Evidence from Peridotite Xenoliths, Knippa, Texas

Making the Southern Margin of Laurentia themed issue

Composition of the mantle lithosphere beneath south-central Laurentia: Evidence from peridotite xenoliths, Knippa, Texas

Urmidola Raye1,*, Elizabeth Y. Anthony2, Robert J. Stern1, Jun-Ichi Kimura3, Minghua Ren2, Chang Qing3, and Kenichiro Tani3 1Department of Geosciences, University of Texas at Dallas, 800 W. Campbell Road, Richardson, Texas 75080, USA 2Department of Geological Sciences, University of Texas at El Paso, 500 W. University Avenue, El Paso, Texas 79968-0555, USA 3Institute for Research on Earth Evolution (IFREE), Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, 237-0061, Japan

ABSTRACT Calculated mean seismic velocities Vs = GEOLOGIC SETTING 4.44 km/sec and Vp =7.87 km/sec show no Mantle xenoliths in ~83 Ma basanites systematic difference between lherzolites The continental crust of Texas—and much from south-central Texas provide a rare and harzburgites, and agree with present of southern Laurentia—was generated as part of opportunity to examine the lithospheric geophysical measurements of upper mantle the ~1.37 Ga southern granite-rhyolite province mantle beneath southern Laurentia. These velocity beneath Texas. The seismic veloci- (Fig. 1) (Anthony, 2005; Barnes et al., 2002; peridotites represent lithosphere at the ties calculated for these samples will provide Bickford et al., 2000; Reese et al., 2000; Whit- boundary between Mesoproterozoic con- important constraints for interpretation of meyer and Karlstrom, 2007). In the southeast- tinental lithosphere and transitional Gulf EarthScope and other geophysical data sets. ern part of the Llano uplift (Fig. 1), dioritic and of Mexico passive margin. Here we report tonalitic gneiss are inferred to represent a 1.33– petrographic, mineral, and major element INTRODUCTION 1.30 Ga allochthonous magmatic arc (Mosher, data for 29 spinel peridotite xenoliths from 1993; Roback, 1996). This arc is thought to Knippa and use these to characterize the Mantle xenoliths entrained in alkaline mag- have accreted to Laurentia during the Grenville lithospheric mantle beneath south central mas are an important source of information orogeny at ~1.1 Ga due to N-dipping subduction Texas. The xenoliths comprise spinel-bearing about the composition and physical state of beneath Laurentia (Mosher, 1998). Young and lherzolites and harzburgites with coarse, subcontinental mantle. Abundant localities Lee (2009) studied trace-element compositions equigranular textures. Some peridotites con- are found in western North America (Wilshire of Knippa peridotites and found enrichments in tain veins of lizardite. There are no pyrox- et al., 1990). The majority of these samples fluid-mobile trace elements (e.g., La) relative to enites or eclogites. The peridotites contain are located in the Basin and Range Province fluid immobile trace elements (e.g., Nb). They

olivine (Fo89-92), orthopyroxene (En89-92), clino where Mesozoic and Cenozoic tectonic ele- concluded that these trace-element patterns

pyroxene (Wo40-45En45-49Fs3-5), and spinel. ments have overprinted Precambrian litho- were caused by subduction-related fluid meta Spinel Cr# (Cr/(Cr+Al)) distinguishes lherzo spheric formation. Very few sample localities somatism that modified previously melt-depleted lites (Cr# = 0.14–0.21) and harzburgites are found in tectonic provinces that represent continental lithosphere. They suggested that the (Cr# = 0.25–0.36). Mineral and major ele- the southern edge of Laurentia. The Knippa continental lithospheric mantle represented by ment compositions indicate that the lherzo- locality, located in central Texas, is one such these xenoliths may have been the upper plate lites are residues after <10% melt extraction example. The locality lies within the Bal- during Mesoproterozoic subduction. from primitive upper mantle and the harz- cones igneous province (Fig. 1) at the nexus of Following Mesoproterozoic subduction and burgites formed by <15% melt extraction. Mesoproterozoic and transitional lithosphere the Grenville orogeny, the lithosphere of southern Calculated oxygen fugacities indicate equili- of the Gulf coastal plain. The lithosphere was Laurentia was affected by three major tectonic bration of the harzburgites at –1 to +0.61 affected by the Mesoproterozoic accretion and events during Phanerozoic time—two episodes and lherzolites at 0 to –2.6 log units with subsequent Paleozoic tectonism. Character- of rifting and ocean opening separated by con- respect to fayalite-magnetite-quartz (FMQ) ization of these samples therefore informs us tinental collision (Thomas, 2006). The first rift- buffer, similar to lightly metasomatized of a very different set of events and permits ing episode in Early Cambrian time (~530 Ma) spinel peridotites elsewhere. The degree of comparison of subcontinental mantle across a was associated with opening of the Iapetus melt depletion and oxidation of the Knippa broad region. Ocean. During or shortly after this, a continental peridotites are consistent with present data This study reports petrographic descriptions, sliver that ultimately became the Precordillera sets for slightly metasomatized lithospheric major-element whole-rock analyses, and min- of Argentina rifted away (Thomas and Astini, mantle and/or backarc samples rather than eral compositions for these spinel-peridotite 1996). Associated with early Paleozoic ocean forearc settings. Equilibration temperatures xenoliths. We characterize the depletion his- formation, a passive continental margin formed range from 824 to 1058 °C (mean= 916 °C), tory, thermometry, oxidation state, and seismic and persisted throughout most of Paleozoic time. calculated at reference pressure of 2.0 GPa. velocities. These results complement a recent Collision of Laurentia with Gondwana during trace-element study from the same locality by the final stages of Pangea assembly resulted in *[email protected] Young and Lee (2009). the Ouachita orogeny during Pennsylvanian

Geosphere; June 2011; v. 7; no. 3; p. 710–723; doi:10.1130/GES00618.1; 13 figures; 3 tables; 1 supplemental table file.

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Composition of the mantle lithosphere beneath south-central Laurentia

102°W 100° 98° 96° 94° 92° mafic crust of Jurassic age (Mickus et al., 2009); 36°N and (4) thinned crust of southern Laurentia. The N Oklahoma Arkansas underlying lithospheric mantle may have a simi- S. OK larly complex history. aulac Knippa peridotite xenoliths are hosted by ogen basanites of the Balcones Igneous Province, which is characterized by isolated, monogenetic 34° Orogen APM Mesoproterozoic igneous centers that formed numerous small plugs, laccoliths, sills, tuff rings, and lava lakes craton (1.4 Ga) (Barker et al., 1987; Spencer, 1969). Balcones Igneous Province basanites give 40Ar/39Ar ages of ~81.5–83.5 Ma (Griffin et al., 2010). The 32° Texas xenoliths are spinel peridotites. The absence Ouachita of garnet- and plagioclase-peridotite indicates derivation from a depth range of 40 to 85 km, Llano Grenville Front (approx.) i.e., within the upper part of the subcontinental uplift ~1.1 Ga deformation Louisiana lithospheric mantle. l crust 30° P na SAMPLE DESCRIPTION BI itio ns ra Spinel-peridotite xenoliths were collected Knippa T from the Vulcan quarry, situated in Knippa, Mexico Uvalde County, Texas (29.278°N, 99.657°W). 28° 200 km Host basanites are dense, black, and fresh. Splitting time These lavas are aphyric to sparsely phyric in Gulf of Mexico hand specimen, and in thin section have scat- Jurassic oceanic tered phenocrysts (90% olivine and 10% clino lithosphere pyroxene). The black groundmass contains finely 0.5 s 1.0 s 1.5 s dispersed clinopyroxene, plagioclase, olivine, 26° nepheline, titaniferous magnetite, melilite, zeo- Figure 1. Location of Knippa mantle xenolith locality in south-central Texas, showing sim- lite, amphibole, phlogopite, and apatite. Mg# plified crustal provinces. Knippa peridotite xenoliths are hosted by ~83 Ma basanites that ( = Mg/Mg + Fe) of the basanites range from erupted along the lithospheric discontinuity separating Mesoproterozoic lithosphere of the 0.67 to 0.75, have low Ni and Cr abundances, Texas craton and the Jurassic transitional lithosphere of the NW Gulf of Mexico passive and show strong, light rare-earth element margin. The Ouachita orogen approximates the boundary between the North American (LREE) enrichment, with chondrite-normalized craton to the north and west and transitional crust to the east and south. Geophysical stud- (La/Yb)N = 19–24 (Griffin et al., 2010). ies show that orientation and magnitude of splits correlate to crustal provinces (Gao et al., The ultramafic xenoliths are dispersed in 2008). The rapid variation in splitting delay times from Llano uplift to southeastward might the basanite, where they comprise 5%–10% of be either due to different degree of alignment of the crystals’ fast axes or to difference in the total rock volume. They have subspherical thickness of the anisotropic layer (Satsukawa et al., 2010). APM—apparent plate motion; or ellipsoidal forms, with long axes ranging BIP—Balcones Igneous Province. from 1 to 6 cm. They are usually well rounded, although some are polygonal. Host rock-xenolith contacts are sharp in most cases, but some show reaction rims suggesting interaction with time (~350 Ma). Late Triassic uplift (~225 Ma; coastal plain, a result of the Jurassic opening of basanite magma. Dickinson et al., 2010) heralded rifting in what the Gulf of Mexico (Mickus et al., 2009; Sawyer was then the interior of Pangea. This culminated et al., 1991). This lithospheric transition was PETROGRAPHY in Late Jurassic seafloor spreading to form the most recently reactivated by Miocene faulting Gulf of Mexico (165–140 Ma; Bird et al., 2005; (~25–10 Ma) (Galloway et al., 1991). Knippa xenoliths are Type 1 peridotites (Frey Stern et al., 2011). The Gulf of Mexico opening We do not have a clear picture of the nature and Prinz, 1978). They consist of olivine, also established the present passive continental of transitional crust, largely because it is buried orthopyroxene, clinopyroxene, and the alumi- margin of SE Texas (Fig. 1). beneath thick sediments of the Gulf of Mexico nous phase is spinel (Fig. 2). All constituent The Balcones Igneous Province (Fig. 1) lies coastal plain. This crust could be composed phases are homogeneous, without chemical above the transition that separates Mesoprotero of one or more of the following components: zoning. Spinels are brown, elongate, and are zoic crust from the Pennsylvanian Ouachita (1) metamorphosed sediments and crust of the mostly in contact with clinopyroxene. Mineral orogenic lithosphere. The Balcones Igneous Paleozoic passive margin, deformed during the boundaries vary from straight to gently curved, Province trend also approximates the southern Ouachita orogeny (Mickus and Keller, 1992); and commonly form 120° triple junctions, indi- limit of cratonic North America and its bound- (2) fragments of Gondwana, left behind when cating recrystallization under equilibrium con- ary with the southern and western limit of atten- Gondwana separated from Laurentia in Jurassic ditions. Olivines are typically fractured, and uated transitional crust beneath the Texas Gulf time (Rowley and Pindell, 1989); (3) juvenile some grains display kink banding and undulose

Geosphere, June 2011 711

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Raye et al.

