ÔØ ÅÒÙ×Ö ÔØ

Dating quaternary raised coral terraces along the Saudi Arabian Red Sea coast

Ammar A. Manaa, Brian G. Jones, Helen V. McGregor, Jian-xin Zhao, David M. Price

PII: S0025-3227(16)30016-0 DOI: doi: 10.1016/j.margeo.2016.02.002 Reference: MARGO 5417

To appear in: Marine Geology

Received date: 24 August 2015 Revised date: 1 February 2016 Accepted date: 7 February 2016

Please cite this article as: Manaa, Ammar A., Jones, Brian G., McGregor, Helen V., Zhao, Jian-xin, Price, David M., Dating quaternary raised coral terraces along the Saudi Arabian Red Sea coast, Marine Geology (2016), doi: 10.1016/j.margeo.2016.02.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT

Ammar Manaa

1 Dating Quaternary raised coral terraces along the Saudi Arabian Red Sea

2 coast

3 Ammar A. Manaaa, Brian G. Jonesa, Helen V. McGregora, Jian-xin Zhaob, and David

4 M. Pricea

5

6 a GeoQuEST Research Centre, School of Earth and Environmental Sciences, University of

7 Wollongong, NSW 2522, Australia

8 b Radiogenic Isotope Facility, School of Earth Sciences, The University of Queensland,

9 Brisbane, QLD 4072, Australia

10

11

12

13

14

ACCEPTED MANUSCRIPT

1

ACCEPTED MANUSCRIPT

Ammar Manaa

15 Abstract

16 Late Pleistocene raised coral reef terraces form extensive outcrops up to 5 km wide along the Saudi

17 coast. Porites coral were dated using U/Th while clastic sediment from Jeddah was dated using

18 thermoluminescence. The pooled mean age for the coral samples is 121.5±0.2 ka suggesting MIS 5e,

19 even for the uplifted 16-20 m high terrace in the north at Haql. In Jeddah the MIS 5e back-reef

20 succession is overlain by fluvial sediment that gave a TL age of 66±13 ka. The structure and faunal

21 composition of the coral terraces suggests that they accumulated in broad shallow embayments

22 following the last interglacial transgression. The consistent elevation of these terraces suggests that

23 the central and southern Saudi coast has been tectonically stable for at least the past 125,000 years and

24 the coral reef terraces (at 3.5-5.5 m elevation) are consistent with the MIS 5e sea level high-stand that

25 peaked at 6-9 m above present sea level. The Saudi coastal coral terrace north of Duba shows

26 progressive uplift to 16-20 m near Haql since 108-120 ka as a result of ongoing transform faulting in

27 the Gulf of Aqaba.

28

29 Keywords: MIS 5e, U-Th dating, tectonism, sea level.

30

ACCEPTED MANUSCRIPT

2

ACCEPTED MANUSCRIPT

Ammar Manaa

31 1. Introduction

32 Late Pleistocene raised coral terraces are well preserved along much of the Red Sea coast

33 (Fig. 1) varying in width from 500 m to 5 km with heights up to 25 m above present sea level

34 (apsl). Due to the complex relationship between eustatic sea-level changes and local

35 tectonics, numerous geological and geomorphogical studies on these Red Sea coral terraces

36 have concentrate on either sea level history (Behairy, 1983; Dullo, 1990; Sheppard et al.,

37 1992; Scholz et al., 2004) and/or neotectonic activity (Gvirtzman and Buchbinder, 1978;

38 Sneh and Friedman, 1980; Jado and Zötl, 1984;, Plaziat et al., 2008).

39 Most of the early studies were concentrated in the northern region of the Red Sea (north of

40 latitude 27⁰ N, Fig.1) where the Red Sea is divided into the Gulf of Aqaba and the Gulf of

41 Suez. This is because of the presence of abundant coral reefs and terraces, tectonic instability

42 and easy access to the region. Farther south, both the Saudi Arabian and Egyptian margins

43 are rimmed by several levels of Pleistocene and Holocene terraces, but they do not show

44 significant uplift and generally suggest relative vertical stability (Plaziat et al., 1998, 2008).

45 Studies from the northern region of the Red Sea have assigned a wide range of ages to the

46 raised coral terraces: from 4 to 150 ka for terraces with heights between 1-100 m (Friedman,

47 1965; Gvirtzman and Friedman, 1977; Dullo, 1984; Al-Rifaiy and Cherif, 1988; Dullo, 1990;

48 Hoang and Taviani, 1991; Gvirtzman et al., 1992; El-Asmar, 1997; Dullo and Montaggioni, 49 1998; Scholz et al, 2004;ACCEPTED Vincent, 2008; Parker et al., 2012 MANUSCRIPT). 50 A few coral reefs were dated from the west coast of the Red Sea before 1980 (Berry et al., 51 1966; Butzer and Hansen, 1968; Veeh and Giegengack, 1970; Faure et al, 1980) while more

52 published dates have appeared within the last three decades. These ages ranged from 50 to

53 150 ka for the lower raised reefs (1-9 m) previously referred to late Pleistocene times

3

ACCEPTED MANUSCRIPT

Ammar Manaa

54 (Gvirtzman et al, 1992; Gvirtzman, 1994; El-Moursi, 1992; El-Moursi et al, 1994; Walter et

55 al., 2000).

56 The eastern coast of the Red Sea (the coastline of Saudi Arabia) is about 1,840 km long,

57 accounting for 79% of the eastern seaboard of the Red Sea (Fig. 1). Early 14C dating

58 (Nesteroff, 1959; Behairy, 1983) provided a minimum age for the raised coral terraces

59 because the sequence is beyond the limits of radiocarbon dating. Later U/Th dating for two

60 raised (6-10 m) coral reef samples north of Duba gave an age of 95-120 ka (Jado et al., 1989)

61 while Dullo (1990) dated a terrace 4-6 m above present sea level (apsl) near Umm Lajj (105

62 ka), a terrace 8-12 m apsl south of Aqaba (96 ka) and a terrace at 18-25 m apsl north and

63 south of Maqna (118 ka). Despite the paucity of geochemical and mineralogical evidence for

64 possible diagenetic modification of the corals, these ages reflect MIS 5 reefs or terraces.

65 Dullo (1990) considered that the terrace consisted of three on-lapping reef cycles representing

66 Marine Isotopic Stage (MIS) 5e, 5c and 5a sea level high stands that were significantly higher

67 than the currently accepted global MIS 5 high-stand, thus indicating a strong tectonic uplift in

68 the Aqaba area. A more recent study by Scholz et al. (2004) using Porites corals determined

69 an age of 121-122 ka for a 7-10 m apsl terrace and 106 ka for a 4-5 m apsl terrace at the north

70 end of the Gulf of Aqaba. They noted that diagenetic addition of uranium was an important

71 factor in this area and based their conclusions on isochron ages.

72 In contrast to determining the age of the terraces in the northeastern Red Sea region, dating

73 on the central Saudi coastACCEPTED is sparce and suggests a wide rangeMANUSCRIPT of ages for the terraces.

74 Dawood et al. (2013) applied the alpha-counting U/Th dating method to three whole rock

75 samples within a single terrace 3-4 m high in the Rabigh area, resulting in ages ranging from

76 235±1 to 128±0.2 ka while Bantan et al. (2015) dated a 5 m high reef terrace on the Jeddah

77 coastal plain that revealed a range of ages between 121 and 299 ka representing possible

4

ACCEPTED MANUSCRIPT

Ammar Manaa

78 composite MIS 5-MIS 9 reef deposits. However, these studies only considered isolated areas

79 along the central Saudi coast.

80 The present study provides a more comprehensive assessment of the age and composition of

81 the raised coral terraces along the length of the Saudi Arabian Red Sea coast. All the main

82 accessible exposures of coral terraces between the Gulf of Aqaba and Al-Qattan were

83 investigated (Fig. 2). Sampling from the southernmost exposures on the Farasan Islands was

84 restricted at the time of this study but had been seen previously by the first author. The main

85 aims of this study are to assess these terraces in terms of their age, their relationship to late

86 Quaternary sea level changes and to determine the extent of neotectonic influences along the

87 Saudi coast.

88

89 2. Regional setting

90 2.1 Climate

91 The Saudi coast of the Red Sea (eastern Red Sea coast; Fig. 1) can be divided, on the basis of

92 the type, diversity and distribution of coastal environments, into three regions: 1) northern

93 region of the Gulf of Aqaba; 2) the northern-central region from south of the Gulf of Aqaba

94 through to Jeddah; and 3) the central-southern region from Jeddah south to the border with

95 Yemen (which includes the Farasan Bank and Islands; PERSGA/GEF, 2003).

96 The mean annual temperatureACCEPTED in the Jeddah area of Saudi MANUSCRIPTArabia is 28°C (PME, 2005) while 97 the Red Sea surface sea water temperature ranges between 22 and 32°C. Humidity is

98 moderate on the Red Sea coast of Saudi Arabia where a hot and arid climate is combined

99 with minimal levels of precipitation and significant evaporation (averages of 60 mm/y-1 and

100 2050 mm/y-1, respectively; Behairy et al., 1991). The western Red Sea coast is also

101 characterized by localized and infrequent rainfall.

5

ACCEPTED MANUSCRIPT

Ammar Manaa

102 The tidal range along the Red Sea varies from 0.6 m in the north, near the Gulf of Suez, to 0.9

103 m in the south at the Gulf of Aden (Edwards, 1987). Data from five tide stations along the

104 Saudi Red Sea coast provided by Marine Environment Protection Unit (2011) all show that

105 the coast is microtidal with an average daily tidal range of 0.25 m rising to a maximum of 0.6

106 m. Major storm activity and wave set-up in the Red Sea is also infrequent.

107 2.2 Geological setting

108 The study area is located along the narrow Red Sea coastal plain which is known as Tihamah.

109 The coastal plain marks the eastern edge of the north-south Red Sea graben extending for

110 about 1800 km from the Gulf of Aqaba and the border with Jordan in the north, to the border

111 with Yemen in the south (Fig. 1). The breadth of the plain varies from place to place; it is

112 very narrow in the north, especially north of Al Wajh, but it widens irregularly toward the

113 south reaching 40 km wide in the region of Jizan (Chapman, 1978). The plain of Tihamah is

114 composed of fluviatile alluvium in its inner part whereas its outer part consists of Pleistocene

115 coral reefs. In the middle area of the Red Sea (Jeddah, Rabigh and Yanbu) the coastal plain is

116 about 20 km wide, and is underlain by thick sequences of sedimentary rocks, primarily Late

117 Mesozoic and Cenozoic (Vincent, 2008). They consist of continental and marine sediments

118 with some volcanic cones and many salt marshes and sand dunes spread along the surface.

119 Although, no rivers occur in the area, many wadies descend from the western highlands to the

120 Red Sea and the coastal plain is commonly cut through at the shelf edge by lagoons locally

121 known as sharms (BrownACCEPTED et al., 1989). It is not clear whether MANUSCRIPT the remains of Oligocene

122 deposits in the coastal plain are left over from pre-rifting tectonic activity or were once more

123 commonly distributed and now represent the last preserved remains of rift zone deposit.

124 The coastal plain is bordered to the east by older escarpments and Oligocene-Recent

125 sedimentary rocks (Vincent, 2008). These form a series of mountains parallel to the Al-Hejaz

126 and Asir Highlands that extend all the way from Jordan to Yemen, and they vary in width

6

ACCEPTED MANUSCRIPT

Ammar Manaa

127 from a few kilometres to 140 km (Sagga, 2004). These western highlands, composed of

128 granite, gneiss and schist, formed as a result of Triassic geological movements that led to the

129 separation of the Arabian plate from the African shield. These basement rocks are covered by

130 flood basalt (Al-Harrat) ranging in age from late Triassic to Quaternary (Al-Shanti, 1993).

