Monsoon triggered formation of Quaternary alluvial megafans in the interior of Oman

Geomorphology 110 (2009) 128–139

Contents lists available at ScienceDirect

Geomorphology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / g e o m o r p h

Monsoon triggered formation of Quaternary alluvial megafans in the interior of Oman Ingo Blechschmidt 1, Albert Matter, Frank Preusser ⁎, Dirk Rieke-Zapp Institut für Geologie, Universität Bern, Baltzerstrasse 1+3, 3012 Bern, Switzerland

a r t i c l e

i n f o

Article history: Received 2 October 2008 Received in revised form 2 April 2009 Accepted 3 April 2009 Available online 9 April 2009 Keywords: Alluvial fans Monsoon Quaternary Luminescence dating Arabia Oman

a b s t r a c t A vast bajada consisting of coalescing low-gradient (b 0.3°) alluvial fans exceeding 100 km in length formed along the southwestern margin of the Oman Mountains. It comprises an old fan sequence of inferred Miocene to Pliocene age termed Barzaman Formation, diagenetically highly altered to dolomitic clays, and a thin veneer of weakly cemented Quaternary gravels. A combination of remote sensing, lithological analyses and luminescence dating is used to interpret the complex aggradation history of the Quaternary alluvial fans from the interior of Oman in the context of independent regional climate records. From satellite imagery and clast analysis four fans can be discerned in the study area. While two early periods of fan formation are tentatively correlated to the Miocene–Pliocene and the Early Pleistocene, luminescence dating allows the distinction of five phases of fan aggradation during the Middle–Late Pleistocene. These phases are correlated with pluvial periods from Marine Isotope Stage (MIS) 11 through 3, when southern Arabia was affected by monsoonal precipitation. It is concluded that the aggradation of the alluvial fans was triggered by the interplay of increased sediment production during arid periods and high rainfall with enhanced erosion of hillslopes and transport rates during strong monsoon phases. However, the lack of fine-grained sediments, bioturbation and organic material implies that although the Quaternary fans are sourced by monsoonal rains they formed in a semi-arid environment. Thus, it appears that, in contrast to the Oman Mountains, the interior was not directly affected by monsoonal precipitation. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Numerous geological records such as deep-sea sediments and polar ice cores document important changes in climatic conditions during the Quaternary. However, knowledge on the impact of this variability in environmental conditions on geological processes on the continents is still limited. This is particularly true for desert areas where terrestrial records are scarce, indeed a better understanding of the relationships between palaeoclimate variability and the hydrological cycle, groundwater recharge and geomorphic processes is essential in order to evaluate the possible impact of global change on society in these areas. Terrestrial palaeoclimate information on the Quaternary evolution of southern Arabia has so far mainly been constrained by speleothems, aeolian sediments and lake deposits. All of these proxies monitor the effect of past climate variability on Earth surface processes in the area. While aeolian sediments record periods of aridity and, in particular, sediment availability (e.g., Goudie et al., 2000; Preusser et al., 2002; Radies et al., 2004), palaeolake deposits indicate pluvial periods (e.g.,

⁎ Corresponding author. E-mail address: [email protected] (F. Preusser). 1 Present address: Nationale Genossenschaft für die Lagerung radioaktiver Abfälle (NAGRA), Hardstrasse 73, 5430 Wettingen, Switzerland. 0169-555X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.geomorph.2009.04.002

McClure 1976; Lézine et al., 1998; Radies et al., 2005). Oxygen-isotope ratios of ancient groundwater and speleothem-calcite combined with δD values of fluid inclusion water provide information on the isotopic composition of rainfall and thus changes of monsoon precipitation (e.g., Burns et al., 1998, 2001; Fleitmann et al., 2003, 2004; Weyhenmeyer et al., 2000). The present climate of southern Arabia is arid and only the southern part of the region is influenced by the Indian Ocean summer monsoon. This is due to the location of the Intertropical Convection Zone (ITCZ) over the southernmost part of the Arabian Peninsula (Fig. 1). However, lake deposits and speleothems reveal that much larger areas of Arabia were repeatedly affected by monsoon-type climate in the past. These humid phases, triggered by a northward shift of the ITCZ and thus of the monsoon-belt, occurred at about 6– 10.5 ka, 78–82 ka, 120–135 ka, 180–200 ka, and 300–325 ka according to uranium-series dated speleothem growth (Burns et al., 2001), hence, mainly during past interglacial times. It has been assumed that the formation of the vast (ca. 40 000 km2) braidplain (bajada), consisting of coalescing alluvial fans flanking the southern and western piedmont of the Oman Mountains, developed during such wetter phases (Beydoun, 1980; Maizels, 1987; Maizels and McBean, 1990). This assumption has, however, not yet been supported by any independent dating results. The bajada of the Oman Interior shows great lateral thickness variations from a few tens to

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

129

tion of the Quaternary alluvial fans and the palaeoclimate of southern Arabia in a geochronological framework. 2. Materials and methods 2.1. Remote sensing and field analyses

Fig. 1. Overview map of the Arabian Peninsula showing generalised modern, summer surface wind pattern, approximate present position of the Intertropical Convergence Zone (ITCZ). WS = Wahiba Sands.

more than one thousand metres. It consists, according to Béchennec et al. (1993), of two distinct units; the Miocene–Pliocene Barzaman Formation forming the bulk of the bajada and thin Quaternary gravels. Rodgers and Gunatilaka (2002) interpret the bajada as a deposit on the distal Oman Mountains forebulge. The length of the exposed relict alluvial fans exceeds 100 km and may be up to 250 km and their longitudinal gradient is less than 0.3°, i.e. much flatter than classical alluvial fans. Their distal parts are covered by the Rub al-Khali and Wahiba Sand seas (Rodgers and Gunatilaka, 2002; Radies et al., 2004). The megafans show a complex pattern of superimposed sinuous conglomeratic ridges representing exhumed palaeochannel thalwegs rising 5 to 20 m above the plain (Glennie, 2005). Decreasing sinuosity from older meandering to the younger braided channels is interpreted by Maizels and McBean (1990) to reflect reduction of continental pluviality through time associated with a change from possibly perennial to ephemeral stream flow. This change in palaeohydraulic conditions is further supported by the oxygen isotopic composition of early carbonate cements in the alluvial conglomerates. The earlier palaeochannel generations have more negative values than later generations suggesting, based on the inverse relation between δ18O and rainfall amount, decreasing monsoon intensity (Burns and Matter, 1995). However, in the absence of geochronological data, except for a few initial luminescence ages (Juyal et al., 1998), the link between palaeohydrology, channel styles, geomorphological evolution of the alluvial fans and palaeoclimatic phases remains to be proven. In this paper we firstly discriminate different fans and their channel generations from satellite images using superimposition and intersection as main criteria. Secondly, we analyse clast composition of individual fans and, thirdly, present a geochronology of palaeochannel generations and identify phases of fan formation based on the results of luminescence dating. Palaeodrainage systems respond most sensitively to changes in climate, which affect precipitation, runoff and fluvial styles and vegetation. The ultimate aim therefore, is to reconstruct the relationships between the geomorphological evolu-

