIS grid cells intersecting acarbonate system were retained for analysis. This process yielded20,505 cells across the extent of the Saudi Arabian Red Sea. The GISgrid was overlain on the satellite imagery and carbonate systems werethen classified using the typology (Fig. 2). Distance attributes criticalto decision points and thresholds were measured in the GIS. A numeri-cal code specific to the type of carbonate system was then appended toeach GIS grid cell. When a GIS grid cell overlaid multiple morphologicalend-members, the numerical code of the systemoccupying themajorityof that grid cell was appended as an attribute. This situation was rare,affecting only 4.3% of attached system grid cells and 0.7% detachedsystem cells.

The abundance (km2) and distribution of carbonate systems wasassessed through a tally of the GIS grid cells. An analysis of the distribu-tion of Saudi Arabian Red Sea carbonate systemswas structured accord-ing to width of the coastal shelf. The coastal shelf is defined as lyinginboard of the 500 m depth contour as measured from nautical charts(Figs. 4 and 5). On the basis of width, the shelf breaks into three logical‘precincts’, which are denoted A, B, and C. The coastal shelf in precinct Ahas a mean width of 4.4 km (SD: 1.0 km); precinct B has a mean widthof 27.9 km (SD: 12.8 km); precinct C has a mean width of 129.2 km(SD: 10.2 km). Each shelf-width precinct was sub-divided into ‘sectors’of approximately equal N–S length (75 km ± 3 km). The selected N–Slength reflects a compromise that allows most carbonate systems tobe encompassed by a sector in their entirety, yet small enough to iden-tify distinct latitudinal patterns. Sectors extend from the coastline to theseaward edge of the coastal shelf. Patterns in the distribution of carbon-ate systems across the coastal shelf were assessed bymeasuring the dis-tance of each GIS grid cell in a carbonate system from the coastline.Satellite imagery was consulted to digitize the coastline, a process thatyielded a ‘coastline vector’. The coastline vector was resampled withnodes placed every 250 m along its length. The distance of each GIS



Fig. 4.Minimumwidth of the coastal shelf (500mbathymetric contour), assessed normal to a ~1700 km transect running north to south down the central axis of theGulf of Aqaba and RedSea (upper x-axis). To assess the latitudinal and cross-shelf distribution of carbonate systems, the Saudi Arabian Red Sea is split into three precincts (A, B, C) on the basis of thewidth of thecoastal shelf. Within these precincts are nested sectors of approximately equal N–S length (lower x-axis).


grid cell from the coast was calculated as the minimum distancebetween the centroid of a GIS grid cell and a node of the coastline vectorand appended as an attribute to each GIS grid cell.

2.3. Factors creating shallow-water carbonate depositional environmentsin the Red Sea

The Red Sea is a rift basin flanked by mountains and the climate ofthe region is hyper-arid. In common with other tropical seas, carbonateproduction is most productive where the seabed is shallower than ~30m (Kleypas et al., 1999). Bosence (2005) identifies three seabed config-urations as important in the formation of Red Sea carbonate systems;(1) fault-block, (2) salt diapir and (3) delta-top.

As the Red Sea rifts, “fault-blocks” slip back towards the axial troughand tilt (Bosence, 2005), producing bathymetric highs. “Salt diapirism”

describes the subsurface movement of evaporite rock formations. RedSea evaporites, known as the Zeit and South Gharib formations, werelaid down during the Mid- and Upper-Miocene (16.4–5.3 Ma) during

Fig. 5. a) Interpolated thickness of the sub-seafloor evaporite layer based on borehole data (Micovers the entire coastal shelf of the Saudi Arabian Red Sea.Mainland Saudi Arabia is shown in gsectors of approximate equal length (x-axis). b) Distribution of dolines and carbonate systemsQuickBird satellite imagery that illustrates different dolines.


a period when the Red Sea basin was restricted at its southern limit,the Bab el Mandeb (Mitchell et al., 1992). These evaporites aresandwiched beneath Pliocene siliciclastics, above which sit the Pleisto-cene and Holocene carbonates (Brown, 1970; Mitchell et al., 1992;Bosence et al., 1998). Evaporites, which are less dense than the rocklayers above them, tend to rise upwards and distort or pierce the over-lying strata, generating bathymetric highs. These highs, termed“diapirs” tend to have a domed morphology and may be topped byislands (Bosence, 2005). Areas where seabed morphology is controlledby salt diapirs may consist of complex arrangements of shallow waterseparated by deep structural lows (Bosence et al., 1998). In the caseswhere salt diapirs have pierced the seabed and are subaerially exposed,as is common during sea-level low-stands, meteoric dissolution of theevaporite rock may lead to the formation of a void through collapse.The resultant depressions, termed “dolines”, have characteristic circularor semi-circular shapes (Fig. 5b; Bosence, 2005). Boreholes have beendrilled throughout the Red Sea that extend into, and in most casesthrough, the Miocene evaporite layer (Fig. 5a; Mitchell et al., 1992;

tchell et al., 1992; Bosence et al., 1998). The interpolation takes the form of a GIS layer andray. The Saudi Arabian Red Sea is split into three shelf-width precincts (A, B, C) divided intoin the Tower Group relative to the shoreline and the edge of the coastal shelf. c) Subsets of




Bosence et al., 1998). Mitchell et al. (1992) arrange cores from theseboreholes chronologically based on ageing of samples and the depthof the pre-evaporite “basement”.