131 2.3 Tectonic History

132 During the Pliocene and early Pleistocene the southern borders of the Red Sea

133 underwent continued uplift and erosion (Ross et al., 1972). Late Pleistocene uplift in the

134 north-eastern Red Sea (Gvirtzman et al., 1992; Strasser et al., 1992; Gvirtzman, 1994;

135 Bosworth and Taviani, 1996; Plaziat et al., 1998) is concentrated in the central part of the

136 Gulf of Aqaba where uplift can be related to local faulting and the locus of recent seismic

137 activity as shown by the last large earthquake in 1995 (Shaked et al, 2004; Shalaby and

138 Shawky, 2014). Otherwise, the remaining sequences along the southern Red Sea coasts show

139 no major uplift and generally suggest relative vertical tectonic stability during the late

140 Pleistocene (Arvidson et al., 1994; Plaziat et al., 1998, 2008).

141 2.4 Sample site descriptions

142 Samples were collected from six major raised coral reef terraces along the Saudi Red Sea

143 coast from Haql in north to Al-Qattan in south (Fig. 1). The terrace heights ranged between 3-5

144 m apsl in most of the sites except farther north at Haql where it reached 20 m apsl (Fig. 2) 145 and near Maqna (25 m apsl).ACCEPTED MANUSCRIPT 146 Haql 147 The coral terrace at Haql has been uplifted to between 15-20 m apsl (Figs. 3a, 4a). The

148 terrace is 7-8 m thick and overlies poorly consolidated fossiliferous calcareous muddy sand.

149 The terrace is poorly cemented, covered by sand and contains significant amounts of coral

7

ACCEPTED MANUSCRIPT

Ammar Manaa

150 including Porites lutea, Porites nodifera and Acropora humilis. In the back-reef area

151 coralline algae are common and numerous gastropods are present (Table 1).

152 Duba

153 About 225 km farther south at Duba, the coral terrace stretches along the shoreline for about

154 0.5 km. The height and continuity of the terrace is variable; in some parts it disappears

155 whereas in other places it rises to a maximum height of 4 m apsl (Fig. 3b). About 70 km

156 south of Duba a small portion of the coral terrace has been upfaulted in a horst block to a

157 height of 12 m apsl before returning to about 4 m apsl 1 km farther south.

158 The lower part of the terrace is a coral-rich unit with a thickness of about 150 cm. Many coral

159 genera are found in this layer, including Favia lacuna, Acropora, Porites lutea and Fungia

160 scruposa, that accumulated with bivalves, sponges and echinoids (Table. 1). The base of this

161 layer is black (about 20 cm) and is partly dolomitised. The upper part of the terrace is about 2

162 m high, consisting of a slightly friable coral-rich unit containing much coral rubble. This

163 layer is very rich in the long branching coral Acropora and it contains some of the other

164 species that occur in the lower layer plus numerous echinoid spines.

165 Yanbu

166 At this location (430 km south of Duba), a coral terrace extends for about 10 km along the

167 coast separated by some small sandy beaches. The terrace reaches a maximum height of 3.7 168 m apsl at the coast (Fig. ACCEPTED3c). It has similar features to the coralMANUSCRIPT terraces at Duba with the 169 lower part being well-cemented and the upper part being poorly cemented. The terrace 170 contains many coral genera, such as Acropora, Fungia scruposa, Porites lutea, Porites

171 nodifera (Fig. 4b), with common gastropods, Tridacna gigas, calcareous worm tubes and

172 some echinoid spines (Table 1).

173 Rabigh shoreline terrace

8

ACCEPTED MANUSCRIPT

Ammar Manaa

174 The terrace in the Rabigh area (about 180 km south of Yanbu) provides the same heights and

175 features seen farther north with a coral terrace appearing along the coast for about 8 km,

176 except in some areas where it is obscured by coastal infrastructure. The lower layer at Rabigh

177 ranges from 1.1-1.4 m in thickness, is well cemented with a darker colour and incorporates

178 corals, bivalve shells and gastropods (Fig. 4c). The upper layer is about 2 m thick, poorly

179 cemented and dominated by coralline algae and corals including Acropora, Favia lacuna,

180 Goniastrea edwardsi, Porites lutea, Porites nodifera, Fungia, Tubipora musica, and

181 Stylophora. Bivalves, gastropods, sponges and echinoids (Table. 1) are also important

182 constituents in this terrace. The lithostratigraphy and petrography of these terraces have been

183 studied by numerous authors (Basaham, 1998; Mandurah, 2009, 2010; Manaa, 2011;

184 Mandurah and Aref, 2012).

185 The same coral terrace is revealed in a large excavation area 100-400 m east of the beach

186 (Fig. 3f). The exposed terrace is 4.8 m thick and shows the same reef structure as the

187 shoreline terrace but it appears as one unit with no major cementation in the lower portion

188 (Fig. 4d). Porites is the main coral type in this reef and is found in all parts. The top part of

189 the excavation (at 5.4-5.5 m apsl) shows a typical reef structure but the reef is not continuous

190 and sediment containing common shells and clay filled several gaps within the reef crest (Fig.

191 4e).

192 Rabigh back-reefACCEPTED area MANUSCRIPT 193 The coral terrace extends 3-4 km inland from the Rabigh coastal exposures with an almost 194 horizontal surface at about 4.5-5.5 m apsl. A back-reef terrace along the inner margin of this

195 coral sheet in the Rabigh area ranges in height from 1.5 m to 4.5 m apsl (Figs. 3e-f, 4f-g). The

196 lower part of this eroded terrace rises about 2 m above an exposed flat area of calcareous

197 mudstone containing a few 0.3 m Tridacna gigas. The upper part of this unit contains well-

9

ACCEPTED MANUSCRIPT

Ammar Manaa

198 cemented mud with occasional gastropods, Clypeaster humilis and regular echinoderms (Fig.

199 4g, Table 1). No in situ corals were found in this layer but they occur as coral rubble in the

200 upper part. The top part of the terrace is 1+ m thick, poorly cemented and contains coralline

201 algae, gastropods, some calcareous worm tubes and some Fungia scruposa, Tubipora musica

202 and Faviidae corals (Table 1). Some corals are found inverted which indicates these corals

203 may have been transported as a coral rubble apron to overlie a back-reef lagoon (Fig. 4f).

204 Al-Ruwais area, Jeddah

205 The Al-Ruwais area is located in the southern part of Jeddah (21°30.145‟N, 39°10.752‟E),

206 about 3 km from the present shoreline and about 140 km south of Rabigh. A 4 m high terrace

207 was exposed stretching laterally for about 100 m (Fig. 4h). The lower part of the sequence

208 has a thickness of about 1.7 m (2.6-4.3 m apsl, Fig. 3g) and consists of poorly cemented coral

209 rubble filled with sediment, shell fragments and intact coral reef species dominated by

210 Faviidae and small Porites nodifera (Table 1). The upper part of the outcrop consists of

211 fluvial sediments (channel deposits) with a maximum thickness of about 2.3 m (Fig. 3g).

212 Al-Shoaibah (Al-Qattan)

213 The coral terrace in the Al-Shoaibah area, about 100 km south of Jeddah, stretches along the

214 shoreline for about 5.5 km and is cut by the Al-Shoaibah lagoon entrance. The height and

215 continuity of the terrace is variable, in some parts it disappears while in other areas it rises to 216 maximum heights of 4-5ACCEPTED m apsl (Fig. 3h). The terrace can MANUSCRIPT be divided into lower and upper 217 portions and it shows the same features as the terraces at Duba, Yanbu and Rabigh with many 218 corals and other fauna (Table. 1).

219 2.5 Coastal coral terraces

220 The coral terraces along the Saudi coast commonly extend inland for 500 m to 5 km.

221 Sporadic outcrops are available along the coast and inland margins of the terrace but lateral

10

ACCEPTED MANUSCRIPT

Ammar Manaa

222 variability across the terrace usually cannot be assessed. Along most of the coast the exposed

223 terraces are 4-5 m thick but the base is usually covered or below sea level. However along the

224 southern margin of the Gulf of Aqaba, where the terraces have been uplifted, their thickness

225 ranges between 7-8 m and they overlie a fossiliferous muddy sand substrate. Near wadies and

226 sharms along the coast the landward part of the coral terraces is exposed in places and

227 generally overlies or passes laterally into sandy back-reef lagoonal facies. All the coastal

228 terrace heights have been measured from the modern carbonate platform that equates

229 approximately to mean low water level, and since the tidal range is up to 60 cm mean sea

230 level is taken as 30 cm above the platform.

231 Most of the coral genera that occur in these terraces, including Acropora humilis, Porites

232 lutea, Porites nodifera, Fungia scruposa and Tubipora musica, are currently living in shallow

233 water (0-25 m) environments in the modern Red Sea coral reef system whereas other corals

234 such as Favia lacuna, Platygyra daedalea, Stylophora pistillata and Goniastrea edwardsi

235 extend down to shallow slopes (Lieske and Myers, 2004). The associated bivalves (Anadara

236 antiquata, Anodontia edentula, Trachycardium flavum, Tridacna gigas, Tucetona

237 pectunculus), gastropods (Nerita albicillo, Cerithium eburneum, Turbo petholatus), sponges

238 (Crella cyathophora, Callyspongia viridis, Pione cf. vastifica) and echinoids (Clypeaster

239 humilis, Heterocentrotus trigonaris, Schizaster lacunosus) characterise a similar depth range

240 (Table. 1). This supports the concept that these coral terraces ranging from 4-5.5 m apsl along 241 the Saudi coast representACCEPTED shallow water sequences that formed MANUSCRIPT during the MIS 5e high stand 242 of sea level that has been recorded globally at up to 6-9 m apsl (Hearty, 2002; Hearty et al., 243 2007; Rohling et al, 2008; O‟Leary et al., 2013; Dutton et al., 2015b). Although all the dated

244 Porites corals come from within the middle portions of the terrace the whole reef facies has

245 been considered when estimating the elevation of the MIS 5e sea level.

11

ACCEPTED MANUSCRIPT

Ammar Manaa

246 An equivalent MIS 5e shell-rich calcareous clay in the northern part of Jeddah has been

247 described by Bantan et al. (2015) at 3-5.5 m apsl. This assemblage has a restricted coral

248 assemblage dominated by small branching corals but it also contains common bivalves and

249 gastropods that are common in intertidal to shallow subtidal conditions including seagrass

250 beds (e.g. Anodontia edentula, Anadara antiquate, Trachycardium flavum, Strombus

251 fasciatus, S. tricornis and Cypraea historio). The associated foraminiferal and ostracod

252 assemblages also indicate very shallow water back-reef lagoonal conditions (e.g. Agglutinella

253 robusta, Bolivina sp., Nonion fabum, Spiroloculina communis and Elphidium

254 striatopunctatum). This fauna all points to a very shallow marine embayment or lagoonal

255 environment with sea level not very far above a current elevation of 5.5 m.

256 Apart from the lower portion of the terrace in the coastal exposures the inland and raised

257 terraces are only poorly cemented and contain relatively minor amounts of coralline algae.

258 Thus they do not represent well cemented fore-reef or reef crest deposits but they probably

259 represent prolific growth of corals in a sheltered broad platform environment rather than in a

260 prograding reef system. This accounts for the abundance of Porites, Acropora, Fungia and

261 Tubipora forming tall and branching colonies and the occasional patches of shelly sand

262 between the coral thickets. Towards the back-reef areas smaller colonies and broken rubble

263 become more common overlying the sandy lagoonal successions.

