Licenciate thesis

Geology

New Zircon geochronological and Nd isotopic evidence for Neoproterozoic crust reworking events in the Abas terrane, Yemen

Fitsum G. Yeshanew

Stockholm 2014  

 

 Abstract The Arabian-Nubian Shield is an excellent natural laboratory to study crust formation processes during the Neoproterozoic. It is one of the largest juvenile tracts of continental crust formed during this time. It diachronously evolved between the breakup of Rodinia (c.780 Ma) and amalgamation of Gondwana (c.550 Ma). New SIMS zircon U-Pb, whole-rock Sm-Nd isotopic and geochemical data are presented. The results are used to establish the geochronology of the Abas terrane and constrain its crustal evolution. The U-Pb data show bimodal age distribution: an older age group c. 790-760 Ma, which corresponds to the arc-forming stage of the ANS and a younger group c. 625-590 Ma, belonging to post-collisional episode in the region. The oldest sample in the post-collisional tectonomagmatic group is slightly deformed indicating that pervasive deformation in the area was decaying by c. 625 Ma. The inherited zircons documented range in age from Meso-to-Paleoproterozoic. Although few, these inherited zircons indicate that crust material of that age was assimilated during the Neoproterozoic magmatic events. The U-Pb geochronologic also resolved the temporal transition from high-K calc-alkaline to alkaline magmatism in the post-collisional suits. This transition commonly marks the end of orogeny. Almost all samples studied exhibit a strongly enriched initial εNd compositions and Nd model ages that predate the crystallization ages of the rocks by several hundred million years. These features highlight the sharp contrast in the magmatic sources between the Abas granitoids and the rest of the ANS, which is dominantly juvenile except that of the Afif terrane of Saudi Arabia. This suggests that the Abas terrane of Yemen had a distinctive crustal evolution history compared to the rest of the shield and these features provide evidence for the presence of pre-Neoproterozoic crust at depth in this region.  

 

The thesis contains a background to this PhD project followed by the following manuscript: Fitsum G. Yeshanew*, Victoria Pease, Martin J. Whitehouse and Salah Al-Khirbash., New zircon U-Pb geochronology and Nd isotope systematics of the Abas terrane, Yemen: Implications for Neoproterozoic crust reworking events                  

 

Table  of  Contents  

1. Introduction ............................................................................................................... 1  2. The Earth’s crust ........................................................................................................ 3  2.1. General features of the continental crust  .............................................................................................  3  2.2. Composition of the continental crust  ....................................................................................................  4  2.3. Generation and growth of the continental crust  ................................................................................  4  3. Applied analytical methods ....................................................................................... 5  3.1. U-Th-Pb zircon geochronology  ..............................................................................................................  5  

3.1.1. U-Th-Pb data presentation  .................................................................................................................  7  3.1.2. Sample preparation  ................................................................................................................................  8  3.1.3. Textural information  ..............................................................................................................................  8  3.1.4. Measuring techniques  ...........................................................................................................................  9  

3.2. Sm-Nd isotope system  .............................................................................................................................  11  4. Summary of manuscript .......................................................................................... 12  Acknowledgement ....................................................................................................... 14  References ................................................................................................................... 15  Manuscript ................................................................................................................... 23                  

  1  

 

1. Introduction  The Arabian-Nubian Shield (ANS) consists of Precambrian rocks on the flanks of the Red Sea in western Arabia and northeastern Africa (Johnson and Woldehaimanot, 2003). It represents one of the largest Neoproterozoic crustal growth events on Earth and was exposed after uplift and erosion during Oligocene and younger times (Stern, 1994; Stern et al., 2006). The ANS is dominantly juvenile continental crust, i.e.- crust formed from mantle-derived melts (Stern, 1994; Abdelsalam and Stern, 1996; Stern, 2002; Johnson and Woldehaimanot, 2003; Stoeser and Frost, 2006). Much of the ANS formed in the Neoproterozoic by processes indistinguishable from those of modern plate tectonics: It is a collage of well identified intra-oceanic tectonostratigraphic arc terranes which are separated by recognizable sutures commonly marked by ophiolites (Hargrove et al., 2006). Gondwanaland amalgamated during the Neoproterozoic along the East African Orogen (EAO) (Stern, 1994; Johnson and Woldehaimanot, 2003; Stoeser and Frost, 2006). The ANS forms the northern segment of the EAO and its diachronous evolution encompasses the rifting of Rodinia and subsequent assembly of Gondwana (Stern, 1994; Johnson and Woldehaimanot, 2003). The ANS is an excellent natural laboratory to study plate tectonics processes during this major curst-forming period. The main objectives of this PhD project are to : I) understand the crustal evolution of the southern Arabian Shield and eastern Ethiopian basement, II) evaluate the relative contributions of mantle-derived magmatic additions vs. crustal reworking, and III) facilitate correlation among the various terranes in the Arabian Shield. Towards these goals the project is conducted in two phases: Phase 1. Plutonic rocks from Saudi Arabia, Yemen and Ethiopia were collected. These plutons are diverse in composition, represent different magmatic affinities and span the time period from the arc-forming stage to post-collisional stage in the Arabian-Nubian Shield. Secondary ion mass spectrometry (SIMS) U-Th-Pb geochronology, multiple isotope (Hf-O-Nd-Pb) and geochemical analyses are performed to tie their temporal evolution with geochemical evolution. O and Hf analyses are performed on the same spots/zircon domains, enabling us to provide age control while decoding the magmatic histories of polyphase gneiss genesis. Phase 2. Detrital zircons effectively sample large areas of the continental crust exposed at the time of sediment deposition. We have collected samples of Cambrian sandstone across 1590 km, from sections in Jordan and across Saudi Arabia. These sandstones were deposited following the peneplanation of the ANS basement. The U-Pb, O, and Hf analyses on detrital zircons from these sandstones will shed light on their provenance (possible source regions) and crustal evolution. The results will then

  2  

be compared with the known basement character constrained during phase 1 of this study.

 Figure 1. (A) Inset shows the EAO sandwiched between East and West Gondwana (after Abdelsalam et al., 2008). (B) Simplified geologic map of the ANS in NE Africa and Arabia (adapted from Worku and Schandelmeier, 1996). The study areas are marked with red rectangles and golden yellow filled circles.    Modern geochemical techniques are employed in this project. The manuscript presented utilizes only U/Pb SIMS geochronology and Sm-Nd systematics. These are described below (Methods).  

Cover

Neoproterozoic crust Ophiolite-bearing sutureLocations of sandstone

Post-accretionary structure

samples (Paper IV)

East GondwanaWest Gondwana

India

Antarctica

Australia

Arabian-Nubian Shield (Fig.1B)

Africa

S. America

Mozambique Belt

Ethiopia

South EthiopianShield

Gulf of AdenRed Sea

Indian Ocean

Kenya

Arabian ShieldNu

bian

Shi

eld

0°

5°N

10°N

15°N

20°N

25°N

35°E 40°E 45°E 50°E

500 Km

N

Paper I

Paper II

Paper III

AB

Mesoproterozoic and older crust

  3  

2. The Earth’s crust  

2.1. General features of the continental crust  The Earth is unique in the Solar System in that its crust is different from the other terrestrial planets. The earth’s crust exhibits a bimodal (continental and oceanic) distribution of contrasting compositional and mechanical properties (Fig.2). This is due to the operation of plate tectonics on Earth and the way in which the earth’s mantle cools (Condie, 2012). The continental crust extends vertically from the surface to the crust-mantle boundary. This transition is marked by an abrupt increase in compressional wave velocity (Vp) > c. 7.6 kms−1 and is known as the Mohorovičić discontinuity (Moho). However, petrologic studies of xenoliths have demonstrated that the lithologically defined crust-mantle boundary (felsic-mafic crustal rocks and ultramafic upper mantle) is not always coincident with the seismically defined Moho (Griffin and O’Reilly, 1987). Laterally, the continental crust extends to the continent-ocean transition (Cogley, 1984; Levander et al., 2006). The continental crust varies in thickness between 30 km and 80 km, with an average thickness of 39 km (Christensen and Mooney, 1995). It can also be as thin as 16 km, as in the Afar rift system within the Main Ethiopian Rift (Hayward and Ebinger, 1996). Due to its higher density, the oceanic crust is gravitationally unstable and ultimately founders into the mantle at subduction zones. Consequently, there is no oceanic crust older than 200 Ma (Rudnick, 1995; Kemp and Hawkesworth, 2003). Continental crust; however, is buoyant and difficult to destroy, therefore it represents the archive of the Earth’s geological history (Hofmann, 1988; Kemp and Hawkesworth, 2003; Klein, 2003). Although the continental crust represents only 0.57% of the mass of the silicate Earth, it represents over 60% SiO2, 40% K, and 4.7% MgO (Hawkesworth and Kemp, 2006). It dominates the Earth’s budget of incompatible and heat-producing elements such as U, Th and K (Taylor, 1967; Taylor and McLennan, 1985, 1995; Rudnick, 1995; Rudnick and Fountain, 1995; Rudnick and Gao, 2003; McLennan et al., 2006; Taylor and McLennan, 2009). This extreme enrichment of incompatible elements in the upper crust came about due to intracrustal melting and differentiation, and this process is also responsible for the stratification into felsic (granitic/granodioritic) upper crust and mafic (gabbroic/dioritic) lower crust (Taylor and McLennan, 1985, 1995; Condie, 1993; Rudnick and Gao, 2003; McLennan et al., 2006; Taylor and McLennan, 2009). The formation of the continental crust has depleted the mantle with respect to these incompatible elements and the two reservoirs are considered geochemically complementary (Hofmann, 1988; Rudnick, 1995). Therefore, understanding the overall composition and temporal development of the continental crust is fundamental to understanding processes of crustal growth and the evolution of the complementary depleted mantle (Jacobsen and Wasserburg, 1979; Allègre et al., 1983; Taylor and McLennan, 1985; Rudnick, 1995; 1995; Hofmann, 1997; McLennan et al., 2006; Taylor and McLennan, 2010).

  4  

 Figure 2. Bimodal surface elevation features on Earth with reference to sea level. It depicts the difference in composition and average residence age of the continental and oceanic crust (adapted from Taylor and McLennan, 1996; Cawood et al., 2013)  

 

2.2. Composition of the continental crust  The composition of the upper continental crust is well established by averaging the concentrations of insoluble elements in fine-grained sediments, which effectively sample and average upper crustal lithologies, and sampling Precambrian shields and exposed crustal sections (Taylor and McLennan, 1985; Taylor and McLennan, 1991; Condie, 1993; Rudnick and Gao, 2003; Condie, 2012). The average composition of the upper continental crust is akin to granodiorite. Due to its inaccessibility, the composition of the lower crust is less understood and one must rely on indirect methods to constrain its composition. These include (1) measurement of seismic velocities in the laboratory under appropriate temperature and pressure conditions; (2) by directly studying exhumed middle and lower crustal sections, and (3) by analyzing lower crustal xenoliths brought to the surface during volcanic eruptions. All of these approaches suggest that the lower crust is mafic in composition. The bulk composition of the continental crust is similar to arc-related intermediate rocks or andesites (Taylor and McLennan, 1985; Wedepohl, 1995; Rudnick and Gao, 2003; McLennan et al., 2006; Condie, 2012).  

2.3. Generation and growth of the continental crust  There is general agreement that subduction-related magmatism is responsible for the generation of post-Archean continental crust, however, the process giving rise to this in arc-systems is poorly understood (Condie, 2012). Trace element compositions of the continental crust broadly resemble that of magmatic rocks formed in subduction

Continental crust – andesite(average age = 2000 Ma)

Oceanic crust – basalt(Maximum age ≤ 200 Ma

5 10 15 200% of surface

Elev

atio

n (k

m)

–6

–4

–2

0

2

4

  5  

zones (Taylor and McLennan, 1985; Rudnick, 1995). However, differences do exist in some trace element characteristics of the continental crust and convergent margin magmatic rocks. For instance, continental crust has a lower Sr/Nd ratio than typical arc magmas. This is interpreted to be due to the transformation of plagioclase cumulates to plagioclase-free eclogites in thickened crust which ultimately founders into the mantle, but this mechanism fails to explain the discrepancy in the La/Nb ratio, which is insensitive to intracrustal fractionation, between the continental crust and arc rocks (Rudnick, 1995). The bulk composition of the continental crust resembles subduction-zone andesites and this was the basis for the ‘‘Andesite Model’’ for post-Archean continental growth (Taylor, 1966; Taylor 1977; Taylor and McLennan, 1985; Rudnick, 1995; McLennan et al., 2006). This model was hampered by the fact that subduction-zone magmas ultimately issue from the mantle and are therefore expected be basaltic in composition rather than andesitic (Arculus, 1981; Taylor and McLennan, 2010). Consequently, the ‘’delamination’’ model, which suggests that basalts are the main constituents of the lower continental crust, generated a felsic upper crust from the delamination of the residual dense lower continental into the underlying mantle (Kay and Kay, 1991; Nelson, 1991). However, the delamination model itself fails to explain some critical problems. For example, there is no observable change between pre- and post-Archean bulk composition of the bulk continental crust and this requires striking a balance between the amount of material added to the crust and the loss of material from the subcontinental region (Taylor and McLennan, 2010). Additionally, widespread delamination seems to be uncommen; it has only been identified in a few localized cases (Jull and Kelemen, 2001; Taylor and McLennan, 2010). The other major problem with the delamination model is the chemical and isotopic legacy of the lower crust such as positive Eu anomalies appears to be lacking in mantle-derived rocks (Taylor and Mclennan, 2010). Seismological studies of the Izu-Bonin arc, in the Western Pacific, have shown that the bulk composition of the crust beneath the basaltic volcanoes is more mafic than the average continental crust, although both felsic and intermediate rocks have also been formed beneath the arc (Koidara et al., 2007). Therefore, it seems that the transformation of more mafic arc crust into typical ‘andesitic’ continental crust must involve foundering of olivine-pyroxene cumulates into the mantle (Oliver et al., 2003; Koidara et al., 2007). This foundering process in subduction zones, although unobservable, remains more realistic than delamination beneath stable cratons as it evades the problems faced by delamination (Taylor and McLennan, 2010).  