Using the GIS database, several lines of evidence were developed toassess the distribution of evaporite in the Saudi Arabian Red Sea. Stratalevaporite thickness underlying the mapped carbonate systems was es-timated using an inverse distanceweighting interpolation of the verticalthickness of these rock formations measured within published explor-atory bores (Mitchell et al., 1992; Bosence et al., 1998). This interpola-tion produces a continuous map layer within the GIS (Fig. 4a). Boresin Yemen were included to extend the evaporite thickness map layerto the south and through the entire Saudi Arabian Red Sea. As furtherevidence for the spatial distribution of evaporite, dolineswere identifiedin the satellite imagery and their position entered into the GIS.

The “delta-top” carbonate systems defined by Bosence (2005) occurwhere terrestrial siliciclastics accumulate in the coastal zone to formtopographic highs that extend into the photic zone. If sufficiently undis-turbed (10s–100s years; Hayward, 1982), these highs may host consid-erable carbonate deposits. Siliciclastics are mobilized during heavyrainfall and runoff events and are transported seaward within alluvialflows. The word ‘wadi’ is an Arabic term used to describe a valley ordry ephemeral riverbed. In the hyper-arid Arabian climate, wadis typi-cally only contain water during heavy rain events. Rainfall within adrainage basin is channeled through a network of wadis on its way tothe sea (Fig. 6a). Neighboring drainage basins can vary substantially interms of area, relief and bedrock type (Blair and McPherson, 2009). Toestimate the potential for sedimentary input, drainage basins of wadisfeeding into the Saudi Arabian Red Sea were mapped into the GIS. Thistask was accomplished using the ESRI ArcGIS hydrology tool boxapplied to ASTER GDEM v2 satellite-derived elevation data (product of

Fig. 6. a) Diagram of a drainage basin feeding an alluvial fan (adapted from Blair and McPhersriverbed; b) Examples of alluvial fans found just inland of the Saudi Arabian Red Sea coast. Blackage basinsmapped in the study GIS. Those drainage basins associated with Dendritic Systems (Sea is split into three shelf-width precincts (A, B, C) divided into sectors of approximate equal


METI and NASA; Fig. 6c). Within the GIS, drainage basins were definedfor wadi systemsmeasuring N5 km in length. The planar area, gradients,and elevations encompassed by each drainage basin were assessed andtallied as attributes within the GIS.

Lineated Systems are comprised of repeating linear coral reefsaligned parallel to the coastline (Fig. 1b-E). Analysis of satellite and ae-rial imagery in the North of Saudi Arabia (28°09′N, 31°48′ E) revealed aterrestrial morphology, dune ridges, with similar repeating linear units.Dune ridges occur sporadically throughout the length of Western SaudiArabia (Jado and Zötl, 1984), andwhen submerged, have been shown tobe the nucleating point for coral reefs in other seas (Banks et al., 2007).To test the hypothesis that submerged sand dunes might underlieLineated Systems, the crest-to-crest distance of the parallel coral reefswasmeasured and compared to the crest-to-crest distance of terrestrialdune ridges. These measurements were made along transects, overlaidon satellite imagery, and running normal to the long axis of the dunesand coral reefs. Four, 5 km long, transectswere laid across dune systems,while eleven transects, extending from the coastline to the outermostcoral reef crest were placed across Lineated Systems.

3. Results

3.1. Geological influences on shallow-water carbonate depositionalenvironments in the Saudi Arabian Red Sea

The evaporite layer is distributed unevenly beneath the coastal shelfof the Saudi Arabian Red Sea (Fig. 5a). The thickest evaporites are locat-ed beneath the Salif-2 borehole offshore of Yemen, to the south of SaudiArabia (Bosence et al., 1998). An approximately 1 km thick layer ofevaporite underlies precinct C, in the south of Saudi Arabia. A 2 km

on, 2009). The word ‘wadi’ is an Arabic term used to describe a valley or a dry ephemeraldashed outline depicts separate alluvial fan systems. c)Distribution of Saudi Arabiandrain-e.g. Fig. 1b image C) are highlighted as shown in key.Within the GIS the Saudi Arabian Redlength (x-axis).




thick evaporite layer underlies the Al Kurmah borehole in sector B1(Mitchell et al., 1992). In sector B1, between the Al Kurmah andBarquan-1 boreholes, the evaporite layer is b100 m thick. South of theBarquan-1 borehole, in sectors B2–B13, the evaporite layer ranges inthickness from 100 m to 500 m. The distribution of dolines record-ed in the GIS is in good accordance with the interpolated distribu-tion of sub-seabed evaporite (Fig. 5b). The greatest number ofdolines is seen in the mid-coastal shelf in sectors C2–C8 (FarasanBanks and Farasan Islands). Though a doline is evident in the satel-lite imagery in sector B1 near the Al Kurmah borehole, dolines areotherwise absent in precinct B where the evaporite layer is thinnerthan 500 m.