264 The exposed tops of the coral terraces show a remarkably uniform elevation of 4-5.5 m apsl

265 across the whole terrace ACCEPTEDwidth with no sign of higher broken MANUSCRIPT off coral debris. This suggests

266 that the top of the terraces may well reflect an approximation of sea level at the time of coral

267 growth since the very dry climate and high evaporation rates along the Saudi Red Sea coast

268 would have severely limited any dissolution and erosion of these slightly raised terraces. Also

269 if sea level had been significantly higher than the exposed terraces coral terraces or associated

12

ACCEPTED MANUSCRIPT

Ammar Manaa

270 back-reef deposits should extend higher and farther inland but no such deposits have been

271 recognised.

272

273 3. Sample preparation methods

274 Twenty nine Porites corals were dated using U/Th methods while three sediment cores were

275 dated using thermoluminescence. Sample locations are given in Table 2 and shown

276 diagrammatically in Fig. 3.

277 Coral samples for dating were mainly collected from exposed cliffs using a hammer and

278 chisel while some coral samples were cored using an electrical drill to collect clean, un-

279 weathered parts of the corals. Corals from reef top exposures were too altered to be

280 considered for dating. The coral samples were cut into thin slabs and cleaned in an ultrasonic

281 bath and by using an ultrasonic probe in milli-Q water before being crushed by hand to a fine

282 powder in an agate mortar. X-ray diffraction (XRD) analyses were conducted on all coral

283 samples to define the percentage of aragonite in order to determine if it was pure enough to

284 be potentially useful for U/Th dating. The percentages of minerals were calculated using

285 SiroquantTM software. The 29 coral samples with high percentages (>97%) of aragonite were

286 U/Th dated at the Radiogenic Isotope Laboratory, University of Queensland following the

287 analytical methods reported in Zhao et al. (2009) and Clark et al. (2012, 2014). Sample 288 aliquots (50–100 mg) wereACCEPTED dissolved in nitric acid, spiked MANUSCRIPT with 229Th-233U and chemically 289 separated to create a U–Th solution with a final U concentration of less than10 ppb. The U– 290 Th isotopic ratio was measured by MC-ICP-MS using Hellstrom‟s (2003) method with minor

291 modifications proposed by Zhou et al. (2011). A mass fractionation correction for both U and

292 Th isotopic ratio measurements was based on a 238U/235U value of 137.82. The 230Th memory

293 was consistently less than 0.1 count s-1, and the memory for all other isotopes was negligible.

13

ACCEPTED MANUSCRIPT

Ammar Manaa

294 Open-system model ages for these samples were calculated by using Thompson et al.‟s

295 (2003) model since open-system model ages provide a realistic estimate of the true age for a

296 sample when the initial 234U/238U is elevated above the seawater value of 1.1455 ± 0.0023

297 (Cheng et al., 2000) or 1.1468 ± 0.0001 (Andersen et al., 2010). Frank et al. (2006) found

298 that the Thompson et al. (2003) and Villemant and Feuillet (2003) open system model ages

299 gave essentially the same results (within uncertainty) provided the amount of post-

300 depositional alteration is small. Since these conditions are met for many of the samples in this

301 study the Thompson et al. method was adopted here. However, as pointed out by Stirling and

302 Andersen (2009) the techniques and constants used in U-series dating need to be improved to

303 increase the accuracy of age determinations. In addition the quoted errors probably represent

304 minimum errors since they are based on the analytical method alone.

305 Three sediment samples collected from the upper and lower parts of a 4 m high terrace in the

306 Al-Ruwais area, 3 km inland in Jeddah, were dated in the University of Wollongong

307 thermoluminescence laboratory. Core samples C1 and C2 were collected in open ended tubes

308 from the lower back-reef part of the outcrop at heights of 3.35 m and 4.25 m apsl,

309 respectively, while sediment sample C3 was collected from the overlying fluvial sediment (at

310 about 5.3 m apsl). These samples were analysed by combined additive and regenerative

311 methods using the 90 to 125 µm quartz grain fraction separated by wet sieving and suitable

312 chemical treatment followed by heavy liquid separation. This method provides a means of

313 checking for possible TLACCEPTED sensitivity change caused by the MANUSCRIPT laboratory procedure. As a modern 314 sample analogue was not available the TL starting level at the time of final deposition for

315 each sample was assumed to be that reached following a 24-hour prepared sample exposure

316 beneath a laboratory ultraviolet lamp (Philips MLU300W). In order to correct for sample

317 aliquot variations all TL outputs recorded were nomalised using a second glow procedure

318 following a standard irradiation of approximately 19 Grays. Each sample analysis utilised 28

14

ACCEPTED MANUSCRIPT

Ammar Manaa

319 sample aliquots in total, eight of these were used to determine the natural TL accumulated

320 since the time of deposition, six to check for sensitivity change and fourteen in the

321 preparation of a TL growth curve to which the mean natural TL value was fitted. By this

322 means the equivalent radiation dose accumulated since the time of deposition was

323 determined.

324 The annual radiation dose for each TL sample was determined by thick source alpha counting

325 (TSAC) to determine the uranium and thorium specific activity and x-ray fluorescence (XRF)

326 to measure the sample potassium and rubidium contents. Corrections were made for cosmic

327 radiation and measured as collected sample moisture contents. The presence of moisture

328 moderates the radiation received by the sample thus increasing the age determined by

329 approximately 1% for each 1% increase in moisture. Full details of sample preparation and

330 TL measurement procedures were given by Shepherd and Price (1990).

331

332 4. Results

333 4.1 U/Th dating of coral

334 The calculated open system model ages have low analytical uncertainties of ±1 ka or less (Table

335 2; Supplementary Table A1). These age uncertainties are probably less than the true uncertainties,

336 as noted by Scholz and Mangini (2007a,b). The results show that the upper parts of the terrace 337 over the whole study areaACCEPTED give a range of ages between 1 19MANUSCRIPT to 125 ka and a pooled mean age 338 of 121.1±0.2 ka (calculated following Murray-Wallace and Woodroffe, 2014), which 339 indicates MIS 5e. At Al-Qattan and Duba the lower parts of the same terraces give ages

340 between 119 to 125 ka. In contrast, the ages for the lower parts of the terraces at Haql, Yanbu

341 and Rabigh range between 42 to 118 ka. Clearly these younger ages contravene the law of

342 superposition, and are not within uncertainties of the upper level ages. These samples all

15

ACCEPTED MANUSCRIPT

Ammar Manaa

343 contain high uranium contents and low 230Th/238U indicating probable diagenetic addition of

344 uranium. They have therefore been eliminated from the pooled mean age calculations. In

345 addition the uppermost age at Al-Qattan has been disregarded since it equates to a time when

346 sea level was at least 10 m below present sea level. Pooled mean ages at each site all along

347 the Saudi coast show a restricted range from 119.9±0.5 to 122.0±0.2 ka, lying within the last

348 interglacial maximum (MIS 5e at 128-116 ka) when sea level was about 6 m above present sea

349 level (Murray-Wallace and Woodroffe, 2014).

350 4.2 TL dating of sediment

351 The results from the two core samples (C1 and C2) from lower carbonate terrace portion of

352 the Al-Ruwais area gave TL ages of 93±13 ka for C1 (3.35 m apsl) and 71±12 ka for C2

353 (4.25 m apsl). The overlying fluvial sediment, sample C3 at 5.3 m apsl gave a TL age of

354 66±13 ka (Supplementary Table A2).

355

356 5. Discussion

357 5.1 Age of the coral terraces

358 Early 14C dating results for the Saudi reef terraces had ages between 40-53 ka (Nesteroff,

359 1959; Behairy, 1983), but these ages are at the limits of 14C dating and are not reliable results.

360 Reefs aged 40-53 ka would be well below present sea level, and thus the radiocarbon dates 361 for the Saudi reef terracesACCEPTED probably reflect minor contamination MANUSCRIPT by modern carbon and give 362 minimum ages for the reef sequences. Reefs of this age could only occur above present sea 363 level if tectonic uplift had occurred in the central area of the Red Sea coast but this has not

364 been evident since at least MIS 5e (Hearty, 1987). Behairy (1983) had suggested four major

365 marine transgressions since 31,000 years B.P in three terraces between Jeddah and Yanbu

366 based on radiocarbon dating. These findings are now considered to be inaccurate using

16

ACCEPTED MANUSCRIPT

Ammar Manaa

367 current Quaternary sea-level variation models based on global examples correlated with the

368 deep sea oxygen isotope data (Lambeck and Chappell, 2001).

369 The new U/Th results from the Saudi coral samples show that most of the ages ranged

370 between 113-132 ka (Fig. 5), with a total pooled mean age of 121.5±0.2 ka indicating a MIS

371 5e age for the reef terrace along the Saudi coast. Even the terrace at a height of about 16-20 m

372 in the north at Haql showed the same range of ages. This strongly suggests that these terraces

373 were formed during the last interglacial MIS 5e when global sea level is believed to have

374 been about 6-9 m above present sea level (Chappell and Shackleton, 1986; Woodroffe et al.,

375 1995; Lisiecki and Raymo, 2005; Murray-Wallace and Woodroffe, 2014; Dutton et al.,

376 2015a,b). Recently, Parker et al. (2012) dated >99% aragonite Tridacna shells from an 2.5 -

377 12 m high raised coral terrace on the eastern coastline at the southern tip of the Sinai

378 Peninsula which gave U/Th ages ranging from 77 to 129 ka suggesting correlation with MIS

379 5e for this terrace. Furthermore, recent U/Th dating by Bantan et al. (2015) of Porites,

380 Tridacna, Strombus and Clypeaster samples from a reef terrace 3-5.5 m apsl on the Jeddah

381 coastal plain revealed a range of ages between 125±1 and 121±1 ka indicating MIS 5e for the

382 upper layer of the terrace and the top of the underlying white limestone layer.

383 A number of similar last interglacial reef terraces have been observed and analysed around

384 the world. They include the Rendevous Hill „III‟ terrace in Barbados (Fairbanks and

385 Matthews, 1978; Schellmann and Radtke, 2004); the 'VIIIb' coral reef terrace in New Guinea 386 (Aharon and Chappell, 1986ACCEPTED); terraces 1-4 m high and 81 -MANUSCRIPT140 ka age at A1 Aqaba, Jordan 387 (El-Rifaiy and Cherif, 1988); Devonshire/Spencer's Point in Bermuda (Vacher and Hearty, 388 1989); terrace '112' on Sumba Island, Indonesia (Pirazzoli et al., 1991); and the Grotto Beach

389 Formation in the Bahamas (Hearty and Kindler, 1993).

390 Nine of the dated Porites samples, mainly from the lower part of the terrace in the Haql,

391 Yanbu and Rabigh areas, gave ages younger than MIS 5e. Three of these ages were less than

17

ACCEPTED MANUSCRIPT

Ammar Manaa

392 10 ka younger than MIS 5e while the remaining six ages are significantly younger ranging

393 from 42-89 ka. Similar young ages have been recorded from elsewhere around the Red Sea

394 (Plaziat et al., 2008) and globally. Most of these younger sample ages are from the lower part

395 of the same terrace beneath MIS 5e dated corals. They are also adjacent to the ocean and have

396 been quite strongly cemented for up to about 1 m above present mean sea level, probably

397 during the Holocene high-stand about 1000-3000 years ago (Siddall et al., 2003). Even

398 though the samples showing young dates were all selected with >98% aragonite some of this

399 aragonite may be in the form of younger cement (e.g. Fig. 6a). Thus they could have been

400 subjected to isotopic resetting by young cements or uranium mobility since they all have high

401 uranium contents. In addition, most of these corals with apparent young ages also suffer some

402 degree of diagenesis and open-system behaviour, evidenced by elevated initial 234U/238Uand

403 low 230Th/238U ratios. This suggestion of Holocene cementation is further vindicated by the

404 uncemented lower portion of the terrace in the excavation 100-400 m landward from the

405 present sea cliffs at Rabigh which were not affected by the raised sea level. Plaziat et al.