3. Applied analytical methods  

3.1. U-Th-Pb zircon geochronology  Several features of zircon (ZrSiO4) make it a excellent mineral for geochronology. It is common in crustal rocks, it survives reprocessing, it is radiogenic and therefore can be used for age dating. Zircon incorporates modest amounts of U and Th into its

  6  

crystal structure when it forms, but excludes Pb. The radioactive decay of U and Th causes the accumulation of radiogenic Pb, thereby establishing the basis for accurate and precise measurement of zircon’s isotopic age. Diffusion rates of Pb, U and Th in zircon are found to be extremely low under c. 1000°C (Lee et al., 1997; Cherniak and Watson, 2001), therefore it retains its original isotopic signature or integrity even when exposed to extreme temperatures and through multiple episodes of sedimentary and magmatic recycling (Hawkesworth et al., 2010). Zircon often preserves different growth zones (Fig.3) within a single grain and may contain information about different stages in the subduction and exhumation cycle (Harley and Kelly, 2007; Hawkesworth and Kemp, 2006b; Ireland and Wlotzka, 1992; Rubatto and Hermann, 2007; Scherer et al., 2007). Of the many radioactive isotope systems utilized in geochronology, U-Pb zircon dating is unique owing to its dual decay scheme of two radioactive and chemically identical uranium isotopes (235U and 238U). These two isotopes independently decay at different rates to their respective stable daughter isotopes of 207Pb and 206Pb (Williams, 1998). Lead has four naturally occurring isotopes: 204Pb, 206Pb, 207Pb, and 208Pb. Among these only 204Pb is stable (non-radiogenic) and 208Pb is produced by the decay of 232Th. Each of these decay schemes involves many intermediate steps and several short-lived intermediate isotopes. These three decay schemes can be described using the following isochron equations: 206Pb204Pb

=  206Pb204Pb 0 +  

238U204Pb

eλ238t − 1                                                                                   1 207Pb204Pb

=  207Pb204Pb 0 +  

235U204Pb

eλ235t − 1                                                                                   2 208Pb204Pb

=  208Pb204Pb 0 +  

232U204Pb

eλ232t − 1                                                                                   3 However, because the final step in the decay chain is many orders of magnitude slower than the intermediate steps, the whole decay process can be described mathematically by a single decay equation relating the number of ultimate parent atoms (e.g. 235U, 238U) remaining and the number of final radiogenic daughter atoms (e.g. 207Pb, 206Pb* - the asterisk signifies that it is radiogenic) to time (Harley and Kelly, 2007; Schoene, 2014). Equations [1] to [3] can be reduced to: 206Pb*238U

=   𝑒λ238! − 1                                                                                                                                                       4 207Pb*235U

=   𝑒λ235! − 1                                                                                                                                                       5 208Pb*232U

=   𝑒λ232! − 1                                                                                                                                                       6

  7  

 

3.1.1. U-Th-Pb data presentation  There are several ways of graphically displaying U-Th-Pb data. The two most commonly used, the Concordia and the Inverse concordia, are described below. The Wetherill (Concordia) diagram. This is also known as the ‘conventional’ concordia diagram and is constructed by plotting 207Pb*/235U is plotted against 206Pb*/238U. The reference concordia curve is constructed using the solutions that satisfy equations [4] and [5]. All data points that lie on this curve yield concordant U-Pb ages and are consistent with being closed isotopic systems (Ahrens, 1955; Wetherill, 1956a; Wetherill, 1963). All points that lie off of the curve are discordant and signify open system behavior (Fig.3). There are several causes of discordance in the U-Th-Pb system. The two most common geological causes of discordance are due to i) analyzing a mixture of older and younger domains (Vavra et al., 1999) and ii) subsequent Pb-loss after crystallization of the zircon. An excellent review on U-Pb discordance can be found in Corfu, 2013.

 Figure 3. A Wetherill concordia diagram showing the history of zircon growth after its crystallization (redrawn from Shoene, 2014).

 The Tera-Wasserburg (Inverse) diagram An alternative way of plotting U-Pb data is the Tera-Wasserburg concordia or inverse concordia developed by Tera and Wasserburg (1972). In this diagram 207Pb/206Pb is plotted against 238U/206Pb. Lunar rocks contain excess radiogenic 206Pb, 207Pb and 208Pb that was incorporated during the time of crystallization. This causes a discrepancy in dates of lunar samples calculated by U, Th-Pb methods and other isotopic dating techniques. For instance, Tatsumoto and Rosholt (1970) and Tera and Wasserburg (1972) reported U-Th-Pb ages for lunar samples, which were considerably older than the Rb-Sr and K-Ar dates of the same rocks. To overcome this problem, Tera and Wasserburg (1972) developed a concordia diagram that does

3000

2500

2000

1500

1000

tc = age of zircon after c. 1700 My

t’ = discordant dates

t2 = Pb-loss or growth of new zircon rim

Pb-Pb dates of the cyan analysesdefine the true date

5 10 15 20 25

0.1

0.2

0.3

0.4

0.5

0.6

0.7

206 Pb

/238 U

207Pb/235U

  8  

not use the initial 206Pb/204Pb and 207Pb/204Pb ratios. Another advantage of the Tera-Wasserburg diagram is that the measured 207Pb/206Pb and 238U/206Pb errors are much more weakly correlated than those of 207Pb/235U and 206Pb/238U, thus regression analysis of the data is simpler (Williams, 1998). In Tera-Wasserburg diagram recent Pb-loss can easily be discerned (Fig.4).

 Figure 4. Tera-Wasserburg concordia diagram showing reverse discordance and recent Pb-loss  

3.1.2. Sample preparation  The concentration and purification of zircon crystals from rock samples is a time-consuming process that requires patience and care. After the rock samples are crushed and sieved, the zircon fraction is recovered by means of hydrodynamic separation using a Wilfley table and heavy liquids that separate minerals by their specific gravity. Further purification of the zircon concentrate is done using magnetic separation because of compositional differences and differing magnetic susceptibilities. Krogh (1982a) showed that magnetic sorting of zircons to select the least paramagnetic grains tends to move data points upwards along discordia towards the upper intercept with the concordia curve. Mechanical abrasion of U-bearing minerals by a jet of dry air inside air-abrasion vessels was also found to improve the precision of U-Pb dates considerably (Goldich and Fischer, 1986; Krogh, 1982b). Abrasion removes highly metamict (radiation-damaged) grains, as these tend to be softer than the pristine and crack-free grain domains. This process also reduces discordance of U-Pb dates by removing the outter surface layer of zircons where Pb-loss is most dominant. Thermal annealing and partial-dissolution techniques are used to remove high-U zones, which are prone to Pb-loss (Ramezani et al., 2007). Zircon crystals are then handpicked and sorted on the basis of their color, size and morphology under binocular microscope and cast into a circular epoxy. Mounts are polished to reveal internal structures of zircon grains. The polished samples are coated with gold to facilitate acceleration of the sputtered ions and to provide an equipotential surface to the secondary-ion extraction and avoid charging during sputtering (Hinthorne et al., 1979; Hinton and Long, 1979; Ireland and Williams, 2003).  

3.1.3. Textural information  

820

700

780

780

6 7 8 9 10 110.062

0.063

0.064

0.065

0.066

0.067

recent Pb-loss

207 Pb

/206 Pb

238U/206Pb

reversely discordant

concordia age = 786 ± 7 Ma(MSWD = 1.5; n = 4)

  9  

It is common practice to perform detailed imaging of internal zircon structure using optical microscopy, cathodoluminescence (CL) and back-scattered electrons prior to SIMS or LA-ICP-MS analysis on zircon. CL guided studies demonstrate that zircons are far more variably zoned than previously thought (e.g., Whitehouse et al., 1999). These detailed imaging techniques are essential for the correct interpretation of U-Pb data (Hanchar and Miller 1993; Vavra, 1990) and enable us choose discrete portions of different domains within zircon for isotopic analysis, avoiding inclusions, cracks or radiation damages.  

3.1.4. Measuring techniques  The secondary ion mass spectrometry (SIMS) is applied in this thesis. SIMS ability to perform in situ chemical and isotopic analyses on solid materials of few tens of microns in diameter make it a powerful tool in geochronology (Ireland and Williams, 2003). This method allows in-situ isotopic analysis of separate mineral grains or multiple analyses within a single grain, retaining textural information and avoiding dissolution or chemical treatment of samples. The main components of an ion microprobe include the primary ion source, the secondary ion extraction systems, the mass spectrometer and secondary ion detectors. Different primary ion beams are used for various purposes. The commonly used ones in isotope geochemistry are 133Cs+, 16O− and 16O2

−. The yield of the secondary ions is strongly influenced by the properties of the primary species and sometimes strongly electropositive species such as Cs+ are used when a greater negative secondary ion yield is desired for electronegative elements such as S and O (Reed, 1989). In SIMS analysis, primary ion beams of a few microns in diameter are used see Fig.5). Bombardment by these primary ions erodes atoms from the surface of the specimen by a process called sputtering (MacRae, 1995; Reeds, 1989; Williams, 1998). The secondary atoms are ionized and then accelerated into the entrance slit of the mass spectrometer (Reed, 1989). This process generates all the elements present in the target mineral and their oxides and hydroxides, and therefore requires high mass resolution to resolve isobaric interferences on the desired ions. This is elaborated in more detail below.

  10  

 Figure 5. Primary ions impacting a sample surface and progressively ejecting secondary ions.

 The accuracy of SIMS measurements can be affected by instrumental mass fractionation (IMF), which is a mass dependent bias introduced during measurement of isotope ratios. In ion microprobe analysis, IMF is corrected by comparing measurements of a chemically and isotopically homogeneous mineral reference material (RM). The RM should be a matrix-match (compositionally similar to the unknown). The similarity in composition between the RM and the unknown is of crucial importance since there is a compositionally dependent fractionation known as matrix effect (Fletcher et al., 2010). Different processes account for the observed IMF during SIMS analysis. Some of these are detection (Lyon et al., 1994; Valley and Graham, 1991), secondary ion transmission (Shimizu and Hart, 1982b), sputtering and ionization (Schroeer et al., 1973; Yu and Lang, 1986), matrix effect (Fletcher et al., 2010; Gurenko et al., 2005; Reed, 1989). In zircon U-Pb analyses by SIMS, the fractionation between Pb isotopes is considered insignificant (Ireland and Williams, 2003) and the major use of a zircon U-Pb reference material is to determine the inter-element, U/Pb isotopic ratio. Quantification of the mass-dependent fraction of Pb isotopes is very difficult because of the low abundance of Pb in zircon and the difficulty of discriminating mass interferences on Pb from hydrides (Ireland and Williams, 2003). Mass resolution and isobaric interferences The secondary ion yield from the sputtering process contains both atomic and molecular ions (oxide, multi-element, and hydroxide species) and all of them are accelerated into the mass spectrometer (Hinton and Long, 1979). Consequently, a mass interference known as isobaric interference is introduced whenever another ion that has the same nominal mass as the analyte ion occurs. High mass resolution is therefore required to overcome these interference problems. Mass resolution (R) is expressed as R = M/ΔM, where ΔM is the difference in mass of the two ions having

e–

e–

e–

e–

+

+

+

+

+

+

+

+

––

–

Primary ion beam

Sample

  11  

mass number M. On the other hand, elemental interference occurs when an isotope of one element has the same nominal mass as an isotope of another. Discriminating these requires ultra-high mass resolution and cannot be achieved by commercial SIMS instruments. In ion microprobe analysis high mass resolution can be attained by using a double focusing system, involving both an electrostatic filter as well as a magnet in the mass spectrometer (Reed, 1989). The mass resolution capacity of small ion microprobes does not usually exceed 1000, which is not adequate for zircon U-Pb geochronology due to interferences (Williams, 1998). Therefore, high mass resolution is used to separate some of the molecular interferences from the Pb isotopes but at a cost of sensitivity loss (Hinthorne et al., 1979). The largest ion microprobes were built with these problems in mind. The first sensitive high-resolution ion microprobe (SHRIMP) was designed and built at the Australian National University (Compston et al., 1982) in order to achieve a mass resolution of 10 000, although 5000 is enough to resolve the main isobaric interferences that are encountered in zircon U-Pb dating. Depth Profiling The progressive erosion of the sample by the bombardment of primary ions enables the depth distribution of an element or isotope of interest to be determined. This can be done by measuring the appropriate peak intensity as a function of time and converting the time scale to depth profile (Reed, 1989).  

3.2. Sm-Nd isotope system  The decay of 147Sm to 143Nd by α-emission, with a decay constant of 6.54 × 10−12yr−1 (Lugmair and Marti, 1978), forms the basis for this isotopic system (Patchett and Samson, 2003; DePaolo, 2012). This decay process happens at an extremely sluggish rate with a half-life of 147Sm being 106 billion years (Dickin, 2005; Faure and Mensing, 2005; DePaolo, 2012; Patchett and Samson, 2003). Therefore the accumulation of radiogenic 143Nd is very slow, resulting in the change of 143Nd/144Nd from c. 0.50687 at 4.5 Ga to a present-day value of c. 0.51264 (Jacobsen and Wasserburg, 1984). The long half-life of this decay process makes the Sm-Nd isotopic system useful for studying crust-mantle differentiation events. Both samarium and neodymium are light REEs and refractory. Their refractory nature suggests that they are likely to form early during the condensation of the Solar System (Grossman and Larimer, 1974). This implies the parent-daughter fractionation during condensation of the solar nebula and a variety of other processes were insignificant (McSween and Huss, 2010; DePaolo, 2012). Therefore, this coupled with the lithophile character of both elements imply variations in measured Sm/Nd ratios are solely due to Earth’s internal differentiation and are insensitive to planetary formation and planetary core formation (DePaolo, 2012).

  12  

The variations in 143Nd/144Nd are quite small and are expressed by comparison to a standard reservoir known as Chondritic Uniform Reservoir (CHUR). It is denoted by the parameter εNd(t) (DePaolo and Wasserburg, 1976):

εNd(t) = 104143Nd144Nd Sample t −

143Nd144Nd CHUR t

143Nd144Nd CHUR t

The use of Sm-Nd method in crustal evolution draws upon the Sm/Nd fractionation between the crust and its mantle source. Nd preferentially partitions into mantle derived melt whereas Sm is enriched in the depleted mantle. Through time their initial Nd isotopic composition, εNd(t), diverge. Depleted mantle attains more radiogenic (positive) εNd(t)) and the continental crust will have enriched (negative) εNd(t) compositions (Arndt and Goldstein, 1987). Therefore DM derived juvenile magmas attain positive εNd(t) values whereas crust formed by melting or assimilation of preexisting crustal material will have negative εNd(t). Depleted mantle model (TDM) ages assume the magma from which the rock has crystallized was derived from the DM reservoir. TDM is completely dependent upon various assumptions such as the assigned initial value (depleted mantle vs. chondritic mantle), single-stage vs. two-stage models, which result in Nd-TDM differences of a few hundreds of million years and the model used (e.g. DePaolo, 1981 vs. Goldstein et al., 1984). Generally, if Nd-TDM approximates the crystallization age of the rocks, determined by independent method (e.g. U/Pb zircon age), the source is regarded as juvenile. If Nd-TDM is much older than the crystallization age of the rock by over c. 300 Ma, the source is interpreted as contaminated.  

4. Summary of manuscript  In the current study we have employed an integrated high-precision SIMS U-Th-Pb zircon geochronology, whole-rock Sm-Nd isotopic and geochemical techniques to decipher the crustal evolution of the Abas terrane, Yemen, in the SE Arabian Shield. We have established the time span of the older granitic gneisses, which belong to the arc-formation stage in the ANS, determined when post-collisional magmatism commenced and when the transition from high-K calc-alkaline to alkaline occured in this region. Furthermore, our geochronological data show c. 135 Ma magmatic shut down in the Abas terrane, which is slightly younger than the magmatic gap documented in other parts of the shield.

  13  

The Nd isotope data shows a clear distinction between the older gneisses and post-collisional granites. The later show a more depleted initial Nd isotopic signatures and Nd model ages that are slightly but consistently younger than the gneisses. These temporal variations are interpreted to be due to the involvement of lesser amount of older crustal component and more juvenile component during the genesis of the post-collisional granites. Whereas, the older magmatic group were formed in a compressional regime, which favors a more pronounced crustal reworking hence they have a more enriched initial Nd signatures (Fig.6). The data also show the Abas terrane had a different evolutionary history compared to other parts of the shield, which are dominantly juvenile crust of Neoproterozoic age.  