(~0.1 mm) than those in lherzolite. Table 1 A B lists modal analyses of Knippa peridotites, 70% of which are lherzolites, with up to 12% clinopyroxene. Modes were also calculated Opx Sp using the method of Lee (2003) (Calculated Ol Mode, Table 1). Bulk chemical compositions Opx Cpx and mineral compositions were used to deter- Ol mine mineral mass proportions. Calcium oxide,

MgO, FeO, Al2O3, and SiO2 were used for the Cpx Triple inversion. In these calculations, homogeneous 1 mm Opx junction 1 mm Opx four-phase mineral compositions (i.e., orthopyroxene, clinopyroxene, olivine, and spinel) were assumed. Accessory phases were not considered. Mineral mass proportions (X ) were determined C D i Ol Lizardite by matrix inversion via X = (CTC)–1 CTB, where Veins Ol X was the column matrix consisting of mineral mass proportions (X ), C was the mineral com- Opx Ol i position matrix, CT was the transpose of C, and B was the bulk composition column matrix. The Lizardite calculated modes were also checked by MINSQ Ol veins program of Herrmann and Berry (2002), which Ol produced similar results. Agreement between 200 µm visual and calculated modes is excellent. 1 mm ANALYTICAL PROCEDURES

E F The xenoliths were trimmed to remove all adhering basanite and cut into two similar Lizardite Opx halves. One part was cut into rectangular slabs veins Ol 0.5 cm thick and sent for thin-section prepa- Sp ration. Visual modes (Table 1) are estimates from thin-section examination. The other half Ol Cpx was crushed into rock chips to be pulverized. Ol Considering xenolith grain size, texture, and 500 µm homogeneity, 5–10 g of the crushed rock pro- 1 mm vided a representative whole-rock sample of each xenolith. Aliquots of the crushed mate- Figure 2. Photomicrograph of Knippa peridotites showing (A) elongated spinel (Sp) grain rial were ground to a fine powder in agate jars surrounded by orthopyroxene (Opx) and clinopyroxene (Cpx); (B) olivine (Ol) grains show- using a shatter box. Mineral analyses were ing triple junctions along with Opx and Cpx; (C) lizardite veins along grain boundaries carried out on 29 polished thin sections using and within cracks of olivine grains; (D) backscattered-electron image of thick lizardite the wavelength-dispersive Cameca SX-100 veins within Ol grains and thin veins within Opx grains; (E) zoom-in image of lizardite veins electron microprobe at the University of within Ol grains; (F) X-ray scan image (including Ca, Fe, Si, and Cr elements) showing Texas at El Paso. The machine was operated grain-size variations between Ol, Opx, Cpx, and Sp, respectively. using an accelerating voltage of 15kV, a beam current of 20 nA, a focused beam diameter of 5–10 µm, and a counting time of 10 s. Natural extinction. Sparse spinel grains are distributed Following the International Union of Geo- standards from the Smithsonian Institute were around margins of large silicate minerals. Some logical Sciences (IUGS) classification for ultra- used for the analysis of all phases. Analytical clinopyroxene and spinel show reaction rims mafic rocks (LeBas and Streckeisen, 1991), results reported in Tables S1–S4 in the Sup- along common grain boundaries. None of the Knippa xenoliths are lherzolites (modal clino- plemental Table File1 generally represent the Knippa xenoliths contain accessory amphibole, pyroxene >5%) and harzburgites (<5% clinopy phlogopite, or silicate glass. Serpentine veins roxene) as shown in Figure 4. Lherzolites are 1Supplemental Table File. Excel file of four (Fig. 3) are dominated by lizardite (Satsukawa more abundant than harzburgites. Lherzolites tables: Table S1: Representative Microprobe Analysis of Olivine Compositions; Table S2: Representative et al., 2010); these are found in some of the are predominantly equigranular (Fig. 2), and Microprobe Analysis of Orthopyroxene Compositions; peridotite xenoliths. These veins follow grain olivine is 1–4 mm, orthopyroxene is 1–2 mm, Table S3: Representative Microprobe Analysis of Clino- boundaries as well as cracks in olivine grains. clinopyroxene is 0.25–1 mm, and spinel is pyroxene Compositions; and Table S4: Representa- Serpentine-rich veins also contain apatite, pent- ~0.15 mm. This variation in mineral grain size tive Microprobe Analysis of Spinel Compositions. If you are viewing the PDF of this paper or reading landite, and pyrrhotite (Fig. 3), but most olivines is also shown by thin-section X-ray scan map it offline, please visit http://dx.doi.org/10.1130/ and orthopyroxenes are unaffected away from (Fig. 2F). Harzburgites are moderately frac- GES00618.S1 or the full-text article on www.gsapubs the reaction boundaries. tured and have clinopyroxenes that are smaller .org to view the supplemental table file.

712 Geosphere, June 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Composition of the mantle lithosphere beneath south-central Laurentia

Lizardite 2

1 Apatite 2

3 1 Lizardite

Figure 3. Backscattered-electron image 2 Pentlandite of accessory phases within lizardite veins. 1 Common accessory phases are apatite and pentlandite. Numbers show repre- 50 µm sentative microprobe analyses listed in the table below.

Lizardite Pentlandite Apatite 1 2 Avg(n=5) Std.dev At% 1 2 3 Avg Std.dev 1 2 3

SiO2 37.09 37.95 37.73 0.56 Fe 26.94 26.88 24.31 26.05 1.50 SiO2 0.46 0.46 0.46

TiO2 0.00 0.00 0.00 0.00 Ni 40.40 38.85 40.35 3 9.87 0.88 TiO2 0.00 0.00 0.08

Al2O3 0.10 0.00 0.14 0.17 S 33.56 34.89 31.78 33.41 1.56 Al2O3 0.02 0.00 0.00

Cr2O3 0.00 0.00 0.06 0.10 Cr2O3 0.00 0.00 0.12 FeO 7.28 7.41 7.62 0.49 FeO 0.09 0.06 0.55 MnO 0.01 0.00 0.06 0.10 MnO 0.00 0.00 0.04 NiO 0.27 0.14 0.24 0.09 MgO 2.66 0.45 0.44 MgO 39.66 38.39 37.82 2.17 CaO 53.48 53.66 54.34

CaO 0.00 0.00 0.15 0.27 Na2O 0.22 0.23 0.16

Na2O 0.01 0.00 0.01 0.01 K2O 0.00 0.00 0.02

K2O 0.00 0.00 0.00 0.00 P2O5 38.95 39.46 38.18 Total 84.42 83.89 83.85 0.59 F 2.67 3.00 2.51 Cl 0.40 0.37 1.26 Total 98.95 97.69 98.16

Figure 4. Modal mineralogy of Knippa Ol peridotites obtained by visual estimates. Following International Union Dunite Peridotite of Geological Sciences classification 10 90 for peridotites (LeBas and Streckeisen, 1991), Knippa xenolith suite consists 20 80 Pyroxenite HarzburgiteWehrlite of lherzolites (Lhz; modal Cpx >5%) and harzburgites (Hzb; <5% Cpx). 30 Opx Cpx 70 Data from Young and Lee (2009) also 40 shown. Ol—olivine; Opx—orthopy- 60 Hzb Lherzolite Serp Hzb roxene; Cpx—clinopyroxene; Serp— 50 50 Lhz serpentine. Serp Lhz Young and Lee(2009) Opx Cpx

Geosphere, June 2011 713

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Raye et al.

average of five or more analyses of each grain Minerals in Knippa peridotite xenoliths are Major-element composition of Knippa xeno- and of several grains from different parts homogeneous with little grain-to-grain chem- liths (Table 1) was determined in two separate of the same sample; these results are sum- ical variation except near contacts with host labs. Nine samples were sent to Activation Lab marized in Table 2. In order to examine the basalt where some reaction is evident. All of oratories Ltd., Lancaster, Ontario, and analyzed equilibrium among mineral phases, attention the compositions we report were measured far by tetraborate fusion–inductively coupled plasma was paid to evaluating phase homogeneity. from these contacts. (ICP) method, while 20 samples were analyzed by

TABLE 1. WHOLE-ROCK COMPOSITION OF KNIPPA PERIDOTITES Rock type Lherzolites Sample Kn1 Kn3 Kn4 Kn6 Kn7 Kn12Kn13 K2F3 RR1 RR2 K2R3 K2GK2R5K2B Kn31 Kn34 Kn36

SiO2 43.70 42.05 42.6543.15 42.7642.40 42.59 41.60 43.3344.12 42.0945.1240.84 42.77 44.01 42.76 44.56 TiO2 0.08 0.03 0.02 0.03 0.07 0.04 0.07 0.03 0.030.05 0.020.020.04 0.040.05 0.05 0.05 Al2O3 3.21 3.63 4.06 3.27 3.24 2.39 2.83 1.49 3.831.91 0.931.991.55 2.532.34 1.98 1.91 FeO 7.48 7.86 7.45 7.61 8.04 7.73 8.21 7.50 7.568.047.185.407.607.667.957.937.22 MnO 0.12 0.12 0.11 0.12 0.12 0.11 0.12 0.12 0.120.120.110.120.12 0.120.13 0.12 0.11 MgO 40.54 42.64 42.2941.85 41.8844.04 42.10 45.47 41.1342.83 45.9144.8244.71 43.17 40.83 42.8642.29 CaO 2.66 1.70 1.53 2.17 1.91 1.81 2.12 1.052.111.340.661.861.322.492.532.041.70