406 (2008) also suggested that isotopic resetting of coral dates could occur during diagenesis by

407 the addition of 234U into the coral lattice enhanced by the decomposition of organic sheaths

408 around the aragonite needles in scleractinian corals. This would invalidate using a “closed

409 system” assumption in the isotopic date calculation and such changes would not be

410 recognised by microscopic examination for later cements.

411 The U/Th data suggest thatACCEPTED the upper and lower parts of theMANUSCRIPT reef represent the same terrace 412 with the only difference being the more widespread cementation in the lower part of the reef

413 along the present coast. This concept of a single terrace is strongly supported by seeing the

414 same terrace 100-400 m inland in the Rabigh area where there is no difference in

415 cementation, colour or facies between the upper and lower parts of the terrace sequence (Fig.

416 4f-g). Younger dates (107-109 ka) in some samples from the raised portions of this terrace

18

ACCEPTED MANUSCRIPT

Ammar Manaa

417 probably reflect the effects of diagenesis, with replacement carbonate, secondary carbonate

418 cement or additional 234U being incorporated into the sample. For example, blocky calcite

419 cement in the raised terrace is distributed equally around most of the grains leaving an

420 average porosity of 14.8% (Fig. 6b). Some of the calcite features are typical of a fresh water

421 phreatic environment, as mentioned by Meyers (1978).

422 A petrographic study of the lower 1-1.5 m of the terrace in the Rabigh and Yanbu areas

423 shows more cementation and diagenesis than in the upper part (Fig. 6a). Aragonite

424 cementation is a characteristic feature of the lower part of the reef, and is similar to that

425 observed in a variety of other carbonate depositional environments including the Great

426 Barrier Reef (Marshall and Davies, 1981; Marshall, 1983), the Red Sea (Friedman et al.,

427 1974) and Barbados (Macintyre et al., 1968).Thin sections from the lower part of the terrace

428 also show some leaching of aragonite accompanied by calcite replacement and low porosity.

429 The inclusion of, or replacement by, younger aragonite cement may account for the low

430 numerical ages obtained from some samples in this part of the succession.

431 5.2 Older terraces

432 Recent studies along the Saudi Red Sea coast have also determined some older ages within

433 the coral terrace successions. Dawood et al. (2013) applied whole rock U/Th dating

434 methodology to the coral terrace in the Rabigh area of Saudi Arabia, resulting in ages of

435 128±0.2 ka at about 3.5 m, 212±1 ka at about 2.5 m , and 235±1 ka at about 0.9 m apsl in the

436 reef terrace. They suggestACCEPTEDed that deposition occurred during MANUSCRIPT periods of higher sea levels

437 linked to MIS 5 and 7. However, it is worth noting that the significantly larger uncertainty

438 associated with 234U/238U ratios obtained using the alpha-counting method prevent effective

439 assessment of diagenesis and open-system behaviour, and thus the reliability of the ages is

440 unknown. The age of the upper part of the reef terrace reflects MIS 5e of the last interglacial

441 and is consistent with previously mentioned references and the new data presented here. The

19

ACCEPTED MANUSCRIPT

Ammar Manaa

442 212 ka sample from Rabigh provided by Dawood et al. (2013) is shown coming from the

443 upper MIS 5e terrace in their figure 2 but from the lower terrace in their figure 3. However

444 our terrace mapping suggests that exposures 100 m inland show that there is only a single

445 terrace at Rabigh and the junction between their terraces actually represents a notch cut

446 during the Holocene high-stand of sea level with the inclusion of beach rock both in this

447 upper notch and in areas between the cliff exposures.

448 Dawood et al.‟s (2013) measurements from the middle and lower part of the reef are

449 inconsistent with samples from similar levels presented in Table 2 at Rabigh and along the

450 Saudi coast but could be related to older emergent reefs on the west coast of the Red Sea in

451 Egypt and Sudan (Plaziat et al., 2008). Also, the recent U/Th dates of 144±2 ka (Tridacna)

452 and 300±6 ka (Porites) at 2 m and 1.2 m from the white limestone terrace in the northern

453 Jeddah area (Bantan et al., 2015) could also represent reset ages from MIS 7 or MIS 9 reef

454 deposits even though the top of the limestone terrace gave an age of 121±1 ka linked to MIS

455 5e and no erosion break was detected between the upper and lower part of this layer. This

456 white limestone is much more indurated than the MIS 5e terraces investigated during this

457 study and it may represent a MIS 9 deposit, since MIS 7 probably did not reach present sea

458 level (Shackleton, 1987; Murray-Wallace and Woodroffe, 2014).

459 In the Rabigh area there is no indication of a break in sedimentation between the coral terrace

460 and back-reef area which suggests a continuous reef to back-reef succession (Fig. 3f). 461 Additional sampling andACCEPTED dating from this area would be necessaryMANUSCRIPT to clearly define the age of 462 this backreef succession. 463 5.3 Post-terrace sequences

464 The TL dating of the fluvial sediment that unconformably overlies the lower carbonate part of

465 the terrace in the Al-Ruwais area in Jeddah showed a much younger age (66±13 ka) for this

466 sediment than the coastal terraces. With only a few corals found in the lower carbonate

20

ACCEPTED MANUSCRIPT

Ammar Manaa

467 sequence and no coral or shell suitable for dating, it could be suggested that this terrace

468 represents an MIS 5e back-reef area (Bantan et al., 2015) covered by fluvial sediment coming

469 through wadis from the mountains. Although the TL ages overlap at the 1σ level, the two

470 samples from within the back-reef succession gave ages younger than MIS 5e and probably

471 reflect infiltration of younger fluvial sand into the porous back-reef deposits. This could also

472 account for the upward decrease in TL age through the back-reef sequence with a greater

473 proportion of younger quartz present towards the top of the unit.

474 5.4 Neotectonism

475 The dating results presented in this research from south of Duba, Yanbu, Rabigh, Al-Ruwais

476 and Al-Qattan suggest tectonic stability along most of the Saudi Red Sea coast. The MIS 5e

477 age for these 4-5.5 m high coral terraces is consistent with similar elevation MIS 5e coral

478 terraces in northern Jeddah (Bantan et al., 2015) and along the western Red Sea coast from

479 Egypt south to Ethiopia and Djibouti (Hoang and Taviani, 1988, 1991; Andres et al., 1988;

480 Reyss et al., 1993; Walter et al., 2000). It is also consistent with the dating of a reef terrace 2-

481 6 m apsl on the Sudanese coast, which gave ages of 125-142 ka (Hoang et al., 1996). These

482 dates suggest that Saudi Arabian reef growth at MIS 5e during the last interglacial is

483 consistent with the results obtained from the western Red Sea coast from Egypt south to

484 Ethiopia and Djibouti (Andres et al., 1988; Hoang and Taviani, 1988, 1991; Reyss et al.,

485 1993; Plaziat et al., 2008). This strongly supports the stability of most of the Saudi and

486 western Red Sea coast sinceACCEPTED at least the last interglacial, apartMANUSCRIPT from the northern part around

487 the Gulf of Aqaba. Since MIS 9 coral reefs and terraces also occur at similar altitudes around

488 the Red Sea coast this tectonic stability may extend back to at least MIS 9.

489 Farther south at the Farasan Islands, even though samples were not collected for this study

490 (due to restricted access in this area), previous field work showed a sequence of coral terraces

491 2-6 m apsl with a very similar internal coral-dominated structure (Fig.7; Bantan and Abu-

21

ACCEPTED MANUSCRIPT

Ammar Manaa

492 Zied, 2014). These terraces are probably also of MIS 5e age and could support the tectonic

493 stability of most of the Saudi coast for at least the last 125,000 years.

494 In the northern part of Saudi Arabia, the results show a progressive uplift north of Duba to the

495 Gulf of Aqaba (Fig.8). The dated 6-10 m high MIS 5 coral terraces exposed north and south

496 of Duba increase in height to 18-25 m around Maqna and 16-20 m near Haql in the centre of

497 the gulf of Aqaba before decreasing to a 7-10 m high terraces at the north end of the gulf

498 (Jado et al., 1989; Scholz et al., 2004; Parker et al., 2012). A coral terrace of possible MIS 5

499 age has also been reported from 15 m apsl just south of Dahab on the Sinai Peninsula

500 (Shalaby and Shawky, 2014). These findings are consistent with previous studies in the Gulf

501 of Aqaba (Gvirtzman et al., 1992; Strasser et al., 1992; Gvirtzman, 1994; Bosworth and

502 Taviani, 1996; Plaziat et al., 1998; Shalaby and Shawky, 2014). The uplift of this last

503 interglacial reef is comparable to other reefs in the northern Red Sea. For example, the height

504 of a reef terrace from Um Seed at the southern tip of the Sinai Peninsula (northern Red Sea),

505 3-6 m apsl, was dated by U/Th and gave an age of 118-125 ka (El-Asmar, 1997) and a raised

506 reef, about 6-12 m, between Sharm al Harr and Sharm al Bad north of Duba was dated at 95-

507 112 ka (Jado et al., 1989). It also consistent with higher elevation stage MIS 5e coral terraces

508 in some more highly tectonic areas such as Barbados, Bahamas, New Guinea and Indonesia

509 (Fairbanks and Matthews, 1978; Aharon and Chappell, 1986; Vacher and Hearty, 1989;

510 Pirazzoli et al., 1991; Hearty and Kindler, 1993).

511 5.5 Sea level implicationsACCEPTED MANUSCRIPT

512 The definition of past sea level heights has been determined from around the world,

513 especially in areas with tectonic stability. For example, the relative sea level height during

514 much of the MIS 5e along the Western Australian coast (129-120 ka) was 3-4 m apsl, while a

515 substantial sea-level rise to about 9 m apsl has been suggested at the end of the last

516 interglacial at 118 ka (O‟Leary et al., 2013). Also, a sea level rise to about 5 m apsl during

22

ACCEPTED MANUSCRIPT

Ammar Manaa

517 stage 5e has been observed in the Bahamas, Bermuda Island and Florida (Stirling et al., 1998;

518 Chen et al., 1991; Muhs et al., 2002, 2004) while in locations such as Haiti and Barbados, sea

519 level was at 2.7-5 m and 6 m apsl at approximately 132 ka and 128 ka, respectively

520 (Schellmann and Radtke, 2004; Dumas et al., 2006). Thompson et al. (2011) revised the ages

521 for part of the Bahamas succession and confirmed the presence of two highstand events

522 between 123 ka and 114 ka. In the eastern Atlantic Ocean, stratigraphic data from the Canary

523 Islands show MIS 5e points at between 0-2 m apsl (Zazo et al., 2002). These latter values

524 may be related to neotectonism since during the early and middle Pleistocene the Canary

525 Islands had been rising but over the past 300 ka they were probably stable or subsiding (Zazo

526 et al., 2002). Frank et al. (2006) used a MIS 5e height of 6±3 m apsl in their estimation of

527 subsidence rates on the west coast of New Caledonia. PALSEA (2010) suggested that the

528 MIS 5e peak sea level of 3-6 m apsl was attained by 126±1 ka while Dutton et al. (2015a)

529 dated raised in situ corals from the Seychelles islands that indicated a sea level of 7.6±1.7 m

530 at 125 ka.