The different alluvial fan systems and their spatial extent and the morphology of palaeodrainage networks have been investigated using satellite images. The main focus of these investigations was to identify the morphological characteristics, the spatial arrangement and relative age relationships of different channel systems. The contributing upland area draining towards the fans in modern times was calculated in a geographic information system (GIS). The GIS analysis was based on digital elevation data collected during the Shuttle RADAR Topography Mission (SRTM; Rabus et al., 2003). SRTM data are provided to the public by the United States Geological Survey for download (http://edcsns17. cr.usgs.gov/EarthExplorer/). Gaps in the data were filled with elevation data from other sources, i.e. topographic maps, or were filled by interpolation to create a consistent Digital Elevation Model (DEM) of Northern Oman with a cell size of 90× 90 m2. Flow direction and flow accumulation within the DEM were calculated in order to delineate the drainage basins. Channels were derived from the flow accumulation data and match with channels that can be identified on satellite imagery of the same area. A contributing drainage area threshold of 16.2 km2 was found appropriate. Sedimentary characteristics were determined on 20 measured outcrop sections. Clast analyses (size, roundness, major composition) were performed on a randomly chosen 1.0 × 1.0 m surface by counting all components, generally about 400, with the long axis N2 cm. Five lithological groups were distinguished: cherts, ophiolites (harzburgite, dunite, gabbro, peridodite, basalt), limestones, siliciclastics and whitish-pink, fine-grained, dolomitic and clay-rich rocks termed barzamanite by Maizels (1987). 2.2. Luminescence dating 2.2.1. Samples and laboratory procedures Luminescence dating of sediments uses a light-sensitive signal in quartz and feldspar grains that is zeroed by daylight during sediment transport and accumulates during burial, when the grains are sealed from daylight (cf., Aitken, 1998; Duller, 2004; Preusser et al., 2008). For dating, the dose accumulated by the grains during burial (equivalent dose—De) and the amount of natural radioactivity within the sediment (dose rate—D) have to be determined. The alluvial sediments from interior Oman are a special challenge in this context. First, sand layers that are suitable for luminescence dating are only rarely found within the usually coarse-grained gravel. Secondly, the sand layers bear only small amounts of quartz grains needed for dating due to the abundance of limestones and ophiolites in the catchment area. This forced us to sample large amounts of sand (several kilograms) at the few suitable locations. Due to the moderate cementation of the sand, massive blocks were recovered from the outcrops during two field surveys. All preparation work for De determination was carried out under subdued red-light conditions in the laboratory. The outer lightexposed part of the sand blocks was first carefully removed using a water-cooled rock saw. The removed material was used for dose rate determination. From the remaining material carbonates were dissolved using hydrochloric acid that caused a volumetric reduction of up to 90% in the sample material. The dried material was sieved and a Frantz magnetic separator was used to extract the non-magnetic minerals (i.e. quartz). Quartz was separated using heavy liquids and subsequently etched by 40% hydrofluoric acid to remove plagioclase and the outer part of the quartz grains that is affected by external alpha radiation. De measurements were made on small aliquots (some

130

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

dozens of grains) applying the single-aliquot regenerative-dose (SAR) protocol for quartz (Murray and Wintle, 2000, 2003) with preheating at 230 °C for 10 s prior to all measurements. Optically stimulated luminescence (OSL) was recorded during a 60 s exposure to blue diodes using a Hoya U340 detection filter. The performance of the SAR protocol was routinely checked by dose recovery tests. The OSL signal is usually bright and is dominated by the fast component, recycling ratios are mainly within 10% of unity and recuperation is negligible (Fig. 2). For most samples, the amount of extracted quartz was so low that only relatively few repeated measurements could be carried out (Table 1). For a few samples, no grains remained after etching. From the aliquots that passed through the rejection criteria described by Murray and Wintle (2000), median De and standard error were calculated. As discussed below, partial bleaching seems to be of negligible significance. Median De was favoured instead of mean or central age approaches as it compensates for the effect of microdosimetry that can produce positively skewed distributions and is highest in sediments with low dose rate, as were the samples investigated here. Beta dose inhomogeneity also explains the rather large scatter observed for the present samples (cf. Mayya et al., 2006). Dose rates were calculated from the contents of K, Th and U derived from gamma spectrometry (cf. Preusser and Kasper, 2001) using the conversion factors by Adamiec and Aitken (1998). No evidence for disequilibrium in the uranium decay chain has been observed. For all samples, it has been assumed that the average water content during burial has been between 0 and 8%. While the contribution by cosmic radiation is only of minor importance (b10% of total

dose rate) in most luminescence dating studies, it has a rather large impact on the samples investigated here due to the fact that the sediments have low dose rates. This particularly applies for the oldest of our samples (Fan B: OMF22–29), for which cosmic radiation contributes up to 67% to the total dose rate. Cosmic dose rates were calculated by ADELE software (Kulig, 2005) following the approach of Prescott and Hutton (1994). We used present day depth and concluded from field observations that sediment overburden was accumulated as one contemporary event and that no significant erosion of the surface occurred after deposition for Fans C and D. A 10% uncertainty of cosmic dose contribution has been generally assumed. For Fan B, the situation is more complex and consequently we use the OSL ages as rough estimates only. A summary of the relevant information on luminescence dating is given in Table 1. 2.2.2. Zeroing of OSL signals An important issue in the context of dating fluvial deposits that undergo short and/or rapid transport, such as the sediments investigated here, is potentially incomplete resetting of the signal prior to deposition. Incompletely bleached OSL signals on deposition will result in overestimation of the true deposition age. Previous experience regarding zeroing of OSL signals in alluvial deposits from arid environments is limited to a few case studies. Initial work was carried out by Porat et al. (1996, 1997) on fault-related alluvial sediments from the Arava Valley, southern Negev, Israel. In a later study, Porat et al. (2001) calculated residual ages of the order of a few thousand years for modern fluvial deposits of the Arava Valley. For deposits older than the Holocene, age overestimation due to incomplete bleaching will only have a small effect for most samples. Bourke et al. (2003) demonstrated that although incomplete bleaching is an important factor for many samples from slack water facies sediments from the Kuiseb River, Namibia, luminescence dating can be used to establish a reliable time framework for such kinds of deposit. We investigated the amount of resetting in four modern samples from interior Oman (OMF 1–4) that were deposited as the result of a rain storm that occurred in the area during the time of the first field survey. For one sample (OMF 2), five sub-samples representing different grain-sizes were prepared (OMF 2a–e). All these samples most probably underwent similar transport distances and presumably experienced similar transport modes as the samples from ancient alluvial fan systems. The degree of zeroing in the modern samples may thus be considered as useful information regarding zeroing of the OSL signal in this particular environment. The De in the modern samples is below 2 Gy for 126 out of 128 aliquots measured (Fig. 3). Calculating median and standard error for the individual samples and using a typical dose rate of 0.5 Gy ka− 1 (dose rates of modern samples were not measured), results in residual ages of less than 1 ka. From the modern analogue it appears that age overestimation due to incomplete bleaching will be of negligible significance for pre-Holocene deposits. Indication for complete resetting is supported by the broad but Gaussian-like shape of dose distributions of the ancient alluvial sediments (Fig. 4). We therefore conclude that the dating results are not overestimated due to incomplete bleaching of the luminescence signal prior to deposition. 3. Results 3.1. Geological setting and geomorphology

Fig. 2. Typical OSL decay curve dominated by fast component (a) and dose response curve (b) revealing little recuperation and the absence of any saturation effect up to 200 Gy.