A total of 2120 drainage basins were mapped within the GIS, with acombined area of 258,574 km2 (Fig. 6c). 80% of this area is drained byonly 10% of the drainage basins. In precinct A (Gulf of Aqaba), the north-ern mountain range abuts the coastline. Drainage basins enteringprecinct A are small (mean area: 15 km2; SD: 75 km2). However, moun-tains within these drainage basins attain heights of 1863 m and theirslopes are steep, with gradients of up to 56°. In precincts B and C, 87%of the drainage basins have an area b15 km2, but these basins are re-stricted to the coastal plain where slopes are gentle (mean gradient:3°; SD: 1°). The remaining 13% of drainage basins in precincts B and Care much larger (mean area: 930 km2; SD: 6416 km2). These largedrainage basins extend into the coastal mountain belt, and elevationsattain 2989 m, while slopes have gradients of up to 69°. The largestdrainage basin has an area of 104,977 km2 and feeds into the SaudiArabian Red Sea in sector B5, approximately 200 km north of the portof Yanbu, in a region named Al Wajh.

The distance between terrestrial dune crests ranged from 80 m to426 m, and had a mean of 231 m (SD: 91 m). By contrast, the distancebetween submerged reef crests in Lineated Systems ranged from 322m to 2300 m, with a mean of 1260 m (SD: 662 m).

Fig. 7. The latitudinal distribution of Carbonate Systems. The Red Sea basin, with key sitesmentinto three shelf-width precincts (A, B, C) divided into sectors of approximate equal length (x-axwhile non-shaded boxes denote absence.


3.2. Abundance and distribution of carbonate systems in the Saudi ArabianRed Sea

The latitudinal distribution of carbonate systems through the SaudiArabian Red Sea is highly variable (Fig. 7). Carbonate systems in theNarrow Group (Fringing; Fringing with embayment) are found in allsectors of all precincts. Indeed, carbonate systems in precinct A are en-tirely of the Narrow Group. Other morphological end-members are re-stricted to distinct regions of the Red Sea. For instance, DendriticSystems are restricted to the northern half of precinct B, while LineatedSystems are found in the southern half of precinct B, and the first sector(C1) of precinct C. Unorganized Systems are found in two clusters in thenorth of precinct B (sectors B1 and B5–B7) and throughoutmost of pre-cinct C. Over 97% of carbonate systems in the Tower Group are found inprecinct C (sectors C2–C5). The few carbonate systems in the TowerGroup found outside of precinct C (sectors B6 and B7) are exclusivelyof the Planar type. Partially-aggraded Platform Systems are only foundin precinct C in the very south of the Saudi Red Sea (sectors C3–C8).Planar Platform Systems are found throughout much of precincts Band C. Crescentic, Lagoonal with Patches, Lagoonal, and Ribbon PlatformSystems are spread in small clusters throughout precincts B and C, as farsouth as sector C6.

The area occupied by each of the sixteen morphological end-members is variable (Fig. 8). Attached Forms are slightlymore abundantthan Detached. The Expansive Group accounts for 66% of carbonatesystemswith anAttached Form.However, this area abundance is spreadacross fewer distinct systems. Carbonate systems in the Platform Groupaccount for 96% of Detached Forms. PlatformswithNon-aggraded Edgesaccount for 67% of systems in the Platform Group. Three of the sixteenmorphological end-members account for half of all carbonate systems,(1) Unorganized Systems, (2) Partially-aggraded Platform Systemsand (3) Non-aggraded Platform Systems.

ioned in the text is shown above the plot.Within the GIS, the Saudi Arabian Red Sea is splitis).Within each sector, shaded boxes denote presence of a given type of Carbonate System,



Fig. 8. The area (vertical bars) occupied by each of the sixteenmorphological end-memberSaudi Arabian Red Sea typology is highly variable. The Attached Forms ofcarbonate systemoccupy a similar area (10,554 km2) to the Detached Forms (9951 km2). However, ifGroups of carbonate systems are considered, trends are apparent. For instance, theTower Group of carbonate systems occupies the least area (352 km2), while the PlatformGroup occupies the greatest (9599 km2). The Narrow Group of carbonate systems(3565 km2) occupies around half the area of the Expansive Group (6989 km2). Thenumber of distinct systems comprised by each morphological end-member is shown inthe right hand column (n).

Fig. 9. a) Primary y-axis: bars represent the combined area occupied by all types of SaudiArabian Red Sea carbonate system across three shelf-width precincts (A, B, C) divided intosectors of approximate equal length. Secondary y-axis: line plot of the area occupied bycarbonate systems in each sector expressed as a percentage of the area of the coastalshelf in each sector. b) Distance of GIS grid cells used to map carbonate systems fromcoastline, plotted for Groups of carbonate systems in each latitudinal sector; position ofouter edge of the coastal shelf (500 m bathymetric contour) also plotted.