531 Dutton and Lambeck (2012) concluded a MIS 5e highstands occurred between 5.5 and 9 m

532 apsl based on isostatic modelling and stratigraphic evidence from a wide range of global

533 localities. However there was considerable variability between the different locations.

534 Vyverberg et al. (2014) recorded at least two MIS 5e highstand signatures on the Seychelles

535 islands. A later review paper by Dutton et al. (2015b) has indicated that there is no clear 536 consensus on the preciseACCEPTED height and age of MIS 5e highstands MANUSCRIPT but there appear to be one of 537 more highstands between 129-116 ka that reached heights of 6-9 m above present sea level. 538 The current data from the Saudi Red Sea coast is consistent with these findings.

539 Since the central and southern portions of the Saudi Red Sea coast appear to have been

540 tectonically stable for the last 125 ka, and possibly since MIS 9, the information from the

541 Saudi terraces can be used to place limits on the last interglacial sea level in the Red Sea area.

23

ACCEPTED MANUSCRIPT

Ammar Manaa

542 The consistent elevation of these terraces at 4-5.5 m apsl over a lateral extent of up to 5 km

543 inland, and their juxtaposition with respect to back-reef lagoonal deposits, suggests that the

544 last interglacial maximum sea level must have been at least 6 m above present sea level. The

545 molluscan fauna associated with the coral terraces is consistent with a very shallow subtidal

546 environment and the foraminiferal fauna recorded in the terrace 3-5.5 m apsl in northern

547 Jeddah by Bantan et al. (2015) is equivalent to modern faunas abundant at 1-3 m water

548 depths in Shuaiba Lagoon (Abu-Zied and Bantan, 2013). The combination of these factors

549 suggests that locally the last interglacial maximum sea level was at least 5-6 m above present

550 sea level at about 120 ka. The timing of this highstand is very similar to the MIS 5e sea level

551 curve estimated by Siddall et al. (2003) and Rohling et al. (2008) from δ18O data derived

552 from cores in the central Red Sea. However, as noted by Lambeck et al. (2012), although the

553 Red Sea region is tectonically stable isostatic rebound is likely to have occurred following

554 melting of the MIS 6 Eurasian icesheet. This would have resulted in uplift in the Red Sea

555 region being most pronounced in the northern part near the Gulfs of Suez and Aqaba but this

556 cannot be distinguished around the Gulf of Aqaba because of the active tectonism in this area.

557

558 6. Conclusions

559 Dating of the coral terraces along the Saudi coast indicate a major period of coral reef growth 560 between 122-119 ka duringACCEPTED MIS 5e. These subhorizontal reefMANUSCRIPT facies are up to 5 km wide and 561 have upper elevations of 3.5-5.5 m apsl along the central and southern parts of the Saudi Red 562 Sea coast. These results are supported by the same range of ages for similar terraces up to1 m

563 high from along the western coast of the Red Sea (Butzer and Hansen, 1968; Veeh and

564 Giegengack, 1970; El-Moursi, 1992; Arvidson et al., 1994; El Moursi et al., 1994; El-Moursi

565 and Montaggioni, 1994; Plaziat et al., 1998). Since this region has been tectonically stable for

566 at least the last 125,000 years (Lambeck et al., 2012) the internal structure and fauna in these

24

ACCEPTED MANUSCRIPT

Ammar Manaa

567 coral terrace deposits can be related to a MIS 5e sea level high stand that was at least 6 m apsl

568 at about 120 ka.

569 Nevertheless, the Saudi coast north and south of Duba and along the Gulf of Aqaba has been

570 uplifted by faulting to 16-20 m at Haql since 108-120 ka. Our estimate of the degree of uplift

571 is comparable to other results north of Duba and along the Gulf of Aqaba, where a 6-10 m

572 reef terrace provided an age of 95-120 ka (Jado et al., 1989), a 7-10 m reef terrace gave an

573 age of 121 ka (Scholz et al., 2004) and a 2.5 -12 m reef terrace set an age of 77 -129 ka

574 (Parker et al., 2012). This uplift can be associated with the late Pleistocene rifting along the

575 Gulf of Aqaba transform fault (Bosworth and McClay, 2001; Shalaby and Shawky, 2014).

576

577 7. Acknowledgements

578 This study was supported by a scholarship and field expenses provided by King Abdulaziz

579 University, Saudi Arabia, and by facilities and funding from Geoquest Research Centre at the

580 University of Wollongong, Australia. HVM acknowledges support from Australian Research

581 Council Future Fellowship FT140100286. The authors acknowledge the valuable comments

582 from Paul Hearty and an anonymous reviewer of an earlier version of this manuscript.

583

584 8. References ACCEPTED MANUSCRIPT 585 Abu-Zied, R.H., Bantan, R.A., 2013. Hypersaline benthic foraminifera from the Shuaiba 586 Lagoon, eastern Red Sea, Saudi Arabia: their environmental controls and usefulness in sea

587 level reconstruction. Marine Micropaleontology, 103, 51-67.

25

ACCEPTED MANUSCRIPT

Ammar Manaa

588 Aharon, P., Chappell, J., 1986. Oxygen isotopes, sea level changes and the temperature

589 history of a coral reef environment in New Guinea over the last 105 years. Palaeogeography,

590 Palaeoclimatology, Palaeoecology, 56, 337-379.

591 Al-Rifaiy, I.A., Cherif, O.H., 1988. The fossil coral reefs of Al-Aqaba, Jordan. Facies, 18,

592 219-230.

593 Al-Shanti, A., 1993. Geology of the Arabian Shield (Arabic): King Abdulaziz University

594 Publishing, Jeddah.

595 Andersen, M.B., Stirling, C.H., Zimmermann, B., Halliday, A.N., 2010. Precise

596 determination of the open ocean 234U/238U composition. Geochemistry, Geophysics,

597 Geosystems, 11, Q12003.

598 Andres, W., Radtke, U., 1988. Quartäre strandterrassen an der küste des Gebel Zeit (Golf von

599 Suez/Ägypten). Erdkunde, 42, 7-16.

600 Arvidson, R., Becker, R., Shanabrook, A., Luo, W., Sturchio, N., Sultan, M., Lofty, Z.,

601 Mahmood, A.M., El Alfy, Z., 1994. Climatic, eustatic, and tectonic controls on Quaternary

602 deposits and landforms, Red Sea Coast, Egypt. Journal of Geophysical Research, 99, B6,

603 12175-12190.

604 Bantan, R.A., Abu-Zied, R.H., 2014. Sediment characteristics and molluscan fossils of the

605 Farasan Islands shorelines, southern Red Sea, Saudi Arabia. Arabian Journal of Geosciences,

606 7, 773-787. ACCEPTED MANUSCRIPT

607 Bantan, R.A., Abu-Zied, R.H., Haredy, R.A., 2015. Lithology, fauna and environmental

608 conditions of the Late Pleistocene raised reefal limestone of the Jeddah coastal plain, Saudi

609 Arabia. Arabian Journal of Geosciences, 8, 9887-9904.

26

ACCEPTED MANUSCRIPT

Ammar Manaa

610 Basaham, A.S., 1998. The composition and diagenetic features of the inland Quaternary

611 coralline limestone, south Sharm Ubhur, Red Sea coastal plain of Saudi Arabia. Journal of

612 King Adulaziz University, Marine Science, 9, 75-87.

613 Behairy, A.K.A., 1983. Marine transgressions in the west coast of Saudi Arabia between mid-

614 Pleistocene and present. Marine Geology, 52, 25-31.

615 Behairy, A.K.A., Durgaprasada-Rao, N.V.N., El-Shater, A., 1991. A siliciclastic coastal

616 sabkha, Red Sea coast, Saudi Arabia, Journal of King Abdulaziz University, Marine Science,

617 2, 65-77.

618 Berry, L., Whiteman, A.J., Bell, S.V., 1966. Some radiocarbon dates and their

619 geomorphological significance, emerged reef complex of the Sudan. Zeitschrift für

620 Geomorphologie, 10, 119-143.

621 Bosworth, W., McClay, K., 2001. Structural and stratigraphic evolution of the Gulf of Suez

622 rift, Egypt: a synthesis. In: Ziegler, P.A., Cavazza, W., Robertson, A.H.F., Crasquin-Soleau,

623 S. (eds.), Peri-Tethys Memoir 6: Peri-Tethyan Rift/Wrench Basins and Passive Margins.

624 Mémoires du Muséum National d'Histoire Naturelle, 186, 567-606.

625 Bosworth, W. Taviani, M., 1996. Late Quaternary reorientation of stress field and extension

626 direction in the southern Gulf of Suez, Egypt: evidence from uplifted coral terraces,

627 mesoscopic fault arrays, and borehold breakouts. Tectonics, 15, 0278-7407.

628 Brown, G.F., Schmidt, D.L.,ACCEPTED Huffman, A.C., 1989. Geology MANUSCRIPT of the Arabian Peninsula, shield 629 area of western Saudi Arabia. U.S. Geological Survey Professional Paper, 560-A, 188.

630 Butzer, K.W., Hansen, C.L., 1968. The coastal plain of Mersa Alam. In: Butzer, K.W.,

631 Hansen, C.L. (eds.), Desert and River in Nubia: Geomorphology and Prehistoric

632 Environments at the Aswan Reservoir. University of Wisconsin Press, Madison, p. 395-432.

27

ACCEPTED MANUSCRIPT

Ammar Manaa

633 Chapman, R.W., 1978. Geomorphology. In Al-Sayari, S.S., Zoti, J.G., (eds) Quaternary

634 Period in Saudi Arabia, 19-30. Springer, Vienna.

635 Chappell, J., Shackleton, N.J., 1986. Oxygen isotopes and sea level. Nature, 324, 137-140.

636 Cheng, H., Adkins, J., Edwards, R.L., Boyle, E.A., 2000. U-Th dating of deep-sea corals.

637 Geochimica et Cosmochimica Acta, 64, 2401–2416.

638 Chen, Y., Brumby, S., Jacobson, G., Beckwith, A.J.L., Polach, H.A., 1991. A novel

639 application of the ESR method: dating of insular phosphorites and reef limestone. Quaternary

640 Science Reviews, 11, 209-217.

641 Clark, T.R., Zhao, J.-X., Feng, Y.X., Done, T.J., Jupiter, S., Lough, J., Pandolfi, J.M., 2012.

642 Spatial variability of initial 230Th/232Th in modern Porites from the inshore region of the Great

643 Barrier Reef. Geochimica et Cosmochimica Acta, 78, 99-118.

644 Clark, T.R., Zhao, J.-X., Roff, G., Feng, Y.X., Done, T.J., Nothdurft, L.D., Pandolfi, J.M.,

645 2014. Discerning the timing and cause of historical mortality events in modern Porites from

646 the Great Barrier Reef. Geochimica et Cosmochimica Acta, 138, 57–80.

647 Dawood, Y.H., Aref, M.A., Mandurah, M.H., Hakami, A., Gameil, M., 2013. Isotope

648 geochemistry of the Miocene and Quaternary carbonate rocks in Rabigh area, Red Sea coast,

649 Saudi Arabia. Journal of Asian Earth Sciences, 77, 151-162.

650 Dullo, W.C., 1984. Progressive diagenetic sequence of aragonite structures: Pleistocene coral

651 reefs and their modern counterpartsACCEPTED on the eastern Red Sea MANUSCRIPT coast, Saudi Arabia.

652 Palaeontographic Americana, 54, 254-260.

653 Dullo, W.C., 1990. Facies, fossil record, and age of Pleistocene reefs from the Red Sea

654 (Saudi Arabia). Earth and Environmental Science, 22, 1-45.

