1. The Earth III: Lithosphere Lecture 47

2. Basalts from the Lithosphere The lithosphere is the part of the Earth through which heat is conducted rather than convected. Mantle lithosphere (the subcontinental lithospheric mantle: SCLM) tends to have fast seismic velocities, suggesting is cold compared with the convective mantle. Xenoliths derived from these regions) are often harzburgitic, harzburgite being a comparatively low-density peridotite. This subcontinental lithosphere is of variable thickness: it is only 10’s of km under tectonically active areas such as the Great Basin but is more than 200 km thick under the South African craton. Basalts from the lithosphere reveal it to be very chemically heterogeneous, with some areas apparently incompatible element- enriched.

3. Xenoliths in Continental Magmas

4. SCLM can be old Inclusions in diamonds can be time capsules. Sulfides in this one from a South African kimberlite was Re-Os date at 2.9 billion years. Indeed, in many cases it seems the SCLM mantle is about the same age as the crust above it.

5. The Crust

6. The Oceanic Crust The oceanic crust is layered: basaltic lava flows (buried by sediment most of the time) underlain by sheeted dikes (magma conduits) that are in turn underlain by gabbro (basaltic magma crystallized at depth).

7. Compositional Variation The composition of the oceanic crust is controlled by two processes: partial melting and fractional crystallization. o In this plot, variable extents of melting produce parental magmas of variable Na concentration. o The magmas then evolve along subparallel trends through fractional crystallization.

8. What Controls Melting? Na concentrations of the parental magmas correlate with ridge depth: elevated ridges are Na-poor, indicative of high extents of melting. They are also Fe-rich, suggesting deep melting.

9. Melting, Depth, and Temperature Temperature controls the extent and depth of melting: o hotter rising mantle crosses the solidus deeper and ultimately melts more than colder mantle. At the same time, ridges above hot mantle are shallow because the hot mantle is expanded and buoyant. Ridges close to mantle plumes (e.g., Iceland, Galapagos, Azores) are elevated, consistent with the idea that mantle plumes are hot.

10. The Continental Crust The continental crust is the part of the Earth that is most readily sampled and the part with which we are most familiar. It is, however, very likely the most variable part of the Earth in every respect, including compositionally. It is the part of the Earth where geology reveals the planet’s history. In this respect, the continents are arguably the most interesting part of the planet. We’ll first consider the composition of the continental crust, then we’ll see what geochemistry can reveal about its creation and evolution. The continental crust is extremely heterogeneous, thus the task of estimating its overall composition is a difficult one. Furthermore, only the upper part of the continental crust is exposed to direct sampling: the deepest scientific borehole, drilled by the Russians in the Kola Peninsula, reached only 12 km, compared to an average thickness of ~35km.

11. Structure of the Continental Crust Seismic velocity and heat flow indicate that the continental crust is compositionally stratified, with the lower part being distinctly denser and more mafic – richer in Mg and Fe and poorer in SiO 2 and incompatible elements, including the heat producers K, U, and Th. We can divide the problem of estimating crustal composition into two parts: o The “upper”, accessible parts of the crust. Direct observations provide the most important constraints on the composition of this part of the crust. o The lower (sometimes divided between “middle” and “lower”) crust, not readily accessible. Xenoliths, tectonically emplaced portions, seismic velocities, and heat flow data provide constraints. The division of the continental crust into layers is done arbitrarily for convenience. The continental crust does not have a systematic layered structure that resulting from its creation the way that the oceanic crust does.

12. The Upper Crust Several approaches to estimating the composition of the upper continental crust: o Average analyses of samples taken over a large area (first done by F.W. Clarke in 1889). o Mix sample powders to form composites of various rock types and thus reduce the number of analyses to be made. A third approach, pioneered by V. Goldschmidt is to let Nature do the averaging by focusing in sedimentary materials.

13. Upper Crustal Composition from Averages and Composites The first two approaches produce an average upper crustal composition similar to that of granodiorite, with the concentrations of major oxides agreeing within ±10-20%). This is encouraging since granodiorite is the most common igneous rock in the crust.