Na2O 0.15 0.110.100.130.160.110.13 0.06 0.10 0.11 0.030.090.09 0.170.160.110.08 K2O 0.02 0.01 0.01 0.02 0.04 0.02 0.04 0.01 0.040.02 0.010.030.01 0.030.01 0.02 0.01 P2O5 0.03 0.02 0.02 0.02 0.02 0.02 0.04 0.01 0.010.02 0.020.010.02 0.030.07 0.02 0.01 LOI 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.69 0.000.001.970.753.580.870.000.000.00 Total 97.98 98.16 98.2498.35 98.2498.67 98.26 100.0298.2698.56 98.93100.2299.87 99.8898.07 97.8997.93 VISUAL MODE ESTIMATE Rock type Lherzolites Sample Kn1 Kn3 Kn4 Kn6 Kn7 Kn12Kn13 K2F3 RR1 RR2 K2R3 K2GK2R5K2B Kn31 Kn34 Kn36 Ol 63 69 60 65 69 76 70 68 65 70 76 75 80 70 66 65 69 Opx 25 20 28 23 20 13 20 21 23 22 15 12 12 22 22 25 20 Cpx 10 891099810 1077107610 89 Sp 2 3322221212312222 CALCULATED MODE ESTIMATE Rock type Lherzolites Sample Kn1 Kn3 Kn4 Kn6 Kn7 Kn12Kn13 K2F3 RR1RR2 K2R3 K2GK2R5K2B Kn31 Kn34 Kn36 Ol 61 67 63 64 66 76 68 68 61 65 72 76 83 68 ___ Opx 25 21 26 24 23 15 20 17 26 28 18 13 10 23 ___ Cpx 12 871097912 10 6910 68___ Sp 24432133411211___ (continued)

TABLE 1. WHOLE-ROCK COMPOSITION OF KNIPPA PERIDOTITES (continued) Rock type Serpentine lherzolites HarzburgitesSerpentine harzburgites Sample Kn43 Kn44 Kn2 K2DKn9 Kn28 Kn29 Kn30 K2CK2F6K2R2

SiO2 43.57 43.5242.60 41.50 42.5444.27 44.5943.07 43.5042.76 43.12 TiO2 0.04 0.04 0.04 0.03 0.070.050.010.040.020.030.04 Al2O3 2.15 2.45 2.45 1.12 1.271.041.651.101.871.123.23 FeO 7.81 7.83 7.51 7.37 8.59 6.87 7.20 7.47 7.48 6.69 7.69 MnO 0.12 0.12 0.11 0.11 0.13 0.10 0.12 0.11 0.13 0.12 0.12 MgO 42.36 41.2144.24 44.76 44.79 44.9842.58 45.4742.68 46.1341.73 CaO 1.98 2.56 1.12 0.69 0.64 0.46 1.27 0.72 1.92 0.75 2.27

Na2O 0.10 0.16 0.07 0.04 0.050.040.080.060.080.070.14 K2O 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.17 0.02 P2O5 0.02 0.04 0.05 0.01 0.030.020.030.020.010.020.03 LOI 0.00 0.00 0.00 2.34 0.00 0.00 0.00 0.00 2.42 2.03 0.00 Total 98.16 97.9498.21 97.98 98.13 97.8397.53 98.06100.1299.89 98.38 VISUAL MODE ESTIMATE Rock type Serpentine lherzolites HarzburgitesSerpentine harzburgites Sample Kn43 Kn44 Kn2 K2DKn9 Kn28 Kn29 Kn30 K2CK2F6 K2R2 Ol 64 63 73 80 82 80 78 78 80 80 77 Opx 24 23 24 15 15 15 16 18 16 15 15 Cpx 10 12 2324433 46 Sp 22 1211211 12 CALCULATED MODE ESTIMATE Rock type Serpentine lherzolites HarzburgitesSerpentine harzburgites Sample Kn43 Kn44 Kn2 K2DKn9 Kn28 Kn29 Kn30 K2CK2F6 K2R2 Ol __ 70 81 ____817981 Opx __ 22 15 ____16 17 14 Cpx __ 53____ 235 Sp __ 41____ 111 Note: Cpx—clinopyroxene; LOI—loss on ignition; OL—olivine; Opx—orthopyroxene; Sp—spinel.

714 Geosphere, June 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Composition of the mantle lithosphere beneath south-central Laurentia ) d v. X-ray fluorescence (XRF) spectrometry at Insti- v. 0 2 1 0 3 .0 .2 .0 .0 tute for Research on Earth Evolution (IFREE), .3 0.01 0.35 0.05 0.00 0.00 0.03 0.17 0.13 0.54 0.39 0.00 0.00 0.00 0.05 0.02 0.00 0.00 0.15 0.04 St. de St. de continue Japan Agency for Marine-Earth Science and h ( ig

Technology (JAMSTEC), using high-dilution High 00 00 40 00 90 .0 .0 .0 0.03 1.09 0.29 0.00 0.00 0.61 8.52 5.35 0.96 0.00 0.00 0.00 0.33 0.01 0.00 0.01 (10:1) fused glass beads. Prior to the fusion, the 2.64 92 91 35.0 56.0 K2F5 K2F5 100.07 whole-rock powders were weighed and ignited 100.90 in ceramic crucibles for ≥4 h at 900°C. The v. ignited powders were then weighed together with v. 95 10 00 00 64 .1 .0 .0 .0 lithium tetraborate and fused in a platinum-gold .1 0.00 0.05 0.02 0.01 0.03 0.25 0.06 0.24 0.14 0.22 0.00 0.01 0.01 0.01 0.01 0.00 0.00 0.12 0.08 erpentine harzburgite St. de St. de alloy crucible at 1180 °C to produce glass beads, Serpentine harzburgite wH ow

which were analyzed using ultramafic reference Lo 00 40 00 30 00 11 rock samples at external standards. The XRF .0 .0 0.00 0.33 0. 0.06 0.56 9.07 5.68 3.79 8.78 9.68 5.10 0.0 0.00 0.01 0.01 0.38 0.04 0.00 0.00 3.13 91 91 50.0 40.1 K2R2 technique was designed to make high accuracy K2R2 analyses of major elements in peridotites such as Si, Mg, Fe, Al, and Ca. v. v. 63 09 00 00 49 15 .0 .4 .0 .0 .1 .2 0.01 0.24 0.07 0.00 0.00 0.01 0.12 0.09 0.00 0.00 0.00 0.00 0.02 0.02 0.00 0.01 0.03 0.14 t. de t. de

RESULTS hL ig S High 00 50 20 20 00 2S 2S .00 Mineral Chemistry .0 60 .0 0.03 1.42 0.21 0.00 0.00 0.62 8.34 5.19 0.00 0.00 0.00 0.35 0.02 0.00 0.02 2.66 1.17 92 92 Kn Kn 34.9 00.1 01.2 56.3 eS Major phases in Knippa peridotites are com-

positionally homogeneous and do not show arzburgit v. v. COMPOSITION Harzburgite 75 01 21 00 10 00 34 L .2 .5 .4 .0 .0 .0 significant grain-to-grain chemical variations. .2 0.06 0.00 0.00 0.03 0.04 0.35 0.08 0.27 0.00 0.01 0.02 0.03 0.05 0.00 0.00 0.28 0.42 t. de t. de Olivine compositional range is Fo89 to Fo92 with Lo wH one exception of Fo88 (Table 2; Fig. 5). The Fo Low 70 90 20 00 30 00 90 CS DS content of Knippa olivines falls in the range .0 .0 0.06 0.00 0.00 0.13 0.48 9.29 5.89 4.32 0.0 0.00 0.00 0. 11 0.34 0.04 0.00 0.00 2.95 5.89 91 91 49.6 00.1 99.7 40.6 K2 K2 expected for Proterozoic (Fo92 to Fo93) and espe-

cially Phanerozoic (Fo91to Fo92) lithospheric YSIS OF MINERA mantle (Gaul et al., 2000). Harzburgite olivines v. v. ANAL 13 91 20 00 are slightly more magnesian (Fo ) than lher- 45 .2 .3 .0 .0 .16 90–92 .2 0.03 0.21 0.04 0.00 0.00 0.03 0.12 0.15 0.33 0.00 0.00 0.00 0.01 0.05 0.02 0.00 0.00 0.13 St. de zolite olivines (Fo88–91). St. de Orthopyroxenes are enstatite with a compo- High High 70 30 30 00 80 70 sition of Wo En Fs (Table 2; Fig. 5). eH .0 .0 1–2 88–91 8–11 1 0.04 0.60 0.32 0.00 0.00 0.51 9.17 5.90 0.00 0.00 0.00 0.00 0.28 0.02 0.00 0.00 2.93 2+ 91 34.5 99.8 40.7 55.5 K2F1 Orthopyroxene Mg# ( = 100*Mg/[Mg + Fe ]) K2F1 100.93

correlates with coexisting olivine Mg# (Fig. 6A). IVE MICROPROBE AT Lherzolite orthopyroxenes have higher Al2O3 v. v. 75 (3%–4%) and lower Cr O (0.1%–0.4%) con- 20 00 .3 2 3 .0 .0 0.01 0.00 0.00 0.00 0.04 0.20 0.18 0.16 0.35 0.81 0.05 0.00 0.01 0.04 0.03 0.03 0.00 0.01 0.14 0.28 0.40 Serpentine lherzolite Serpentine lherzolit St. de tent than harzburgite orthopyroxenes (2%–3% St. de

Al O ; 0.2%–0.4% Cr O ) (Fig. 6B). ow

2 3 2 3 Low 80 7 90 00 Clinopyroxenes are apple-green diopsides, 9 .0 .0 19 0.05 0.00 0.00 0.00 0.71 9.37 5.78 3.17 9.92 9.89 0.0 0.00 0.00 0.15 3.72 0.42 0.04 0.00 0.02 5.86 90 49.2 41.0 K2F2 with a composition of Wo40–45En45–49Fs3–5 (Table 2; K2F2

Fig. 5) with Mg# between 90 and 93. Al2O3 and ABLE 2. REPRESENT T Cr2O3 in lherzolite clinopyroxene range from v. v. 93 69 29 10 00 2% to 6% and 0.5% to 1.1%, respectively, while 55 .0 .1 .3 .0 .0 .18 .1 0.01 0.20 0.04 0.02 0.03 0.02 0.23 0.07 0.00 0.00 0.01 0.01 0.07 0.02 0.03 0.00 those in harzburgite clinopyroxene range from 0.01 St. de St. de hL h