This study focuses on three alluvial fan systems B to D of Quaternary age according to Béchennec et al. (1993), which are located in the interior of central Oman south of the Adam mountains and to the west of the Wahiba sand sea (Fig. 5). They lie unconformably on the Mio-Pliocene Barzaman Formation. Outcrops of this formation are widespread and a large area has been stripped of

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

131

Table 1 Summary of OSL data for samples from Oman Fans. Sample code

Grain size (μm)

Depth (m)

n

Cosmic dose (Gy ka− 1)

K (%)

Th (ppm)

U (ppm)

D (Gy ka− 1)

De (Gy)

Age (ka)

OMF 1 OMF 2a OMF 2b OMF 2c OMF 2d OMF 2e OMF 3 OMF 4 OMF 5 OMF 6 OMF 7 OMF 8 OMF 9 OMF 10 OMF 11 OMF 12 OMF 13 OMF 14 OMF 15 OMF 16 OMF 17 OMF 18 OMF 19 OMF 20 OMF 21 OMF 22 OMF 23 OMF 24 OMF 25 OMF 26 OMF 27 OMF 28 OMF 29

100–150 300–355 250–300 200–250 150–200 100–150 100–150 100–150 100–200 150–200 150–200 150–200 150–200 150–200 150–250 150–250 200–300 150–200 300–355 150–200 150–200 150–200 200–250 100–150 200–250 150–250 150–250 150–250 150–250 150–250 150–250 150–250 150–250

0.20 0.20 0.20 0.20 0.20 0.20 0.20 0.20 1.80 1.00 1.80 1.80 0.80 0.80 2.00 2.75 2.00 1.80 0.20 3.00 2.00 1.50 2.80 2.80 0.50 2.25 2.00 0.50 7.50 7.50 2.50 2.50 1.50

20 12 14 14 14 14 20 19 21 13 18 14 18 14 16 18 – 16 6 21 10 3 – – 14 20 21 21 20 16 19 13 15

0.19 ± 0.02 0.19 ± 0.02 0.19 ± 0.02 0.19 ± 0.02 0.19 ± 0.02 0.19 ± 0.02 0.19 ± 0.02 0.19 ± 0.02 0.16 ± 0.02 0.17 ± 0.02 0.15 ± 0.02 0.15 ± 0.02 0.18 ± 0.02 0.18 ± 0.02 0.15 ± 0.01 0.14 ± 0.01 0.19 ± 0.02 0.15 ± 0.02 0.19 ± 0.02 0.15 ± 0.02 0.15 ± 0.02 0.16 ± 0.02 0.13 ± 0.01 0.13 ± 0.01 0.18 ± 0.02 0.15 ± 0.02 0.15 ± 0.02 0.18 ± 0.02 0.08 ± 0.01 0.08 ± 0.01 0.14 ± 0.01 0.14 ± 0.01 0.16 ± 0.01

– – – – – – – – 0.18 ± 0.01 0.15 ± 0.01 0.17 ± 0.01 0.15 ± 0.01 1.33 ± 0.05 0.21 ± 0.01 0.11 ± 0.01 0.08 ± 0.01 0.15 ± 0.01 0.14 ± 0.01 0.12 ± 0.01 0.13 ± 0.01 0.14 ± 0.01 0.15 ± 0.01 0.13 ± 0.01 0.13 ± 0.01 0.18 ± 0.01 0.11 ± 0.01 0.06 ± 0.01 0.05 ± 0.01 0.08 ± 0.01 0.05 ± 0.01 0.08 ± 0.01 0.05 ± 0.01 0.09 ± 0.01

– – – – – – – – 0.62 ± 0.04 0.62 ± 0.03 0.83 ± 0.04 0.76 ± 0.03 1.34 ± 0.06 0.94 ± 0.04 0.31 ± 0.02 0.26 ± 0.01 0.33 ± 0.02 0.59 ± 0.03 0.37 ± 0.02 0.26 ± 0.02 0.38 ± 0.02 0.37 ± 0.02 0.43 ± 0.02 0.43 ± 0.02 0.55 ± 0.03 0.37 ± 0.01 0.20 ± 0.01 0.17 ± 0.01 0.57 ± 0.02 0.14 ± 0.01 0.28 ± 0.01 0.20 ± 0.01 0.23 ± 0.02

– – – – – – – – 1.69 ± 0.06 0.63 ± 0.02 0.73 ± 0.02 0.59 ± 0.02 0.52 ± 0.01 1.24 ± 0.04 0.94 ± 0.06 1.02 ± 0.06 0.54 ± 0.02 1.30 ± 0.04 0.47 ± 0.02 0.54 ± 0.02 0.52 ± 0.02 0.50 ± 0.02 0.21 ± 0.01 0.21 ± 0.01 0.47 ± 0.02 0.19 ± 0.01 0.27 ± 0.01 0.13 ± 0.01 0.20 ± 0.01 0.19 ± 0.01 0.11 ± 0.01 0.27 ± 0.03 0.69 ± 0.03

0.50* 0.50* 0.50* 0.50* 0.50* 0.50* 0.50* 0.50* 0.77 ± 0.06 0.50 ± 0.04 0.54 ± 0.04 0.56 ± 0.05 1.06 ± 0.08 0.72 ± 0.06 0.48 ± 0.05 0.46 ± 0.04 0.43 ± 0.04 0.68 ± 0.06 0.43 ± 0.04 0.40 ± 0.04 0.44 ± 0.02 0.44 ± 0.04 0.33 ± 0.03 0.31 ± 0.03 0.51 ± 0.04 0.31 ± 0.03 0.28 ± 0.03 0.27 ± 0.03 0.23 ± 0.02 0.17 ± 0.02 0.26 ± 0.03 0.26 ± 0.03 0.41 ± 0.04

0.18 ± 0.06 0.32 ± 0.10 0.65 ± 0.34 0.56 ± 0.28 0.13 ± 0.04 0.19 ± 0.05 0.31 ± 0.08 0.42 ± 0.07 164.3 ± 6.5 116.5 ± 8.1 121.2 ± 6.5 128.8 ± 3.9 205.6 ± 12.1 164.5 ± 14.7 196.2 ± 23.0 145.2 ± 13.5 No quartz # 219.6 ± 13.5 95.0 ± 5.9 135.0 ± 7.7 97.9 ± 13.3 94.2 ± 18.5 No quartz # No quartz # 22.8 ± 1.7 194.6 ± 8.7 170.4 ± 13.6 201.3 ± 13.9 141.5 ± 13.5 161.4 ± 14.6 207.7 ± 17.2 119.0 ± 20.5 171.5 ± 16.6

~ 0.4 ~ 0.6 ~ 1.3 ~ 1.1 ~ 0.3 ~ 0.4 ~ 0.6 ~ 0.7 214 ± 19 233 ± 26 225 ± 22 232 ± 21 194 ± 18 227 ± 27 407 ± 61 317 ± 42 – 323 ± 34 220 ± 24 340 ± 36 229 ± 37 213 ± 46 – – 45 ± 5 621 ± 65 613 ± 81 746 ± 104 607 ± 83 923 ± 139 809 ± 110 458 ± 97 415 ± 55

* Assumed average dose rate for modern samples. # Amount of quartz gathered from sample insufficient for dating.

its Quaternary cover exhuming a fan surface with channel traces in the Barzaman Formation (Fan A in Fig. 5). The stratigraphic relationships, clast composition and numerical age of the channels have been investigated in three key areas on the fans (Fig. 5). Originating in the parautochthonous carbonates of the Jabal Akhdar and the Samail Ophiolite Nappe, the sediment routing system of the central Oman fans runs southwards across the low-lying hills formed by the Hawasina Nappes before fanning out onto the piedmont plain beyond narrow gaps in the Adam mountain range.

Fig. 3. Des determined for modern alluvial samples from interior Oman have values mainly below 2 Gy, corresponding to residual mean ages of less than 1 ka.