14. Crustal Composition from Loess The primary problem with sediments is that chemical fractionations are involved in producing sediments from their parents. Glacial loess is less susceptible to this kind of chemical fractionation, though some fractionation nevertheless occurs. Loess is enriched in SiO 2, Hf, and Zr as a consequence of its enrichment in mechanically and chemically stable minerals, such as quartz and zircon. That results from clays being carried further from their site of origin by wind and water. Loess is also depleted in Na, Mg, and Ca, reflecting loss by leaching.

15. Rare Earth Patterns When rock weathers to produce a sediment, the rare earth pattern of the parent is usually preserved in the sediment. This is because the rare earths are concentrated in the clay fraction, which ultimately forms shale. Other Group 3 elements (Sc and Y), as well as Th, behave similarly to the rare earths during weathering. Furthermore, rare earth patterns are remarkably similar in different shales, suggesting shales are indeed good averages of crustal composition.

16. Upper Crustal Composition Most recent estimates of crustal composition are based on a combination of both approaches, together with assumptions about the ratios between elements.

17. Middle & Lower Crust Rocks from the middle and lower crust are typically in amphibolite and granulite metamorphic facies. o Amphibolites are, as their name implies, metamorphic rocks relatively rich in amphibole, a mineral that contains t less water than mica-bearing rocks. o Granulites, on the other hand, are anhydrous, with pyroxene replacing amphibole and biotite. Such rocks are sometimes tectonically exposed and sometimes brought to the surface as xenoliths. o These granulite terranes have often been subjected to retrograde meta­ morphism, which compromises their value. o Furthermore, questions have been raised as to how typical they are of lower continental crust. These questions arise because granulite terranes are generally significantly less mafic than xenoliths from the lower crust. o Xenoliths perhaps provide a better direct sample of the lower crust, but they are rare. Any estimate of the composition of the middle and lower crust will have to depend on indirect inference and geophysical constraints as well as analysis of middle and lower crustal samples.

18. Geophysical Constraints Heat flow. o A portion of the heat flowing out of the crust is produced by radioactive decay of K, U, and Th within the crust. Mantle heat flow also contributes, as does cooling (older crust is cooler). o From heat flow measurements, we conclude that the deep crust must be poorer in K, U, and Th than the upper crust. Seismic velocities. o Seismic velocities depend upon density, compressibility and the shear modulus.

19. Seismic Structure of the Continental Crust Increasing seismic velocities suggest more mafic compositions at depth.

20. Lab Experiments relate seismic velocity to lithology

21. Estimating Deep Crustal Composition Estimates of crustal composition assigned compositions to seismic velocity, then computed weighted averages based on crustal seismic surveys. The estimated composition of the lower crust corresponds to that of tholeiitic basalt; in metamorphic terminology it would be a mafic granulite. The composition of the middle crust corresponds to that of an andesite. At the prevailing pressures and temperatures this rock would be an amphibolite, consisting mainly of amphibole and plagioclase. R & G Middle Wedepohl Lower R & G Lower SiO 2 63.559.053.4 TiO 2 0690.850.82 Al 2 O 3 15.015.816.9 FeO6.07.478.57 MnO0.100.120.10 MgO3.595.327.24 CaO5.256.929.59 Na 2 O3.392.912.65 K2OK2O2.301.610.61 P2O5P2O5 0.150.200.10

22. Rudnick & Gao REE patterns

23. Rudnick & Gao Trace Elements

24. The Total Crust R & GT & MW & TWedeShaw SiO 2 60.657.363.261.563.2 TiO 2 0.70.90.60.680.7 Al 2 O 3 15.9 16.115.114.8 FeO6.79.14.95.675.60 MnO0.10.180.080.100.09 MgO4.75.32.83.73.15 CaO6.47.44.75.54.66 Na 2 O3.1 4.23.23.29 K2OK2O1.81.12.12.42.34 P2O5P2O5 0.1 0.190.180.14

25. Comparing Oceanic & Continental Crust Trace elements provide important hints as to how the crust was made.