   Figure 6. A Nd isotope evolution diagram of the Abas granitoids along with data from other parts of the ANS (Hargrove et al., 2006; Stoeser and Frost, 2006 and references therein.)

     

   

εNd

(t)

0 500 1000 1500 2000-15

-10

-5

0

5

10

Crystallization age (Ma)

CHUR

Abas terrane

Bi’r Umq suture

Arabian Shield

terranes

DM

Granitic gneissesPost-collisional granites

  14  

Acknowledgement  I express my gratitude to my superb supervisors Victoria Pease and Martin Whitehouse for giving me the change to work on this interesting project and for their constant support throughout the project. I thank the Lord Jesus Christ for all He has done in my life. I would like to thank my wife Meron and my daughter Johanna (Haset) for their encouragement and support and for their understanding. My families and friends both in Ethiopia and Sweden deserve special thanks for all they have done for me and for their constant support. Eve Arnold is specially thanked for her invaluable advice throughout my project. At NRM my thanks go to Lev Ilyinsky, Kerstin Linden, Karin Wallner, Hans Schöberg, Per-Olof Persson, Heejin Jeon and Ulf Hålenius for their assistance during my lab work and for being kind people. I also thank Per Andersson for his advise and help and Kjell Billström for helping me settle in Stockholm and with various practical things.

       

  15  

References  Abdelsalam, M.G., and Stern, R.J. 1996. Sutures and shear zones in the Arabian-Nubian Shiled. Journal of African Earth Sciences, 23, 289-310. Abdelsalam, M.G., Tsige, L., Yihunie, T. & Hussien, B. 2008. Terrane rotation during the East African Orogeny: Evidence from the Bulbul Shear Zone, south Ethiopia. Gondwana Research, 14, 497-508. Ahrens, L.H., 1955. Implications of the Rhodesia age pattern. Geochim. Cosmochim Acta, 8, 1-15. Allegre, C.J., Hart, S.R & Minster, J.-F. 1983. Chemical structure and evolution of the mantle and continents determined by inversion of Nd and Sr isotopic data, II. Numerical experiments and discussion, Earth and Planetary Science Letters, 66, 191-213. Arculus, R.J. 1981. Island arc magmatism in relation to the evolution of the crust and mantle, Tectonics, 75, 113-133. Arndt, N.T. and Goldstein, S.L. 1987. Use and abuse of crust-formation ages. Geology, 15, 893-895. Cawood, P.A., Hawkesworth, C.J. & Dhuime, B. 2013. The continental record and the generation of continental crust, The Geological Society of America Bulletin, 125, 4-32 Cherniak, D.J. and Watson, E.B. 2001. Pb diffusion in zircon, Chemical Geology, 172, 5-24. Christensen, N.I., Mooney, W.D. 1995. Seismic velocity structure and composition of the continental crust: A global view. Journal of Geophysical Research, 100, 9761-9788. Cogley, J.G. 1984. Continental margins and the extent and number of the continents. Reviews of Geophysics, 22, 101-122. Compston, W., Williams, I.S. and Clement, S.W., 1982. U-Pb ages within zircons using a sensitive high mass-resolution ion microprobe. 30th American Society for Mass Spectrometry Conference, Honolulu, 593-595 Condie, K.C. 1993. Chemical composition and evolution of the upper continental crust: Contrasting results from surface samples and shales, Chemical Geology, 104, 1-37.

  16  

Condie, K.C. 2012. Earth as an Evolving Planetary System. Elsevier, Amsterdam. Corfu, F. 2013. A century of U-Pb geochronology: The long quest towards concordance. Geological Society of America Bulletin, 125, 33-47. DePaolo, D.J. and Wasserburg, G.J. 1976. Inferences about magma sources and mantle structure from variations of 143Nd/144Nd. Geophysical Research Letters. 3, 743-746. DePaolo, D.J. 1981. Neodymium isotopes in the Colorado Front Range and crust–mantle evolution in the Proterozoic. Nature, 291, 193–196. DePaolo, D.J. 2012. Neodymium isotope geochemistry; An introduction. Springer-Verlag, 187pp. Dickin, A.P. 2005. Radiogenic Isotope Geology. Press Syndicate of the University of Cambridge, 492 pp. Faure, G. and Mensing, T.M. 2005. Isotopes: Principles and Applications 3rd edition. Wiley, 897pp. Fletcher, I.R., McNaughton, N.J., Davis, W.J., Rasmussen, B., 2010. Matrix effects and calibration limitations in ion probe U-Pb and Th-Pb dating of monazite. Chemical Geology. 270, 31-44. Goldich, S.S. & Fischer, L.B., 1986. Air-abrasion experiments in U-Pb dating of zircon. Chemical Geology. 58, 195-215. Goldstein, S.L., O’Nions, R.K. and Hamilton, P.J. 1984. A Sm-Nd isotopic study of atmospheric dusts and particulates from major river systems. Earth and Planetary Science letters, 70, 221-236. Griffin, W.L. & O'Reilly, S.Y. 1987. Is the continental Moho the crust-mantle boundary? Geology, 15, 241-244. Grossman, L. and Larimer, J.W. 1974. Early chemical history of the Solar System. Rev. Geophys Space Phys. 12, 71-101. Gurenko, A.A., Veksler, I.V., Meixner, A., Thomas, R. and Dorfman, A.M., 2005. Matrix effect and partitioning of boron isotopes between immiscible Si-rich and B-rich liquids in the Si-Al-B-Ca-Na-O system: A SIMS study of glasses quenched from centrifuge experiments. Chemical Geology. 222, 268-280.

  17  

Hanchar, J.M. and Miller C.F., 1993. Zircon zonation patters as revealed by cathodoluminescence and back-scattered electron images- implication for interpretation of complex crustal histories. Chemical Geology, 110, 1-13. Hargrove, U.S., Stern, R.J., Kimura, J.I., Manton, W.I. & Johnson, P.R. 2006. How juvenile is the Arabian–Nubian Shield? Evidence from Nd isotopes and pre-Neoproterozoic inherited zircon in the Bi'r Umq suture zone, Saudi Arabia. Earth and Planetary Science Letters, 252, 308–326. Harley, S.L. and Kelly, N.M., 2007. Zircon Tiny but Timely. Elements, 3, 13-18. Hawkesworth, C.J. and Kemp, A.I.S., 2006b. Using hafnium and oxygen isotopes in zircon to unravel the record of crustal evolution. Chemical Geology, 226, 144-162. Hawkesworth, C.J., Dhuime, B., Pietranik, A.B., Cawood, P.A., Kemp, A.I.S. & Storey, C.D. 2010. The generation and evolution of the continental crust. Journal of the Geological Society, 167, 229-248 Hayward, N.J., Ebinger, C.J. 1996. Variations in the along-axis segmentation of the Afar Rift system. Tectonics, 15, 244-257. Hawkesworth, C.J. & Kemp, A.I.S. 2006. The differentiation and rates of generation of the continental crust. Chemical Geology, 226, 134-143. Hinthorne, J.R., Andersen, C.A., Conrad, R.L. and Lovering, J.F., 1979. Single-grain 207Pb/206Pb U/Pb age determinations with a 10-µm spatial resolution using the ion microprobe mass analyzer (IMMA). Chemical Geology, 25, 271-303. Hinton, R.W. and Long, J.V.P., 1979. High-resolution ion-micro-probe measurement of lead isotopes variations within single zircons from Lac Seul, northern Ontario. Earth and Planetary Science Letters. 45, 309-325. Hofmann, A.W. 1988. Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters, 90, 297-314. Hofmann, A.W. 1997. Mantle geochemistry: the message from oceanic volcanism, Nature, 385, 219-229. Ireland, T.R., Wlotzka, F., 1992. The oldest zircons in the solar system. Earth and Planetary Science Letters, 109, 1-10.

  18  

Ireland, T.R. and Williams, I.S., 2003. Considerations in zircon geochronology by SIMS. In: Hanchar, J.M. and Hoskin, P.W.O. (eds.): Zircon. Reviews in Mineralogy and Geochemistry, 53, 215-241. Jaconsen, S.B. and Wasserburg, G.J. 1984. Sm-Nd isotope evolution of chondrites and achondrites, II. Earth and Planetary Science Letters, 67, 137-150. Jacobsen, S.B. and Wasserburg, G.J. 1979. The mean age of the mantle and crustal reservoirs, Journal of Geophysical Research, 84, 7411-7427. Johnson, P. R. & Woldehaimanot, B. 2003. Develop- ment of the Arabian – Nubian Shield: Perspectives on accretion and deformation in the northern East African orogen and the assembly of Gondwana. In: Yoshida, M., Windley, B. F. & Dasgupta, S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 289–325. Jull, M. & Kelemen, P.B. 2001. On the conditions for lower crustal convective instability, Journal of Geophysical Research, 106, 6423-6446. Kay, R.W. & Kay, S.M. 1991. Creation and destruction of continental crust. Geol. Rundsch, 80, 259-278. Kemp, A.I.S. & Hawkesworth, C.J. 2003. Granitic perspectives on the generation and secular evolution of the continental crust. In: R.L., Rudnick (Ed.), The Crust, Treatise in Geochemistry, 3, 349-410. Klein, E.M. 2003. Geochemistry of the igneous oceanic crust. In: R.L., Rudnick (Ed.), The Crust, Treatise in Geochemistry, 3, 433-463. Koidara, S., Sato, T., Takahashi, N., Miura, S., Tamura, Y,. Tatsumi, Y. and Kaneda, Y. 2007. New Seismological constraints on growth of continental crust in the Izu-Bonin intra-oceanic arc, Geology, 35, 1031-1034. Krogh, T.E., 1982a. Improved accuracy of U-Pb zircon dating by selection of more concordant fractions using a high gradient magnetic separation technique. Geochim. Cosmochim. Acta. 46, 631-635. Krogh, T.E., 1982b. Improved accuracy of U-Pb zircon ages by the creation of more concordant systems using an air ablation technique. Geochim. Cosmochim. Acta. 46, 637-649. Lee, J.K.W., Williams, I.S. and Ellis, D.J. 1997. Pb, U and Th diffusion in natural zircon, Nature, 390, 159-162.

  19  

Levander, A., Lenardic, A. and Karlstrom, K. 2006. Structure of the continental lithosphere. In: M., Brown & T., Rushmer (eds), Evolution and Differentiation of the continental crust, Cambridge University Press, 21-66.

Lugmair, G.W. and Marti, K. Lunar initial 143Nd/144Nd; differential evolution of the lunar crust and mantle. Earth and Planetary Science Letters, 39, 349-357. Lyon, I.C., Saxton, J.M. and Turner, G., 1994. Isotope fractionation in secondary ionization mass spectrometery. Rapid Commun. Mass Spectrometry, 8, 837-843. MacRae, N.D., 1995. Secondary-ion mass spectrometry. The Canadian Mineralogist, 33, 219-236. McLennan, S,M., Taylor, S.R. and Hemming, S.R. 2006. Composition, differentiation, and evolution of the continental crust: constraints from sedimentary rocks and heat flow. In: M., Brown & T., Rushmer (eds), Evolution and Differentiation of the continental crust, Cambridge University Press, 92-134. McSween, H.Y. and Huss, G. 2010. Cosmochemistry. Cambridge, 549 pp. Nelson, K.D. 1991. A unified view of craton evolution motivated by recent deep seismic reflection and refraction results. Geophysical Journal International, 105, 25-35. Oliver, S.B., Jones, C.H. and Sheehan, A.F. 2003. Foundering lithosphere imaged beneath the southern Sierra Nevada, California, USA, Science, 305, 660-662. Patchett, P.J and Samson, S.D. 2003. Ages and growth of the continental crust from radiogenic isotopes. In: R.L., Rudnick (Ed.), The Crust, Treatise on Geochemistry, 3, 321-348. Ramezani, J., Schmitz, M.D., Davydov, V,I., Bowring, S.A., Snyder, W.S. and Northrup, C.J., 2007. High-precision U-Pb zircon age constraints on the Carboniferous-Permian boundary in the southern Urals stratotype. Earth and Planetary Science Letters, 256, 244-257. Reed, S.J.B., 1989. Ion microprobe analysis- a review of geological applications. Mineralogical Magazine, 53, 3-24. Rubatto, D. and Hermann, J., 2007. Zircon Behaviour in Deeply Subducted Rocks. Elements, 3, 31-35. Rudnick, R. L. 1995. Making continental crust. Nature ,378, 571-578.

  20  

Rudnick, R. L & Fountain, D. 1995. Nature and composition of the continental crust: a lower crustal perspective. Reviews of Geophysics, 33, 267-309. Rudnick R.C. & Gao,S. 2003. Composition of the continental crust. In: R.L., Rudnick (Ed.), The Crust, Treatise on Geochemistry, 3, 1-64. Scherer, E.E., Whitehouse, M.J. and Munker, C., 2007. Zircon as a Monitor of Crustal Growth. Elements, 3, 19-24 Schoene, B. 2014. U-Th-Pb geochronology. In: R.L., Rudnick (Ed.), The Crust, Treatise on Geochemistry 2nd edition, 4, 341-378. Schroeer, M.J., Rhodin, T.N. and Bradley, R.C., 1973. A quantum-mechanical model for the ionization and excitation of atoms during sputtering. Surface Science, 34, 571-580. Shimizu, N. and Hart, S.R., 1982b. Application of the ion microprobe to geochemistry and cosmochemistry. Annual Review of Earth and Planetary Science. 10, 483-526. Stern, R.J. 1994. Arc-assembly and continental collision in the Neoproterozoic African orogen: implications for the consolidation of Gondwanaland. Annual Review of Earth and Planetary Sciences. ,  22,  319–351. Stern, R.J. 2002. Crustal evolution in the East African Orogen: a neodymium isotopic perspective. Journal of African Earth Sciences, 34, 109–117. Stern, R.J., Avigad, D., Miller, N.R. & Beyth, M. 2006. Evidence for the Snowball Earth hypothesis in the Arabian-Nubian Shield and the East African Orogen. Journal of African Earth Sciences, 44, 1-20. Stoeser, D.B & Frost, C.D. 2006. Nd, Pb, Sr, and O isotopic characterization of Saudi Arabian Shield terranes. Chemical Geology, 226, 163–188. Tatsumoto, M. & Rosholt, J.N., 1970. Age of the Moon: An isotopic study of U-Th-Pb systematics of lunar samples. Science, 167, 461-463. Taylor, S.R. 1967. The origin and growth of the continents. Tectonophysics, 4, 17-34. Taylor, S.R. 1977. Island arc models and the composition of the continental crust. American Geophysical Union Maurice Ewing Series, 1, 325-335. Taylor, S.R., & McLennan, S.M. 1985. The Continental Crust: Its Composition and Evolution. Blackwell, 312pp.