3% to 5% and 0.67% to 1.08%. ig Hig 5 30 80 30 50 30 40 Spinel compositions are summarized in 0 00 .0 .0 .0 19 0.01 0.30 0.06 0.07 0.46 8.62 5.68 0.00 0.00 0.01 0.03 3.26 0.35 0.03 0.01 91 K2B K2B 50.4 34.3 00.7 00.6 41.1 Table 2. Based on Cr# ( = Cr/Cr + Al), there are 56.5 two distinct spinel compositions (Fig. 5). Harz- e burgite spinels have higher Cr# (mean = 0.29) v. v. Lherzolit Lherzolite 01 71 20 than lherzolites (mean = 0.17). 00 00 .4 .0 .0 .0 .0 0.03 0.25 0.00 0.00 0.00 0.12 0.47 0.36 0.52 0.00 0.00 0.00 0.00 0.16 0.03 0.00 0.00 0.30 Serpentine is lizardite, and thin serpen- 0.24 St. de St. de wH tine veins contain small grains of pentlandite w Lo Lo 5 0 0 90 40 90 ([Fe,Ni]9S8), apatite (Ca5[PO4]3), and trace 00 00 .0 .0 .0 .0 .0 .0 99 7.91 0.00 0.00 2.86 0.74 7.03 0.00 0.00 0.00 0.00 3.77 0.12 0.00 88 00.5 00.7 12.02 40.75 55.97 K2 G amounts of pyrrhotite (Fe1 –xS). Pentlandite con- K2G tains 29.88–40.39 atomic% Ni, 24.31–35.18 atomic% Fe, and 31.78–34.89 atomic% S. The

contents of CaO and P2O5 in apatite are 33–34 3 3 3 3 2 2 O O0 2 2 O and 38–39 wt%, respectively, and are fluorapatite O 2 2 O O ta l1 ta l1 2 2 O0 O0 2 2 O O 2 2 Na Ca O0 Orthopyroxene Si O Mg O4 Sample Ni O0 Ni O0 Na Rock type Mn O Mn O Ca O Mg #8 Fo Olivine SiO Al Cr FeO Al FeO Mg O3 K To To K Ti Cr Ti Sample (Ca5PO4)3F) with 2.5–3 wt% fluorine (Fig. 3). Rock type

Geosphere, June 2011 715

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Raye et al. v. v. Whole-rock Geochemistry 8 6 0 0 0 4 7 0 8 4 4 0 2 0 1 11 .1 .2 .0 .0 .0 .1 .0 .0 .0 .0 .0 0. 0.13 0.0 0.02 0.00 0.00 0.2 0.07 0.59 0.02 0.3 0.38 0.0 St. de St. de h h Whole-rock major-element composition of Hig Hig Knippa xenoliths (Table 1) agree with those 50 6 7 00 10 00 40 00 40 70 60 20 60 .0 .0 .0 .0 .9 .6 .1 .2 2.43 0.01 0.3 0.00 0.00 6.92 0.01 2.20 6.06 0.71 0.0 calculated by Young and Lee (2009) from 93 17.44 99.5 99.48 15.1 53.1 30.88 K2F6 K2F5 mineral modes and compositions. The whole-

rock contents of Al2O3 and CaO in Knippa v. v. peridotite xenoliths anticorrelate with MgO 41 00 00 00 10 72 60 00 00 73 12 1 0 .1 .2 .0 .0 .0 .0 .0 .0 .0 .0 .0 content (Fig. 7). For comparison, the compo- 0.1 0.14 0.02 0.01 0.00 0.00 0.17 0.04 0.06 0.24 0.02 0.04 erpentine harzburgite St. de St. de Serpentine harzburgite sition of primitive (undepleted) upper mantle ow

Low (Hart and Zindler, 1986; Jagoutz et al., 1979; 3 90 00 00 10 00 00 90 00 50 00 80 .0 .0 .0 McDonough and Sun, 1995), forearc perido- 0.05 8.94 0.27 0.06 0.17 0.00 2.17 2.50 0.82 0.0 3.6 0.8 0.1 0.2 92 21.99 16.81 98.88 12.9 17.9 23.9 43.5 K2R2 K2R2 tite (Parkinson and Pearce, 1998; Parkinson and Arculus, 1999; Pearce et al., 2000) and backarc basin (BAB) peridotite (Michibayashi v. v. 00 00 10 10 19 15 11 ) .0 .0 .0 et al., 2009) are also shown. The Knippa 0. 0.1 0. 11 0.1 0.08 0.01 0.00 0.00 0.00 0.41 0.10 0.17 0.03 0.02 0.09 0.07 0.02 0.30 0.04 0.39 0.0 t. de t. de d

hL peridotites have high whole-rock MgO and High Hig low CaO and Al2O3 compared to fertile model– 00 10 continue 2S 2S ( primitive mantle compositions. Negative cor- 1.33 0.00 0.32 0.00 0.00 0.00 9.99 2.29 1.10 0.0 0.0 10 0.0 3.25 8.34 0.16 3.37 0.88 0.17 0.19 0.02 0.03 93 Kn Kn 16.95 99.62 12.76 18.41 eS relations of Al and Ca oxides with MgO are consistent with Knippa peridotites represent- arzburgite arzburgit ing the residue of previous melt depletion. v. v. 22 59 1 05 12 24 .02 .02 .00 .01 .1 .06 .05 .00 .3 .2 .05 .29 .13 .04 .00 .2 .01 Both lherzolites and harzburgites follow melt- 0.1 0.19 0.26 0.2 0.08 0.01 0.00 t. de t. de

COMPOSITIONS depletion trends. L Low Low 20 90 00 10 30 30 80 00 90 20 90 70 40 40 00 10 80 DS CS .25 .0 .0 .01 .00 Seismic Velocity 0.0 0.0 2.8 1.1 0.0 0.4 4.5 0.9 0.2 0.0 0.0 91 21.72 15.76 99.60 99.4 15.86 17.70 51.9 21.5 44.0 K2 K2 Seismic velocities were calculated at standard temperature and pressure (STP) (25 °C v. v. YSIS OF MINERA 00 10 1 00 10

11 and 1 atm) using the method described by Lee .00 .0 .0 .28 .49 .54 0. 0.19 0.23 0.02 0.0 0.0 0.28 0.35 0.09 0.6 0.06 0.00 0.01 0.29 0.12 0.03 0.00 0.01 St. de St. de (2003) (Table 3). Vp (Primary wave velocities) ANAL for Knippa xenoliths range from 8.24 to 8.31 Hig h High 10 00 0 00 20 90 60 00 eH eH km/sec and Vs (Secondary wave velocities) .01 .2 .0 .86 .01 .09 .65 .68 .27 .00 .06 0.00 0.01 2.40 1.42 0.0 0.0 93 22.34 16.76 99.27 99.1 19.06 52.4 18.9 49.3 K2F2 K2F1 range from 4.77 to 4.84 km/sec. There is no systematic difference in seismic velocities between Knippa lherzolites and harzburgites. v. v. IVE MICROPROBE 00 00 0 01 20 00 50 93 30 10 20 10 The velocities calculated by the method of .0 .0 .0 .5 .0 .51 .51 .2 .2 .2 .0 .0 .36 .0 AT erpentine lherzolit erpentine lherzolit 0.91 0.43 0.00 0.43 0.54 0.26 0.3 0.22 0.01 0.00 St. de St. de Lee (2003) correspond to temperatures that are lower than those in the upper mantle. For Low Low 00 00 00 8 3 60 00 70 00 70 20 40 60 20 50 90

90 this reason, seismic velocities were also cal- .96 .0 .01 .00 0.0 0.0 0.00 2.7 1.1 0.0 0.6 5.7 1.1 0.0 0.3 0.0 .1 91 11 20.18 15.92 99.59 99.1 19.69 52.0 17.3 49.7 K2F1 K2F2 culated using the method of Hacker and Abers (2004) and mineral equilibration temperatures between 900 and 1060 °C. Table 3 shows that v. v. ABLE 2. REPRESENT 30 9 10 10

T Vp range from 7.80 to 7.97 km/s and Vs range 0 .43 .32 .02 .0 .24 .0 .33 .09 .0 .0 .01 .03 0.01 0.01 0.49 0.27 0.43 0.39 0.23 0.36 0.13 0.00 0.38 from 4.34 to 4.52 km/s. No systematic differ- St. de St. de h ences exist between the seismic velocities of High Hig 90 50 20 40 1 20 50 70 4 60 30 10 10 10 30 lherzolites and harzburgites calculated by this .0 .2 .1 .2 .4 .0 .0 .0 .2 0.03 0.03 9.49 0.78 0.39 2.97 0.80 0.07 9.66 93 23.0 16.5 99.4 99.6 19.1 52.4 method. Kn13 K2F4 eS eS

DISCUSSION v. v. Lherzolit Lherzolit 10 20 3 52 20 00 10 00 72 30 34 00 .0 .0 .1 .0 .0 .0 .0 .0 .0 .0 .4 0.13 0.15 0.00 0.01 0.39 0.64 0.07 0.20 0.01 0.17 0.02 0.00 The following section discusses the implica- St. de St. de w w tions of the compositional data reported above. Lo Lo 70 20 50 9 90 2 60 90 20 10 00 00 We focus on the equilibration temperatures of 0.0 0.0 0.0 0.0 0.0 2.8 0.8 0.0 0.0 0.0 0.66 5.91 0.34 0.17 5.0 80 0.3 0 90 21.87 15.59 99.23 10.34 20.83 50.50 64.4 30 Knippa xenoliths, the oxidation state of the Kn12 Kn12 101.2 mantle beneath Knippa, and estimates of partial melting and depletion of the peridotites. Finally, we compare the calculated seismic velocities 3 3 3 3 2 2 O O 2 2

O O to geophysical seismic-velocity models for this 2 2 O O ta l 2 2 O O 2 2 O O 2 2 otal Na CaO Spinel SiO Sample MgO Rock type NiO Na Mg# Cr# MnO MnO CaO Clinopyroxene SiO K To K T Ti Al FeO Al Cr FeO Mg O Cr Ti Sample NiO Rock type part of southern Laurentia.