The combined area of all carbonate systems is highest in precinct C ofthe Saudi Red Sea (Fig. 9a). This shelf precinct corresponds to a widen-ing of the coastal shelf (Fig. 4). The proportion of the continental shelfoccupied by carbonate systems is variable by latitude and shows notrend of increasing to the south (Fig. 8a); depending on the sector, car-bonate systems occupy between 22% and 85% of shelf area. Sectors witha high proportion of the shelf occupied tend also to have large carbonatesystems in the Expansive Group (Fig. 7). In precincts A and B, the outer-most carbonate systems are found closer to the seaward edge of thecoastal shelf (Fig. 9b). In precinct C, the shelf is wide, extending tobetween 70 km and 125 km offshore. Though carbonate systems inprecinct C are found up to 111 km offshore, the majority of systemsare located mid-shelf between 25 km and 75 km from the coast.

System morphology stratifies with distance offshore. Trends andpatterns in cross-shelf distribution are most apparent when carbonatesystems are analyzed in terms of Groups rather than the sixteen mor-phological end-members of the Saudi Arabian Red Sea typology(Fig. 2). As is expected by their definition, the Narrow Group of carbon-ate systems is found closest to the coastline inmost sectors (Fig. 9b). Theexception to this rule is found in sector C2where, due to the presence ofsmall offshore islands, narrow systems tend to be found further fromthe coastline. Moving offshore in precincts B and C, the Narrow Groupis followed by the Expansive Group, which is in turn followed by car-bonate systems in the Platform Group with Non-aggraded Edges. Inmost sectors, carbonate systems in the Platform Group with AggradedEdges are situated furthest offshore. The exception to this trend occursin sectors B6–B7 and sectors C2–C5 (Farasan Banks; Fig. 7), where theoffshore progression ends with the Tower Group of carbonate systems.In sectors C7 and C8 (Farasan Islands; Fig. 7), the Narrow and ExpansiveGroups of carbonate systemare found30 km to 60 kmoffshore (Fig. 9b).Carbonate systems in the Tower Group, and those of the PlatformGroup with Aggraded Edges, are absent from the two southernmostsectors of the Red Sea. In sectors C7 and C8 the offshore progressiontherefore ends with carbonate systems in the Platform Group withNon-aggraded Edges.


4. Discussion

This study identifies several trends in the spatial distribution andabundance of Saudi Arabian Red Sea carbonate systems. This insightis achieved by developing a typology of sixteen morphological end-members and then mapping the occurrence of these systems withina GIS (Figs. 1 and 2). Many of the systems mapped for Saudi Arabiahave a similar morphology to those identified in the GBR and TorresStrait, Australia (Hopley, 1982; Hopley et al., 1989; Leon andWoodroffe, 2013). The greater number of morphological end-members in the Red Sea, as compared to Australia, can be accountedfor by a higher level of diversity of Attached Forms, aswell as our recog-nition of the Tower Group of carbonate systems as distinct from thePlatform Group. In Hopley's (1982) terms, a similar spectrum of matu-rity is visible in this study from juvenile (non-aggraded systems), tomature (crescentic and lagoonal platform/tower systems), and senilemorphological end-members (planar platform/tower systems).

Variation in relative sea-level over geological time-scales facilitatesor resets the evolutionary development of a carbonate system(Hopley, 1982). If carbonate budgets are positive and relative sea-levelis stable, then systems can be expected to fill available accommodationspace and transition towards more mature morphologies. Records ofsea-level along the Red Sea margins are incomplete, but often departfrom a simple eustatic model by tens of meters (Sheppard et al., 1992;




Siddall, 2004; Lambeck et al., 2011). In the Red Sea, the spatial variabilityof sea-level through the basin is usually interpreted in terms of tecton-ics, but as Lambeck et al. (2011) show, isostatic variation is also animportant consideration. While there is evidence in the literature thatsea-level variation has influenced aspects of morphological develop-ment in the Red Sea (Purkis et al., 2010) a full reconstruction of carbon-ate system growth in response to sea-level is beyond the scope of thecurrent work. Further discussion therefore focuses on other geologicalprocesses that deliver differences in antecedent topography.

Of the sixteen morphological end-members, many are rare (Fig. 8).The occurrence of carbonate systems varies through the latitudinalrange described in the GIS (Fig. 7), while the areal abundance of carbon-ate systems increases from north to south in the Saudi Arabian Red Sea(Fig. 9a). Lowwinter temperature has been proposed as a reason for de-clining reef growth to the north in other high latitude coral reef settings(Kleypas et al., 1999). Periodic cold winds blowing off Eastern Europelower sea surface temperatures in the Red Sea (Sheppard et al., 1992;Purkis et al., 2010). On tidal flats in the Gulf of Aqaba (precinct A), tem-perature lows of 7 °C have been recorded, while a low of 14.5 °C hasbeen recorded as far south as Jeddah (sector B12; Sheppard et al.,1992). However, unlike the much shallower Arabian Gulf where coralreef growth is reset by cold water anomalies on a decadal basis(Purkis and Riegl, 2005), cold water events in the Red Sea tend to beshort in duration. The thermal environment of the Red Sea is bufferedby vertical mixing from perennially warm deep waters (Sheppardet al., 1992). As Hopley et al. (2007) documented for the GBR, the abun-dance of Saudi Arabian Red Sea carbonate systems is closely associatedwith the width of the coastal shelf (Fig. 9).