The source region comprises the major structural units of the Oman Mountains defined by Glennie et al. (1974) and shown in Fig. 5: (1) Parautochthonous sediments of the Jabal Akhdar comprise prePermian carbonate and siltstone formations and Mid-Permian to Cretaceous shallow-marine carbonates. Cretaceous carbonates also form the Adam Mountains which are interpreted as a frontal ramp structure of the Oman Mountains. The parautochthonous is overthrusted by structural units 2 and 3. (2) Hawasina Nappes, forming an imbricate thrust belt consisting of folded deep water radiolarian cherts and shales as well as siliciclastic and limestone turbidites of Late Permian to Cretaceous age. (3) Samail Ophiolite Nappe, an obducted slab of Late Cretaceous oceanic lithosphere. It comprises an upper mantle sequence made up of mainly variably tectonized and serpentinized harzburgites which is overlain by a crustal sequence made up of gabbros and pillow basalts. (4) Neoautochthonous Late Maastrichtian to Tertiary shallow-marine carbonates and clastics that overlie the other structural units. The DEM (Fig. 6) reveals that the geomorphology of the catchment of the piedmont fans closely correlates with these geotectonic units. The Jabal Akhdar, reaching an altitude of 3030 m, shows steep slopes towards the south and southeast, flanked by foothills formed by Samail Ophiolite with elevations of less than 1300 m. The broad low relief terrain between the foothills and the Adam Mountains is occupied by large hilly areas of Hawasina sediments separated by alluvial plains with Wadi networks and small alluvial fans (Figs. 5 and 6). The southwestern central Oman Mountains are drained by five major fluvial networks whose drainage basins vary in size from 3300 km2 to 960 km2 (Fig. 6). The Wadi al Umayri, Wadi Adam and Wadi Halfayn network start in the Jabal Akhdar whereas the Wadi

132

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

Fig. 4. Dose distribution plot for sample OMF 5 reveals a Gaussian-like distribution indicating a negligible effect of incomplete bleaching on De determination (n = 21, mean: 164.3, relative standard deviation = 18%).

Andam network penetrates deep into the Samail Ophiolite (Figs. 5 and 6). However, the largest part of all drainage basins is in Hawasina terrain (Figs. 5 and 6). The gradient of the major channels between the base of the foothills and the gaps in the Adam Mountains and to the west and east of Jabal Madar is b0.3°, matching the gradient of the studied megafans. 3.2. Channel morphology and lithological composition The Barzaman Formation Fan A represents the oldest (Miocene– Pliocene) exposed part of the Wadi Andam drainage system, which extends today from its headwaters in the Samail Ophiolite to the western edge of the Wahiba Sands over about 220 km (Figs. 5 and 6). According to Maizels (1987) and Maizels and McBean (1990), NNE– SSW-trending meandering gravel ridges on the fan surface represent exhumed single-thread palaeochannels that frequently intersect each other. The inverted multistorey channel complex described by these authors comprises at least five superimposed major palaeochannel generations (A1–5) characterised by decreasing sinuosity and channel width. The younger channel courses also appear to decrease in width down-fan, reflecting smaller discharge, higher transmission losses and evaporation. However, the individual channel generations can be followed down-fan over a limited distance of about 10 km only. The lithological composition also changes systematically through channel generations A1 to A5 from up to 80% to 13–18% Hawasina cherts compensated by increasing percentages of ophiolite clasts (Maizels and McBean 1990), heralding a significant reorganization of the drainage network. Fan B is located in the northern part of the present Wadi Andam piedmont (Figs. 5 and 6). It truncates, and partly overlies with erosional contact, the barzamanite of Fan A (Fig. 7). Nine successive palaeochannel generations (B6–14) consisting of anabranching channels typical of a braided stream environment were distinguished in the small Barzaman-Pink Cliffs area by Maizels and McBean (1990). The oldest generation B6 forms broad gravel sheet terraces (N1.5 km) rising up to 15 m above the present Wadi floor (Fig. 7). Younger generations lying at successively deeper topographic levels are represented by NW–SE-trending wide low-sinuosity gravel ridges with a relief of up to 10 m that taper off down-current from widths of N400 m and disappear after a few kilometres before reaching the Wahiba sands. The youngest channel generation B14 is a low terrace following the present course of Wadi Andam. The major facies of Fan B are coarse-grained, cross-bedded gravels with ≥90% ophiolite clasts and cross-bedded gravelly sands. This clast suite, consisting almost exclusively of ophiolitic source rocks, is unique and reflects the fact

that only the Wadi Andam catchment drains large areas of ophiolite (Table 2, Figs. 5 and 6). Fan C, situated to the west of Fan A, is part of the Wadi Adam/Wadi Halfayn and the Wadi al Umayri drainage systems and, as a result, it shows a more complex morphology. It has a typical radial distributary network extending from the Adam Mountains over 200 km to the sea (Figs. 5 and 6). Its geomorphological features are overprinted by numerous modern Wadi courses that formed through headward incision, entrenching up to 20 m into the fan deposits and the underlying Barzaman Formation. Three palaeochannel generations, C1 to C3, were identified on the lower fan in key area II (Fig. 5). Channel generation C1 forms a complex and dense network of NW–SEtrending low-sinuosity (average 1.15) single channels ranging in width from 20 to 500 m. The channels of generation C2 form belts of low-sinuosity anabranching ridges with very low relief and widths up to 1000 m running NE–SW. These are intersected by NE–SW-directed, more linear single channels representing the youngest generation C3 (Fig. 5). The lithological composition of the gravels analysed in key area II is dominated by micritic limestones and cherts and is marked by a very low concentration of ophiolites (Table 2, Fig. 5). The cherts are derived from the Hawasina Nappes whereas the limestones are primarily sourced from the Cretaceous formations exposed on the southern slopes of the Jabal Akhdar and possibly partly from the Adam mountain range (Fig. 5). Fan D represents part of the Wadi Andam/Wadi ath Thali drainage system (Figs. 5 and 6). However, in the distal part where key area III is located, Fan D interferes with Fan C (Figs. 5 and 8a, b). Fan D overlies the Barzaman Formation with an unconformable contact and disappears beneath the Wahiba sands. Four major palaeochannel generations (D1–4) are distinguished based on cross-cutting relationships shown in Fig. 8b: NNW–SSE-running, low-sinuosity (1.20) single channels (D3) intersected by WNW–ESE-directed, straight single channels (D4) and a short relict channel (D2). The oldest generation D1 is poorly represented by a relict palaeochannel fill (OMF-11 in Fig. 8a, b) underneath erosive D3 channel deposits. Whereas the age relationship between D1 and the adjacent D2 channel cannot be established unambiguously based on geometric criteria alone, they differ in their clast composition (Fig. 8c). As the channels can generally be followed over rather short distances only, the sampled channels outside Fig. 8b had to be attributed to the respective channel generations according to their absolute age rather than geometric relationship (Table 2). The weakly cemented clast-supported gravels of channel generation D1 and D3 are characterised by a clast suite dominated by subrounded chert and well-rounded ophiolite cobbles and pebbles (Fig. 8c). They are derived from the widespread Triassic to Cretaceous chert formations of the Hawasina Nappes (Blechschmidt et al., 2004) and the Samail Ophiolite Nappe, respectively (Figs. 5 and 6). In contrast, the WNW–ESE-running palaeochannel generations D4 and D2 in Fig. 8b show a different clast suite with mainly limestone and subordinate chert typical of Fan C clast composition (Fig. 5 and 8c). However, D2 and D4 channel generations in the northern part of the study area outside Fig. 8b display the chert/ophiolite suite of Fan D shown on Fig. 5, indicating that both Fans D and C were active and aggrading simultaneously (Table 2). White dolomite clasts up to 40 cm in size, generally present in minor amounts in all samples, are locally derived from the underlying Barzaman Formation (Table 2). The clasts lying on the channel surfaces show desert varnish and ventifacts, particularly amongst the ophiolite clasts, while limestones clasts show advanced solution rilling. These phenomena indicate long periods of exposure to weathering. 3.3. Stratigraphic relations and numerical dating The stratigraphic relationship of the fans and their channels was studied in three key areas (I–III) shown in Fig. 5. The relative age