716 Geosphere, June 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Composition of the mantle lithosphere beneath south-central Laurentia

Di 50 Knippa xenoliths have equilibrium tempera- Hzb tures of 824–1058 °C for T/BKN, except for 60 Serp Hzb Lhz two samples with temperatures of 760 and 70 Serp Lhz 1150 °C (Table 3). The T/BKN and T/Wells geothermometers give similar results that 2 80 closely correlate, with r = 0.917 (Fig. 8), but with somewhat higher minimum temperature of 90 890 °C. These temperatures denote thinner and warmer lithosphere than typical cratonic litho- 100 sphere but cooler than Phanerozoic lithospheric En 100 90 80 70 60 mantle as discussed by Young and Lee (2009). No significant temperature differences exist between Knippa lherzolites and harzburgites. 92 91 90 89 88 87 Fo in olivine These temperature estimates pertain to the Late Cretaceous time of xenolith entrainment. The thermal structure of the upper mantle may 40 30 20 10 0 have been disturbed by Balcones Province igne- Cr# in spinel ous activity, in which case the temperature esti- Figure 5. Composition of primary minerals in Knippa perido mates may be higher than conditions that exist tites. Olivine compositional range is generally restricted to today. Alternatively, the lithospheric thermal structure may not have changed significantly (Fo89–92). Orthopyroxenes are enstatite (En) with a composition of Wo En Fs . Clinopyroxenes are diopside (Di) with a composi- since that time, due to the brevity of the Bal- 1–2 88–91 8–11 cones Igneous Province event and the thermal tion of Wo40–45En45–49Fs3–5. Spinel in harzburgites (Hzb) have higher Cr# (100*Cr/[Cr + Al]) ranging from 25 to 36 than those in lherzolites inertia of the lithosphere. (Lhz) ranging from 15 to 21. Serp—serpentine. Oxygen Fugacity

The oxidation state of the upper mantle, Temperature peratures were calculated assuming a pressure which can be calculated from the equilib- of 2.0 GPa. This pressure was chosen because rium reaction involving coexisting olivine, Mantle temperatures can be estimated it falls within the spinel-peridotite stability orthopyroxene, and spinel, has been exten- using thermometric techniques on equilibrated field and is commonly used for spinel perido- sively applied in spinel peridotites (Ballhaus samples. Equilibrium between mineral phases tites, facilitating comparison with other studies. et al., 1991; Mattioli and Wood, 1988; Nell and is essential for such estimates. Attainment of equilibrium among mineral phases of the Knippa xenoliths and absence of disturbance 94 during serpentinization and after entrainment in basanite melt are inferred from the correlation Figure 6. (A) Mg# ( = 100*Mg/ 92 between olivine and orthopyroxene Mg# with [Mg + Fe2+]) of orthopyroxene a nearly constant slope (Fig. 6A). In addition, (Opx) and Mg# of olivine (Ol) 90 as described in the sections on petrography and showing a positive correlation mineral chemistry, homogeneity among con- indicating equilibrium between Mg# Ol 88 Hzb stituent phases is also evident from absence of the mineral phases. Forearc Serp Hzb chemical zoning and grain-to-grain chemical peridotite data used for com- Lhz variations as well as by the recognition of typi- parison are from Parkinson 86 Serp Lhz Young and Lee (2009) cal equilibrium textures in thin section. and Pearce (1998), Parkinson Forearc peridotite BAB peridotite Several methods have been proposed for esti and Arculus (1999), and Pearce 84 A mating the equilibrium temperatures of spinel- et al. (2000), and backarc basin peridotite mineral assemblages. Four of the peridotites (BAB) are from 4 most commonly used geothermometers (with Arai and Ishimaru (2008) and

abbreviations in parentheses) are: (1) the two- Ohara et al. (2002). (B) Al2O3 pyroxene thermometer of Brey and Kohler in orthopyroxene decreases x 3

(1990) (T/BKN), (2) the Ca-in-orthopyroxene and Mg# increases with melt

in Op in 3

thermometer of Brey et al. (1990) (T/BK_Ca), depletion. Negative correlation O 2 2

(3) the thermometer of Witt-Eickschen and Seck between these two parameters Al

(1991) based on Cr-Al partitioning between probably reflects melt deple- 1 orthopyroxene and spinel coexisting with tion. Hzb—harzburgite; Lhz— olivine (T/WS), and (4) the thermometer of lherzolite; Serp—serpentine. 0 B Wells (1977) based on iron solubility in coexist 86 88 90 92 94 ing pyroxenes (T/Wells). Knippa xenolith tem- Mg# opx

Geosphere, June 2011 717

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Raye et al.

5 5 Hzb Al2O3 CaO Batch 1.5 GPa PUM Serp Hzb Fractional 2.0 GPa 4 Lhz 4 PUM Fractional 2.5 GPa Serp Lhz Young and Lee (2009) 3 3 Forearc peridotite BAB peridotite 2 2

1 1

AB 0 0 35 40 45 50 35 40 45 50 MgO (wt%) MgO (wt%)

Figure 7. Major-element compositions of Knippa peridotites. (A) Al2O3 plotted as a function of MgO. (B) CaO plotted as a function of MgO. Experimental batch and fractional melt extraction curves at 1.5 GPa (dashed and long line) and 2.5 GPa (short line) of Asimow (1999) are plotted, starting from average Primitive Upper Mantle (PUM) of Jagoutz et al. (1979), McDonough and Sun (1995), and Hart and Zindler (1986). Knippa harzburgites experienced more melt extraction than the lherzolites. Forearc peridotite data are from Parkinson and Pearce (1998), Parkinson and Arculus (1999), and Pearce et al. (2000), and backarc basin (BAB) peridotites are from Michibayashi et al. (2009).

Wood, 1991; O’Neill and Wall, 1987; Wood, ity for unserpentinized harzburgite (0.00 ± 0.64) Comparing Knippa peridotite oxygen- 1990; Wood and Virgo, 1989). The equilibrium is indistinguishable from that of serpentinized fugacity estimates with the oxygen-fugacity reaction is as follows: harzburgite (–0.62 ± 0.49). Similarly, serpen- range of global Proterozoic and Phanerozoic tinized lherzolite (–1.6 ± 1.45) has an oxygen subcontinental lithospheric spinel peridotites

6 Fe2SiO4 + O2 = 3Fe2Si2O6 + 2Fe3O4. fugacity that is indistinguishable from that of (Ballhaus at al., 1991; Ballhaus, 1993), the Olivine Orthopyroxene Spinel unserpentinized lherzolite (–0.94 ± 0.70). Knippa peridotites plot in the field of lightly metasomatized peridotites (Fig. 9). Knippa The oxygen-fugacity values for Knippa xeno liths were calculated using electron microprobe analyses, which yield total iron content TABLE 3. TEMPERATURE, OXYGEN FUGACITY, AND SEISMIC VELOCITY OF SELECTED SAMPLES CALCULATED AT P = 20 KBAR of spinels. The distribution of Fe2+ and Fe3+ in Sample T/Wells T/BKN Vp (STP) Vs (STP) Vp (T) Vs (T)

spinel was inferred from stoichiometry. The no. (°C) (°C) ∆logfO2 (km/s) (km/s) (km/s) (km/s) mantle fugacity values obtained by this method Kn1 934 889– 8.25 4.81 7.82 4.42 (Ballhaus, 1993; Mattioli et al., 1989) differ Kn2 969 950– 8.30 4.84 7.88 4.46 Kn3 991 987−0.46 8.30 4.83 7.83 4.42 little from those obtained by direct determina- Kn4 949 937−0.46 8.29 4.83 7.85 4.44 tions of Fe3+. Kn6 910 858−2.60 8.27 4.82 7.88 4.45 Kn7 960 949− 8.29 4.82 7.82 4.41 Calculated oxygen-fugacity values (Table 3) Kn8 939 939−1.13 −−7.85 4.43 are given relative to the fayalite-magnetite- Kn9 968 975−0.69 8.30 4.83 7.88 4.44 quartz (FMQ) buffer at 2.0 GPa. The oxygen Kn10 856 760−0.99 −−7.92 4.48 Kn12 920 870− 8.27 4.80 7.87 4.42 fugacity relative to FMQ buffer is little affected Kn13 890 825−0.57 8.27 4.81 7.91 4.46 by changes in estimated pressure within the RR1 941 909−2.03 8.28 4.82 7.84 4.43 range of the spinel-lherzolite stability field RR2 947 926−1.50 8.25 4.82 7.80 4.41 K2B 910 855−0.44 8.27 4.82 7.88 4.46 (Wood, 1990). Changing the pressure to 1.5 K2C 897 8600.198.284.827.944.48 GPa causes a shift of oxygen fugacity of <0.001 K2D 894 8460.228.304.827.934.47 K2E 917 840−0.47 −−7.86 4.44 log units. Oxygen-fugacity values of Knippa K2F1 914 840−0.58 −−7.87 4.45 xenoliths fall in a relatively narrow range from K2F2 10231058−2.63 −−7.72 4.35 below the FMQ buffer (Dlog fO = –2.6 to K2F3 10681151−0.64 8.31 4.83 7.72 4.34 2FMQ K2F4 927 860– −−7.88 4.45 0.61; Fig. 9). The harzburgites show slightly K2F5 948 885−0.92 −−7.92 4.48 higher oxygen fugacity than the lherzolites. K2F6 876 781−0.05 8.30 4.83 7.99 4.52 The Dlog fO values for lherzolite are 0 to K3G 926 876−0.24 8.24 4.77 7.84 4.39 2FMQ K2H 926 9170.57− −7.884.45 –2.6; whereas for harzburgite, the values are –1 K2R2 946 891−0.88 8.29 4.82 7.92 4.46 to +0.61 (Fig. 9). Mean oxygen-fugacity value K2R3 966 967−2.48 8.28 4.82 7.82 4.40 K2R4 860 7700.62− −7.974.50 and standard deviation (relative to FMQ) for K2R5 984 960−0.74 8.30 4.82 7.88 4.42 unserpentinized harzburgite is 0.00 ± 0.64 and Note: STP—standard temperature and pressure; T/BKN—two-pyroxene thermometer of Brey and Kohler for lherzolite is –0.94 ± 0.70. Serpentinization (1990); T/Wells—thermometer of Wells (1977); T(936 °C)—Mean Knippa temperature; Vp—primary wave did not affect oxygen fugacity: oxygen fugac- velocity; Vs—secondary wave velocity.

718 Geosphere, June 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Composition of the mantle lithosphere beneath south-central Laurentia

Figure 8. Equilibration temper- 10%–15% melting, which is consistent with the ature for Knippa peridotites. estimate based on Cr# versus TiO2 plot. As noted Temperature was calculated at above, spinel Cr# is a much more sensitive index 2 GPa using the two-pyroxene 1100 y = 0.5291x + 459.97 of melt extraction than olivine Mg# (Hellebrand thermometer (T/BKN) of Brey R2 = 0.91 et al., 2001). Hellebrand et al. (2001) calculated and Kohler (1990) and the Mean Wells = 936 °C fractional melt percentage as a function of spinel Mean BKN = 916 °C thermometer of Wells (1977) 1000 Cr#, yielding the relationship: F = 10*ln (Cr#) (T/Wells). Knippa xenoliths +24, where F = melt percentage. This relation- have equilibrium temperatures ship is thought to be valid for spinel Cr# between of 824–1058 °C (T/BKN). The 10 and 60. Based on Hellebrand et al.’s (2001) / 2 T BKN (°C) two approaches agree with r = 900 model, Knippa lherzolites reflect 5%–9%, and 0.91. There are no signifi- Hzb harzburgites reflect 11%–14% melt extraction Serp Hzb cant temperature differences Lhz from a primitive mantle source (Fig. 12). These between lherzolites (Lhz) and Serp Lhz melt-depletion signatures are less than current harzburgites (Hzb) or between 800 Young and Lee (2009) data sets for forearc peridotite (Parkinson and serpentinized versus fresh perido 800 900 1000 1100 Pearce, 1998; Parkinson and Arculus, 1999; tites, suggesting that these T/Wells 1977(°C) Pearce et al., 2000) but quite consistent with lithologies are mixed in the backarc basin peridotite (Arai and Ishimaru, 0lithosphere beneath Texas. 2008; Ohara et al., 2002). Serp—serpentine. The whole-rock compositions are consistent with the results obtained from mineral composition and reflect moderate extraction of partial peridotites overlap the abyssal and backarc (Bonatti and Michael, 1989; Frey et al., 1985; melts. As melt depletion increases, whole-rock

basin peridotite field (Arai and Ishimaru, 2008) Frey and Prinz, 1978; Hauri and Hart, 1994). compositions show decreasing Al2O3 and CaO and have lower oxygen fugacities compared to Knippa xenoliths plot within the olivine-spinel and increasing MgO. With progressive melt-

forearc peridotites. mantle array (OSMA; Fig. 11), interpreted as a ing, Al2O3 in orthopyroxene decreases (Fig. mantle-peridotite restite trend (Arai, 1994). With 6B), olivine Fo content increases, modal olivine Partial Melting and Depletion of the greater extents of partial melting, olivine Fo increases, and modal clinopyroxene decreases.

