The Orinoco Flow: sediment subduction and the Lesser Antilles volcanic arc

Radiogenic isotopes and the origin of volcanic island arcs

Isotopes are amazing geochemical tools. They can be used not only to date minerals radiometrically but to constrain small and large-scale geological processes. The subduction of oceanic sediments into the mantle is one such process, about which geologists want to know: what is the ultimate fate of ocean-bottom sediments? For example, what portion of these sediments is taken down into the mantle? How much is scraped off onto the accretionary prism? And finally, what portion of oceanic sediments are melted and returned to the surface through volcanism?

Some of the most important isotopic ratios used to answer these questions are also used in radiometric dating: 87Sr/86Sr, 143Nd/144Nd, 206Pb/204Pb, 207Pb/204Pb, 208Pb/204Pb, 176Hf/177Hf, and even 10Be/9Be. In each of these ratios, one isotope is radiogenic (produced by radioactive decay) and the other is stable. This ratio changes over time, therefore, based on the concentration of the radioactive parent element (Fig. 1). Since the isotope 87Rb decays into 87Sr, for example, the 87Sr/86Sr ratio will increase at a faster rate in minerals with a higher concentration of Rubidium than in minerals with zero Rubidium (i.e. if no Rb is present, the 87Sr/86Sr ratio will never change).

Figure 1: Isotopic evolution of 87Sr/86Sr over time in various Earth reservoirs.

The application of this principle to geochronology is straightforward: if the original isotopic ratio and decay rate can be determined, then so can the age of the mineral. But what does this have to do with plate tectonics, subduction zones, and island arcs in particular?

Consider the Lesser Antilles volcanic arc—that parabola of paradise off the north coast of Venezuela. These islands have been forming through volcanic eruptions over the past tens of millions of years, as Cretaceous-aged basalts and sediments were subducted beneath the Caribbean tectonic plate (Fig. 2). A portion of these rocks/sediments then began to melt upon reaching a depth of ~100 km, causing magma to ascend toward the surface and mix with the upper mantle (check out a visual explanation here). Since the subducted crust/sediment and the upper mantle are of different age and composition, their isotopic ratios will also be distinct (remember how and why these isotopic ratios change?). The final product (surface volcanic rocks) depends of the relative contribution of each, much like mixing yellow and blue paint in varying quantities to produce different hues of green. This geochemical variation is the key to answering the aforementioned questions for each volcanic arc around the world.

Figure 2: Age of basaltic oceanic crust (K-Ar method) currently being subducted beneath the Lesser Antilles island arc. Note that radiometric dates increase systematically from north to south, as expected from the geometry of the crust (distance from the spreading center) and the current plate movement of ~2 cm/yr. In other words, rocks that are 400 km further from the spreading center (the star marked 'Barbados' vs. the northernmost age) are about 20 million years older (~400 km divided by 2 cm/yr is 20 Ma; 105 Ma - 83 Ma = 22 Ma). From Carpentier et al. (2008).

From mantle to crust

Now take a step back in time, to an ancient Earth where massive continents do not yet exist. Much of the Earth's upper mantle would have had a very similar isotopic composition for elements such as Sr, Nd, U, and Pb. As early as ~4.3 billion years ago, basaltic crust began to crystallize at the surface. In the 'short' interval between ~3.5 and 2.5 billion years ago, a majority of continental crust also formed, and the mantle and crust became geochemically isolated reservoirs. Since the process of crystallization always prefers certain elements over others (smaller elements over bigger ones), each reservoir developed unique relative abundances of parent/daughter pairs. For example, the relative abundance of Rb vs. Sr is higher in the crust than in the mantle, so the isotopic ratio 87Sr/86Sr increases faster over time in crustal rocks than in the mantle. On the other hand, Sm vs. Nd is higher in the mantle than in the crust, and so the isotopic ratio 143Nd/144Nd increases faster in the mantle than in crustal rocks. Since 238U decays to 206Pb, relatively high abundances of Uranium (as in crustal rocks) contribute to higher 206Pb/204Pb ratios in continental crust.

In short, high 87Sr/86Sr and 206Pb/204Pb ratios accompanied by low 143Nd/144Nd ratios are good geochemical signatures of very old continental crust. Unless, of course, one can propose a viable alternative mechanism to explain their origin and evolution.