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

133

Fig. 5. Landsat 7 ETM+ image showing the study area located in the interior of Oman with major structural units, different fan systems and key areas of investigation. Band 7 (red), Band 4 (green), Band 2 (blue) were used. Pie charts display the clast composition of the different fans (C = chert, O = ophiolite, L = limestone, S = siliciclastics). Barzamanite clasts are not included as they are locally reworked.

sequence based on geomorphic and stratigraphic criteria is confirmed by quartz OSL ages. The best exposures showing the relation of Fans A and B occur in key area I at Pink Cliffs (Fig. 5). The section shows about 9 m of Barzaman Formation (Fan A) separated by a sharp erosional uncon-

formity from the overlying weakly cemented gravels and gravelly sandstones of Fan B measuring ca. 6 m (Fig. 7). The Barzaman Formation consists mainly of white to pale red barzamanite and channelized cross-bedded conglomerates with a high amount of fines which are poorly recognizable as a result of pedogenic and early

134

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

Fig. 6. Digital elevation model of the southwestern central Oman Mountains derived from SRTM data showing major modern Wadi systems with their drainage basins and alluvial megafans A–D. Vertical exaggeration ×3.

Fig. 7. Outcrop photograph of the Pink Cliffs area located in the northern part of the present Wadi Andam drainage system and OSL ages. Note the sharp erosive contact between the Barzaman Formation (Fan A) and the overlying channel deposits of Fan B.

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

135

Table 2 Overview of the OSL samples from different key areas and channel generation, UTM position and lithological composition of the gravel distinguishing cherts, ophiolites (harzburgite, dunite, gabbro, peridodite, basalt), limestones, barzamanite (dolomite) and siliciclastics. Key area

OSL sample

Channel generation

UTM coordinates

Chert (%)

Ophio. (%)

Limes. (%)

I I I I I I I I II II II II III III III III III III III III III III

OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF OMF

B14 B6 B6 B6 B6 B6 B6 B6 C1 C2 C2 C2 D4 D4 D1 D3 D2 D4 D2 D4 D4 D1

0606700/2454416 0605980/2454752 0605983/2454775 0605991/2454808 0605991/2454825 0606005/2454832 0606191/2454681 0607402/2454190 0574504/2358935 0581339/2353304 0578361/2346332 0578361/2346332 0632070/2382902 0625560/2386122 0626230/2387199 0625231/2387510 0624932/2387120 0622628/2390932 0622842/2391063 0621314/2392675 0628772/2392167 0627474/2393967

5 6 – 3 – 4 – 4 65 38 32 32 32 32 60 56 41 57 68 55 38 83

89 92 – 94 – 93 – 90 3 5 4 4 2 3 20 28 5 37 26 20 46 9

3 1 – b1 –

21 22 23 24 25 26 27 28 5 6 7 8 9 10 11 12 14 15 16 17 18 29

diagenetic alterations. Most of the ophiolitic and limestone clasts have been transformed to a mixture of clay minerals and dolomite. The original bedding is obliterated by vertisol formation and mottling. Fan A, i.e. the Barzaman Formation, turned out to be too old for OSL dating. However, Béchennec et al. (1993) postulate a Middle Miocene to Pliocene age of the Barzaman Formation based on poorly supported stratigraphic evidence and lacking datable fossils (see Rodgers and Gunatilaka, 2002, for details). The gravels overlying Fan A barzamanite at Pink Cliffs shown in Fig. 7 represent the highest terrace level of Fan B which is formed by palaeochannel generation B6 of Maizels and McBean (1990). Scattered blocks of lithified barzamanite up to 1 m in size were locally reworked into basal gravels of Fan B. The ancient and modern gravels in the active Wadi show an identical clast composition with N90% ophiolite. OSL dating of quartz from the highest terrace of Fan B revealed ages of 923 ± 139 ka (OMF 26) and 809 ± 110 ka (OMF 27) for samples from the lowest sandstone directly above the unconformity, and ages ranging from 613 ± 81 ka (OMF 23) to 746 ± 104 ka (OMF-24) for beds higher up in the sequence (Table 1, Fig. 7). Sample OMF 28 from palaeochannel B6 deposits forming the high terrace E of Wadi Andam was dated to 458 ± 97 ka. The topographically lowest terrace about 2 m above the present Wadi floor corresponding to palaeochannel generation B14 of Maizels and McBean (1990) revealed an OSL age of 45 ± 5 ka (OMF 21). Good exposures of Fan C deposits occur in key area II (Wadi Tarban) (Fig. 5). Above 16 m of fine-grained barzamanite (dolomite) with four thick vertisol horizons and separated by an erosional unconformity, lie dark grey terrace-forming gravels deposited in palaeochannels. The thin (b3.5 m) succession consists of matrix- and clast-supported amalgamated gravel units with erosive contacts, planar cross-bedding and generally normal grading. Lenses of coarse to medium grained cross-bedded sands representing migrating dunes are abundant. Three samples from the Wadi Tarban sections representing channel generation C2 range in age from 225 ± 22 ka (OMF 7) to 233 ± 26 ka (OMF 6) whereas sample OMF 5 from a single channel C1 in the adjacent Wadi was dated to 214 ± 19 ka (OMF 5) (Table 1). Fan D is characterised by low relief exposing up to 6 m of Barzaman Formation followed by generally less than 2 m of weakly cemented gravels with cross-bedded sandstone lenses. OSL dating confirms the relative age sequence based on geomorphic criteria (Table 1). The oldest palaeochannel generation D1 is dated to 415 ± 55 (OMF 29) and

– 2 29 57 63 63 64 63 17 7 51 3 4 3 7 2

Barz. (%) 1 – 2 – 2 – 3 3 – – – 1 1 3 8 1 3 2 22 7 5

Silic. (%)

OSL age (ka)

2 b1 – b1 – b1 – b1

45 ± 5 621 ± 65 613 ± 81 746 ± 104 607 ± 83 923 ± 139 809 ± 110 458 ± 97 214 ± 19 233 ± 26 225 ± 22 232 ± 21 194 ± 18 227 ± 27 407 ± 61 317 ± 42 323 ± 34 220 ± 24 340 ± 36 229 ± 37 213 ± 46 415 ± 55