28

ACCEPTED MANUSCRIPT

Ammar Manaa

655 Dullo, W.C., Montaggioni, L.F., 1998. Modern Red Sea coral reefs: a review of their

656 morphologies and zonation. In: Purser, B.H., Bosence, D.W.J., (Eds), Sedimentation and

657 Tectonics in Rift Basins: Red Sea–Gulf of Aden, 583-594. London: Chapman and Hall.

658 Dumas, B., Hoang, C.T., Raffy, J., 2006. Record of MIS 5 sea-level highstands based on

659 dated coral terraces at Haiti. Quaternary International, 145-146, 106-118.

660 Dutton, A., Lambeck, K., 2012. Ice volume and sea level during the Last Interglacial.

661 Science, 337, 2016-2019.

662 Dutton, A., Webster, J. M., Zwartz, D., Lambeck, K., Wohlfarth, B., 2015a. Tropical tales of

663 polar ice: evidence of Last Interglacial polar ice sheet retreat recorded by fossil reefs of the

664 granitic Seychelles islands. Quaternary Science Reviews, 107, 182-196.

665 Dutton, A, Carlson, A. E., Long, A. J., Milne, G. A., Clark, P. U., DeConto, R., Horton, B. P.,

666 Rahmstorf, S., Raymo, M. E., 2015b. Sea-level rise due to polar ice-sheet mass loss during

667 past warm periods. Science, 349(6244), aaa4019 1-9.

668 Edwards, F.J., 1987. Climate and oceanography. In: Edwards, A.J., Head, S.M., (eds) Key

669 Environments, Red Sea, 45-68. Pergamon, Oxford.

670 El-Asmar, H.M., 1997. Quaternary isotope stratigraphy and paleoclimate of coral reef

671 terraces, Gulf of Aqaba, south Sinai, Egypt. Quaternary Science Reviews, 16, 911-924.

672 El-Moursi, M., 1992. Évolution quaternaire de la plaine côtière de la Mer Rouge entre

673 Hurghada et Marsa Alam,ACCEPTED Égypte.- PhD thesis, Université MANUSCRIPT d'Aix-Marseille II, n°

674 92AIX22087, 245 p.

675 El-Moursi, M., Montaggioni, L.F., 1994. Diagenesis of Pleistocene reef-associated sediments

676 from the Red Sea coastal plain, Egypt. Sedimentary Geology, 90, 49-59.

29

ACCEPTED MANUSCRIPT

Ammar Manaa

677 El-Moursi, M., Hoang, C.T., Fahmy El Fayoumi, I., Hegab, O., Faure, M., 1994. Pleistocene

678 evolution of the Red Sea coastal plain, Egypt: evidence from uranium-series dating of

679 emerged reef terraces. Quaternary Science Reviews, 13, 345-359.

680 El-Rifaiy, I., Cherif, O., 1988. The fossil coral reefs of A1 Aqaba, Jordan. Facies, 18, 219-

681 230.

682 Encyclopedia of Life (EOL), http://eol.org/

683 Fairbanks, R., Matthews, R.K., 1978. The marine oxygen isotope record in Pleistocene coral,

684 Barbados, West Indies. Quaternary Research, 10, 181-196.

685 Faure, H., Hoang, C.T., Lalou, C., 1980. Datations 230Th/234U des calcaires coralliens et

686 mouvements verticaux à Djibouti. Bulletin de la Société géologique de France, (7° série), 22,

687 959-962.

688 Frank, N., Turpin, L., Cabioch, G., Blamart, D., Tressens-Fedou, M., Colin, C., Jean-

689 Baptiste, P., 2006. Open system U-series ages of corals from a subsiding reef in New

690 Caledonia: implications for sea level changes, and subsidence rate. Earth and Planetary

691 Sciences Letters, 249, 274-289.

692 Friedman, G.M., 1965. A fossil shoreline reef in the Gulf of Elat (Aqaba), Israel. Earth

693 Science, 14, 86-90.

694 Friedman, G.M., Amiel, A.J., Schneidermann, N., 1974. Submarine cementation in reefs:

695 examples from the Red Sea.ACCEPTED Journal of Sedimentary Petrology, MANUSCRIPT 44, 816-825.

696 Gvirtzman, G., 1994. Fluctuations of sea-level during the past 400,000 years: the record of

697 Sinai, Egypt (northern Red Sea). Coral Reefs 13, 203-214.

30

ACCEPTED MANUSCRIPT

Ammar Manaa

698 Gvirtzman, G., Buchbinder, B., 1978. Recent and Pleistocene coral reefs and coastal

699 sediments of the Gulf of Aqaba. 10th International Sedimentological Congress, Jerusalem,

700 Post Congress Excursion, Y4, 163-191.

701 Gvirtzman, G., Friedman, G.M., 1977. Sequence of progressive diagenesis in coral reefs.

702 Studies in Geology 4, 357-380.

703 Gvirtzman, G., Kronfeld, J., Buchbinder, B., 1992. Dated coral reefs of southern Sinai (Red

704 Sea) and their implication to late Quaternary sea levels. Marine Geology, 108, 29-37.

705 Hearty, P., 1987. New data on the Pleistocene of Mallorca. Quaternary Science Reviews, 6,

706 245-257.

707 Hearty, P.J., 2002. A revision of the Late Pleistocene stratigraphy of Bermuda. Sedimentary

708 Geology, 153, 1-21.

709 Hearty, P., Kindler, P., 1993. New perspective on Bahamian geology: San Salvador Island,

710 Bahamas. Journal of Coastal Research, 912, 577-594.

711 Hearty, P.J., Neumann, A.C., Hollin, J.T., O‟Leary, M.J., McCulloch, M.T., 2007. Global

712 sea level fluctuations during the last interglaciation (MIS 5e). Quaternary Science Reviews,

713 26, 2090-2112.

714 Hellstrom, J., 2003. Rapid and accurate U/Th dating using parallel ion-counting multi-

715 collector ICP-MS. Journal of Analytical Atomic Spectrometry, 18, 1346-1351.

716 Hoang, C.T., Taviani, M.,ACCEPTED 1988. 230Th/234U ages of raised MANUSCRIPT coral reefs from tectonically

717 uplifted Red Sea islands. Tectonophysics, 150, 251

718 Hoang, C.T., Taviani, M., 1991. Stratigraphic and tectonic implications of uranium-series

719 dated coral reefs from uplifted Red Sea islands. Quaternary Research, 35, 264-273.

31

ACCEPTED MANUSCRIPT

Ammar Manaa

720 Hoang, C.T., Dalongeville, R., Sanlaville, P., 1996. Stratigraphy, tectonics and palaeo-

721 climatic implications of uranium-series-dated coral reefs from the Sudanese coast of the Red

722 Sea. Quaternary International, 31, 47-51.

723 Jado, A.R., Zotl, J.G., (eds.), 1984. Quaternary Period in Saudi Arabia, vol. 2. Springer:

724 Berlin.

725 Jado, A.R., Hötzl, H., Roscher, B., 1989. Development of sedimentation along the Saudi

726 Arabian Red Sea coast. In: First Saudi Symposium on Earth Science, Journal of King

727 Abdulaziz University - Earth Science, Special Issue, 3, 47-69.

728 Khalil, H.M., 2012. Pliocene-Pleistocene stratigraphy and marcofauna of the Farasan Islands,

729 south east Red Sea, Saudi Arabia. Arabian Journal of Geosciences, 5, 1223-1245.

730 Lambeck, K., Chappell, J., 2001. Sea level change through the last glacial cycle. Science 292,

731 679-686.

732 Lambeck, K., Purcell, A., Dutton, A., 2012. The anatomy of interglacial sea levels: the

733 relationship between sea levels and ice volumes during the Last Interglacial. Earth and

734 Planetary Science Letters, 315-316, 4-11.

735 Lieske, E., Myers, R.F., 2004. Coral Reef Guide Red Sea. Harper Collins Publishers Ltd,

736 London.

737 Lisiecki, L.E., Raymo, M.E., 2005. A Pliocene-Pleistocene stack of 57 globally distributed

738 benthic δ18O records. PaleoceanographyACCEPTED 20, PA1003, doi:10.1029/ MANUSCRIPT 2004PA001071.

739 MacIntyre, I.G., Mountjoy, E.W., D‟Anglejan, B.F., 1968. An occurrence of submarine

740 cementation of carbonate sediments off the west coast of Barbados, W. I. Journal of

741 Sedimentary Petrology, 38, 660-664.

32

ACCEPTED MANUSCRIPT

Ammar Manaa

742 Manaa, A., 2011. Late Pleistocene raised coral reefs in the eastern Red Sea – Rabigh, Saudi

743 Arabia. M Sc thesis, University of Wollongong. http://ro.uow.edu.au/theses/3501

744 Mandurah, M.H., 2009. A preliminary paleontologic study on some Pleistocene corals,

745 Rabigh area, Red Sea Coast, Saudi Arabia. Egypt. Journal of Paleontology, 9, 213-228.

746 Mandurah, M.H., 2010. Paleontological studies on the Neogene (Miocene and Quaternary)

747 carbonate rocks of Rabigh-Ubhur areas, Red Sea coast, Saudi Arabia. Journal of King

748 Abdulaziz University Earth Science, 21, 105-143.

749 Mandurah, M.H., Aref, M.A.M., 2012. Lithostratigraphy and standard microfacies types of

750 the Neogene carbonates of Rabigh and Ubhur areas, Red Sea coastal plain of Saudi Arabia.

751 Arabian Journal of Geoscience, 5, 1317-1332.

752 Marine Environment Protection Unit (MEPU), 2011. Tide Tables 2012 Red Sea, Saudi

753 Aramco, Jeddah.

754 Marshall, J.F., 1983. Submarine cementation in a high-energy platform reef: One Tree Reef,

755 southern Great Barrier Reef. Journal of Sedimentary Petrology, 53, 1133-1149.

756 Marshall, J.F., Davies, P.J., 1981. Submarine lithification on windward reef slopes:

757 Capricorn-Bunker Group, southern Great Barrier Reef. Journal of Sedimentary Petrology, 51,

758 953-960.

759 Meyers, W.J., 1978. Carbonate cements: their regional distribution and interpretation in

760 Mississippian limestonesACCEPTED of southwestern New Mexico. SMANUSCRIPTedimentology, 25, 371-400.

761 Muhs, D.R., Simmons, K.R., Steinke, B,. 2002. Timing and warmth of the Last Interglacial

762 period: new U-series evidence from Hawaii and Bermuda and a new fossil compilation for

763 North America. Quaternary Science Reviews, 21, 1355-1383.

33

ACCEPTED MANUSCRIPT

Ammar Manaa

764 Muhs, D.R., Wehmiller, J.F., Simmons, K.R., York, L.L., 2004. Quaternary sea-level history

765 of the United States. In: Gillespie, A.R., Porter, S.C., Atwater, B.F., (Eds.), The Quaternary

766 Period in the United States. Elsevier, Amsterdam, 147-183.

767 Murray-Wallace, C.V., Woodroffe, C.D., 2014. Quaternary Sea-level Changes: a Global

768 Perspective. Cambridge University Press, Cambridge.

769 Nesteroff, W.D., 1959. Age des derniers mouvements du graben de lamer Rouge déterminé

770 par la méthode du C14 appliquée aux récifs fossiles (Age of the last movements of the Red

771 Sea graben determined by the carbon-14 method applied to fossil reefs). Societe Géologie

772 France Bulletin, 7, 415-418.

773 Obura, D., Fenner, D., Hoeksema, B., Devantier, L. and Sheppard, C. 2008. Tubipora musica.

774 The IUCN Red List of Threatened Species. Version 2015.1.

775 downloaded on 02 June 2015.