  21  

Taylor, S.R. & McLennan, S.M. 1995. The geochemical evolution of the continental crust. Reviews of Geophysics, 33, 241-265. Taylor, S.R. & McLennan, S.M. 2009. Planetary Crusts, Cambridge University Press, 387pp. Taylor, S.R. & McLennan, S.M. 1996. The evolution of the continental crust, Scientific American, 274, 76-81. Tera, F.& Wasserburg, G.J., 1972. U-Th-Pb systematics in three Apollo 14 basalts and the problem of initial Pb in Lunar rocks. Earth and Planetary Science Letters. 14, 281-304. Valley, J.W. and Graham, C.M., 1991. Ion microprobe analysis of oxygen isotope ratios in granulite facies magnetites: diffusive exchange as a guide to cooling history. Contributions to Mineralogy and Petrology, 109, 38-52. Vavra, G. 1990. On the kinematics of zircon growth and its petrogenetic significance- A cathodoluminescence study. Contributions to Mineralogy and Petrology, 106, 90-99. Vavra, G., Schmid R. and Gebauer, D. 1999. Internal morphology, habit and U-Th-Pb microanalysis of amphibolite-to-granulite facies zircons: geochronology of the Ivrea zone (Southern Alps). Contributions to Mineralogy and Petrology, 134, 380-404. Wetherill, G.W., 1956a. Discordant uranium-lead ages. Trans. Amer. Geophys. Union, 37, 320-326. Wetherill, G.W., 1963. Discordant uranium-lead ages. Journal of Geophysical Research, 68, 2957-2965. Whitehouse, M.J., Kamber, S.B. and Moorebath, S., 1999. Age significance of U-Th-Pb zircon data from early Archaean rocks of west Greenland- a reassessment based on combined ion-microprobe and imaging studies. Chemical Geology, 160, 201-224 Williams, I.S., 1998. U-Th-Pb Geochronology by Ion Microprobe. In: McKibben, M.A., Shanks III, W.C., and Ridley, W.I. (eds.): Application of microanalytical techniques to understand mineralizing processes. Reviews in Economic Geology. 7, 1-35. Worku, H. and Schandelmeier, H. 1996. Tectonic evolution of the Adola Belt of southern Ethiopia: evidence of a Wilson Cycle process and implications for oblique plate collision. Precambrian Research, 77, 179-210.

  22  

Yu, M.L. and Lang, N.D., 1986. Mechanisms of atomic ion emission during sputtering. Nuclear Instruments and Methods in Physics Research, B14, 403-413.                                                                                      

  23  

Manuscript

  24  

New zircon U-Pb geochronology and Nd isotope systematics of the Abas terrane, Yemen: Implications for Neoproterozoic crust reworking events. Fitsum G. Yeshanew1, 2,*, Victoria Pease1, Martin J. Whitehouse2 and Salah Al-Khirbash3 1- Department of Geological Sciences, Stockholm University, SE-106 91 Stockholm,

Sweden 2- Swedish Museum of Natural History,

Department of Geosciences, Box 50007, SE-104 05 Stockholm, Sweden 3- Sultan Qaboos University, Al Khoudh, Muscat 123, Oman *- Corresponding author: Phone: +46 8 51955102, Fax number: +46 8 51955102, E-mail: Fitsum.Girum@nrm.se Abstract The Abas terrane in Yemen represents Precambrian continental crust of the southeast Arabian Shield. Its Precambrian tectonomagmatic evolution is poorly constrained. New high-spatial-resolution secondary-ion mass spectrometry (SIMS) U-Pb zircon ages, Nd isotopic and geochemical data are reported for granites and granitic gneisses from a traverse across the northern part of the Abas terrane. SIMS U-Pb analyses define two age groups at c. 790−730 Ma and 630−590 Ma, the latter belonging to the post-collisional stage. The oldest sample in the post-collisional group is slightly deformed while younger samples are undeformed; thus penetrative deformation ceased at c. 630 in the Abas terrane. Xenocrystic zircons range in age from c. 1860 to 2340 Ma and, together with εNd(t) values between +0.8 and -11, indicate a significant contribution of evolved continental material in the genesis of the Abas granitoids. The Abas granitoids are high-K calc-alkaline, predominantly metaluminous to weakly peraluminous, granites. The combination of subduction zone chemistry, absence of older rocks and paucity of xenocrysts, suggests either assimilation of i) evolved crust hidden at depth, or ii) subducted sediments derived from older continental crust. Keywords: Precambrian, Arabian Shield, magmatism, U-Pb, zircon, post-collisional magmatism, Nd isotopes

  25  

1. Introduction The Arabian-Nubian Shield (ANS) is a region of Precambrian basement exposed in the Arabian Peninsula and northeast Africa (Fig. 1). The shield covers more than 6 ·106 km2 (Stein and Goldstein, 1996; Kusky et al., 2003) and represents one of the largest tracts of juvenile Neoproterozoic continental crust on Earth. The ANS evolved as the northern sector of the East African Orogen (EAO) (Stern, 1994) and is a collage of juvenile oceanic island arc terrains that amalgamated when East and West Gondwana collided to form the Gondwana supercontinent (Stern, 1994, 2002; Bentor, 1985; Kusky et al., 2003; Meert 2003; Stoeser and Frost 2006). Although the ANS is predominantly comprised of Neoproterozoic juvenile crust (e.g., Stein and Goldstein, 1996; Stern, 1994, 2002), geological (Stoeser et al., 2001), geochronological (Sultan et al., 1990; Kröner et al., 1992; Whitehouse et al., 1998; Ali et al., 2009b; Be’eri-Shlevin et al 2009a) and radiogenic isotopic (Stacey and Hedge, 1984; Windley et al., 1996; Whitehouse et al., 2001b; Hargrove et al., 2006; Stoeser and Frost, 2006) studies have identified pre-Neoproterozoic continental crust within the shield and along its periphery, including in the part of the shield exposed in Yemen. The Precambrian rocks of Yemen lie east of the Saudi Arabian shield, between the ANS to the north and the Mozambique Belt (MB) to the south. The latter is a high-grade, mostly gneissic terrane of Mesoproterozoic to Archean age (Kröner and Stern, 2004). The Precambrian basement of Yemen occurs in the critical transitional area between the ANS and MB, and thus is likely to hold important clues for understanding the timing and amalgamation history within the East African Orogen (EAO) of northern Gondwana (Kröner 1985; Stern, 1994; Whitehouse et al 2001b). A particularly intriguing feature of the Abas terrane in Yemen, arising from the reconnaissance studies of Windley et al. (1996) and Whitehouse et al. (1998), is the appearance of evolved initial Nd isotopic signatures indicating the involvement of pre-Neoproterozoic continental crust in their petrogenesis despite the absence of older zircon cores or known older rocks in the region. This contrasts with the long and polyphase history of the Al Mahfidh terrane to the east and raises the questions: 1) How is such a crustal component introduced into the parental melts (e.g. by intrusion into pre-existing old crust or transport of sediments to the melt emplacement region) ; 2) Why are no old zircons indicative of this crust preserved (e.g. dissolution in high-T melts or simply an absence if the old Nd signature comes from pelagic or fine-grained sediments)? In order to investigate these questions, samples were collected from a traverse across the northernmost exposures of the Abas terrane (Fig. 2), which represent a more diverse range of rock types than seen in earlier studies (Windley et al., 1996; Whitehouse et al., 1998). Samples include Precambrian granitoid gneisses that temporally belong to the arc-forming stage of the ANS, as well as post-collisional granites. We present new secondary ion mass spectrometer U-Pb zircon geochronology and whole rock Sm-Nd isotopes and geochemical data from these

  26  

samples. The results are used to constrain the time of magma genesis and crustal growth in the Abas terrane, evaluate the extent to which older continental material is involved in this crustal growth, and to refine models for crustal evolution of the region. 2. Background geology and sampling The Precambrian basement of Yemen exposes the southeastern part of the Arabian Shield (Fig.1) and contains the oldest known rocks of the shield. The Precambrian basement of Yemen comprises alternating gneissic terranes, the Abas and Al-Mahfid, and late Proterozoic volcanic arcs, Al-Bayda and Al-Mukalla (Windley et al., 1996). These were amalgamated to form an arc-gneiss collage during the Pan-African orogeny (Whitehouse et al., 2001). Gneisses and metamorphosed supracrustal rocks dominate the Abas terrane (Fig.1). The former are mainly homogeneous, hornblende-biotite orthogneisses, some of which show extensive migmatization. A major belt of metamorphosed supracrustal rocks, the Rada Group, occurs within the Abas terrane and it consists of rhyolite, biotite schist, chlorite schist, amphibolite, metatuff, metasandstone, and marble (Windley et al., 1996). Numerous discordant intermediate to mafic dykes intrude these supracrustal rocks (Windley et al., 1996). Whitehouse et al. (1998) reported ages of 754 ± 13 Ma and 765 ± 6 Ma from two gneiss samples. Present-day εNd(0) values from the Abas rocks range from -10.6 to -18.2, with depleted mantle Nd model ages (Nd-TDM) of c. 1.7 to 2.3 Ga, indicating the involvement of older continental crust (Windley et al., 1996). No Archaean rocks have been found in the Abas terrane although the geochronological data is limited. Unlike the Abas terrane, the Al-Mahfid gneiss terrane exhibits a protracted and complicated history (Heikal et al., 2014). Migmatized orthogneisses, which contain several generations of abundant amphibolitic dykes, alternate with belts of two distinct groups of supracrustal rocks consisting of biotite schist, chlorite, amphibolite, rhyolite, meta-tuff, meta-arkose and marble (Windley et al., 1996; Whitehouse et al., 1998). The gray gneisses have been migmatized by several sets of granitic veins, particularly by swarms of folded tourmaline-bearing aplite-pegmatite sheets, which locally comprise up to 90% of the rock mass (Windley et al., 1996). These granitic veins are in turn intruded by two generations of amphibolite dykes. The Al-Mahfid gneiss terrane exhibits distinctly different Sm-Nd isotopic compositions from the Abas terrane. The oldest known gneisses in the Al-Mahfid terrane (2.55 Ga, U-Pb zircon) have Nd-TDM of 2.73 to 3.03 Ga and present-day εNd(0) values ranging from -25.6 to -39.8 (Windley et al., 1996). The Al-Bayda island arc terrane separates the Abas and Al-Mahfid gneiss terranes, and includes greenschist-grade metavolcanic rocks. Ophiolites occur at both its eastern and western margins, and it is intruded by post-tectonic granitic plutons (Windley et al., 1996; Whitehouse et al., 1998). The Al-Bayda arc is also intruded by

  27  

undeformed rhyolite dykes along its eastern margin adjacent to the suggested terrane boundary (Ba-Bttat et al., 1991). Thirteen representative samples were collected from a traverse across the northern part of the Abas terrane for zircon U-Pb geochronology, whole rock Sm-Nd isotopes and geochemical investigations. The samples include 9 granitic gneisses and 4 post-tectonic granites. All rocks are medium to coarse grained and the gneisses show a well-developed foliation. A sample summary with locations and petrography is provided in Table 1. Sample locations are shown on Figure 2. 3. Analytical methods 3.1. Zircon U-Pb dating Zircon crystals were separated using Wilfley water table. A hand magnet was used to remove magnetic minerals. Zircon was handpicked under the binocular microscope and cast into an epoxy puck with the 1065 Ma Geostandard reference zircon 91500 (Wiedenbeck et al., 1995). Zircon grains were polished to reveal potentially older cores. Detailed imaging of the polished zircons was carried out with transmitted light microscopy and cathodoluminescence (CL) imaging using a Hitachi S4300 Scanning Electron Microscope (SEM) equipped with a ETP Semra Robinson-CL-detector at the Swedish Museum of Natural History. U-Th-Pb analyses of zircon was conducted at the Swedish Museum of Natural History using a large geometry CAMECA IMS 1280 Secondary Ion Mass Spectrometer (SIMS). Prior to U-Pb analysis mounts were ultrasonically cleaned and coated with c. 30nm of gold. Analytical procedures closely followed that of Whitehouse et al. (1999) and Whitehouse and Kamber (2005). The intensity of the primary O2

- ion beam was c. 9nA with a spot size of 20 µm. Pre-sputtering with a 25 µm raster for 120 seconds, centering of the secondary ion beam in the 4000 µm field aperture, mass calibration, and optimization of the secondary beam energy distribution were performed automatically for each run. Measured 204Pb counts were used to assess the common Pb present and, when appropriate, corrections were made using the present-day terrestrial Pb compositions of Stacey and Kramers (1975). Data plotting was performed using Isoplot/Ex (Ludwig, 2008). All ages are reported at 2σ uncertainties unless otherwise stated. Decay constants follow the recommendations of Steiger and Jäger (1977). 3.3 Nd isotope analysis Powdered whole rock samples were analyzed on a Thermo Scientific TRITON TIMS using total spiking with a mixed 147Sm/150Nd spike. Total Nd blank for this whole rock preparation was 0.10 ng. Concentrations and ratios were calculated assuming exponential fractionation. Samarium concentrations were determined in multi-

  28  

collector static mode on rhenium double filaments. Samarium ratios were normalized to 149Sm/148Sm = 1.22937. Neodymium was run in static mode on double rhenium filaments. Calculated ratios were normalized to 146Nd/144Nd = 0.7219. The external precision for 143Nd/144Nd as judged from values for La Jolla standard was 11 ppm. Accuracy correction was not necessary since the mean 143Nd/144Nd ratio was 0.51185±46 (n=32), within error of the accepted value of 0.511847. 3.2 Whole-rock geochemistry Major and trace element analysis were determined by X-ray fluorescence (XRF) at Stockholm University. Fresh portions of samples were pulverized using a tungsten carbide or stainless steel swing mill for 2 minutes. Two grams of the milled sample were mixed with a 66% lithium tetraborate: 34% lithium metaborate flux (5g) weighed to a precision of ±0.0002g. The mixture was fused in platinum crucibles for 10 minutes at 1100°C using a Phoenix VFD automated fusion machine to obtain a homogeneous 32 mm glass disk with a lower surface of analytical quality. The concentrations of 10 major elements were determined by comparison of X-ray intensities for each element with calibration lines constructed from the analysis of known concentrations in 24 international standards of known concentrations. Operating conditions are similar to other facilities (e.g. Johnson et al., 1997). USGS standards AGV 2 and RGM 1 were analyzed as unknowns during the analytical session for data quality control. 4. Analytical results 4.1. Zircon morphology and U-Pb results The U-Pb analytical results are presented in the Supplementary file, cathodoluminescence (CL) images in Figure 3, and inverse concordia diagrams (Tera and Wasserburg, 1972) in Figure 4. Post-analytical evaluation of spot locations on CL images permitted recognition of mixed analytical domains – these were omitted from the final age determinations. Inherited cores, also recognized from CL images, were excluded from the final age calculation. Analyses >10% discordant (1σ) and/or those with unusually large errors due to a large common lead correction were excluded from the final age calculations. Inverse concordia diagrams were used to distinguish between ancient and modern Pb-loss. When Pb-loss was indicated, 207-corrected ages were evaluated for younging trends; in this case, successively younger ages were excluded until an acceptable MSWD was obtained. When a large number of xenocrystic cores were present, successively older ages were excluded until an acceptable mean square weighted deviation (MSWD) was obtained. Concordia ages were calculated for each sample.