Knippa Lithospheric Mantle increases slightly, but the Cr# of spinel increases The whole-rock contents of Al2O3 and CaO, greatly, defining the olivine-spinel mantle array. which reflect variations in melt extraction from

The Cr# versus TiO2 in spinel (Fig. 10) dis- Peridotites from different tectonic settings primitive (undepleted) upper mantle (PUM; tinguishes between spinels that experienced occupy distinct parts of the array. Melting curves Hart and Zindler, 1986; Jagoutz et al., 1979; melt-rock interaction and those defining a partial from Pearce et al. (2000) based on experimen- McDonough and Sun, 1995), are plotted as a melting trend (Arai, 1992; Zhou et al., 1996). tal studies of Jaques and Green (1980) show function of MgO (Fig. 7). Experimental batch Figure 10 shows a partial melting trend, start- that Knippa lherzolites were produced by <10% and fractional melting curves of Asimow (1999) ing from fertile mid-ocean ridge basalt (MORB) melting, whereas harzburgites were produced by are plotted starting from PUM compositions,

mantle with 0.18% TiO2, and is superimposed on published experimental results of Johnson et al. (1990). Details of the method are explained in Figure 9. Oxygen fugacity Pearce et al. (2000). Ti, being an incompatible of Knippa peridotites. Oxy- element, should rapidly decrease as the degree gen fugacities fall near the of melting increases and Cr# increases. The fayalite-magnetite-quartz buf-

trend for Knippa peridotites toward higher Ti, fer (Dlog fO2FMQ = –2.6–0.61). 2 especially for harzburgites, therefore implies Knippa lherzolites and harz melt-mantle interaction through reaction or melt burgites plot in the field of 1 Strongly impregnation by a Ti-rich melt (Edwards and lightly metasomatized spinel metasomatized Malpas, 1996; Kelemen et al., 1995). This reac- peridotites (Ballhaus, 1993).

) 0 Q

tion would have followed depletion of the host In contrast, forearc perido- M

F (

harzburgite as origin of the Ti-rich trend point to tites have distinctly higher Cr# 2

O Hzb f –1

and oxygen fugacity. Forearc g

the most Cr#-rich melting trend of the peridotite. ve

o Serp Hzb l Lightly peridotite data are from Par- Lhz No significant difference is seen between serpen- metasomatized Abyssal rimiti Serp Lhz tinized and unserpentinized peridotites, again kinson and Pearce (1998), –2 P spinel peridotite Young and Lee (2009) confirming that the serpentinization is probably Parkinson and Arculus (1999), Forearc peridotite BAB peridotite a late-stage, low-temperature phenomenon. and Pearce et al. (2000), and –3 Spinel Cr# correlates with Fo content of backarc basin (BAB) peridotite 0.1 0.3 0.50.7 0.9 coexisting olivines and Cr contents of coexisting data are from Arai and Ishi- Cr # in spinel orthopyroxene and clinopyroxene. These com- maru (2008). The less oxidized positional variations are interpreted to have been forearc samples are from, for caused by extraction of a basaltic partial melt and example, Pali Aike. See text for have been observed in xenolith suites worldwide further discussion.

Geosphere, June 2011 719

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Raye et al.

1.0 Hzb 1 Hzb Serp Hzb Serp Hzb Lhz % Lhz 0.8 + 40 Serp Lhz ++ Serp. Lhz + Young and Young and Lee (2009) + Partial + 30% Lee (2009) FMM = Fertile + + + Forearc peridotite 0.6 melting pinel + MORB Mantle s ++++ BAB peridotite + + + +++ FMM = Fertile 0.5 ++ 20% MORB Mantle 20% +++ Cr# in in Cr# + 0.4 + crystallFractional Cr# in spinel 15% 10% Partial ization 10% melting

0.2 5% Continental Melt-rock Peridotite FMM reaction 0 FMM 95 90 85 80 0.0 0.2 0.4 0.6 0.8 1.0 Mg# of olivine TiO2 wt% in spinel

Figure 10. Chromium is compatible in spinel and increases with melt Figure 11. Knippa peridotite compositions extraction, while titanium is incompatible and decreases with plotted on the olivine-spinel mantle array (OSMA) of Arai (1994). OSMA is a mantle- melt extraction. Increase of Ti content in the Cr# versus TiO2 wt% of spinel is thus indicative of an early melt-extraction event followed peridotite restite trend. Peridotites from by subsequent melt-rock reaction. The harzburgites appear to have different tectonic settings plot in differ- undergone more melt-rock reaction than the lherzolites. ent parts of this trend. Knippa peridotites plot within the continental field. Melting curve (dashed line with % melting) from which again indicate that lherzolites reflect less (1975) using different stations in Texas (Cor- Pearce et al. (2000) based on experimental melt depletion than do harzburgites. Note that pus Christi, Edinburg, Laredo, San Marcos, and studies of Jaques and Green (1980) indi-

the harzburgite samples with TiO2-rich spinel Houston, Fig. 13A) show that crustal structure cate that the lherzolites were produced by that follow the melt-rock reaction trend in Fig- is generally similar along all profiles extending <10% melting, whereas the harzburgites ure 10 plot around the highest MgO-end of the from the Llano uplift southeastward to the Gulf were produced by 10%–15% melting. All whole-rock trend (Fig. 7), implying that melt of Mexico. A generalized crustal structure model forearc peridotite data are from Parkinson depletion formed the harzburgite first, and melt proposed by Keller and Shurbet (1975) is shown and Pearce (1998), Parkinson and Arculus impregnation occurred later. Effect of melt reac- in Fig. 13B. Based on Rayleigh wave-dispersion (1999), and Pearce et al. (2000), and back- tion to the depleted harzburgites shown by Cr# data, the upper layers (Vp ≤5.2 km/s) are inter- arc basin (BAB) peridotite data are from

versus TiO2 in Figure 10 is not identifiable in preted as Mesozoic and Cenozoic sedimentary Arai and Ishimaru (2008) and Ohara et al. other plots, suggesting negligible effect on the rocks, the upper crustal layer (Vp >5.2 km/s) (2002). FMM—fertile MORB mantle; estimates of melt extraction based on major is interpreted to consist primarily of Paleozoic MORB—mid-ocean ridge basalt; Serp— and compatible elements. However, the effect metamorphic rocks, and the lower crustal layer serpentine. should not be underestimated for incompat- (Vp ≤6.9 km/s) is interpreted to comprise mafic ible elements such as Ti or rare-earth elements (REEs). According to the tectonic model shown in figure 7 of Mosher (1998), the mantle beneath Figure 12. Melt depletion of 0.9 Knippa would have underlain a forearc during Knippa peridotites, based on late Mesoproterozoic time. However, Knippa Cr# of spinel. Chromium being 0.7 peridotites have major-element and mineral compatible in spinel, increases 23% compositions that are different from forearc in the residue with increas- peridotites and are consistent with metasoma- ing melt extraction. Hellebrand 0.5 tized continental lithosphere (Young and Lee, et al. (2001) calculated fractional Hzb 2009) or backarc peridotites (Arai and Ishimaru, melt percentage as a function 11% Cr# in spinel in Cr# Serp Hzb 2008). However, neither melt depletion nor oxi- of spinel Cr#, yielding the rela- Lhz 0.3 5% dation state is specific to tectonic setting. Iso- tionship F = 10*ln (Cr#) +24, Serp Lhz Young and Lee (2009) topic and trace-element data (in progress) may where F = melt fraction. This 16% Forearc peridotite BAB peridotite help constrain these models. relationship is valid for spinel 0.1 Cr#s between 0.1 and 0.6. 010 20 30 Seismic-Velocity Structure beneath Texas Knippa lherzolites (Lhz) are F (% melt fraction) formed by <10% melt extrac- Calculated velocities were compared with tion and harzburgites (Hzb) by ~15% melt extraction. All forearc peridotite data are from geophysical measurements of seismic velocity Parkinson and Pearce (1998), Parkinson and Arculus (1999), and Pearce et al. (2000), and beneath the region. Rayleigh wave-dispersion backarc basin (BAB) peridotite data are from Arai and Ishimaru (2008) and Ohara et al. experiments carried out by Keller and Shurbet (2002). Serp—serpentine.