776 O‟Leary, M.J., Hearty, P.J., Thompson, W.G., Raymo, M.E., Mitrovica, J.X., Webster, J.M.,

777 2013. Ice sheet collapse following a prolonged period of stable sea level during the last

778 interglacial. Nature Geoscience, 6, 796–800.

779 PALSEA (Paleo Sea level working group), 2010. The sea-level conundrum: case studies from

780 palaeo-archives. Journal of Quaternary Science, 25, 19-25.

781 Parker, J.H., Gischler, E., Eisenhauer, A., 2012. Biodiversity of foraminifera from Late

782 Pleistocene to HoloceneACCEPTED coral reefs, South Sinai, Egypt. MarMANUSCRIPTine Micropaleontology, 86-87,

783 59-75.

784 PERSGA/GEF, 2003. Coral reefs in the Red Sea and Gulf of Aden, surveys 1990 to 2000

785 summary and recommendations. PERSGA Technical Series No. 7, PERSGA, Jeddah.

34

ACCEPTED MANUSCRIPT

Ammar Manaa

786 Pirazzoli, P.A., Radtke, U., Hantoro, W.S., Jouannic, C., Hoang, C.-T., Causse, C., 1991.

787 Quaternary raised coral reef terraces on Sumba Island, Indonesia. Science, 252, 1834-1836.

788 Plaziat, J.-C., Baltzer, F., Choukri, A., Conchon, O., Freytet, P., Orszag- Sperber, F.,

789 Raguideau, A., Reyss, J.-L., 1998. Quaternary marine and continental sedimentation in the

790 northern Red Sea and Gulf of Suez (Egyptian coast): influences of rift tectonics, climatic

791 changes and sealevel fluctuations. In: Purser, B.H., Bosence, D.W.J., (Eds), Sedimentation

792 and Tectonics of Rift Basins: Red Sea–Gulf of Aden. Chapman and Hall, London, pp. 537-

793 573.

794 Plaziat, J.-C., Reyss, J.-L., Choukri, A, Cazala, C., 2008. Diagenetic rejuvenation of raised

795 coral reefs and precision of dating. The contribution of the Red Sea reefs to the question of

796 reliability of the uranium-series datings of Middle to Late Pleistocene key reef-terraces of the

797 world. Carnets de Géologie 4, 1-35.

798 PME (Presidency of Meteorology and Environment), 2005. Saudi Arabia. First national

799 communication, Kingdom of Saudi Arabia, submitted to the United Nations Framework

800 Convention on Climate Change (UNFCCC).

801 Reyss, J.L., Choukri, A., Plaziat, J.C., Purser, B.H., 1993. Datations radiochimiques des

802 récifs coralliens de la rive occidentale du nord de la mer Rouge, premières implications

803 stratigraphiques et tectoniques. Comptes Rendus de l‟Académie des Sciences, 317, 487-491.

804 Rohling, E.J., Grant, K.,ACCEPTED Hemleben, C., Siddall, M., Hoogakker MANUSCRIPT, B.A.A., Bolshaw, M., 805 Kucera, M., 2008. High rates of sea-level rise during the last interglacial period. Nature 806 Geoscience, 1, 38-42.

807 Ross, D.A., Whitmarsh, R.B., Ali, S., Boudeauz, J.E., Coleman, R.G., Fleischer, R.L.,

808 Girdler, R., Manheim, F., Matter, A., Nigrini, C., Stoffers, P. and Supko, P., 1972. Deep sea

809 drilling project in the Red Sea. Geotimes, 17, 24-26.

35

ACCEPTED MANUSCRIPT

Ammar Manaa

810 Sagga , A.M., 2004. Physical Geography of the Kingdom of Saudi Arabia (Arabic), 3rd ed,

811 Kunoz Al Maarefa Publishing: Jeddah.

812 Saifullah, S.M., 1996. Mangrove ecosystem of Saudi Arabian Red Sea coast - an overview.

813 Journal of King Abdulaziz University, Marine Science, 7, 263-270.

814 Schellmann, G., Radtke, U., 2004. A revised morpho- and chronostratigraphy of the late and

815 middle Pleistocene coral reef terraces on southern Barbados (West Indies). Earth-Science

816 Reviews, 64, 157-187.

817 Scholz, D., Mangini, A., 2007a. Estimating the uncertainty of coral isochron U-Th ages.

818 Quaternary Geochronology, 1, 279-288.

819 Scholz, D., Mangini, A., 2007b. How precise are U-series coral ages? Geochimica et

820 Cosmochimica Acta, 71, 1935-1948.

821 Scholz, D., Mangini, A., Felis, T., 2004. U-series dating of diagenetically altered fossil reef

822 corals. Earth and Planetary Science Letters, 218, 163-178.

823 Sea life base, http://www.sealifebase.org/

824 Shackleton, N.J., 1987. Oxygen isotopes, ice volume and sea level. Quaternary Science

825 Reviews, 6, 183-190.

826 Shaked, Y., Agnon, A., Lazar, B., Marco, S., Avner, U., Stein. M., 2004. Large earthquakes 827 kill coral reefs at the northACCEPTED‐west Gulf of Aqaba. Terra Nova MANUSCRIPT, 16, 133-138. 828 Shalaby, A., Shawky, M., 2014. Morphotectonics of Kid drainage basin, southeastern Sinai: a 829 landscape evolution coeval to Gulf of Aqaba – Dead Sea rifting. Journal of African Earth

830 Sciences, 100, 289-302.

831 Sheppard, C.R.C., Price, A.R.G., Roberts, C., 1992. Marine Ecology of the Arabian Region.

832 Patterns and Processes in Extreme Tropical Environments. Academic Press, London.

36

ACCEPTED MANUSCRIPT

Ammar Manaa

833 Shepherd, M.J., Price, D.M., 1990. Thermoluminescence dating of Late Quaternary dune

834 sand, Manawata/Horowhenua area, New Zealand: a comparison with 14C age determinations.

835 New Zealand Journal of Geology and Geophysics, 33, 535-539.

836 Siddall, M., Rohling, E.J., Almogi-Labin, A., Hmeieben, Ch., Meischner, D., Schmeizer, I.,

837 Smeed, D.A., 2003. Sea-level fluctuations during the last glacial cycle. Nature, 423, 853-858.

838 Sneh, A., Friedman, G.M., 1980. Spur and groove patterns on the reefs on the northern gulf

839 of the Red Sea. Journal of Sedimentary Petrology, 50, 981-986.

840 Stirling, C. H., Andersen, M. B., 2009. Uranium-series dating of fossil coral reefs: extending

841 the record beyond the last glacial cycle. Earth and Planetary Science Letters, 284, 269-283.

842 Stirling, C.H., Esat, T.M., Lambeck, K., McCulloch, M.T., 1998. Timing and duration of the

843 Last Interglacial: evidence for a restricted interval of widespread coral reef growth. Earth and

844 Planetary Science Letters, 160, 745-762.

845 Strasser, A., Strohmenger, C., Davaud, E., Bach, A., 1992. Sequential evolution and

846 diagenesis of Pleistocene coral reefs (south Sinai, Egypt). Sedimentary Geology, 78, 59-79.

847 Thompson, W.G., Spiegelman, M.W., Goldstein, S.L., Speed, R.C., 2003. An open-system

848 model for the U-series age determinations of fossil corals. Earth and Planetary Science

849 Letters, 210, 365-381.

850 Thompson, W. G., Curran, H. A., Wilson, M. A., White, B., 2011. Seqa-level oscillations

851 during the last interglacialACCEPTED highstand recorded by Bahamas MANUSCRIPT corals. Nature Geoscience, 4, 684-

852 687.

853 Vacher, H., Hearty, P., 1989. History of stage 5 sea level in Bermuda: review with new

854 evidence of a brief rise to present sea level during substage 5a. Quaternary Science Reviews,

855 8, 159-168.

37

ACCEPTED MANUSCRIPT

Ammar Manaa

856 Veeh, H.H., Giegengack, R., 1970. Uranium-series ages of corals from the Red Sea. Nature,

857 226, 155-156.

858 Villemant, B., Feuillet, N., 2003. Dating open systems by the 238U-234U-230Th method:

859 application to Quaternary reef terraces. Earth and Planetary Science Letters, 210, 105-118.

860 Vincent, P., 2008. Saudi Arabia: an Environmental Overview. Taylor and Francis: London.

861 Vyverberg, K.; Dutton, A.; Dechnik, B.; Webster, J.; and Zwartz, D., 2014. Evidence from

862 the Seychelles of Last Interglacial Sea Level Oscillations. American Geophysical Union, Fall

863 Meeting 2014, abstract #PP33C-1251.

864 Walter, R.C., Buffler, R.T., Bruggemann, J.H., Guillaume, M.M.M., Berhe, S.M., Negassi,

865 B., Libsekal, Y., Cheng, H., Edwards, R.L., Von Cosel, R., Néraudeau, D., Gagnon, M.,

866 2000. Early human occupation of the Red Sea coast of Eritrea during the last interglacial.

867 Nature, 405, 65-69.

868 Woodroffe, C., Murray-Wallace, C., Bryant, E., Brooke, B., Heijnis, H., 1995. Late

869 Quaternary sea-level highstands in the Tasman Sea: evidence from Lord Howe Island.

870 Marine Geology, 125, 61-72.

871 World Register of Marine Species, www.marinespecies.org/

872 Zazo, C., Goy, J.L., Hillaire-Marcel, C., Gillot, P.Y., Soler, V., Gonzalez, J.A., Dabrio, C.J.,

873 Ghaleb, B., 2002. Raised marine sequences of Lanzarote and Fuerteventura revisited – a

874 reappraisal of relative seaACCEPTED level changes and vertical movements MANUSCRIPT in the eastern Canary Islands

875 during the Quaternary. Quaternary Science Reviews, 21, 2019-2046.

876 Zhao, J-X., Yu, K.F., Feng, Y.X., 2009. High-precision 238U-234U-230Th disequilibrium

877 dating of the recent past – a review. Quaternary Geochronology, 4, 423-433.

38

ACCEPTED MANUSCRIPT

Ammar Manaa

878 Zhou, H., Zhao, J-X., Qing, W., Feng, Y.X., Tang, J., 2011. Speleothem-derived Asian

879 summer monsoon variations in Central China, 54-46 ka. Journal of Quaternary Science, 26,

880 781-790.

881

ACCEPTED MANUSCRIPT

39

ACCEPTED MANUSCRIPT

Ammar Manaa

882 List of Tables

883 Table.1. Faunal distribution and environment in the Saudi Red Sea coast from recent studies, 884 Mandurah, 2010, Khalil, 2012, Bantan and Abu-Zied, 2014, and Bantan et al, 2015. H= Haql, D= 885 Duba, Y= Yanbu, R= Rabigh, RB= Rabigh Back reef, RU= Al-Ruwais, J= Jeddah, Q= Qattan, F = 886 Farasan Islands (Bantan and Abu-Zied, 2014). (a) Lieske and Myers, 2004, (b) AIMS, (c) Obura et al., 887 2008, (d) Sea life base, (e) Encyclopedia of Life, (f)World Register of Marine Species. 888 Table 2. U/Th dating results for coral samples. The pooled mean ages and uncertainties were 889 calculated using formulae in Murray-Wallace and Woodroffe (2014, pp. 152-153) and exclude all 890 dates in italics based on the law of superposition, high uranium contents, high 234U/238U, low 891 230Th/238U and their 2-sigma analytical uncertainties.