  29  

Errors in Supplementary file are 1σ, all age uncertainties in the text and concordia plots are quoted at 2σ.The MSWD is for combined concordance and equivalence following the recommendation of Ludwig (2009). For the Th/U ratio, U concentration and U-Th-Pb data mentioned in the text below the reader is referred to Table 1 and the Supplementary file. 4.1.1 Granitic gneiss samples Location 2, sample BO04A. Zircons extracted from this sample range in length from c. 75−300 µm and 60−120 µm in width. The zircon grains are mostly euhedral transparent to light brownish. In CL the zircons show oscillatory zoning (Fig. 3a). Later overgrowth in a few zircons is observed. Uranium concentrations of zircons from this sample range between 59 and 648 ppm with an average value of 238 ppm and Th/U ratios range from 0.11 to 1.05 (average = 0.82). A total of 30 U-Pb analyses were performed. 207-corrected ages exhibit a tendency towards younger ages, which is interpreted as Pb-loss. Excluding the youngest 7 analyses, the remaining 23 analyses yield a concordia age of 760 ± 4 Ma (MSWD = 1.5; Fig. 4a), interpreted as the crystallization age of this gneissic rock. Location 4, sample BO08. Zircons mounted from this sample range in length from c. 80−160 µm and width from c. 50−100 µm. They are mostly euhedral and elongate prisms but some of them are subhedral. CL imaging reveals a number of xenocrysts. Brighter oscillatory-zoned cores with low U surrounded by darker homogeneous rims with high U dominate the zircon population and sometimes the darker rims extend to inner brighter domains. There are some CL dark and high U cores mantled by bright low U rims (Fig. 3b). U concentrations are in the range 691−804 ppm and Th/U ratios range from 0.17 to 2.8. From a total of 14 analyses, two analyses are interpreted to represent a recent Pb-loss and one older grain is clearly an outlier and is therefore interpreted to be a xenocryst (Fig. 4b). The remaining 11 analyses yield a concordia age of 767 ± 5 Ma (MSWD = 1.4; Fig. 4b), interpreted as the crystallization age of the sample. Location 5, sample BO10. Zircons of this sample are acicular, light brown in color, are 120−150 µm in length and 30−70 µm wide. CL images of some grains show oscillatory-zoned cores mantled by a darker zone, which is in turn surrounded by bright thin rims (Fig. 3c). U concentrations are in the range 77−2700 ppm (Table 1; Supplementary file). A total of 37 analyses were performed. Seven xenocrysts are recognized and 7 younger discordant ages were excluded from the age calculation (Fig. 4c). Of the analyzed inherited cores, grains 2c and 16 give a concordant 207Pb/206Pb age of 1865 ± 12 Ma and a discordant minimum 207Pb/206Pb age of 2252 ± 8 Ma respectively. Twenty-one analyses combine to yield a concordia age of 785 ± 5 Ma (MSWD = 1.7; Fig. 4c), interpreted as the crystallization age of the protolith for this gneissic sample.

  30  

Location 6, sample BO13. Zircon grains from this sample are c. 100−250 long and c. 50−80 µm wide. They are elongate light brownish grains and are inclusion free. In CL most of the grains are moderately dark but a few oscillatory brighter grains are also documented (Fig. 3d). The U concentrations are in the range 148−9168 ppm with average high U concentrations of 2922 ppm. From a total of 14 analyses, six analyses are reversely discordant. Reverse discordance in some high U analyses is attributed to U-Pb calibration artifacts that result from accumulated radiation damage (e.g., McLaren et al., 1994), which would be consistent with the high-U concentrations of these analyses. Four normally discordant analyses are interpreted to represent recent Pb-loss. Normal and reversely discordant analyses are excluded from the final age calculation. The remaining four analyses combine to give a concordia age of 786 ± 7 Ma (MSWD = 1.5; Fig. 4d), which is interpreted to represent the crystallization age of the sample. Location 7, sample BO14. Yellowish brown, euhedral and elongate prismatic to equidimensional zircons were separated from this sample. They are 120−340 µm long and 50−100 µm wide. CL imaging revealed various zircon textures in which zircons with dark homogeneous cores mantled by oscillatory-zoned brighter rims dominating the population (Fig. 3e). Uranium concentrations fall in the range from 36 to 1125 ppm. The analyses define a large age range (Fig. 4e). Of 28 analyses, 11 yield a concordia age of 727 ± 6 Ma (MSWD = 1.7; Fig. 4e), taken to be the emplacement age of the protolith of this gneissic sample. Seventeen analyses were excluded from the age calculation and include xenocrysts clearly recognizable in CL (Fig. 3e), as well as younger ages defining a younging trend associated with Pb-loss. Location 8, sample BO15. Euhedral and prismatic zircon grains were separated from this sample. They have light brownish color and inclusions are common especially along the edges of the grains. The length of these zircons is 150−280 µm and width 50−90 µm. In CL they show variable intensity and oscillatory zoning (Fig. 3f). The average U concentration is 510 ppm and Th/U ratios range from 0.29 to 0.97. A total of 20 analyses were performed, 7 indicate Pb-loss, and 13 analyses combine to yield a concordia age of 783 ± 4 Ma (MSWD = 0.96; Fig. 4f), which is considered to be the crystallization age of the protolith of this granitic gneiss. Location 9, sample BO16. Zircons selected from this sample range in length from 150 to 220 µm and in width from 50 to 110 µm. Acicular bright cores in CL (Fig. 3g) are generally too small for analysis. Darker homogeneous rims surround these. Oscillatory zoning is common. Uranium concentrations are range from 148−1035 ppm (with a high average value of 1224 ppm) and Th/U ratios fall in the range 0.02 to 1.07 (Table 1). A total of 13 analyses were performed and six of these combine to yield a concordia age of 764 ± 7 Ma (MSWD = 1.6; Fig. 4g). This is interpreted as the protolith age of the sample. The youngest nearly concordant ages of c. 600 Ma are interpreted as ancient Pb-loss.

  31  

Location 10, sample BO17. Euhedral zircons with length in the range 100−250 µm and width 60−100 µm were selected for U-Pb analysis. In CL they are invariably dark with no discernible zoning (therefore not shown in Fig. 3). Average uranium concentration is high (4500 ppm) (Table 2). Analyses with a younging tendency are interpreted to represent Pb-loss and are excluded from the age calculation. Four analyses combine to yield a concordia age of 786 ± 11 Ma (MSWD = 1.7; Fig. 4h), which is interpreted to represent the age of the sample. Location 11, sample BO18. Euhedral elongated to equidimensional zircons from this sample were picked. In length they range from 150−400 µm and are 50−110 µm wide. They are brownish to transparent. The zircons are oscillatory zoned (Fig. 3h) which is consistent with their magmatic origin (Corfu et al., 2003). U concentrations are in the range from 67 to 848 ppm. Of 21 U-Pb analyses 13 were combined to give a concordia age of 789 ± 4 Ma (MSWD = 1.4; Fig. 4i). The remaining 8 represent recent Pb-loss and were excluded from the age calculation. This age is interpreted to be the crystallization age of the sample. 4.1.2 Post-collisional granites Location 1, sample BO02A. Zircon grains from this undeformed post-collisional granite are mostly euhedral and between ca. 40−300 µm in length. The crystals are transparent but some brownish grains are occasionally observed. CL images (Fig. 3i) show variable response, with darker crystals generally having higher U content than those with bright CL response. CL imaging also reveals oscillatory zoning of the zircons (Fig. 3i), indicating an igneous origin (e.g., Corfu et al., 2003). Dark narrow rims could not be analyzed because of their small size. Excluding one analysis with high U concentration, the average U concentration of the remaining analyses is low (252 ppm). A total of 13 analyses were performed and 7 of them combine to give a concordia age of 591 ± 6 Ma (MSWD = 1.6) (Fig. 4j). This is interpreted as the crystallization age of the sample. Location 3, sample BO05A. Zircons from this granite are elongate and euhedral to subhedral. They are brownish in color and some of them have black inclusions. These zircons have a width of c. 70−120 µm and length of c. 150−400 µm. In general they show oscillatory zoning (Fig. 3j) but some grains are CL dark with patchy bright domains. U concentrations in zircon from this sample are in the range 61−1076 ppm, with Th/U ratios averaging c. 0.8. A total 31 analyses were performed on 19 grains and 14 of them show a tendency towards younger ages and are discordant –these are interpreted to reflect recent Pb-loss. Three other analyses are excluded from the final age calculation because of their large error. The remaining 13 analyses yield a concordia age of 625 ± 4 Ma (MSWD = 1.6; Fig. 4k), interpreted to be the crystallization age of the sample.

  32  

Location 12, sample BO22A. Zircons from this post-collisional granite are elongate and euhedral to subhedral and range in length from 100 to 400 µm and in width from 60 to 130 µm. CL imaging shows oscillatory zoned zircons, but some grains are homogeneous and dark (Fig. 3k). U concentrations are in the range 409 to 1778 ppm, with an average Th/U ratio of 0.29. A total of 14 analyses were performed. Two younger ages are interpreted to represent Pb-loss, while four older analyses are interpreted as xenocrysts – these are excluded from the final age calculation. The remaining eight analyses combine to give a concordia age of 605 ± 4 Ma (MSWD = 0.9; Fig. 4l). This age is interpreted to be the intrusive age of this post-collisional granite. Location 13, sample BO24. Euhedral to subhedral zircon grains were picked from this sample. In CL the zircons are mostly dark (therefore not shown in Fig. 3). A few cores have been documented. The uranium concentrations are in the range 155−13096 ppm and the elevated U content corresponds to darker areas or grains. A total of 12 analyses were carried out. A single near-concordant grain with a 207Pb/206Pb discordant minimum age of 2346 ± 7 Ma and is interpreted as xenocrystic. The remaining analyses define a younging trend consistent with Pb-loss. Two concordant analyses yield a concordia age of 604 ± 7 Ma (MSWD = 0.75; Fig. 4m), which is interpreted as the intrusive age of this granite. 4.2 Whole-rock Nd isotopic data Whole-rock Nd data are given in Table 2 and plotted in Figure 5, 7. The samples show a wide range of concentrations in Nd (6.48– 89.58 ppm) and Sm (1.62– 21.17 ppm) and measured 147Sm/144Nd (0.1026-0.1961) and initial 143Nd/144Nd (0.51193– 0.51246) ratios. Initial epsilon Nd (εNd(t)) values were calculated for 11 samples using the U-Pb ages reported (section 4.1.1). Initial epsilon Nd calculations were not performed for two samples: One sample (BO24) was over-spiked and the elevated natural ratio indicates an unreliable epsilon value. The second sample is undated. Most samples have negative εNd(t) values in the range of –0.7 to –10.9, while one sample (BO14) has a slightly positive εNd(t) value of +0.8. The initial εNd(t) are plotted against crystallization age (Fig. 5). Model ages were calculated using the model of DePaolo (1981), which gives a reasonable estimate of TDM in arc-dominated environments (Stern, 2002; Dickin, 2005). Higher 147Sm/144Nd values result in large errors and therefore unreliable TDM (Stern 2002). Using a conservative screening value of <0.14 (after Stoeser and Frost, 2006), ten samples were acceptable and yield Nd- TDM in the range 1.13 to 1.83 Ga. These model ages predate the average crustal residence age of 0.85 Ga for the Arabian shield, except for the Afif terrane in Saudi Arabia (Agar et al., 1992; Whitehouse et al., 2001; Stern, 2002; Stoeser and Frost, 2006). 4.3 Major and trace element compositions

  33  

Major and trace element analyses are presented in Table 3 and Figure 6. The granitic gneisses and granites have SiO2 > 68 wt.% and therefore granites (Cox et al., 1979; Wilson. 1989). They define high-K calc-alkaline compositions (Fig. 6a). The granites and granitic gneisses range in SiO2 from 70.68 – 76.92 wt.% and K2O is positively correlated with SiO2 (Fig. 6a) and is generally high (except for sample BO-14). Most of the older gneisses represent calc-alkaline and highly fractionated calc-alkaline compositions. The post-collisional granites represent both the calc-alkaline and alkaline compositions (Fig. 6b; after Sylvester, 1989). The alkali saturation index (A/CNK) values are of dominantly metaluminous with a few being marginally peraluminous (Table 3). Most samples represent volcanic arc granites, while some, including two of the post-collisional alkaline granites, represent within plate compositions (Fig. 6c; Pearce, 1984). 4.4 Zircon saturation temperature Zircon saturation temperatures (TZr) were calculated using the thermometer of Watson and Harrison (1983) and the compositional parameter M = [(Na + K + 2 x Ca)/(Al x Si)] (Table 3). For the older granitic gneisses the TZr ranges from 739 to 927°C with an average of 793 ± 69°C (1σ) whereas the post-collisional granites have TZr from 687 to 915°C with an average of 818 ± 86°C (1σ). The Abas granitoids show the features of both ‘‘hot’’ (>800°C) and ‘‘cold’’ (<800°C) granites of Miller et al., 2003. 5. Discussion 5.1. Timing of events in the Abas terrane The U-Pb data from this study yield ages of 790 – 590 Ma. Two high-K calc-alkaline and alkaline magmatic episodes are recognized: an older group at c. 790 – 760 Ma (the gneissic rocks) and a younger group at 625 – 590 Ma (post-collisional granites). These magmatic events are in good agreement with the Neoproterozoic history established for the ANS (Stern, 1994; Johnson and Woldehaimanot, 2003; Stoeser and Frost, 2006). A pervasive magmatic event at 760 Ma, similar in age to our older magmatic group, was reported for both the Abas terrane and the neighboring Al-Mahfid terrane (Whitehouse et al., 1998) suggesting that the two terranes might have been amalgamated by or at that time. There is no magmatism between 760 and 625 Ma. This 135 Ma magmatic hiatus is slightly longer than in the other parts of the ANS, where a 90 Ma magmatic gap occurs between 740 and 650 Ma (Morag et al., 2012; Yasen et al., 2013). In the ANS this magmatic gap may simply reflect a sampling bias and crust of this age might exist concealed beneath the Phanerozoic sediment cover. Many of the older gneiss samples show considerable Pb-loss, similar to the Pb-loss patterns of the two granitic gneisses from the Abas terrane reported by Whitehouse et

  34  

al. (1998). Samples BO04, BO14, BO16 and BO17 (see Figs. 4a, e, g, h) show possible ancient Pb-loss (indicated by arrows in the concordia diagrams) occurring at about the time of post-tectonic granite intrusion. This ancient Pb-loss is interpreted to be due to thermal perturbation related to the emplacement of the post-collisional granites. Discordant ages displaced horizontally from the concordia curve in many of the samples are interpreted as due to recent Pb-loss. In the ANS, post-collisional magmatism commenced at c. 640 Ma (Stern, 1994). Post-collisional high-K calc-alkaline rocks often transition to alkaline magmatism, the latter marking the end of the orogeny (Liégeois et al., 1998). This common transition is evident in the Abas terrane where post-collisional high-K calc-alkaline (BO24) and alkaline (BO22A) samples are contemporaneous at c. 605 Ma (Fig. 6b). The period from 605 Ma onwards is characterized by alkaline magmatism pointing towards the end of orogeny in the region. This feature is also documented in the northern part of the ANS (Ali et al., 2009; Be’eri-Shlevin et al., 2009b; Eyal et al., 2010). A notable difference between the two regions is that in the northern ANS the early stage (c. 630-620 Ma) of post-collisional magmatism was dominated by gabbro to quartz-diorite plutonism and felsic plutonism became volumetrically significant only at c. 610 Ma (Be’eri-Shlevin et al 2009b). However, in the Abas terrane felsic magmatism dominates the entire time span of c. 625-590 Ma. Of the dated late-to post-tectonic granites, the oldest (BO-05A; 625 Ma) is slightly deformed while the three younger granites dated c. 590-605 Ma are not and this may suggest that pervasive regional deformation was ebbing by 625 Ma in the Abas region. 5.2 Crustal assimilation Pre-Neoproterozoic xenocrystic zircons have been recognized in various parts of ANS (Sultan et al., 1990; Agar et al., 1992; Whitehouse et al., 1998; Hargrove et al., 2006; Ali et al., 2009a,b; Ali, B.H. et al., 2009; Be’eri-Shlevin et al., 2009a; Stern et al., 2010a). The common case of inheritance in the ANS is the occurrence of pre-Neoproterozoic inherited zircons in rather juvenile crust (Hargrove et al., 2006; Ali et al., 2009b; Liégeois and Stern, 2010; Stern et al., 2010a). Temperature and magma chemistry are the principal factors that control the survival of inherited zircons (Watson and Harrison, 1983; Miller et al., 2003; Boehnke et al., 2013). The calculated zircon saturation temperatures for the older granitic gneisses (793 ± 69°C) and the post-collisional granites (818 ± 86°C) encompasse the temperature range of both ‘‘cold’’ and ‘‘hot’’ granites of Miller et al. (2003). Given this wide range in TZr, inherited zircons are expected to be present in some of the samples. However, only a few pre-Neoproterozoic zircons are documented in this study despite evidence of ancient continental material (Section 5.3). Shallow level assimilation or mixing characteristically leaves xenoliths and inherited zircons of the contaminant, which is not observed in the Abas granitoids. Therefore, we suggest assimilation or mixing occurred at a greater depth, where temperature is very high leading to the dissolution of inherited zircons. In addition, in the ANS differential preservation of inherited