Sediment contribution over time to the Lesser Antilles archipelago

In the following graphics, one can see how geologists apply these isotopic systems to understand the contribution of subducted sediment to volcanic eruptions. First, we need to determine the 'hue' of our yellow and blue paints (i.e. the mixing 'end-members'). On the one hand, we have upper mantle that resides directly beneath the volcanic islands. The isotopic composition of the upper mantle is relatively easy to determine, since it is continually sampled and preserved by Mid-Ocean Ridges (MORs). Dredging ships frequently collect samples of Mid-Ocean Ridge Basalt (MORB), which are then analyzed for their isotopic and elemental compositions, as well as their magnetic polarity and age. Not surprisingly, MORBs from around the world are geochemically very similar. Furthermore, their distinct geochemistry is consistent with theories that the chemistry of the upper mantle is due to the formation of oceanic crust early on in Earth history (see previous section).

Our second mixing end-member is formed by clay-sized particles that sink to the bottom of the ocean. The origin of these particles is found on continents, where old continental crust is weathered and carried out to sea by major rivers. Fortunately, the ocean sediments being subducted beneath the Lesser Antilles have also been sampled in many places by the Deep Sea Drilling Project (the stars in Figures 2 and 3 represent such sites) and analyzed geochemically.

To determine the extent of mixing between these end-members, volcanic rocks are sampled along the arc. Figure 3 is taken from one study that focused on the petrologically diverse island of Martinique, which developed in multiple stages of a long eruption history. Since the volcanic island formed over the past 25 million years, the authors were able to address this question for much of the Cenozoic.

Figure 3: From Labanieh et al. (2010). Map of the Lesser Antilles volcanic arc and Martinique island, for which three stages of the island's volcanic history are delineated. Sampling sites of ocean sediments (stars) and volcanic rocks (see inset) are also shown. As an aside, note the width of the accretionary prism (distance between the 'Subduction trace' and the 'Prism front'). How long must this subduction zone have been collecting sediment?

One can predict the composition of the final product by plotting the isotopic compositions of both end-members on a graph with isotopic ratios on each axis (Fig. 4). The more that subducted sediments contributed to magma generation in the subduction zone, the closer those values will plot to modern isotopic ratios of oceanic sediments (dark gray fields):

Figure 4: From Labanieh et al. (2010). Four isotopic cross plots from Martinique Island,  indicating a significant contribution of oceanic sediments during magma generation (up to 20%). Red/Blue lines are mixing lines between MORB values and ocean sediment values. The range of compositions of Mid-Ocean Ridge Basalts between 30°N and 30°S are shown in light gray fields.

It is evident from the examples above that the composition of volcanic rocks on Martinique Island are due to a mixture between the upper mantle and subducted oceanic sediments. One can also see how the extent of mixing is variable in time and space (Fig. 5). For example, note how the mixing lines—blue vs. red—shifted slightly when the arc itself moved west (cf. Fig. 2). This westward shift in volcanism was possibly due to a shallowing of the angle of subduction following the subduction of an aseismic ridge around 6–7 Ma (Labanieh et al., 2010). In any case, the isotopic mixing trends seen at Martinique Island are common to the rest of the islands of the Lesser Antilles and are prime examples of this phenomenon of plate tectonics. These data also corroborate the theory elegantly, since they are predicted directly by it.

Figure 5: From Labanieh et al. (2010). Relative contribution of oceanic sediments over time, based on K-Ar dating of individual volcanic rock samples. After 5.1 Ma, the end-member composition changed, possibly due to the subduction of an aseismic ridge on the Atlantic Plate.

The Orinoco Flow

Figure 6: From White et al. (1985). Isotopic con-
tours show more radiogenic Sr and Pb in front of
the Orinoco River delta. Results are similar when
Nd isotopes are plotted.

What is the source of deep-ocean sediments being subducted between the Caribbean tectonic plate? One key to answering this question is the geographic variation in isotopes along the Lesser Antilles arc. Nearly 30 years ago, White et al. (1985) documented the modern influence of the Orinoco River on the isotopic composition of sediments lying immediately in front of the Caribbean Plate (Fig. 6). Since the Orinoco River watershed drains sediments derived from very old continental crust (the Archean-aged Guiana Highland), 206Pb/204Pb and 87Sr/86Sr ratios become higher as one moves closer to the river delta. We can predict, therefore, that these isotopic ratios are higher in volcanic rocks from the Southern Islands than in those from the Northern Islands.

As it turns out, that is precisely what White et al. (1985) and later researchers found. Although the debate regarding the significance of this mechanism is not yet fully settled (Carpentier et al., 2008), it is generally agreed that the Orinoco River basin has long been in place (since the Cretaceous) and contributed sediments enriched in radiogenic isotopes to the subducting slab.