b1 1 1 1 1 b1 b1 1 1 1 b1 2 b1

407 ± 61 ka (OMF 11). Two channels representing D2 reveal OSL ages of 340 ± 36 ka (OMF 16) and 323 ± 34 ka (OMF 14) similar to the long winding channel D3 (317 ± 42 ka, OMF 12). D4 channels are distinctly younger, ranging from 229 ± 37 ka (OMF 17) to 194 ± 18 ka (OMF 9) and thus matching in age the channels of Fan C. The fact that D2 and D4 channels may both have either Fan C or Fan D clast composition reveals coeval activity of these two fans (Table 2). Sample OMF 14 is from an isolated mesa (Fig. 8a) and therefore cannot be put into the geomorphologic sequence, but with an age of 323 ± 34 ka it is a D2 channel which, according to its clast suite, is part of Fan C. Early diagenetic calcite is a characteristic feature of the weakly to moderately cemented Quaternary conglomerates. According to Burns and Matter (1995) the crystal size and the morphology of the cements vary widely. Finely crystalline calcite at grain contacts often shows a meniscus morphology or forms microstalactites at the underside of grains which, together with internal micritic sediment, are typical fabrics of the vadose zone. However, the more widespread medium crystalline isopachous calcite cement formed in the phreatic zone. Thus the conglomerates originally had a high intergranular porosity and typically no clayey matrix. Although early diagenetic carbonate cements are common, neither calcic nor gypsic soils were observed in the Quaternary sequence, in contrast to the Tertiary Barzaman Formation with common calcrete horizons. 4. Discussion The study demonstrates not only the variable nature of the sedimentary succession of the bajada in the interior of Oman, but also changing channel styles and clast composition. According to the stratigraphic relationships of the fans, the morphology and interrelationships of the palaeochannels, the petrography and the results of OSL dating, eight phases of alluvial fan formation can be distinguished (Table 3). The strong contrast in thickness, facies, degree of lithification and diagenetic overprint between the heavily altered, hard rocks of the thick Barzaman Formation and the overlying poorly consolidated thin deposits of Fan B suggest a change from more or less continuous bajada aggradation during Middle Miocene–Pliocene times, to episodic deposition in the Pleistocene. In addition, the associated marked changes in channel style, clast composition and amount of fines from Barzaman Fan A to the Fan B reveal (i) a shift of the major catchment area away from the Hawasina Nappes to the Samail Ophiolite Nappe, and (ii)

136

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

Fig. 8. (a) Satellite image showing complex channel pattern of key area III (Fig. 5) and location of sites dated by OSL. White line indicates the area covered by panel b. (b) Map with major palaeochannel generations (D1 to D4) of key area III. OSL ages: Fan D4: 194 ± 18 ka (OMF 9), 227 ± 27 ka (OMF 10); Fan D3: 317 ± 42 (OMF 12); Fan D2: 323 ± 34 (OMF 14); Fan D1: 407 ± 61 (OMF 11). (c) Pie charts displaying clast composition of different channel generations. See Fig. 5 for legend.

a higher gradient and discharge and a change from a meandering to a braided river style related to perennial and high-power ephemeral flow regimes, respectively. The oldest set of OSL ages (OMF 22 to OMF 28) has been determined for samples taken from one outcrop in Fan B. Unfortunately, these ages have rather large errors and their reliability is to some extent

questionable due to the fact that dose rates mainly originate from cosmic radiation, with the associated problem of possible changes in sediment overburden. The latter may explain some of the substantial scatter of ages observed for samples taken from the same outcrop (OSL ages between 500 ka and 1000 ka, Fig. 7). However, it is the low dose rate that actually allowed the production of ages that are far beyond the

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139 Table 3 Summary of phases of alluvial fan formation identified in central Oman. Age

Name

Ca. 45 ka

Ca. 220 ka

Fan B14 Northern W. Andam Wahiba Wahiba Sands Fan Fan C West of W. Adam

Ca. 210 ka

Fan D4

Ca. 330 ka

Fan D2

Ca. 410 ka

Fan D1

Ca. 130 ka

Early Fan B Pleistocene Miocene– Fan A Pliocene

Location

Between W. Andam and W. ath Thali Between W. Andam and W. ath Thali Between W. Andam and W. ath Thali Northern W. Andam Between W. Andam and W. Halfayn

Characterisation Topographically lowest terrace (locally developed) Covered by aeolian sediments Radial distributary network, min. three channel generations Cross-cutting relationships within Fan D

Cross-cutting relationships within Fan D

Cross-cutting relationships within Fan D

Anabranching, braided style, min. nine channel generations Barzaman Formation, multistorey channel complex, min. five channel generations

commonly accepted age range of OSL dating. Similar low dose rates and high OSL ages have been reported by Rhodes et al. (2006) for a sediment sequence from Morocco and were proved there to be consistent with independent age constraints. Carefully considering the limitations of OSL in this time range and the particular setting in Oman, allows the conclusion that the formation of Fan B occurred during the Early Pleistocene or earliest Middle Pleistocene. The next phase of fan formation occurred apparently after a long gap in the stratigraphic record. Its initial period is documented by palaeochannel

137

generation D1 and dated by two samples to about 410 ka (OMF 11, OMF 29). According to two OSL samples (OMF 14,16), reactivation of Fan D and the onset of Fan C deposition occurred at around 330 ka (palaeochannel generation D2). Best documented is a further period of fan formation at about 220 ka (nine OSL ages), represented by palaeochannels C1, C2 and D4. Fan D is hence characterised by a complex development, where fluvial deposition was separated by phases of erosion to progressively lower elevations. This is accompanied by a modification of the source area as indicated by a higher percentage of Hawasina chert clasts compared to Fan B. During the later phases, the major depocentre shifted westwards and a new dispersal system was established depositing the huge alluvial Fan C. Its limestone- and chert-dominated clast suite and large fan size suggest an expansion of the catchment northwestwards draining larger areas of the Hawasina Nappes and the parautochthonous carbonates of the Jabal Akhdar. In the western Wahiba Sands, relict gravels are found on top of aeolianites dated to about 160–140 ka (Radies et al., 2004). Fluvial deposits attributed to the Last Interglacial (130–120 ka) have been observed in drill-cores from the central Wahiba Sands, covered by aeolian sediments dated to ca. 115 ka (Preusser et al., 2002). It therefore appears that alluvial fan formation during this time was located further to the east of the present study area and has later been covered by aeolian sediments of the Wahiba Sands. Fig. 9 reveals that the phases of fan build-up most likely correlate with interglacial marine isotope stages (MIS) in the LR04 benthic d18O stack of Liesicki and Raymo (2005), as a measure of global climatic conditions, and pronounced peaks of the Indian Summer Monsoon Index of Leuschner and Sirocko (2003). Relatively few OSL ages are available for MIS 11 (D1) and MIS 9 (D2, D3) but there is striking evidence for substantial fan formation during MIS 7 (C1, C2, D4). Due to the uncertainties associated with OSL dating, it is however not

Fig. 9. Comparison of OSL ages of fan formation in interior Oman plotted (a) versus the LR04 benthic δ18O stack of Liesicki and Raymo (2005) as a measure of global climate conditions with number of odd marine isotope stages, (b) the Indian Summer Monsoon Index of Leuschner and Sirocko (2003), and (c) phases of speleothem growth observed in Oman speleothems (Burns et al., 2001) (blue boxes) and periods of aeolian deposition as recognised in the Wahiba Sands (red shading) (Preusser et al., 2002). Lower δ18O values indicate warmer conditions on a global scale, and larger insolation differences indicate stronger Monsoon circulation. Speleothem growth is interpreted to reflect pluvial periods. Individual OSL ages have been analysed using a density probability plot (a) as well as by calculating weighted mean and standard deviation for OSL ages determined for specific palaeochannel complexes identified from remote sensing (Table 2) (red boxes). Applying the latter approach is justified as OSL ages from the individual fan systems are consistent within errors (cf. Geyh, 2008). Weighted mean and standard deviation indicate the statistically most likely time of deposition. OSL ages for the oldest Quaternary Fan B are not included owing to the limited reliability of the dating results. The age of the Wahiba fan is deduced from OSL dating of associated aeolian deposits (Preusser et al., 2002; Radies et al., 2004).