Bosence (2005) identifies three seabed configurations as importantin the formation of Red Sea carbonate systems; (1) fault-block, (2) saltdiapir and (3) delta-top. The typology of the Saudi Arabian Red Seacarbonate systems presented in this paper (Figs. 1 and 2) is unable todiscriminate between Detached Forms nucleated atop salt diapirs andDetached Forms nucleated atop fault-blocks. Most morphological end-members that are found in precinct C, where sub-seafloor evaporite isthick, are also found in precinct B where the evaporite layer isthin (Figs. 5a and 6b). The Tower Group of carbonate systems andPartially-aggraded Platform Systems are, by contrast, preferentially as-sociated with precinct C where the underlying evaporite is thick.Based on the distribution of evaporite and the presence of dolines(Fig. 5), salt diapirism seems the most likely factor creating the numer-ous topographic highs in precinct C that extend into the photic zone andhost shallow-water carbonate depositional systems (Figs. 4 and 6;Guilcher, 1988; Sheppard et al., 1992). Salt diapirism has not led to, inthe terms of Hopley (1982), systems that are eithermore or lessmature.Rather, in the Saudi Arabian Red Sea it seems to have increased thediversity of carbonate systems present.

The interpolated evaporite thickness GIS layer informs on the latitu-dinal variation in the distribution of sub-seafloor evaporite (Fig. 5a).However, given the spacing between the available boreholes, our inter-polation is not deemed to capture the variability in evaporite thicknessacross the coastal shelf. The cross-shelf patterns in evaporite thicknessare better known for the western Red Sea shelf, where Mitchell et al.(1992) document an area at equivalent latitude to the Farasan Islands(sectors C7–C8). It is worthwhile considering the cross-shelf transectdescribed by Mitchell et al. (1992) that runs from the coastline ofEritrea, beneath the Dahlak Islands (Fig. 7), and on towards the deepRed Sea axial trough. Along this transect, the evaporite layer steeplythickens away from the Eritrean coast, before passing into a “salt-zone” at the main axis of the evaporite basin (Mitchell et al., 1992). Inthis salt-zone, evaporite thickness ranges from 2000 m to 4000 m. To-wards the axial trough, evaporite “thins and interfingers with volcanicsin a transitional zone” (Mitchell et al., 1992). Although there are noboreholes in the mid- and outer-coastal shelf of the Farasan Islandsand Farasan Banks (Fig. 5a; precinct C), the density of dolines in theseareas suggests the presence of a significant salt basin — an ‘eastern


salt-zone’ (Fig. 5b). Most carbonate systems in the Tower Group arelocated along the outer edge of this eastern salt zone and may be asso-ciatedwith a similar thinning of the underlying evaporite layer. Upliftedthrough salt diapirism, the Farasan Islands increase the length of coast-line in sectors C7 and C8 by a factor of four. This increase in the length ofcoastline accounts for an almost doubling of the area occupied bycarbonate systems with an Attached Form in sectors C7 and C8, andthe distribution of these systems tens of kilometers offshore (Fig. 9a).

The threemorphological end-members that comprise the ExpansiveGroup, Dendritic Systems, Unorganized Systems, and Lineated Systemsoccur in different regions of the Saudi Red Sea (Fig. 7). Different process-es of formation can account for these differences in latitudinal distribu-tion. The delta-top model proposed by Bosence (2005) considers thespatially variable delivery of terrestrial siliciclastic sediment into theRed Sea. The delta-top model should not be confused with the “deltaic”systems identified by Maxwell (1970) in the Pompey Complex of theGBR which form atop debris accumulated by ebb and flood tidal deltas.Of all the carbonate systems in the Expansive Group, Dendritic Systemsbear the closest resemblance to alluvial features. These systems projectoffshore in a manner reminiscent of “bird's foot” river deltas (Fig. 1bimage C; Coleman et al., 1998). Bird's foot river deltas tend to be associ-ated with large river channels and the continuous delivery of finesediment (Blair and McPherson, 1994). Such persistent sediment loadshave been shown to retard coral reef development (Fabricius, 2005).However, in the hyper-arid Middle East, alluvial flows are of a differenttype. Red Sea wadi systems only activate ephemerally, typically duringrare but intense rain storms (Dullo andMontaggioni, 1998). Such punc-tuated bouts of activity, with long periods of stasis in between, repre-sent a highly episodic delivery of siliciclastics out of the wadi mouthand into the Red Sea. Coral communities may be temporarily destroyedthrough sediment stress at the time of such outflows, but given that theviolent alluvial disturbance is rare (occurring at time scales of hundredsof years — Hayward, 1982) the coral community is able to recover andpersist (Dullo and Montaggioni, 1998).