720 Geosphere, June 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Composition of the mantle lithosphere beneath south-central Laurentia

N was less than 10% for lherzolite and 10%–15% for harzburgite. This melt-depleted mantle sub- Llano uplift Knippa sequently underwent melt-mantle interaction 0 through reaction or melt impregnation by a Upper Transitional lithospherSedimentes Ti-rich melt. No significant difference between 10 crust Llano System ) serpentinized and unserpentinized peridotites H 20 uplift C Lower confirmed that the serpentinization was a late- 30 Crust stage, low-temperature phenomenon. The com- OuachitaSAM D HOU 4 (km Depth MOHO bination of these characteristics suggests that Knippa 5 40 this mantle was either a slight metasomatized Upper mantle continental lithosphere and/or backarc basin COR 3 1 Coast A B 100 200 300 400 500 rather than a forearc setting. LAR Gulf Distance (km) Calculated seismic-velocity data are consistent with geophysical profiles from the Llano EDN 2 uplift southeastward to the coastline of the Gulf T1 T2 T3 CDH of Mexico. Our velocity calculations will be 0 useful to constrain improved geophysical mod- Sedi- els generated by Earthscope’s new data set. ments ACKNOWLEDGMENTS 10 Upper We thank Carol Frost, Tim Lawton, Cin-ty Lee, Crust and Melanie Barnes for their reviews. W.R. Griffin has been a great help with providing samples and infor- 20 mation on the Balcones Igneous Province. This work Lower is supported by Texas Norman Hackerman Advanced Depth (km) Crust Research Program grant 003661-0003-2006 to Eliza- 30 beth Y. Anthony and Robert J. Stern. MOHO 8.41 Upper 7.8 7.8 8.2 8.0 REFERENCES CITED mantle 40 Anthony, E.Y., 2005, Source regions of granites and their links to tectonic environment: Examples from the C western United States: Lithos, v. 80, p. 61–74, doi: 10.1016/j.lithos.2004.04.058. Arai, S., 1992, Chemistry of chromian spinel in volcanic Figure 13. (A) Index map showing Knippa (star), seismograph stations: LAR—Laredo, rocks as a potential guide to magma chemistry: Lon- COR—Corpus Christi, EDN—Edinburg, SAM—San Marcos, HOU—Houston (Keller and don, Mineralogical Magazine, v. 56, p. 173–184. Shurbet, 1975), C—Cram (1961, 1962), D—Dorman et al. (1972), H—Hales et al. (1970) Arai, S., 1994, Characterization of spinel peridotites by olivine-spinel compositional relationships: Review and refraction line and gravity profiles 1, 2, 3, and 4. T1 (stations 2, 4, and 5), T2 (stations 1, 2, interpretation: Chemical Geology, v. 113, p. 191–204, and 3), and T3 (stations 1, 3, and 4) are tripartite locations. (B) Generalized crustal struc- doi: 10.1016/0009-2541(94)90066-3. Arai, S., and Ishimaru, S., 2008, Insights into petrologi- ture model proposed by Keller and Shurbet (1975). (C) Crustal structure as interpreted cal characteristics of the lithosphere of mantle wedge from seismic velocities. Our calculated upper mantle seismic velocity agrees well with those beneath arcs through peridotite xenoliths: A review: velocities obtained from geophysical experiments. Journal of Petrology, v. 49, p. 665–695, doi: 10.1093/ petrology/egm069. Asimow, P.D., 1999, A model that reconciles major- and trace-element data from abyssal peridotites: Earth and Planetary Science Letters, v. 169, p. 303–319, doi: igneous rocks (Fig. 13C). Gravity and refraction portable array investigations of mantle beneath 10.1016/S0012-821X(99)00084-9. data suggest that the lower crustal layer dips Texas are interpreted. The velocities reported in Ballhaus, C., 1993, Redox states of lithospheric and astheno- spheric upper mantle: Contributions to Mineralogy and to the northwest and that thick Paleozoic sedi- this study will serve as important constraints on Petrology, v. 114, p. 331–348, doi: 10.1007/BF01046536. mentary and metasedimentary rocks are present models generated by the Earthscope experiment. Ballhaus, C., Berry, R.G., and Green, D.H., 1991, High- seaward of the buried Ouachita belt. Seismic pressure experimental calibration of the olivine-ortho pyroxene-spinel oxygen geobarometer: Implications velocities for the crust do not indicate a crustal CONCLUSIONS for the oxygen state of the upper mantle: Contributions granitic layer seaward of the buried Ouachita to Mineralogy and Petrology, v. 107, p. 27–40, doi: 10.1007/BF00311183. belt, but they do indicate that the lower crust is Peridotite xenoliths in Cretaceous basanites Barker, D.S., Mitchell, R.H., and McKay, D., 1987, Late composed of oceanic crust thickened presum- from the Knippa quarry of south-central Texas Cretaceous nephelinite to phonolite magmatism in ably by thrusting and other deformation. are among a few mantle samples from the the Balcones province, Texas, in Morris, E.M., and Pasteris, J.D., eds., Mantle Metasomatism and Alkaline Our seismic velocities calculated at appro- southern margin of Laurentia and thus provide Magmatism: Geological Society of America Special priate temperatures for Knippa peridotites constraints on the nature of this lithospheric Paper, v. 215, p. 293–304. (7.8–8.3 km/sec) shown in Figure 13C are mantle. Based on our petrographic, mineral- Barnes, M.A., Anthony, E.Y., Williams, I., and Asquith, G.B., 2002, Architecture of a 1.38–1.34 Ga granite-rhyolite consistent with the geophysical measurements chemical, and whole-rock chemical studies, we complex as revealed by geochronology and isotopic (7.8–8.4 km/sec) shown in Figure 4C (Keller and estimate that equilibration temperatures at the and elemental geochemistry of subsurface samples from west Texas, USA: Precambrian Research, v. 119, Shurbet, 1975). This velocity structure obtained time of entrainment were 824–1058 °C and that p. 9–43, doi: 10.1016/S0301-9268(02)00116-X. from gravity data and calculated seismic data Dlog fO2FMQ values are 0 to –2.6 for lherzolite Bickford, M.E., Soegaard, K., Nielsen, K.C., and McLelland, will be refined as data from the Earthscope trans- and –1 to +0.61 for harzburgite. Melt depletion J.M., 2000, Geology and geochronology of Grenville-

Geosphere, June 2011 721

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Raye et al.

age rocks in the Van Horn and Franklin Mountains Hauri, E.H., and Hart, S.R., 1994, Constraints on melt Mosher, S., 1998, Tectonic evolution of the Southern Lau- area, west Texas: Implications for the tectonic evolu- migration from mantle plumes: A trace element study rentian Grenville orogenic belt: Geological Soci- tion of Laurentia during the Grenville: Geological of peridotite xenoliths from Savai’i, Western Samoa: ety of America Bulletin, v. 110, p. 1357–1375, doi: Society of America Bulletin, v. 112, p. 1134–1148, Journal of Geophysical Research, v. 99, p. 24,301– 10.1130/0016-7606(1998)110<1357:TEOTSL> doi: 10.1130/0016-7606(2000)112<1134:GAGOGR> 24,321, doi: 10.1029/94JB01553. 2.3.CO;2. 2.0.CO;2. Hellebrand, E., Snow, J.E., Dick, H.J.B., and Hoffmann, Nell, J., and Wood, B., 1991, High-temperature electri- Bird, D.E., Bruke, K., Hall, S.A., and Casey, J.F., 2005, A.W., 2001, Coupled major and trace elements as cal measurements and thermodynamic properties of