138

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139

possible to assess whether fan formation coincides exactly or if it preor post-dates the peaks in Summer Monsoon Index. Weighted mean ages for D1 and D2, in each case for two samples only, are 411 ± 4 ka (monsoon peak at 406 ka) and 331 ± 6 ka (monsoon peak at 331 ka). For C1, C2 and D4 weighted means of 225 ± 8 ka and 211 ± 13 ka, respectively, have been calculated (monsoon peak at 216 ka). Interestingly, MIS 9 and MIS 7 have already been identified as phases of intense monsoonal rains in stalagmites from Oman and Yemen (Burns et al., 2001; Fleitmann et al., 2002), although the U/Th dated speleothem growth does not coincide with maximum Summer Monsoon Index peaks. It is possible that the speleothem record is as yet incomplete. The OSL age of 45 ± 5 ka for the lowest developed terrace of Fan B falls into MIS 3 and may be related to a peak in Summer Monsoon Index located at 50 ka or possibly to Dansgaard/Oeschger event 12 (ca. 46 ka) that has been identified as a period of increased monsoon precipitation in southern Arabia (Burns et al., 2003). Since then, the study area has been subject mainly to deflation, although fluvial dissection has also contributed to landscape evolution possibly during the Holocene wet phase. The exhumed palaeochannels are the result of wind erosion which has blown out finer-grained interchannel deposits leaving the coarser-grained, more cemented channel fills behind as raised winding palaeothalwegs (Maizels, 1987). As shown in Fig. 9, the Pleistocene alluvial record in the Oman Interior is characterised by relatively short depositional phases of increased continental humidity related to northward movement of the monsoonal rainfall belt during interglacials documented by palaeogroundwater data from speleothems (Fleitmann et al., 2003). The intervening long non-depositional arid periods are reflected as phases of aeolian deposition in the Wahiba Sands (Preusser et al., 2002; Fig. 9). These arid periods were most likely characterised by pronounced differences in day and night temperatures, enhanced physical breakdown of rocks and large amounts of rock debris that accumulated in the source areas of the alluvial fans. It is generally agreed that sediment yield in semi-arid environments is governed by the intensity of rainfall rather than the annual total (Allen, 1997). Bookhagen et al. (2005) report a five-fold increase of sediment transport rates in the northwestern Himalaya during millennial-scale monsoonal phases in the Late Pleistocene and Holocene. These relationships may explain why the Pleistocene fans of the Oman Interior were mainly built up during MIS 11, 9, 7 and 5. The stark contrast between the well lithified and diagenetically heavily altered rocks of the Mio-Pliocene Barzaman Formation and the moderately lithified gravels of the Quaternary alluvial fans, together with the change in fluvial style from meandering to braided and irregularly sinuous, indicate a change from a wet-humid climate to a generally drier climate in the Quaternary. However, the Quaternary megafans formed during several successive short-lived pluvial periods when according to regional palaeoclimate data the monsoonal rainfall was high. Under such environmental conditions a wet alluvial fan similar, for example, to the Kosi fan on the southern flanks of the Himalayas (India) is expected. There are indeed similarities between the Kosi fan located in a monsoonal region and the Omani megafans such as their large size and typical low gradient. However, in contrast to the Kosi fan where coarse debris is transported only a few kilometres beyond the fan head and which is vegetated and characterised by broad marshes, oxbow lakes and extensively bioturbated fine-grained sediments (Wells and Dorr, 1987), the Omani megafans B to D are gravel fans lacking soil profiles, clayey sediments and any bioturbation at all. This is more typical for semi-arid fans than for wet fans. Thus, the Quaternary alluvial fans of Oman share aspects of both semi-arid and wet fans. 5. Conclusions In this paper we analyse the OSL-dated geomorphologic evolution of alluvial megafans in the interior of Oman at glacial–interglacial

scales within the context of independent records of climate change. Our findings suggest that monsoonal precipitation was high over the Oman Mountains but that the interior remained rather dry and the rivers delivered the debris to a semi-arid depositional environment. Only vigorous rivers were able to transport the coarse-grained bedload. Such rivers existed during the monsoonal season whereas during the dry season stream runoff was low or the streambeds were dry. This implies that runoff and bedload transport rates were subject to high seasonal variations. Not only did discharge, sediment transport rates and aggradation change seasonally, but more importantly also from pluvial to dry periods during which entrenchment and terrace formation occurred, best documented by Fan B. The thin veneer of the Quaternary gravel sequences and the fact that the depositional centre shifted through time from Fan B to Fan D and Fan C, and finally to the Wahiba Fan, are indications of a non-subsiding basin with no change in accommodation space. This is in contrast with the situation during deposition of the thick Barzaman Formation where the available accumulation space caused the development of stable transport paths fed by single sources for a long period.

Acknowledgements Preparation and measurement of a first set of OSL samples were carried out using the facilities at Geographisches Institut, Universität zu Köln. The authors would like to thank Ulrich Radtke and his team for hospitality and support. A second set of samples was prepared and measured at the newly establish OSL laboratory in Bern. High resolution gamma spectrometry was carried out by Johanna Lomax (Geographisches Institut, Universität zu Köln) and Detlev Degering (VKTA Rossendorf e.V.) for sample sets one and two, respectively. The authors also thank Dirk Leuschner for providing Monsoon Index data. We are indebted to Hilal al-Azri, former Director General of Minerals, Ministry of Commerce and Industry, Sultanate of Oman for logistical help during fieldwork. Markus Fuchs and an anonymous reviewer are thanked for their constructive suggestions on the manuscript.

References Adamiec, G., Aitken, M.J., 1998. Dose-rate conversion factors: update. Ancient TL 16, 17–30. Aitken, M.J., 1998. An Introduction to Optical Dating. University Press, Oxford. 267pp. Allen, P.A., 1997. Earth Surface Processes. Blackwell Science Ltd, Oxford. 404pp. Béchennec, F., Le Métour, J., Platel, J.P., Roger, J., 1993. Geological map of the Sultanate of Oman, scale 1:1,000,000 with explanatory notes. Directorate General of Minerals, Oman Ministry of Petroleum and Minerals, Muscat. Beydoun, Z.R., 1980. Some Holocene geomorphological and sedimentological observations from Oman and their palaeogeological implications. Journal Petroleum Geology 2, 427–437. Blechschmidt, I., Dumitrica, P., Matter, A., 2004. Stratigraphic architecture of the northern Oman continental margin — Mesozoic Hamrat Duru Group, Hawasina complex, Oman. GeoArabia 9, 81–132. Bookhagen, B., Thiede, R.C., Strecker, M.R., 2005. Abnormal monsoon years and their control on erosion and sediment flux in the high, arid northwest Himalaya. Earth and Planetary Science Letters 231, 131–146. Bourke, M.C., Child, A., Stokes, S., 2003. Optical age estimates for hyper-arid fluvial deposits at Homeb, Namibia. Quaternary Science Reviews 22, 1099–1103. Burns, S.J., Matter, A., 1995. Geochemistry of carbonate cements in surficial alluvial conglomerates and their paleoclimatic implications, Sultanate of Oman. Journal Sedimentary Research A65, 170–177. Burns, S.J., Matter, A., Frank, N., Mangini, A., 1998. Speleothem based paleoclimatic record from Northern Oman. Geology 26, 499–502. Burns, S.J., Fleitmann, D., Matter, A., Neff, U., Mangini, A., 2001. Speleothem evidence from Oman for continental pluvial events during interglacial periods. Geology 29, 623–626. Burns, S.J., Fleitmann, D., Matter, A., Kramers, J., Al-Subbary, A.A., 2003. Indian Ocean climate and an absolute chronology over Dansgaard/Oeschger events 9–13. Science 301, 1365–1367. Duller, G.A.T., 2004. Luminescence dating of Quaternary sediments: recent advances. Journal of Quaternary Science 19, 183–192. Fleitmann, D., Burns, S.J., Neff, U., Al-Subbary, A.A., Al-Aowah, M.H., Matter, A., Mangini, A., 2002. Climate change in southern Arabia: speleothems from Yemen and Oman as archives of monsoonal rainfall over the last 330.000 years. Yemen Journal of Science 4, 15–23.