892 /

893 List of Figures

894 Figure 1. Map of the Red Sea showing the location of the principal study areas along the Saudi 895 Arabian coast, salinity profiles and coral reef densities: shading represents areas of 500 m x 500 m 896 quadrat reef coverage (modified from Saifullah, 1996, and PERSGA/GEF, 2003)

897 Figure 2. A profile showing the height of the coastal exposures along the Saudi Red Sea coast.

898 Figure 3. A diagrammatic representation of the coral terraces in each of the main study areas showing 899 the elevations and the dated coral samples locations in relation to the modern coral platform elevation. 900 (a) Haql. (b) Duba. (c) Yanbu. (d) Rabigh. (e) Rabigh backreef. (f) profile from the Rabigh coast to 901 the backreef area showing the location of the quarry excavation. (g) Al-Ruwais. (h) Al-Qattan.

902 Figure 4. (a) The 8 m thick coral terrace at Haql is uplifted to 16-20 m apsl. (b) Corals in the upper 903 part of the terrace at Yanbu. (c) The terrace in the Rabigh area shows a well-cemented lower layer 904 overlain by a porous upper layer. (d) The coral terrace in the excavation area 400 m east of the beach 905 at Rabigh is dominated by Porites coral and appears as one unit with no major cementation. (e) 906 Sediment, clay, shells and coral fragments filling gaps within the coral terrace in the excavation at 907 Rabigh. (f) The back-reef carbonate apron at Rabigh contains carbonate rubble and an inverted 908 Porites coral. Pen is 0.15 m long. (g) Clypeaster humilis in the lower muddy lagoonal deposit in the 909 back-reef area at Rabigh. (h) Back-reef terrace in Al-Ruwais (southern Jeddah) is unconformably 910 overlain by fluvial sandstone. The exposure is 4 m high.

911 Figure 5. Chart showing the dating result of coral, shells and sediments from the current study and 912 dating results from other recentACCEPTED studies. MANUSCRIPT 913 Figure 6. Cross-polarized thin section image of the lower part in Yanbu area showing: a) 914 Aragonite cementation and diagenesis. b) Blocky calcite cement distributed equally around 915 most of the grains leaving an average porosity of 14%.

916 Figure 7. A sequence of coral terraces 3-6 m apsl located in Farasan Islands, southern Red 917 Sea, showing very similar elevation and internal coral-dominated structure to the MIS 5e dated 918 terraces. 919 Figure 8. Northern part of the Red Sea showing the elevations of the MIS 5e dated terraces indicating 920 tectonic uplift around the Gulf of Aqaba.

40

ACCEPTED MANUSCRIPT

Ammar Manaa

Formatted: Font: (Default) Times New Roman, 12 pt

ACCEPTED MANUSCRIPT

921

922 Figure 1

923

41

ACCEPTED MANUSCRIPT

Ammar Manaa

Formatted: Font: (Default) Times New Roman, 12 pt

924

925

926 Figure 2

927

ACCEPTED MANUSCRIPT

42

ACCEPTED MANUSCRIPT

Ammar Manaa

Formatted: Font: (Default) Times New Roman, 12 pt

928 ACCEPTED MANUSCRIPT 929 930 Figure 3

931

43

ACCEPTED MANUSCRIPT

Ammar Manaa

Formatted: Font: (Default) Times New Roman, 12 pt

ACCEPTED MANUSCRIPT

932

933 Figure 4

934

44

ACCEPTED MANUSCRIPT

Ammar Manaa

Formatted: Font: (Default) Times New Roman, 12 pt

935

936

937 Figure 5

938

ACCEPTED MANUSCRIPT

45

ACCEPTED MANUSCRIPT

Ammar Manaa

Formatted: Font: (Default) Times New Roman, 12 pt

939

940

941 Figure 6

942

ACCEPTED MANUSCRIPT

46

ACCEPTED MANUSCRIPT

Ammar Manaa

Formatted: Font: (Default) Times New Roman, 12 pt

943

944

945 Figure 7

946

ACCEPTED MANUSCRIPT

47

ACCEPTED MANUSCRIPT

Ammar Manaa

Formatted: Font: (Default) Times New Roman, 12 pt

947

948

949 Figure 8

950

ACCEPTED MANUSCRIPT

48

ACCEPTED MANUSCRIPT

Ammar Manaa

Species Availability Environment Corals Acropora humilis H, D, Y, R, Q Exposed outer reef flats (a) Acropora maryae D, Y, R, Q Protected reef slopes (a) Acropora pharaonis D, Y, R, Q Sheltered reef slopes 2-30 m (a) Cosinaraea monile Y, R, Q Lagoon and seaward reefs (a) Favia favus H, R, Q Reef back margins (b) Favia lacuna D, R Shallow exposed slopes (a) Favia stelligera Q Reef flats, upper lagoon and seaward slopes (a) Fungia puamotensis Y, Lagoons and reef slopes (a) Fungia scruposa H, D, R, RB, Q Rubble of reef flats and lagoon, to 25 m (a) Goniastrea edwardsi D, Y, R, Q Shallow subtidal communities 0-40 m (b) Lobophyllia corymbosa Y, Lagoon and sheltered reef slopes (a) Oxypora convoluta Y, Protected reef slopes (a) Platygyra daedalea R, Y, Q Reef flats, back reefs and slopes to 30m (a) Porites lutea H, D, Y, R, RU, Q Reef flats and lagoon (a) Porites nodifera H, D, Y, R, RB, Q Shallow bays and seaward to 18 m (a) Stylophora pistillata R, Q Reef flat and lagoons to 80 m (a) Tubipora musica D, Y, R, RB, Q Shallow reef, to 12 m or more(c) Bivalves Anadara antiquata H, R, J, Q, F 0 - 25 m(d) Anodontia edentula R, RB, J 0 - 20 m (d) Barbatia barbata R, J, Q 0 - 200 m (d,e) Antigona reticulata H, R, Q, F 0-12+ m(d) Trachycardium flavum H, Y, R, RB, J, Q, F 0 - 25 m (d) Tridacna gigas H, D, Y, R, J, Q, F 0-35 m (d) Tucetona pectunculus R, J, Q, F 0 - 20 m (d) Gastropods Cerithium eburneum R, J, Q 0 - 18 m (d) Chicoreus virgineus R, J, Q, F Shallow protected area and lagoons (a) Conus sp D, R, RB, Q, F Shallow environment (d) Cypraea historio R, J, Q Shallow environment, reefs (d) Cypraea staphylaea R, Q Shallow environment (d) Nerita albicillo H, R, RB Intertidal – 4 m (e) Strombus fasciatus R, J, Q Shallow environment, sea grass (d) S. tricornis R, J, Q, F Shallow environment, sea grass (d) Turbo petholatus R, Q Shallow lagoons and fore slopes to 10 m (a) Sponges Crella cyathophora H, R, Q To at least 20 m (a) Callyspongia viridis H, R, Q Shallow coral to about 22m (a) Pione cf. vastifica R Shallow reef to 20 m (a) Siphonochalina siphonella D, R, Q Lagoons and reef slopes , 2-35 m (a) Echinoids Clypeaster humilis R, RB, J, Q 0– 216 m (f) Heterocentrotus trigonaris D, Y, R, Q 0 - 25 m (d) Schizaster lacunosus Y, R, RB, Q 4 -15m (f) 951 ACCEPTED MANUSCRIPT 952 Table.1. 953

49

ACCEPTED MANUSCRIPT

Ammar Manaa

Height Open-

Area above system Lab number sea level ±2σ (%) model

Latitude Longitude (m) Terrace

Aragonite Aragonite age (ka)

Haql Pooled mean 119.9 0.5 N 29 11.450 E 034 54.037 FL01-38 H3 16.0 L 98.6 108.8 0.5 N 29 09.674 E 034 53.578 FL01-39 H6 16.1 L 98.8 120.8 0.8 N 29 11.450 E 034 54.037 FL01-37 H2 18.0 U 98.3 118.5 0.9 N 29 11.450 E 034 54.037 FL01-36 H1 19.2 U 98.4 120.2 0.9 Duba Pooled mean 121.3 0.3 N 27 15.377 E 035 46.882 WZ05_11 DA5 0.8 L 98.6 119.1 0.7 N 27 15.378 E 035 46.883 WZ05_12 DA6 0.9 L 99 121.6 0.7 N 27 15.348 E 035 46.907 WZ05_13 DB7 2.15 U 98.3 121.5 0.8 N 27 15.378 E 035 46.883 WZ05_14 3.0 U 99.3 122.5 0.6 DA19 Yanbu Pooled mean 121.7 0.4 N 24 16.910 E 037 30.568 WZ05_01 YA1 0.5 L 99.1 42.3 0.5 N 24 16.752E E 037 30.763 WZ05_02 YG2 0.7 L 98.8 52.6 0.6 N037 24 30.56821.947 E 037 25.969 FL01-27 Y3 1.6 U 97 131.7 1.1 N 24 16.842 E 037 30.627 WZ05_03 YC5 2.0 U 98 112.7 0.7 N 24 16.883 E 037 30.600 WZ05_04 YB4 2.05 U 99 125.3 0.6 Rabigh Pooled mean 122.0 0.2 N 22 48.277 E 038 56.422 WZ05_07 3.1.4 0.5 L 99.3 77.6 0.7 N 22 47.118 E 038 57.463 WZ05_08 KC1 0.8 L 99.4 89.3 0.6 N 22 49.557 E 038 55.755 FL01-33 R10 0.9 L 98.6 75 0.5 N 22 48.071 E 038 56.901 FL01-31 R8 1.0 L 98.4 117.8 0.9 N 22 47.201 E 038 57.300 B21 #21 1.1.7 1.0 L 98.6 71.6 0.1 N 22 48.467 E 038 55.697 WZ05_09 KB6 1.7 U 98.9 125.2 0.9 N 22 48.119 E 038 55.698 WZ05_10 KB7 1.8 U 99.1 124.2 0.9 N 22 47.884 E 038 57.032 FL01-35 RB8 20 U 98.7 115.1 0.7 N 22 47.884 E 038 57.032 FL01-34 RB6 2.3 U 97.6 125.5 0.9 N 22 49.557 E 038 55.755 FL01-32 R9 2.5 U 98.5 122.9 0.7 N 22 47.200 E 038 57.300 B22 #22 1.2.5 2.95 U 98 122.8 0.3 N 2247.109 E 038 57.471ACCEPTED FL01-30 R2 4.0 MANUSCRIPTU 99 114.6 1.0 Al-Qattan Pooled mean 121.7 0.5 N 20 50.542 E 039 24.487 WZ05_05 QE1 0.6 L 98.9 125.3 0.9 N 20 50.543 E 039 24.488 WZ05_06 QE3 1.0 L 99 120.4 0.8 N 20 50.393 E 039 24.637 FL01-28 Q17 1.2 L 98.9 120.6 0.7 N 20 49.773 E 039 25.146 FL01-29 Q19 3.0 U 97.8 107 0.7

954 Table 2.

955

50

ACCEPTED MANUSCRIPT

Ammar Manaa

956

957 Highlights

958  Major coral terrace growth along Saudi Red Sea coast at 122-119 ka during MIS 5e. Formatted: Bulleted + Level: 1 + Aligned at: 0.25" + Indent at: 0.5" 959  Coral terrace represents MIS 5e sea level highstand at least 6 m apsl at 120 ka.

960  Uniform terrace height shows tectonic stability on central to southern Saudi coast.

961  Coast north of Duba uplifted to 25 m apsl by faulting in last 100 ka.

962

963

ACCEPTED MANUSCRIPT

51