  35  

zircons is based on the chemistry of the magma and its mode of emplacement. Inherited zircons are preserved more in mafic rocks than in alkali-rich felsic ones and four times as many inherited zircons have been found in volcanic rocks than in their plutonic equivalents (Stern et al., 2010). In sample BO10 two analyses give older ages: a concordant 207Pb/206Pb age of 1865±12 Ma and a discordant 207Pb/206Pb age of 2252±8 Ma ages (Fig. 4c; Table 2). In sample BO24, a single xenocrystic zircon gives a 207Pb/206Pb age of 2346±7 Ma was documented (Fig. 4m; Table 2). Furthermore, Whitehouse et al. (1998) reported a 207Pb/206Pb age of 2606±15 Ma from a single zircon core from the Abas terrane and five additional xenocrystic zircons having 207Pb/206Pb ages from 913 to 1261 Ma. The few inherited zircons both in this study and in Whitehouse et al. (1998) have oscillatory growth zoning, suggesting that they document magmatic processes (Corfu et al., 2003). 5.3. Origin of the Abas granitoids Various tectonic settings have been proposed for the genesis of high-K magmatic rocks and these include: continental arc, oceanic arc, post-collisional arc and within-plate settings (Muller et al., 1992; Muller and Groves, 1997). In addition, the generation of within-plate potassic igneous rocks could be associated with hot-spot activity or extensional tectonics (e.g., Muller and Groves, 1997). The majority of our samples represent volcanic-arc granites and extending into the within plate granite field (Fig. 6c). Previous Nd isotopic studies have demonstrated that the ANS, with the exception of the Afif terrane in Saudi Arabia, is overwhelmingly comprised of juvenile crust, i.e.- of mantle derivation during the Neoproterozoic in an intra-oceanic environment (e.g. Harris et al., 1990; Stern, 1994; Stern, 2002; Hargrove et al., 2006; Stoeser and Frost, 2006; Liegeois and Stern, 2010). Its early development is represented by juvenile calc-alkaline magmatism (Harris et al., 1990; Stern, 1994; Johnson and Woldehaimanot, 2003). However, like previous reconnaissance geochronologic, Nd and Pb isotopic studies from Yemen (Windley et al., 1996; Whitehouse et al., 1998; Whitehouse et al., 2001a), our data highlight the importance of pre-Neoproterozoic crust in the genesis of the Abas terrane. Most of the older (gneissic) samples have strongly enriched εNd(t) values (c. –3 to –11). This extreme variation in initial epsilon Nd reflects isotopic heterogeneity in the genesis of the Abas gneissic rocks and confirms that significant pre-Neoproterozoic crustal material with low εNd(t) was incorporated during granitic magma genesis. Therefore, the early stage of Abas terrane evolution has a distinct history compared to most other parts of the ANS, with the exception of the Afif terrane in Saudi Arabia that exhibits similarly enriched εNd(t) isotopic compositions (Agar et al., 1992; Whitehouse et al., 2001b; Johnson and Woldehaimanot, 2003). No systematic pattern in εNd(t) values is observed between samples showing minor zircon inheritance and those that do not. In comparison with samples from other parts of the shield (Fig. 5), Abas rocks are more

  36  

strongly enriched (negative εNd(t)) than the predominantly juvenile ANS (Stoeser and Frost, 2006; Hargrove et al., 2006). The post-collisional granites have εNd(t) values within 1−2 εNd units of the Chondritic Uniform Reservoir (CHUR). They consistently have a more radiogenic (less evolved) εNd signature and slightly but distinctly younger Nd-TDM compared to the older gneissic rocks. A similar but cyclic temporal variation in bulk rock εNd and single zircon εHf is observed in the eastern Australian Tasmanides (Kemp et al., 2009b; Hawkesworth et al., 2011). This variation appears to mirror tectonic switching where negative εNd and εHf values are associated with arc compression and lithospheric thickening and therefore more crustal input; positive εNd and εNd are related to extensional phases and therefore enhanced mantle input through time (Kemp et al., 2009b; Hawkesworth and Kemp, 2011). We therefore suggest that a switch in tectonic setting at c. 630 Ma, from collisional to post-collisional settings, controls the temporal variation of the observed Nd isotopic composition of the Abas granitoids (Fig. 5). The older arc-related gneisses were formed in a compressional regime where crustal reworking is enhanced and the post-collisional granites were emplaced during a relaxation phase with more juvenile melt input progressively imparting a more radiogenic εNd signature compared to the older gneisses. In addition, the post-collisional granites of the Abas terrane are markedly different from those in other parts of the ANS in both εNd and Nd model ages. Post-collisional plutons elsewhere in the shield have a dominantly juvenile εNd composition and Nd-TDM close to their crystallization ages (Be-eri-Shlevin et al., 2010; Eyal et al., 2010). This reflects the source variability of the post-collisional granitoids across the shield and supports the assertion that post-collisional rocks could be derived from both crustal and mantle sources (Liégeois, 1998; Sylvester, 1998; Bonin, 2004). Depleted mantle model ages can indirectly fingerprint magma sources (Harris et al., 1990). Windley et al. (1996) reported Nd model ages for the Abas terrain in the range 1.7 to 2.3 Ga, which compares well to the Nd model ages of 1.65 to 2.41 Ga from the Afif terrane of southern Saudi Arabia (Agar et al., 1992). A TDM as old as 3 Ga was found in the Abas terrane (Whitehouse et al., 1998). Nd-TDM values from the Abas terrane in this study range from 1.13 to 1.83 Ga (Table 2; Fig. 7). The significant difference between the U-Pb crystallization ages and Nd- TDM indicates crustal reworking was an important process in the formation of the Abas terrane (Fig. 7). These model ages are also substantially older than the average Neoproterozoic Nd model ages for the ANS (Stern, 2002). The older gneissic rocks of the Abas terrane document the involvement of variable (smaller to larger) amounts of an older crustal component, while the post-collisional granites record only a smaller amount of this crustal component. The implication is that assimilation plays an important role in the genesis of Abas terrane magmatism and that, while more pronounced in the older gneissic rocks, this component is also present in the younger post-tectonic granites as well. The negative εNd(t) values, the high-K calc-alkaline affinity, and the ancient model ages combined with the few inherited xenocrystic zircons of the

  37  

Neoproterozoic Abas terrane magmas are best explained by generation in an Andean-type continental margin. Neoproterozoic magmatism assimilated variable amounts of an older Mesoproterozoic crustal component (Fig. 7). This crustal component may be represented by Mesoproterozoic crust at depth in agreement with accepted models for the formation of EAO (e.g., Stern, 1994). Another possibility is mantle source contamination by subducted sediments or sediment underplating. Integrated stable and radiogenic isotopic studies have demonstrated a sediment component in some modern island arc lavas (White and Dupré, 1986; Carpentier et al., 2008; Chauvel et al., 2009; Nebel et al., 2011). Therefore, subduction or underplating of fine-grained pelagic sediments derived from old continental crust can account for the observed enriched Nd isotopic composition and the general lack of any inherited zircons. 6. Summary and conclusions New SIMS zircon U-Pb geochronology and Sm-Nd isotopic and elemental geochemistry constrain crust-forming and deformational events in the Abas terrain of Yemen. An older episode of granitic gneiss genesis occurred at c. 790 to 760 Ma, followed by post-collisional granite genesis at 625 to 590 Ma, with a magmatic gap of c. 135 Ma in the Abas terrane. The slightly deformed state of the oldest (625 Ma) post-collisional granite indicates that pervasive deformation in the region was ebbing by 625 Ma. The two post-collisional calc-alkaline and alkaline samples that are synchronous at 605 Ma and the younger 590 Ma alkaine granite indicates that from 605 Ma onwards alkaline magmatism dominated presumably in an extensional regime. No pre-Neoproterozoic rocks have been identified and only a few older xenocrystic zircons occur. The Nd model ages are in the range 1130 to 1830 Ma and significantly older than the crystallization ages of the samples. Samples in the older age group define a range of strongly enriched εNd(t) values in the range −3 to −11 indicating variable proportions of evolved continental material have been assimilated. This enriched Nd isotopic composition is in sharp contrast to the rest of the ANS, except that of the Afif terrane. The post-collisional granites have εNd(t) values within 2 epsilon units of CHUR but are much more enriched compared to post-collisional rocks from other parts of the shield and thus indicate source variability for post-collisional magmatism. The Sm-Nd data also suggest that the post-collisional granites of the Abas terrane assimilated lesser amount of evolved continental material compared to the older group. This is consistent with an extensional setting for the post-collisional granites where they assimilate only minor amounts of continental material while quickly traversing to the upper crust. 7. References

  38  

Agar, R.A, Stacey, J.S, & Whitehouse, M.J. (1992). Evolution of the southern Afif terrane- a geochronologic study. In: Jiddah, Saudi Arabia. Directorate General of Mineral Resources Open File Report DGMR-OF-10-15. Ali, B.H., Wilde, S.A. & Gabr, M.M.A. 2009. Granitoid evolution in Sinai, Egypt, based on precise SHRIMP U–Pb zircon geochronology. Gondwana Research, 15, 38–48. Ali, K.A., Stern,R.J., Manton, Kimura, J-I., & Khamees, H.A. 2009b. Geochemistry, Nd isotopes and U–Pb SHRIMP zircon dating of Neoproterozoic volcanic rocks from the Central Eastern Desert of Egypt: New insights into the ∼750Ma crust-forming event. Precambrian Research, 171, 1–22. Ali, K.A., Wilde, S.A., Stern R.J., Abdel-Kader, M.M. & Ameen, S.M.M. 2013. Hf isotopic composition of single zircons from Neoproterozoic arc volcanics and post-collision granites, Eastern Desert of Egypt: Implications for crustal growth and recycling in the Arabian-Nubian Shield. Precambrian Research, 239, 42–55. Ba-Bttat, M.A.O. 1991. Geology, petrochemistry and tectonics of the Lowdar-Mudian area, Abyan Province, Yemen Republic. Ph.D. Thesis, Leicester University, Leicester, UK, 178p. Barazangi, M., & Isacks, B.L. 1976. Spatial distribution of earthquakes and subduction of Nazca plate beneath South America. Geology, 4, 686–692. Be'eri-Shlevin, Y., Katzir, Y., Whitehouse, M.J. & Kleinhanns, I.C. 2009a. Contribution of pre Pan-African crust to formation of the Arabian Nubian Shield: New secondary ionization mass spectrometry U-Pb and O studies of zircon. Geology, 37, 899–902. Be'eri-Shlevin, Y., Katzir, Y. & Whitehouse, M. 2009b. Post-collisional tectonomagmatic evolution in the northern Arabian-Nubian Shield: time constraints from ion-probe U-Pb dating of zircon. Journal of the Geological Society, 166, 71–85. Be’eri-Shlevin, Y., Katzir, Y., Blichert-Toft, J., Kleinhanns, I.C. & Whitehouse, M.J. 2010. Nd–Sr–Hf–O isotope provinciality in the northernmost Arabian–Nubian Shield: implications for crustal evolution. Contributions to Mineralogy and Petrology, 160, 181–201. Bentor, Y.K. 1985. The crustal evolution of the Arabo-Nubian Massif with special reference to the Sinai Peninsula. Precambrian Research, 28, 1-74. Boehnke, P., Watson, E.B., Trail, D., Harrison, T.M. & Schmitt, A.K. 2013. Zircon saturation re-revisited. Chemical Geology, 351, 324-334, doi: 10.1016/j.chemgeo.2013.05.028. Bonin, B. 2004. Do coeval mafic and felsic magmas in post-collisional to within-plate regimes necessarily imply two contrasting, mantle and crustal, sources? A review. Lithos, 78, 1–24. Bonin, B. 2007. A-type granites and related rocks: Evolution of a concept, problems and prospects. Lithos, 97, 1–29. Carpentier, M., Chauvel, C. & Mattielli, N. 2008. Pb–Nd isotopic constraints on sedimentary input into the Lesser Antilles arc system. Earth and Planetary Science Letters, 272, 199–211. Chauvel, C., Marini, J.C., Plank, T. & Ludden, J.N. 2009. Hf-Nd input flux in the Izu-Mariana subduction zone and recycling of subducted material in the mantle. Geochemistry, Geophysics, Geosystems, 10, doi: 10.1029/2008GC002101. Corfu, F., Hanchar, J.M., Hoskin, P.W.O. & Kinny, P. 2003. Atlas of zircon textures. In: Hanchar, J.M. & Hoskin, P.W.O. (eds) Zircon. Mineralogical Society of America, Reviews in Mineralogy and Geochemistry, 53, 469–500.