Gulf of Mexico tectonic history: Hotspot traces crustal indicators of the extent of melting in mid-ocean- Fe3O4-FeCr2O4-MgCr2O4-MgAl2O4-FeAl2O4 spinels: boundaries and early salt distribution: American Asso- ridge peridotites: Nature, v. 410, p. 677–681, doi: The American Mineralogist, v. 74, p. 339–351. ciation of Petroleum Geologists Bulletin, v. 89, p. 311– 10.1038/35070546. Ohara, H., Stern, R.J., Ishi, T., Yurimoto, H., and Yamazaki, 328, doi: 10.1306/10280404026. Herrmann, W., and Berry, R.F., 2002, MINSQ—A least- T., 2002, Peridotites from the Mariana Trough: First Bonatti, E., and Michael, P.J., 1989, Mantle peridotites from squares spreadsheet method for calculating mineral look at the mantle beneath an active back-arc basin: continental rifts to ocean basins to subduction zones: proportions from whole-rock major-element analyses: Contributions to Mineralogy and Petrology, v. 143, Earth and Planetary Science Letters, v. 91, p. 297–311, Geochemistry: Exploration, Environment, Analysis, p. 1–18, doi: 10.1007/s00410-001-0329-2. doi: 10.1016/0012-821X(89)90005-8. v. 2, no. 4, p. 361–368, doi: 10.1144/1467-787302-010. O’Neill, H., and Wall, V., 1987, The olivine-spinel oxygen Brey, G.P., and Kohler, T., 1990, Geothermobarometry in Jagoutz, E., Palme, H., Baddenhausen, H., Blum, K., geobarometer, the nickel precipitation curve, and the four-phase lherzolites: II. New thermobarometers, and Cendales, M., Dreibus, G., Spettel, B., Lorenz, V., and oxygen fugacity of the upper mantle: Journal of Petrol- practical assessment of existing thermobarometers: Wanke, H., 1979, The abundances of major, minor, ogy, v. 28, p. 1169–1192. Journal of Petrology, v. 31, p. 1353–1378. and trace elements in the Earth’s mantle as derived Parkinson, I.J., and Arculus, R.J., 1999, The redox-state Brey, G.P., Köhler, T., and Nickel, K.G., 1990, Geothermo- from primitive ultramafic nodules: Houston, Proceed- of subduction zones: Insights from arc-peridotites: barometry in four-phase lherzolites: I. Experimental ings of the 10th Lunar and Planetary Science Confer- Chemical Geology, v. 160, p. 409–423, doi: 10.1016/ results from 10 to 60 kb: Journal of Petrology, v. 31, ence, Pergamon, p. 2031–2050. S0009-2541(99)00110-2. p. 1313–1352. Jaques, A.L., and Green, D.H., 1980, Anhydrous melt- Parkinson, I.J., and Pearce, J.A., 1998, Peridotites from Cram, I.H., Jr., 1961, A crustal refraction survey in South ing of peridotite at 0–15 kb pressure and the genesis the Izu-Bonin-Mariana forearc (ODP Leg 125): Evi- Texas: Geophysics, v. 26, p. 560–573. of tholeiitic basalts: Contributions to Mineralogy and dence for mantle melting and melt-mantle interaction Cram, I.H., Jr., 1962, Crustal structure of Texas coastal plain Petrology, v. 73, p. 287–310, doi: 10.1007/BF00381447. in a supra-subduction zone setting: Journal of Petrol- region: Am. Assoc. Petroleum Geologists Bull., v. 46, Johnson, K.T.M., Dick, H.J.B., and Shimizu, N., 1990, ogy, v. 39, p. 1577–1618, doi: 10.1093/petrology/ p. 1721–1727. Melting in the oceanic upper mantle: An ion micro- 39.9.1577. Dickinson, W.R., Gehrels, G.E., and Stern, R.J., 2010, probe study of diopsides in abyssal peridotites: Journal Pearce, J.A., Barker, P.F., Edwards, S.J., Parkinson, I.J., Late Triassic Texas uplift preceding Jurassic opening of Geophysical Research, v. 95, p. 2661–2678, doi: and Leat, P.T., 2000, Geochemistry and tectonic sig- of the Gulf of Mexico: Evidence from U-Pb ages of 10.1029/JB095iB03p02661. nificance of peridotites from the South Sandwich arc- detrital zircons: Geosphere, v. 6, no. 5, p. 641–662, doi: Kelemen, P.B., Shimizu, N., and Salters, V.J.M., 1995, basin system, South Atlantic: Contributions to Miner- 10.1130/GES00532.1. Extraction of mid-ocean ridge basalt from the upwell- alogy and Petrology, v. 139, p. 36–53, doi: 10.1007/ Dorman, J., Worzel, J.L., Leyden, R., Crook, T.N., and ing mantle by focused flow of melt in dunite channels: s004100050572. Hatziemmanuel, M., 1972, Crustal section from seis- Nature, v. 375, p. 747–753, doi: 10.1038/375747a0. Reese, J.F., Mosher, S., Connelly, J., and Roback, R., mic regraction measurements near Victoria, Texas: Keller, G.R., and Shurbet, D.H., 1975, Crustal struc- 2000, Mesoproterozoic chronostratigraphy of the Geophysics, v. 37, p. 325–336. ture of the Texas Gulf Coastal Plain: Geological southeastern Llano uplift, central Texas: Geologi- Edwards, S.J., and Malpas, J., 1996, Melt-peridotite inter Society of America Bulletin, v. 86, p. 807–810, cal Society of America Bulletin, v. 112, p. 278–291, actions in shallow mantle at the East Pacific Rise: doi: 10.1130/0016-7606(1975)86<807:CSOTTG> doi: 10.1130/0016-7606(2000)112<278:MCOTSL> Evidence from ODP site 895 (Hess Deep): London, 2.0.CO;2. 2.0.CO;2. Mineralogical Magazine, v. 60, no. 1, p. 191–206. LeBas, M.J., and Streckeisen, A.L., 1991, The IUGS sys- Roback, R.C., 1996, Characterization and tectonic evolu- Frey, F.A., and Prinz, M., 1978, Ultramafic inclusions from tematics of igneous rocks: Journal of the Geological tion of a Mesoproterozoic island arc in the south- San Carlos, Arizona: Petrology and geochemical data Society of London, v. 148, p. 825–833, doi: 10.1144/ ern Grenville orogen, Llano uplift, central Texas: bearing on their petrogenesis: Earth and Planetary gsjgs.148.5.0825. Tectonophysics, v. 265, p. 29–52, doi: 10.1016/ Science Letters, v. 38, p. 129–176, doi: 10.1016/ Lee, C.-T.A., 2003, Compositional variation of density and S0040-1951(96)00145-X. 0012-821X(78)90130-9. seismic velocities in natural peridotites at STP condi- Rowley, D.B., and Pindell, J.L., 1989, End Paleozoic–Early Frey, F.A., Suen, C.J., and Stockman, H., 1985, The Ronda tions: Implications for seismic imaging of composi- Mesozoic western Pangean reconstruction and its high temperature peridotite: Geochemistry and petro tional heterogeneities in the upper mantle: Journal of implications for the distribution of Precambrian and genesis: Geochimica et Cosmochimica Acta, v. 49, Geophysical Research, v. 108, p. 1–20. Paleozoic rocks around Meso-America: Precambrian p. 2469–2491, doi: 10.1016/0016-7037(85)90247-9. Mattioli, G., and Wood, B., 1988, Magnetite activities across Research, v. 42, p. 411–444, doi: 10.1016/0301-9268 Galloway, W.E., Bedout, D.G., and Fisher, W.L., Jr., the MgAl2O4-Fe3O4 spinel join, with application to (89)90022-3. Cabrera-Castro, R., Lugo-Rivera, J. E., and Scott, thermobarometric estimates of upper mantle oxygen Satsukawa, T., Michibayashi, K., Raye, U., Anthony, E.Y., T.M., 1991, Cenozoic, in Salvador, A., ed., The Gulf of fugacity: Contributions to Mineralogy and Petrology, Pulliam, J., and Stern, R.J., 2010, Uppermost mantle Mexico Basin: Geological Society of America, Geol- v. 98, p. 148–162, doi: 10.1007/BF00402108. anisotropy beneath the southern Laurentian margin: ogy of North America, v. J, p. 245–324. Mattioli, G., Baker, M., Rutter, M., and Stolper, E., 1989, Evidence from Knippa peridotite xenoliths, Texas: Gao, S.S., Liu, K.H., Stern, R.J., Keller, R.G., Hogan, J.P., Upper mantle oxygen fugacity and its relationship to Geophysical Research Letters, v. 37, p. L20312, doi: Pulliam, J., and Anthony, E.Y., 2008, Characteristics of metasomatism: The Journal of Geology, v. 97, p. 521– 10.1029/2010GL044538. mantle fabric beneath the south-central United States: 536, doi: 10.1086/629332. Sawyer, D.S., Buffler, R.T., and Pilger, R.H.P., Jr., 1991, Constraints from shear-wave splitting measurements: McDonough, W., and Sun, S., 1995, The composition of The crust under the Gulf of Mexico Basin, in Salvador, Geosphere, v. 4, p. 411–417. the Earth: Chemical Geology, v. 120, p. 223–253, doi: A., ed., The Gulf of Mexico Basin: Geological Soci- Gaul, O.F., Griffin, W.L., O’Reilly, S.Y., and Pearson, N.J., 10.1016/0009-2541(94)00140-4. ety of America, The Geology of North America, v. J, 2000, Mapping olivine composition in the lithospheric Michibayashi, K., Ohara, Y., Stern, R.J., Fryer, P., Kimura, p. 53–72. mantle: Earth and Planetary Science Letters, v. 182, J.-I., Tasaka, M., Harigane, Y., and Ishi, T., 2009, Spencer, A.B., 1969, Alkalic igneous rocks of the Balcones p. 223–235, doi: 10.1016/S0012-821X(00)00243-0. Peridotites from a ductile shear zone within back-arc Province, Texas: Journal of Petrology, v. 10, p. 272– Griffin, W.R., Foland, K.A., Stern, R.J., and Leybourne, lithospheric mantle, southern Mariana Trench: Results 306. M.I., 2010, Geochronology of bimodal alkaline vol of a Shinkai 6500 dive: Geochemistry Geophysics Stern, R.J., Anthony, E.Y., Ren, M., Norton, I., Kimura, canism in the Balcones Igneous Province, Texas: Impli- Geosystems, v. 10, Q05X06, 17 p., doi: 10.1029/ J.I., Miyazaki, T., Hanyu, T., Chang, Q., and Hira- cations for Cretaceous intraplate magmatism in the 2008GC002197. hara, Y., 2011, Southern Louisiana Salt Dome xeno- northern Gulf of Mexico Magmatic Zone: The Journal Mickus, K., Stern, R., Keller, G.R., and Anthony, E., 2009, liths: First glimpse of Jurassic (~160 Ma) Gulf of of Geology, v. 118, p. 1–21, doi: 10.1086/648532. Potential field evidence for a volcanic rifted margin Mexico crust: Geology, doi: 10.1130/G31635.1 (in Hacker, B.R., and Abers, G.A., 2004, Subduction Factory along the Texas Gulf Coast: Geology, v. 37, p. 387– press). 3: An Excel worksheet and macro for calculating the 390, doi: 10.1130/G25465A.1. Thomas, W.A., 2006, Tectonic inheritance at a continen- Mickus, K., and Keller, G.R., 1992, Lithospheric structure tal margin: GSA Today, v. 16, p. 4–11, doi: 10.1130/ densities, seismic wave speeds, and H2O contents of minerals and rocks at pressure and temperature: of the south-central U.S: Geology, v. 20, p. 335–338, 1052-5173(2006)016[4:TIAACM]2.0.CO;2. Geochemistry, Geophysics, Geosystems, v. 5, p. 1–7. doi: 10.1130/0091-7613(1992)020<0335:LSOTSC> Thomas, W.A., and Astini, R.A., 1996, The Argentine Pre- doi:10.1029/2003GC000614. 2.3.CO;2. cordillera: A traveler from the Ouachita embayment Hales, A.L., Helsley, C.E., and Nation, J.B., 1970, Crustal Mosher, S., 1993, Western extensions of Grenville age rocks: of North American Laurentia: Science, v. 273, p. 752– structure study of the Gulf Coast of Texas: Am. Assoc. Texas, in Reed, J.C., Jr., Bickford, M.E., Houston, R.S., 757, doi: 10.1126/science.273.5276.752. Petroleum Geologists Bull., v. 54, p. 2040–2057. Link, P.K., Rankin, D.W., Sims, P.K., and Van Schmus, Wells, P.R.A., 1977, Pyroxene thermometry in simple Hart, S.R., and Zindler, A., 1986, In search of a bulk Earth W.R., eds., Precambrian: Conterminous U.S: Boulder, and complex systems: Contributions to Mineral- composition: Chemical Geology, v. 57, p. 247–267, Colorado, Geological Society of America, The Geol- ogy and Petrology, v. 62, p. 129–139, doi: 10.1007/ doi: 10.1016/0009-2541(86)90053-7. ogy of North America, v. C-2, p. 365–378. BF00372872.

722 Geosphere, June 2011

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021 Composition of the mantle lithosphere beneath south-central Laurentia

Whitmeyer, S.J., and Karlstrom, K.E., 2007, Tectonic model Wood, B.J., 1990, An experimental test of the spinel genic belt: Lithosphere, v. 1, p. 370–381, doi: 10.1130/ for the Proterozoic growth of North America: Geo- peridotite oxygen barometer: Journal of Geophysi- L72.1. sphere, v. 3, p. 220–259, doi: 10.1130/GES00055.1. cal Research, v. 95, p. 15,845–15,851, doi: 10.1029/ Zhou, M.-F., Robinson, P.T., Malpas, J., and Li, Z., 1996, Wilshire, H.G., McGuire, A.V., Noller, J.S., and Turrin, JB095iB10p15845. Podiform chromitites in the Loubusa ophiolite (South- B.D., 1990, Petrology of lower crustal and upper man- Wood, B.J., and Virgo, D., 1989, Upper mantle oxidation ern Tibet): Implications for melt-rock interaction and tle xenoliths from the Cima volcanic field, California: state: Ferric iron contents of lherzolite spinels by chromite segregation in the upper mantle: Journal of Journal of Petrology, v. 32, p. 170–200. 57Fe Mossbauer spectroscopy and resultant oxygen Petrology, v. 37, p. 3–21, doi: 10.1093/petrology/ Witt-Eickschen, G., and Seck, H.A., 1991, Solubility of Ca fugacities: Geochimica et Cosmochimica Acta, v. 53, 37.1.3. and Al in orthopyroxene from spinel peridotite: An p. 1277–1291, doi: 10.1016/0016-7037(89)90062-8. improved version of an empirical geothermometer: Young, H.P., and Lee, C.-T.A., 2009, Fluid-metasomatized Manuscript Received 28 May 2010 Contributions to Mineralogy and Petrology, v. 106, mantle beneath the Ouachita belt of southern Lauren- Revised Manuscript Received 13 January 2011 p. 431–439, doi: 10.1007/BF00321986. tia: Fate of lithospheric mantle in a continental oro- Manuscript Accepted 22 January 2011

Geosphere, June 2011 723

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/7/3/710/3340191/710.pdf by guest on 28 September 2021