The outflow and accumulation of siliciclastics creates bathymetrichighs on the coastal shelf. These highs take the form of channel leveesand debris deposits of alluvial fans (Fig. 6a and b) which represent a vi-able substrate for the initiation of coral growth (Perry and Smithers,2009). The morphological similarity between the Dendritic Systemsidentified in the typology and alluvial fans suggests that the two maybe linked in process-terms (Figs. 1b image C, 5c). For this link to befeasible, there are several conditions that must be satisfied. First, themorphology of the fan must be preserved in a submerged form atopthe coastal shelf, such as would occur if deposition coincides with asea-level low-stand. Second, the drainage basin landward of the fanmust be large and steep, such that it is able to feed sufficient volumeof sediment to build a coherent systemof channel levees. Third, episodicheavy rainfall is required to mobilize quantities of sediment into the fansystem. Finally, the alluvial fan must have sufficient area on the coastalshelf in which to form (Fig. 6a). There are several large (N1000 km2)drainage basins with steep (N25°) slope gradients that feed into theRed Sea along the Saudi Arabian coast (Fig. 6c). The GIS reveals thatDendritic Systems only occur in the northern half of precinct B(Fig. 7). While hyper-arid today, during the Mid Holocene, rainfall inthe Red Sea was greater in the north than the south of Saudi Arabia be-cause of the influence of Mediterranean weather systems (Purkis et al.,2010). This difference goes some way to explaining why Dendritic Sys-temsmay be more prevalent in the north than in the south of the basin.It should be noted, however, that Dendritic Systems are absent in theGulf of Aqaba (precinct A), but here the coastal shelf is narrow and theabsence of the morphology can be explained by a lack of space inwhich it could develop (Tibor et al., 2010).

Earlier in this paper, the similarity between the morphology ofLineated Systems and that of terrestrial dune ridges was recognized.Investigation of this similarity within the GIS reveals that the crest-to-crest distance of parallel coral reefs in Lineated Systems is five times




farther than the distance separating terrestrial dune ridges in thedessert of northern Saudi Arabia. The separation distance of the duneridges, which varied between 80 m and 426 m, is typical for suchmorphologies (Levin et al., 2004) and can therefore be considered asrepresentative. Because of these differences, a morphology other thandune ridges must be sought to explain the formation of Lineated Sys-tems. Bosence (2005) recognized a situation in the Red Sea wherebycarbonate deposition proceeds atop the bathymetric highs created byfault-blocks. Carbonate systems with a Detached Form nucleate atophighs and are separated by areas of deep water (structural lows)defined by large-scale faults (Guilcher, 1988).Widths between offshorestructural highs range from 4 km to 15 km, which is much greater thanthe distance separating reefs within the Attached Lineated Systems.However, the fault-block concept of Bosence (2005) can be extendedto smaller-scale faults associated with horst and graben structures.This may offer an explanation for the morphology of Lineated Systems,as previously suggested by Guilcher (1988) and Sheppard et al. (1992).

5. Conclusions

In the Saudi Arabian Red Sea, processes act across a variety of spatialand temporal scales to deliver diverse morphologies of shallow-watercarbonate depositional systems. Seafloor topography is influenced bythe tectonics of the Red Sea rift basin, and the spatially variable occur-rence of salt diapirs. Changes in relative sea-level can influence theprogression of morphological development. A hyper-arid climate punc-tuated by extreme weather events affects shallow-water topographythrough episodic and spatially restricted sediment delivery to thebasin. Through a GIS analysis, trends and patterns in the arrangementof sixteen morphological end-members are recognized. While somemorphological end-members appear at all latitudes of the Red Sea,others are confined to narrow ranges. By creatingGIS layers that capturethe distribution of sub-seafloor evaporite thickness, dolines, and terres-trial features including drainage basins and wadis, the occurrence ofseveral morphological end-members is explained in process-terms. De-spite the geological differences of the rift basin, many of the SaudiArabian Red Sea morphological end-members are similar to those firstidentified by Hopley (1982) in the GBR, Australia. The presented GISanalysis delivers insight into a suite of subsurface and climate processesthat serve to control the placement of different types of shallow-watercarbonate depositional systems in the Red Sea. As such, the work is rel-evant to understanding themorphological variation of reefal accumula-tions, and their associated biodiversity, in both the modern ocean androck records.

Acknowledgments

Financial support was provided by the National Coral Reef Institute(NCRI) and the Khaled bin Sultan Living Oceans Foundation (KBSLOF).Field support was provided by the crew of the M/Y Golden Shadow,through the generosity of HRH Prince Khaled bin Sultan. We thankBrett Thomassie and DigitalGlobe Inc. for assistance with satelliteimage acquisition. The use of Google Earth data, ESRI ArcGIS base imag-ery, and the United States Geological Survey Landsat satellite imagerywas critical to our work and is acknowledged. This is NCRI contribution152.

References

Andréfouët, S., Cabioch, G., 2011. Barrier reef (Ribbon reef). In: Hopley, D. (Ed.), En-cyclopedia of Modern Coral Reefs: Structure, Form and Process. Springer, Dordrecht,pp. 102–104.

Andréfouët, S., Pages, J., Tartinville, B., 2001.Water renewal time for classification of atollsin the Tuamotu Archipelago (French Polynesia). Coral Reefs 20, 399–408.

Andréfouët, S., Müller-Karger, F.E., Robinson, J.A., Kranenburg, C.J., Torres-Pulliza, D.,Spraggins, S.A., Murch, B., 2006. Global assessment of modern coral reef extentand diversity for regional science and management applications: a view from


space. Proceedings of the 10th International Coral Reef Symposium, Okinawa,Japan, pp. 1732–1745.