  39  

Cox, K.G., Bell, J.D. & Pankhurst, R.J. 1979. The interpretation of igneous rocks. Allen and Unwin, London, 450p. DePaolo, D.J. 1981. Neodymium isotopes in the Colorado Front Range and crust–mantle evolution in the Proterozoic. Nature, 291, 193–196. Dickin, A.P. 2005. Radiogenic Isotope Geology. Press Syndicate of the University of Cambridge, 492 pp. Eyal, M., Litvinovsky, B., Jahn, B.M., Zanvilevich, A. & Katzir, Y. 2010. Origin and evolution of post-collisional magmatism: Coeval Neoproterozoic calc-alkaline and alkaline suites of the Sinai Peninsula. Chemical Geology, 269, 153–179. Hargrove, U.S., Stern, R.J., Kimura, J.I., Manton, W.I. & Johnson, P.R. 2006. How juvenile is the Arabian–Nubian Shield? Evidence from Nd isotopes and pre-Neoproterozoic inherited zircon in the Bi'r Umq suture zone, Saudi Arabia. Earth and Planetary Science Letters, 252, 308–326. Harris, N.B.W., Gass, I.G., Hawkesworth, C.J., Vail, J.R. & Hoffman, P.F. 1990. A Geochemical Approach to Allochthonous Terranes: A Pan-African Case Study [and Discussion]. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 331, 533–548. Haschke, M.R., Scheuber, E., Günther, A. & Reutter, K.J. 2002. Evolutionary cycles during the Andean orogeny: repeated slab breakoff and flat subduction? Terra nova, 14, 49–55. Heikal, M.T.S., Al-Khirbash, S.A., Hassan, A.M., Al-Kotbah, A.M. & Al-Selwi, K.M. 2013. Lithostratigraphy, deformation history, and tectonic evolution of the basement rocks, Republic of Yemen: an overview. Arabian Journal of Geosciences, 7, 2007-2018. Johnson D.M., Hooper P.R. & Conrey R.M. 1999. XRF analysis of rocks and minerals for major and trace elements on a single low dilution Li-tetraborate fused bead. Advances in X-ray Analysis, 41, 843–867. Johnson, P. R. & Woldehaimanot, B. 2003. Develop- ment of the Arabian – Nubian Shield: Perspectives on accretion and deformation in the northern East African orogen and the assembly of Gondwana. In: Yoshida, M., Windley, B. F. & Dasgupta, S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 289–325. Kemp, A.I.S., Hawkesworth, C.J., Collins, W.J., Gray, C.M., Blevin, P.L. & EIMF. 2009b. Isotopic evidence for rapid continental growth in an extensional accretionary orogen: The Tasmanides, eastern Australia. Earth and Planetary Science Letters, 284, 455–466. Kretz, R. 1983. Symbols for rock-forming minerals. American Mineralogist, 68, 277–279. Kröner, A. 1985. Ophiolites and the evolution of tectonic boundaries in the Late Proterozoic Arabian–Nubian shield of northeast Africa and Arabia. Precambrian Research, 27, 277–300. Kröner, A., Todt, W., Hussein, I.M., Mansour, M. & Rashwan, A.A. 1992. Dating of the late Proterozoic ophiolites in Egypt and Sudan using the single grain zircon evaporation technique. Precambrian Research, 59, 15–32. Kröner, A. & Stern, R.J. 2004. Pan-African Orogeny. Encyclopedia of Geology, 1, 1–12. Kusky, T.M., Abdelsalam, M., Stern, R.J. & Tucker, R.D. 2003. Preface: evolution of the East African and related orogens, and the assembly of Gondwana. Precambrian Research, 123, 81–85.

  40  

Liégeois, J-P. & Stern, R.J.. 2010. Sr–Nd isotopes and geochemistry of granite-gneiss complexes from the Meatiq and Hafafit domes, Eastern Desert, Egypt: No evidence for pre-Neoproterozoic crust. Journal of African Earth Sciences, 57, 31–40. Ludwig, K.R. 2008. Manual for Isoplot 3.7: Berkeley Geochronology Center Special Publication No 4, 77p. McLaren, A.C., FitzGerald, J.D. & Williams, I.S. 1994. The microstructure of zircon and its influence on the age determination from Pb/U isotopic ratios measured by ion microprobe. Geochimica et Cosmochimica Acta, 58, 993–1005. Meert, J.G. 2003. A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics, 362, 1–40. Morag, N., Avigad, D., Gerdes, A. & Harlavan, Y. 2012. 1000–580Ma crustal evolution in the northern Arabian-Nubian Shield revealed by U–Pb–Hf of detrital zircons from late Neoproterozoic sediments (Elat area, Israel). Precambrian Research, 208-211, 197–212. Müller, D, Rock, N.M.S, & Groves, D.I. 1992. Geochemical discrimination between shoshonitic and potassic volcanic rocks in different tectonic settings: a pilot study. Mineralogy and Petrology, 46, 259-289. Müller, D., Groves, D.I. (Eds.), 1997. Potassic Igneous Rocks and Associated Gold- Copper Mineralization. Springer-Verlag, Berlin Heidelberg, 225p. Nebel, O., Vroon, P.Z., Westrenen, W.v., Iizuka, T. & Davies, G.R. 2011. The effect of sediment recycling in subduction zones on the Hf isotope character of new arc crust, Banda arc, Indonesia. Earth and Planetary Science Letters, 303, 240–250. Pearce, J.A., Harris, N.B.W. & Tindle, A.G. 1984. Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. Journal of Petrology, 25, 956–983. Pearce, J.A. 1996. Sources and settings of granitic rocks. Episodes, 19, 121–125. Rickwood, P.C. (1989). Boundary lines within petrologic diagrams which use oxides of major and minor elements. Lithos, 22, 247–263. Stacey, J.S. & Agar, R.A. 1985. U-Pb Isotopic evidence for the accretion of a continental microplate in the Zalm region of the Saudi Arabian Shield. Journal of the Geological Society, 142, 1189–1203. Stacey, J.S. & Hedge, C.E. 1984. Geochronologic and isotopic evidence for early Proterozoic crust in the eastern Arabian Shield. Geology, 12, 310–313. Stacey, J.S. & Kramers, J.D. 1975. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters, 26, 207–221. Steiger R.H, & Jäger, E. 1977. Subcommision on geochronology: convention of the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359-362. Stein, M. & Goldstein, S.L.1996. From plume head to continental lithosphere in the Arabian–Nubian shield. Nature, 382, 773–778. Stern, R.J. & Hedge, C.E. 1985. Geochronologic and isotopic constraints on Late Precambrian crustal evolution in the Eastern Desert of Egypt. American Journal of Science, 285, 97–127. Stern, R.J. 1994. Arc-assembly and continental collision in the Neoproterozoic African orogen: implications for the consolidation of Gondwanaland. Annual Review of Earth and Planetary Sciences, 22, 319–351. Stern, R.J. & Abdelsalam, M.G. 1998. Formation of juvenile continental crust in the Arabian–Nubian shield: evidence from granitic rocks of the Nakasib suture, NE Sudan. Geologische Rundschau, 87, 150–160.

  41  

Stern, R.J. 2002. Crustal evolution in the East African Orogen: a neodymium isotopic perspective. Journal of African Earth Sciences, 34, 109–117. Stern, R.J., Ali, K.A., Liegeois, J.P., Johnson, P.R., Kozdroj, W. & Kattan, F.H. 2010. Distribution and significance of pre-Neoproterozoic zircons in juvenile Neoproterozoic igneous rocks of the Arabian-Nubian Shield. American Journal of Science, 310, 791–811. Stoeser, D.B., Whitehouse, M.J. & Stacey, J.S. 2001. The Khida Terrane – Geology of Paleoproterozoic Rocks in the Muhayil Area, Eastern Arabian Shield, Saudi Arabia. Gondwana Research, 4, 192–194. Stoeser, D.B & Frost, C.D. 2006. Nd, Pb, Sr, and O isotopic characterization of Saudi Arabian Shield terranes. Chemical Geology, 226, 163–188. Sultan, M., Chamberlain, K.R., Bowring, S.A., Arvidson, R.E., Abuzied, H. & El Kaliouby, B. 1990. Geochronologic and isotopic evidence for involvement of pre-Pan-African crust in the Nubian shield, Egypt. Geology, 18, 761–764. Sylvester, P.J. 1989. Post-collisional alkaline granites. The Journal of geology, 97, 261–280. Sylvester, P.J. 1998. Post-collisional strongly peraluminous granites, Lithos, 45, 29–44. Tera, F. & Wasserburg, G.J. 1972. U-Th-Pb systematics in three Apollo 14 basalts and the problem of initial Pb in lunar rocks. Earth and Planetary Science Letters, 14, 281–304. White, W.M. & Dupré, B. 1986. Sediment subduction and magma genesis in the Lesser Antilles: isotopic and trace element constraints. Journal of Geophysical Research, 91, 5927–5941. Whitehouse, M.J, Windley, B.F, Ba-Bttat, M.A.O, Fanning, C.M, & Rex, D.C. (1998). Crustal evolution and terrane correlation in the eastern Arabian Shield, Yemen: geochronological constraints. Journal of the Geological Society, London, 155, 281-295. Whitehouse, M.J., Kamber, B.S. & Moorbath, S. 1999. Age significance of U–Th–Pb zircon data from early Archaean rocks of west Greenland—a reassessment based on combined ion-microprobe and imaging studies. Chemical Geology, 160, 201–224. Whitehouse, M.J. & Kamber, B.S. 2005. Assigning Dates to Thin Gneissic Veins in High-Grade Metamorphic Terranes: A Cautionary Tale from Akilia, Southwest Greenland. Journal of Petrology, 46, 291–318. Whitehouse, M.J., Windley, B.F. & Stoeser, D.B., Ba-Bttat, M.A.O., Haider,A. 2001a. Precambrian basement character of Yemen and correlations with Saudi Arabia and Somalia. Precambrian Research, 105, 357-369. Whitehouse, M.J., Stoeser, D.B. & Stacey, J.S. 2001b. The Khida terrane—geochronological and isotopic evidence for Paleoproterozoic and Archean crust in the eastern Arabian Shield of Saudi Arabia. Gondwana Research, 4, 200–202. Wiedenbeck, M., Allé, P., Corfu, F., Griffin, W., Meier, M., Oberli, F., Quadt, A.V., Rodick, J. & Spiegel, W. 1995. Three natural zircon standards for U–Th–Pb, Lu–Hf, trace element and REE analysis. Geostandards Newsletter, 19, 1–23. Wilson M. 1989. Igneous petrogenesis. Chapman & Hall, London. Windley, B.F., Whitehouse, M.J. & Ba-Bttat, M.A.O. 1996. Early Precambrian gneiss terranes and Pan-African island arcs in Yemen: Crustal accretion of the eastern Arabian Shield. Geology, 24, 131–134. Yaseen, J., Pease, V., Jarrar, G.H. & Whitehouse, M. 2013. U–Pb detrital zircon provenance of the Saramuj Conglomerate, Jordan, and implications for the Neoproterozoic evolution of the Red Sea region. Precambrian Research, 239, 6–23.

  42  

8. Figure captions Figure 1. Simplified geological map of the Arabian-Nubian Shield (after Stern et al., 2010). The study area in Yemen is indicated by the rectangle. Figure 2. Detailed geological map of the Abas terrane showing major geological units. Sample locations are numbered and indicated with an asterisk. Figure 3. Cathodoluminescence images of representative zircon grains for each studies samples except BO17 and BO24, which are CL-dark. Scale bar for all the CL images is 100 micrometers and is shown in Fig. 3k. Figure 4. U-Pb inverse concordia diagrams (Tera and Wasserburg, 1972). Recent and ancient Pb-losse is indicated by arrows. All error ellipses and calculated ages are at 2σ. Data excluded from age calculations are indicated as dotted ellipses. Figure 5. εNd(t) vs. crystallization age including data from the literature. Filled black circles and triangles are the granitic gneisses and post-collisional granites respectively from this study. Data for the Bir’Umq suture of Saudi Arabia (circles) are from Hargrove et al. (2006) and references therein; Eastern and Western arc terranes of Saudi Arabia include AdDawadimi, Sawdah, Khida and Hail terranes (diamond shaped symbols) are from Stoeser and Frost (2006). CHUR represents Chondritic Uniform Reservoir reference line. Depleted mantle evolution curve ifrom DePaolo ( 1981). Figure 6. Major element classification. (a) K2O vs. SiO2 (after Rickwood, 1989) (b) Major element discrimination diagram of the Abas granitoids (> 68 wt% SiO2). The darker shaded area encompasses the calc-alkaline field and the lighter one, marked A + HFCA, designates the alkaline plus highly fractionated calc-alkaline field (after Sylvester, 1998). (c) Tectonic discrimination diagram, Rb vs. Y+Nb (after Pearce et al., 1984). Filled circles represent granitic gneiss samples and the triangle symbols are the post-collisional granites. Figure 7. Nd model age vs. crystallization age (after Harris et al., 1990; Hargrove et al., 2006). Data points that are in the field Nd-TDM < t+300 are interpreted to be juvenile crust; those in the t+300<Nd-TDM<t+900 Ma range are formed by mixing of juvenile magma with evolved ancient continental material; data that plot in the field Nd-TDM > t+900 Ma are considered to be evolved continental crust. Abas granitoids occupy the contaminated field attesting to the involvement of evolved ancient continental material in their genesis. 9. Table captions Table 1. Table containing information about the studied samples including their geographic coordinates, age and simplified mineralogy.

  43  

Table 2. Sm-Nd analyses for samples from the current study were conducted on a Thermo Scientific TRITON TIMS instrument at the Swedish museum of Natural History. A mixed 147Sm/150Nd spike was used to monitor instrumental drift. Sm ratios were normalized to 149Sm/148Sm = 1.22938 and Nd ratios to 0.7219. External precision for 143Nd/144Nd evaluated from values for La Jolla standard was 11 ppm. εNd(t) values were calculated using the present-day (147Sm/144Nd)CHUR = 0.1967 and (143Nd/144Nd)CHUR = 0.512638. TDM values were determined using the present-day (147Sm/144Nd)DM = 0.2137 and (143Nd/144Nd)DM = 0.51315. a-data from Hargrove et al. (2006); b-data from Stoeser and Frost (2006); 1-crystallization age. AdDawa = Ad Dawadimi; nr = not reported. Table 3. All Fe represented as Fe2O3; Mg# = molar ratio (MgO/(MgO + FeOT))*100; A/CNK = molar ratio Al2O3/(CaO+Na2O+K2O); b.d.: below detection; TZr = 12,900 ÷ [2.95 + ln(496,000 ÷ Zrmelt)] after Watson and Harrison (1983); Miller et al. (2003).