I. Blechschmidt et al. / Geomorphology 110 (2009) 128–139 Fleitmann, D., Burns, S.J., Neff, U., Mangini, A., Matter, A., 2003. Changing moisture sources over the last 330,000 years in Northern Oman from fluid-inclusion evidence in speleothems. Quaternary Research 60, 223–232. Fleitmann, D., Burns, S.J., Neff, U., Mudelsee, M., Mangini, A., Matter, A., 2004. Palaeoclimatic interpretation of high-resolution oxygen isotope profiles derived from annually laminated speleothems from Southern Oman. Quaternary Science Reviews 23, 935–945. Geyh, M.A., 2008. The handling of numerical ages and their random uncertainties. Eiszeitalter & Gegenwart (Quaternary Science Journal) 57, 239–252. Glennie, K.W., 2005. The Desert of Southeast Arabia. Gulf PetroLink, Bahrain. 215pp. Glennie, K.W., Boeuf, M.G.A., Hughes Clarke, M.W., Moody-Stuart, M., Pilaar, W.F.H., Reinhardt, B.M.E., 1974. Geology of the Oman mountains. Verhandelingen Koninklijk Nederlands geologisch mijnboukundig Genootschap, vol. 31. 423pp. Goudie, A.S., Colls, A., Stokes, S., Parker, A., White, K., Al-Farraj, A., 2000. Latest Pleistocene and Holocene dune construction at the north-eastern edge of the Rub’ Al Khali, United Arab Emirates. Sedimentology 47, 1011–1021. Juyal, N., Glennie, K.W., Singhvi, A.K., 1998. Chronology and paleoenvironmental significance of Quaternary desert sediments in SE Arabia. In: Alsharhan, A.S., Glennie, K.W., Whittle, G.L., Kendall, C.G. (Eds.), Quaternary Deserts and Climatic Change, Balkema, pp. 315–325. Kulig, G., 2005. Erstellung einer Auswertesoftware zur Altersbestimmung mittels Lumineszenzverfahren unter spezieller Berücksichtigung des Einflusses radioaktiver Ungleichgewichte in der 238U-Zerfallsreihe. Unpublished BSc thesis, Technical University Bergakademie Freiberg. Leuschner, D.C., Sirocko, F., 2003. Orbital insolation forcing of the Indian Monsoon — a motor for global climate change? Palaeogeography, Palaeoclimatology, Palaeoecology 197, 83–95. Lézine, A.M., Saliège, J.F., Robert, C., Wertz, F., Inizan, M.L., 1998. Holocene lakes from Ramlat as-Sab´atayn (Yemen) illustrate the impact of Monsoon activity in southern Arabia. Quaternary Research 50, 290–299. Liesicki, L.E., Raymo, M.E., 2005. A Pliocene–Pleistocene stack of 57 globally distributed benthic d18O records. Paleoceanography 20, PA 1003. doi:10.1029/2004PA001071. Maizels, J.K., 1987. Plio-Pleistocene raised channel systems of the western Sharqiya (Wahiba). Oman. In: Frostick, L., Reid, I. (Eds.), Desert Sediments: Ancient and Modern. Geological Society, London, Special Publication, vol. 35, pp. 31–50. Maizels, J.K., McBean, 1990. Cenozoic alluvial fan systems of interior Oman: palaeoenvironmental reconstruction based on discrimination of palaeochannels using remote sensed data. In: Robertson, A.H.F., Searle, M.P., Ries, A.C. (Eds.), The Geology and Tectonics of the Oman Region. Geological Society, London, Special Publication, vol. 49, pp. 565–582. Mayya, Y.S., Mortekai, P., Murari, M.K., Singhvi, A.K., 2006. Towards quantifying beta microdosimetric effects in single-grain quartz dose distribution. Radiation Measurements 41, 1032–1039. McClure, H.A., 1976. Radiocarbon chronology of late Quaternary lakes in the Arabian desert. Nature 263, 755–756.

139

Murray, A.S., Wintle, A.G., 2000. Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Radiation Measurements 33, 57–73. Murray, A.S., Wintle, A.G., 2003. The single aliquots regenerative dose protocol: potential for improvements in reliability. Radiation Measurements 37, 377–381. Porat, N., Wintle, A.G., Amit, R., Enzel, Y., 1996. Late Quaternary earthquake chronology from luminescence dating of colluvial and alluvial deposits of the Arava Valley, Israel. Quaternary Research 46, 107–117. Porat, N., Amit, R., Zilberman, E., Enzel, Y., 1997. Luminescence dating of fault-related alluvial fan sediments in the southern Arava valley, Israel. Quaternary Science Reviews 16, 397–402. Porat, N., Zilberman, E., Amit, R., Enzel, Y., 2001. Residual ages of modern sediments in an hyperarid region, Israel. Quaternary Science Reviews 20, 795–798. Prescott, J.R., Hutton, J.T., 1994. Cosmic ray contribution to dose rates for luminescence and ESR dating: large depths and long-term time variations. Radiation Measurements 23, 497–500. Preusser, F., Kasper, H.U., 2001. Comparison of dose rate determination using highresolution gamma spectrometry and inductively coupled plasma - mass spectrometry. Ancient TL 19, 19–23. Preusser, F., Radies, D., Matter, A., 2002. A 160,000-year record of dune development and atmospheric circulation in Southern Arabia. Science 226, 2018–2020. Preusser, F., Degering, D., Fuchs, M., Hilgers, A., Kadereit, A., Klasen, N., Krbetschek, M., Richter, D., Spencer, J., 2008. Luminescence dating: basics, methods and applications. Eiszeitalter & Gegenwart (Quaternary Science Journal) 57, 95–149. Rabus, B., Eineder, M., Roth, A., Bamler, R., 2003. The shuttle radar topography mission — a new class of digital elevation models acquired by spaceborne radar. ISPRS Journal of Photogrammetry and Remote Sensing 57, 241–262. Radies, F., Preusser, F., Matter, A., Mange, M., 2004. Eustatic and climatic controls on the development of the Wahiba Sand Sea, Sultanate of Oman. Sedimentology 51,1359–1385. Radies, D., Hasiotis, S.T., Preusser, F., Neubert, E., Matter, A., 2005. Paleoclimatic significance of Early Holocene faunal assemblages in wet interdune deposits of the Wahiba Sand Sea, Sultanate of Oman. Journal of Arid Environments 62, 109–125. Rhodes, E.J., Singarayer, J.S., Raynal, J.-P., Westaway, K.E., Sbihi-Alaoui, F.Z., 2006. New age estimates for the Palaeolithic assemblages and Pleistocene succession of Casablanca, Morocco. Quaternary Science Reviews 25, 2569–2585. Rodgers, D.W., Gunatilaka, A., 2002. Bajada formation by monsoonal erosion of a subaerial forebulge, Sultanate of Oman. Sedimentary Geology 154, 127–146. Wells, N.A., Dorr, J.A., 1987. A reconnaissance of sedimentation on the Kosi alluvial fan of India. In: Ethridge, F.G., Flores, R.M., Harvey, M.D. (Eds.), Recent Developments in Fluvial Sedimentology. SEPM Spec. Publ., vol. 39, pp. 51–61. Weyhenmeyer, C.E., Burns, S.J., Waber, H.N., Aeschbach-Hertig, W., Kipfer, R., Loosli, H.H., Matter, 2000. Cool glacial temperature and changes in moisture source in Oman groundwaters. Science 287, 842–845.