Banks, K.W., Riegl, B.M., Shinn, E.A., Piller, W.E., Dodge, R.E., 2007. Geomorphology of theSoutheast Florida continental reef tract (Miami-Dade, Broward, and Palm BeachCounties, USA). Coral Reefs 26, 617–633.

Bellwood, D.R., Hughes, T.P., 2001. Regional-scale assembly rules and biodiversity of coralreefs. Science 292, 1532–1535.

Blair, T.C., McPherson, J.G., 1994. Alluvial fans and their natural distinction from riversbased on morphology, hydraulic processes, sedimentary processes, and facies assem-blages. J. Sediment. Res. 64, 450–489.

Blair, T.C., McPherson, J.G., 2009. Processes and forms of alluvial fans. In: Parsons, A.J.,Abrahams, A.D. (Eds.), Geomorphology of Desert Environments. Springer, Berlin,pp. 413–467.

Bosence, D., 2005. A genetic classification of carbonate platforms based on their basinaland tectonic settings in the Cenozoic. Sediment. Geol. 175, 49–72.

Bosence, D.W.J., Al-Aawah, M.H., Davison, L., Rosen, B.R., Vita-Finzi, C., Whitaker, E., 1998.Salt domes and their control on basin margin sedimentation: a case study from theTihamam Plain, Yemen. In: Purser, B.H., Bosence, D.W.J. (Eds.), Sedimentation andTectonics of Rift Basins: Red Sea — Gulf of Aden. Chapman and Hall, London, pp.441–472.

Brown, G.F., 1970. Eastern margin of the Red Sea and the coastal structures in SaudiArabia. Philos. Trans. R. Soc. Lond. Ser. A Math. Phys. Sci. 267, 75–87.

Bruckner, A., Rowlands, G., Riegl, B., Purkis, S., Williams, A., Renaud, P., 2011. Khaled binSultan Living Oceans Foundation Atlas of Saudi Arabian Red Sea Marine Habitats.Panoramic Press, Phoenix, Arizona (274 pp.).

Coleman, J.M., Roberts, H.H., Stone, G.W., 1998. Mississippi River delta: an overview. J.Coast. Res. 14, 698–716.

Darwin, C.R., 1842. The Structure and Distribution of Coral Reefs. Smith, Elder, London(344 pp.).

Dullo, W.C., Montaggioni, L., 1998. Modern Red Sea coral reefs: a review of their mor-phologies and zonation. In: Purser, B.H., Bosence, D.W.J. (Eds.), Sedimentationand Tectonics of Rift Basins: Red Sea — Gulf of Aden. Chapman and Hall,London, pp. 583–594.

Fabricius, K., 2005. Effects of terrestrial runoff on the ecology of corals and coral reefs:review and synthesis. Mar. Pollut. Bull. 50, 125–146.

Fairbridge, R.W., 1967. Coral reefs of the Australian region. In: Jennings, J.N., Mabbutt, J.A.(Eds.), Landform Studies From Australia and NewGuinea. Australian National Univer-sity, Canberra, pp. 386–451.

Guilcher, A., 1988. A heretofore neglected type of coral reef: the ridge reef. Morphologyand origin. Proceedings of the 6th International Coral Reef Symposium, Townsville,Australia, pp. 399–402.

Harris, P.M., Vlaswinkel, B., 2008. Quantifying facies attributes of the Caicos platform.Developing Models and Analogs for Isolated Carbonate Platforms — Holocene andPleistocene Carbonates of Caicos Platform, British West Indies. Society of EconomoicPaleontologists and Minerologists Core Workshop, 22, pp. 1–10.

Harris, P.M., Purkis, S.J., Ellis, J., 2011. Analyzing spatial patterns inmodern carbonate sandbodies from Great Bahama Bank. J. Sediment. Res. 81, 185–206.

Harris, P.M., Ellis, J., Purkis, S.J., 2013. Assessing the extent of carbonate deposition in earlyrift settings. Am. Assoc. Pet. Geol. Bull. 97, 27–60.

Hart, D.E., Kench, P.S., 2007. Carbonate production of an emergent reef platform,Warraber Island, Torres Strait, Australia. Coral Reefs 26, 53–68.

Hayward, A.B., 1982. Coral reefs in a clastic sedimentary environment: fossil (Miocene, S.W. Turkey) and modern (recent, Red Sea) analogues. Coral Reefs 1, 109–114.

Hopley, D., 1982. Geomorphology of the Great Barrier Reef: Quaternary Development ofCoral Reefs. Wiley, New York (453 pp.).

Hopley, D., Parnell, K.E., Isdale, P.J., 1989. The Great Barrier Reef Marine Park: dimensionsand regional patterns. Aust. Geogr. Stud. 27, 47–66.

Hopley, D., Smithers, S.G., Parnell, K.E., 2007. The geomorphology of the Great BarrierReef. Development Diversity and ChangeCambridge University Press, Cambridge(532 pp.).

Jado, A.R., Zötl, J.G., 1984. Sedimentological, Hydrological, Hydrochemical, Geomorpholog-ical, Geochronological, and Climatological Investigations in Western Saudi Arabia.Quaternary Period in Saudi Arabia, 2. Springer-Verlag, Vienna, Austria.

Kleypas, J.A., McManus, J.W., Menez, L.A.B., 1999. Environmental limits to coral reefdevelopment: where do we draw the line? Am. Zool. 39, 146–159.