  44  

   

   

Table&1Summary&of&sample&location,&U7Th7Pb&zircon&data&and&petrography&of&the&studied&granitoidsSample&ID Location&in& Latitude Longitude Rock&type U&(ppm) U7Pb&zircon Simplified&petrographya

Fig.2 (°N) (°E) age(Ma)BO02A 1 15.13899 45.2558 Granite 31−4526 591 Qtz+Kfs+Pl+Bt+Ttn+Ap+ZrnBO04A 2 15.1038 45.2806 Granitic&gneiss 59−648 760 Qtz+Pl+Kfs+Ep+ZrnBO05A − 16.09625 45.2874 Granite 61−1076 625 Pl+Qtz+Kfs+Bt+ZrnBO08 4 15.06446 45.4105 Granitic&gneiss 16−1035 767 Kfs+Qtz+Pl+Bt+Ttn±Zrn±EpBO10 5 15.0222 45.4321 Granitic&gneiss 77−2699 785 Qtz+Kfs+Bt+Pl+Zrn+ApBO13 6 14.93662 45.5221 Granitic&gneiss 149−9169 786 Qtz+Kfs+Ttn+Pl+ZrnBO14 7 14.9522 45.5348 Granitic&gneiss 36−1124 727 Qtz+Pl+Hbl+Zrn+ApBO15 8 14.9713 45.5380 Granitic&gneiss 177−1893 783 Kfs+Qtz+Bt+Ap±ZrnBO16 9 15.0055 45.5538 Granitic&gneiss 148−4620 764 Kfs+Qtz+Pl+Bt+ZrnBO17 10 15.0287 45.5847 Granitic&gneiss 1664−7247 786 Kfs+Qtz+Bt+Grt+ZrnBO18 11 15.0307 45.59459 Granitic&gneiss 69−848 789 Qtz+Kfs+Bt+Grt+Ep+ZrnBO22A 12 15.0501 45.6670 Granite 409−1526 605 Qtz+Bt+Kfs+Zrc±Apt±EpBO24 − 14.9069 46.25072 Granite 155−13096 605 Qtz+Kfs+Bt+Pl+Zrn

Table&2Nd&isotopic&data&for&the&granitic&gneisses&and&post5collisional&granites&of&the&Abas&terrane&along&with&data&from&the&literature&for&the&various&terranes&in&Saudi&Arabia.&Sample Latitude Longitude Terrane Rock&type U5Pb&age Sm& Nd& 147Sm/144Nd143Nd/144Nd Initial Nd5TDM

(°N) (°E) (Ma) (ppm) (ppm) εNd(t) (Ga)BO02A 15.13899 45.2558 Abas Granite 591 13.36 75.63 0.1068 0.5122541 −0.69 1.13BO04A 15.1038 45.2806 Abas Granitic&gneiss 760 12,86 42.15 0.1845 0,5120159 −10.97 −BO05A 16.09625 45.2874 Abas Granite 625 3.622 19.09 0.1147 0.5122159 −1.68 1.27BO08 15.06446 45.4105 Abas Granitic&gneiss 767 6.089 30.38 0,1211 0.5120105 −4.83 1.7BO10 15.0222 45.4321 Abas Granitic&gneiss 785 8.925 32.56 0.1657 0.5122105 −5.23 −BO14 14.9522 45.5348 Abas Granitic&gneiss 727 20,27 88,14 0.139 0.5124051 0.82 1.3BO15 14.9713 45.5380 Abas Granitic&gneiss 783 5.322 28.82 0.1116 0.5119425 −5.05 1.64BO16 15.0055 45.5538 Abas Granitic&gneiss 764 4.951 26.75 0.1119 0.5120526 −3.14 1.48BO17 15.0287 45.5847 Abas Granitic&gneiss 786 21.17 65.26 0.1961 0.5124632 −3.35 −BO18 15.0307 45.59459 Abas Granitic&gneiss 776 10.83 63.78 0.1026 0.5119319 −4.44 1.52BO22A 15.0501 45.6670 Abas Granite 605 20.07 89.58 0.1354 0.5123817 −0.26 1.29BO24 14.9069 46.25072 Abas Granite 604 1.62 6.48 0.151 0.511995 − −SA015068a 22.4710 39.4892 Jeddah Metatonalite 782 3.98 16.11 0.149 0.512745 7.1 0.70SA035269Ba 23.4154 40.8162 Jeddah Rhayolite 771 10.00 35.80 0.169 0.512799 6.1 0.81SA035270a 23.4770 40.9256 Jeddah Granite 776 2.54 12.00 0.128 0.512642 7.2 0.71SA035271a 23.5031 41.0586 Jeddah Tonalite 785 3.06 13.85 0.134 0.512671 7.2 0.70SA045318a 22.4122 39.8883 Jeddah Diorite 772 2.27 11.12 0.124 0.512594 6.6 0.75SA045322a 22.9033 39.4745 Hijaz Granodiorite 807 2.83 12.47 0.137 0.512680 7.2 0.71SA045373a 22.7768 39.3243 Hijaz Granite 596 6.23 40.09 0.094 0.512503 5.4 0.68SA045375a 22.8331 39.4790 Hijaz Diorite 807 2.72 10.91 0.151 0.512704 6.3 0.81SA045412a 23.4450 41.3943 Jeddah Quartz&diorite 813 3.21 12.97 0.150 0.512708 6.5 0.78184417b 23.8100 44.7700 AdDawadimi nr 600 6.42 28.85 0.135 0.512549 3.02 0.95184418b 23.8100 44.7700 AdDawadimi nr 600 18.42 84.43 0.132 0.512565 3.52 0.90155019b 25.5800 42.7900 Sawdah nr 574 9.38 43.38 0.131 0.512566 3.43 0.88155025b 25.5000 42.7700 Sawdah nr 574 5.98 30.46 0.119 0.512535 3.71 0.83155029b 25.45 42.7500 Sawdah nr 574 4.83 26.92 0.108 0.512501 3.80 0.80155016b 25.7500 42.9600 Sawdah nr 574 17.26 89.96 0.116 0.512509 3.41 0.84175609b 22.2080 44.7480 Khida nr 645 5.07 23.25 0.132 0.512426 1.21 1.14155101b 27.2700 41.3100 Hail nr 566 6.25 29.37 0.129 0.512625 4.67 0.76155116b 27.4100 41.3500 Hail nr 566 12.93 68.01 0.115 0.512586 4.91 0.72155128b 27.5400 41.5300 Hail nr 566 4.46 21.57 0.125 0.512594 4.35 0.78

  45  

       

Table&3Major&and&trace&element&data&of&the&Abas&granitic&gneisses&and&post8collisional&granites.Sample BO02A BO04A BO05A BO08 BO10 BO13 BO14 BO15 BO16 BO17 BO18 BO22A BO24wt#%SiO2 72.9 74.8 73.7 76.6 76.6 76.9 70.7 71.7 72.6 74.5 74.3 74.6 74.8Al2O3 13.4 12.9 14.0 12.5 12.4 12.2 12.6 14.2 14.1 13.7 13.4 12.1 13.8CaO 1.29 0.57 1.40 0.68 0.83 0.87 1.77 2.04 1.53 0.39 1.47 0.80 1.10MgO 0.37 0.04 0.28 0.05 0.18 0.09 0.24 0.58 0.31 0.04 0.32 0.08 0.19MnO 0.06 0.02 0.03 0.03 0.01 0.01 0.13 0.06 0.05 0.07 0.03 0.05 0.02P2O5 0.09 0.02 0.06 0.03 0.04 0.00 0.07 0.10 0.07 0.03 0.06 0.03 0.03Fe2O3 2.52 2.45 1.78 1.29 1.67 1.34 5.62 3.27 2.71 1.14 1.93 3.78 1.11Na2O 3.33 3.81 4.03 3.37 3.19 2.78 4.21 3.40 3.76 3.31 2.66 3.47 3.51K2O 5.44 4.88 4.35 5.14 4.89 5.62 3.93 4.27 4.36 6.72 5.43 4.67 5.29TiO2 0.39 0.18 0.16 0.15 0.09 0.08 0.45 0.36 0.29 0.05 0.23 0.25 0.10Mg# 14.10 1.82 14.65 4.27 11.00 6.74 4.58 16.41 11.17 3.65 15.57 2.23 15.88A/CNK 0.97 1.02 1.01 1.01 1.02 0.98 0.87 1.01 1.03 1.02 1.03 0.98 1.02TZr (°C) 845 879 763 777 758 736 915 798 797 687 791 927 736ppmCe 220 93 41 72 84 23 157 63 59 57 106 167 27La 126 53 20 37 35 16 78 36 31 24 59 94 12Th 36 15 5 10 21 12 15 16 11 25 16 19 24Ni 2 2 4 3 3 3 1 4 4 3 4 4 5V 17 8 14 4 9 4 7 30 9 5 14 4 9Ba 687 2297 1114 777 245 387 1148 1064 1062 255 783 1059 358Cr b.d. 5 b.d. b.d. b.d. b.d. 3 3 b.d. b.d. b.d. b.d. b.d.Cu 1 b.d. 8 b.d. b.d. 1 b.d. b.d. b.d. 3 1 1 b.d.Ga 22 20 18 16 15 17 25 18 17 21 16 24 24Hf 8 9 6 7 6 3 18 1 5 3 5 15 3Nb 39 18 5 11 4 2 33 8 5 6 8 32 19Pb 36 17 15 32 21 35 5 22 21 31 27 32 39Rb 287 131 84 167 143 191 94 135 112 331 121 112 232Sr 163 110 294 57 69 113 108 201 178 49 143 155 102Y 59 40 11 58 39 9 112 19 16 35 23 106 15Zr 331 425 126 142 111 86 782 194 185 46 166 689 88

  46  

Figure 1  

                 

Asir

Jiddah

AfifHijaz

Ha’il

Gebeit

Haya

Midyan

Hulayfah

Khida

Ar Rayn

Ad Dawadimi

Riyadh

Arabian Shield

Sa’anaAbas

Al Bayda

Al MukallaAl Mahfid

Somalia basement

Inda Ad

Jiddah

Red Sea

Gulf of Aden

32°N 26°E

28°N

30°E 34°E 38°E 42°E 46°E

24°N

Juvenile ANS crust

Isotopically delineated Khida terrane

Archean to Mesoproterozoic crust

Archean to early Neoproterozoic crust of the Saharan Metacraton

Terrane boundary

0 250 500 km

Nubian Shield

Fig.2

20°N

16°N

Egypt

Saudi Arabia

Yemen

Sudan

Ethiopia

Asmara

  47  

Figure 2  

             

15°20’ E 15°50’E

Mafic metavolcanics

Late syntectonic or post-tectonic granite

Paragneiss

Amphibolite

Syntectonic diorite

Orthogneiss

N45°10’N

45°02’

Traverse

10 km

4

2

1

5

6 7

8

9

10

11

12

Felsic metavolcanics

Migmatite

Schist

Syntectonic syenogranite

Fault

Cover

Road

  48  

Figure 3    

     

       

  49  

Figure 4    

   

900

9.5 10.5 11.5

0.062

0.066

0.070

recent Pb-loss800

700

600

7.3 7.5 7.7 7.9 8.10.061

0.063

0.065

0.067

0.069BO15

783 ± 4 Ma(MSWD = 0.96, n = 13)

207 Pb

/206 Pb

238U/206Pb

(f )

760

820

800

780

7.0 7.4 7.8 8.2 8.60.058

0.060

720

680

9.0 9.4

0.062

0.064

0.066

0.068

0.070

238U/206Pb

840

800

760

7.3 7.5 7.7 7.9 8.1 8.3 8.50.061

0.063

0.065

0.067

0.069BO08

767 ± 5 Ma(MSWD = 1.4, n = 11)

207 Pb

/206 Pb

(b)

750

790

6 70.062

0.063

8 9 10 11

0.064

0.065

0.066

0.067

7.3 7.5 7.7 7.9 8.10.062

0.063

0.064

0.065

0.066

0.067

0.068BO13

786 ± 7 Ma(MSWD = 1.5, n = 4)

238U/206Pb

207 Pb

/206 Pb

(d)

820

800

780

760

recent Pb-loss820

740

700

780

1 3 50.04

0.06

2400

2000

1600

1200

800

7 9 11

0.08

0.10

0.12

0.14

0.16

7.3 7.5 7.7 7.9 8.1 8.30.061

0.063

0.065

0.067

0.069

0.071BO10(c)

207 Pb

/206 Pb

238U/206Pb

xenocrysts

800

760

785 ± 5 Ma(MSWD = 1.7, n=15)

0.068

0.066

0.064

0.062

0.060

0.0589 10 11

207 Pb

/206 Pb

238U/206Pb7.4 7.6 7.8 8.0 8.2 8.4 8.6

0.066

0.068

0.064

0.062

0.060

760 ± 4 Ma(MSWD = 1.5, n = 23)

BO04A(a)

ancient Pb-loss

600

680

640

800

840

740

780

760

720

6.5 7.5 8.50.057

0.059

880

840

9.5

0.061

0.063

0.065

0.0670.069

0.071

800

760

720 68

064

0

7.8 8.0 8.2 8.4 8.6 8.80.059

0.061

0.063

0.065

0.067

0.069

0.071

727 ± 6 Ma(MSWD = 1.7, n = 11)

BO14

207 Pb

/206 Pb

238U/206Pb

(e)

760

720

8.5

recentPb-loss

  50  

   

     

6.5 7.5 8.5 9.5 10.5 11.50.057

0.059

0.061

0.063

0.065

0.067

0.069

ancient Pb-loss

800

7.2 7.4 7.6 7.8 8.0 8.20.062

0.064

0.066

0.068

786 ± 11 Ma(MSWD = 1.7, n = 4)

BO17

207 Pb

/206 Pb

238U/206Pb

(h)

810

770

700

600

0.054

0.058

7 8

800820

0.066

0.062

0.070

9 10 11 12

9.89.6 10.0 10.2 10.4 10.6

0.057

0.059

0.061

0.063

0.055

605 ± 4 Ma(MSWD = 0.9, n = 8)

BO22A

207 Pb

/206 Pb

238U/206Pb

(l)

740700

630

610

590

60062

0

580

540

700

620

9.99.7

580

16 18

0.055

0.057

0.059

0.061

0.063

0.065

8 10 12 140.053

10.1 10.3 10.5 10.7 10.90.056

0.058

0.060

0.062

0.064

0.066BO02A

238U/206Pb

207 Pb

/206 Pb

(j)

600

500

400

600580

591 ± 6 Ma (MSWD = 1.6; n = 7)

7.0 7.4 7.80.059

0.061 680

720

760

840

800

8.6 9.0 9.4

0.063

0.065

0.067

0.069

7.2 7.4 7.6 7.8 8.0 8.20.062

0.063

0.064

0.065

0.066

0.067

0.068

238U/206Pb

207 Pb

/206 Pb

789 4 Ma(MSWD = 1.4, n=13)

(i) BO18

820

780

0.05

8 10 120.04

720

680

14

0.06

0.07

0.08

9.1 9.3 9.5 9.7 9.9 10.1 10.3 10.50.052

0.056

0.060

0.064

0.068 BO05A

625 ± 4 Ma(MSWD = 1.6, n = 13)

207 Pb

/206 Pb

238U/206Pb

(k)

recent Pb-loss

640

660 620

600

560

520

480

0.054

0.056

0.058

0.060

7.5 8.5

840

800

0.062

0.064

0.066

0.068

6.5 9.5 10.5 11.5

recent Pb-loss

ancient Pb-loss760

72079

0

7.5 7.7 7.9 8.1 8.3 8.50.062

0.063

0.064

0.065

0.067

0.067

0.068BO16

764 ± 7 Ma(MSWD = 1.6, n = 6)

207 Pb

/206 Pb

238U/206Pb

(g)75

0

680

640600

560

790

  51  

   

         

238U/206Pb

207 Pb

/206 Pb

4.00.054

0.056

0.058

0.060

0.062

0.064

0.066

0.068

8.0 12 16 20

604 ± 7 Ma(MSWD = 0.75, n = 2)

BO24(m)

recent Pb-loss

  52  

Figure 5

εNd

(t)

0 500 1000 1500 2000-15

-10

-5

0

5

10

Crystallization age (Ma)

CHUR

Abas terrane

Bi’r Umq suture

Arabian Shield

terranes

DM

Granitic gneissesPost-collisional granites

  53  

Figure 6

0

2

4

6

8

Shoshonite

High-K calc-alkaline

Calc-akaline

Tholeiite

70 71 72 73 74 75 76 77

SiO2 (wt.%)

(a)K2

O(w

t.%)

1 3 5 7 90.8

1.0

1.2

1.4

1.6

1.8

2.0

Calc-alkaline Strongly peraluminous

Alkaline

A + HFCA

100 x ((MgO+FeOT+TiO2)/SiO2)

(Al 2O

3+CaO

)/(F

eOT+N

a 2O+K

2O)

(b)

1000

100

10

1

10 100 1000

Syn-Collision

Granites Within Plate

Granites

Volcanic Arc

GranitesOceanic Ridge

Granites

(c)

Rb (p

pm)

Y+Nb (ppm)

  54  

Figure 7

Nd -TDM (Ma)500

5001000

1000

1500

1500

2000

2000

2500

Juvenile crustEvolved crust

T DM=t

T DM=t+300

T DM=t+900

U–P

b ag

e (M

a)

Contaminated

Granitic gneissesPost-collisional granites