Lambeck, K., Purcell, A., Flemming, N.C., Vita-Finzi, C., Alsharekh, A.M., Bailey, G.N., 2011.Sea-level and shoreline reconstructions for the Red Sea: isostatic and tectonic consid-erations and implications for hominin migration out of Africa. Quat. Sci. Rev. 30,3542–3574.

Leon, J.X., Woodroffe, C.D., 2013. Morphological characterisation of reef types in TorresStrait and an assessment of their carbonate production. Mar. Geol. 338, 64–75.

Levin, N., Ben-Dor, E., Karnieli, A., 2004. Topographic information of sand dunes asextracted from shading effects using Landsat images. Remote Sens. Environ. 90,190–209.

Maxwell, W.G.H., 1970. Deltaic patterns in reefs. Deep-Sea Res. 17, 1005–1018.McLeod, E., Salm, R., Green, A., Almany, J., 2009. Designing marine protected area

networks to address the impacts of climate change. Front. Ecol. Environ. 7,362–370.

Mitchell, D.J.W., Allen, R.B., Salama, W., Abouzakm, A., 1992. Tectonostratigraphic frame-work and hydrocarbon potential of the Red Sea. J. Pet. Geol. 15, 187–210.

Montaggioni, L.F., Behairy, A.K.A., Sayed,M.K.E., Yusuf, N., 1986. Themodern reef complex,Jeddah area, Red Sea: a facies model for carbonate sedimentation on embryonicpassive margins. Coral Reefs 5, 127–150.

Perry, C.T., Smithers, S.G., 2009. Stabilisation of intertidal cobbles and gravels byGoniastreaaspera: an analogue for substrate colonisation during marine transgressions? CoralReefs 28, 805–806.


http://refhub.elsevier.com/S0169-555X(14)00140-8/rf0005





























































































Perry, C.T., Spencer, T., Kench, P.S., 2008. Carbonate budgets and reef production states: ageomorphic perspective on the ecological phase-shift concept. Coral Reefs 27,853–866.

Purdy, E.J., 1974. Reef configurations, cause and effect. In: Laporte, L.F. (Ed.), Reefs in Timeand Space: Selected Examples of the Recent and Ancient. Society of EconomoicPaleontologists and Minerologists, Special Publication, 18, pp. 9–76.

Purkis, S.J., Pasterkamp, R., 2004. Integrating in situ reef-top reflectance spectra withLandsat TM imagery to aid shallow-tropical benthic habitat mapping. Coral Reefs23, 5–20.

Purkis, S.J., Riegl, B., 2005. Spatial and temporal dynamics of Arabian Gulf coral assem-blages quantified from remote-sensing and in situ monitoring data. Mar. Ecol. Prog.Ser. 287, 99–113.

Purkis, S.J., Kohler, K.E., Riegl, B.M., Rohmann, S.O., 2007. The statistics of natural shapes inmodern coral reef landscapes. J. Geol. 115, 493–508.

Purkis, S.J., Rowlands, G.P., Riegl, B.M., Renaud, P.G., 2010. The paradox of tropical karstmorphology in the coral reefs of the arid Middle East. Geology 38, 227–230.

Purkis, S.J., Harris, P.M., Ellis, J., 2012a. Patterns of sedimentation in the contemporary RedSea as an analog for ancient carbonates in rift settings. J. Sediment. Res. 82, 859–870.


Purkis, S., Vlaswinkel, B., Gracias, N., 2012b. Vertical to lateral transitions amongcretaceous carbonate facies — a means to 3-D framework construction via Markovanalysis. J. Sediment. Res. 82, 232–243.

Rankey, E.C., 2002. Spatial patterns of sediment accumulation on a Holocene carbonatetidal flat, Northwest Andros Island, Bahamas. J. Sediment. Res. 72, 591–601.

Rowlands, G., Purkis, S., Riegl, B., Metsamaa, L., Bruckner, A., Renaud, P., 2012. Satelliteimaging coral reef resilience at regional scale. A case-study from Saudi Arabia. Mar.Pollut. Bull. 64, 1222–1237.

Sheppard, C., Price, A., Roberts, C., 1992. Marine Ecology of the Arabian Region. Patternsand Processes in Extreme Tropical Environments. Academic Press, London.

Siddall, M., 2004. Understanding the Red Sea response to sea-level. Earth Planet. Sci. Lett.225, 421–434.

Smithers, S.G., 2011. Fringing reefs. In: Hopley, D. (Ed.), Encyclopedia of Modern CoralReefs: Structure, Form and Process. Springer, Dordrecht, pp. 430–446.

Tibor, G., Niemi, T.M., Ben-Avraham, Z., Al-Zoubi, A., Sade, R.A., Hall, J.K., Hartman, G.,Akawi, E., Abueladas, A., Al-Ruzouq, R., 2010. Active tectonic morphology and subma-rine deformation of the northern Gulf of Eilat/Aqaba from analyses of multibeamdata. Geo-Mar. Lett. 30, 561–573.






































Diversity in the geomorphology of shallow-water...

Documents

Transcript of Diversity in the geomorphology of shallow-water...