Figure 7 represents a cross-plot of Nd and Pb isotopic data from the island arc. As you may recall, the relatively high 206Pb/204Pb and low 143Nd/144Nd ratios are characteristic of old continental crust. Such values (note the Site 543 sediments field) are represented more by the Southern Islands than the Northern Islands. This trend suggests that the geochemistry of volcanic rocks along the Lesser Antilles arc is strongly determined by the geochemistry of oceanic sediments in front of the arc.

Figure 7: From Carpentier et al. (2008). Cross plot of Nd and Pb isotopes showing greater contribution of oceanic sediments in the Southern Islands than in the Northern Islands.

To add some icing to our petrological cake, Carpentier et al. (2008) also demonstrated that Uranium-rich black shale units, which were deposited during the Late Cretaceous Oceanic Anoxic Events (OAE I and II), could account for anomalously high 206Pb/204Pb in the Southern Islands, since they are present only in sedimentary strata to the south. Over time, 238U decays to 206Pb, so the ratio of 206Pb/204Pb will increase at a faster rate in U-enriched sediments relative to the crust or mantle. What does all this mean? Even the islands of the eastern Caribbean can attest that the oceanographic events responsible for these black shales (a long-term reduction in available oxygen to the oceans) occurred more than 80-million years ago, while the Atlantic plate was being subducted slowly (about 2 cm/yr) below the Caribbean plate.

Conclusions and implications for YEC

Corroborative evidence is still being discovered for Plate Tectonic theory, which revolutionized geological disciplines in the 1970's. The Lesser Antilles volcanic island arc is a prime example of the phenomena associated with the subduction of oceanic crust and sediments. Combined isotopic data from numerous elements demonstrate how conventional theories regarding the Earth's age and history can be tested through multiple, independent methods. Simultaneously, one can see how such theories are potentially falsifiable, since they were formulated before the isotopic data were available, yet were able to predict trends in those data. If you were able to follow the examples above, I hope you will better understand how geology has progressed scientifically (rather than through conspiracy and ideology).

Young-Earth Creationism has attempted to keep up with modern geology by transforming Plate Tectonic theory (which describes a notably slow process) into 'Catastrophic' Plate Tectonics—a process that barely works on paper, but so rapidly that it would have boiled off most of the oceans. The latter is hotly debated within YEC and not accepted by many. Personally, I would predict its disappearance in the near future if it weren't so necessary to accepting the obvious evidence for Plate Tectonic theory. It will be interesting to see whether YECs make any attempt to adapt to the flood of new evidence for the theory's more conventional version.

Despite the fact that Catastrophic Plate Tectonics is an abstract, geophysical model (i.e. difficult to grasp or critique by the layman—myself included), one can ask whether it accurately predicts the relevant evidence. For example, how does Catastrophic Plate Tectonics explain the isotopic data presented above from the Lesser Antilles? How did the South American continent crystallize from molten rock (a slow process in itself), decay radioactively to produce distinct Nd, Sr, and Pb isotopic values (a 2.5 billion-year process), deform structurally (to form major river basins like the Orinoco), weather extensively (to fill the North Atlantic basin with fine-grained sediment, which typically needs thousands of years to settle at the ocean bottom), and yet be covered with miles of Phanerozoic sediments? When the big picture is considered as a whole, the YEC mentality becomes a fleeting illusion.

Radiogenic isotope data from the Lesser Antilles arc seem to indicate that oceanic sediments have long been weathering out of the Orinoco River watershed, subducted beneath the Caribbean Plate, and melted and incorporated into the magma that produced the idyllic islands of the eastern Caribbean. Radiometric dating of volcanic rocks and oceanic crust corroborate the picture interpreted from radiogenic isotope data (Fig. 2), and are even predicted by modern rates of subduction and distance from the trench to the spreading center. The Beryllium-10 method (which I will discuss in a later post) provides an additional test to determine that the process of subduction occurred over several million years rather than several hundred.

Catastrophic Plate Tectonics and YEC cannot account for the isotope trends in volcanic arcs, because they provide no mechanism by which distinct isotopic signatures form in a matter of hundreds or thousands of years. Furthermore, these paradigms cannot explain the origin of massive amounts of continentally derived sediment, which were subducted beneath oceanic plates before the modern tectonic picture formed (i.e. before the Flood). I will return to this point in the next post when I discuss Dr. Andrew Snelling's article entitled "The Relevance of Rb-Sr, Sm-Nd, and Pb-Pb Isotope Systematics to Elucidation of the Genesis and History of Recent Andesite Flows at Mt. Ngauruhoe, New Zealand, and the Implications for Radioisotopic Dating", in which he cunningly misrepresents the geochemical methods employed to understanding volcanic arc systems.

If this post was sensible to you, his errors will be obvious.


References Cited:


Carpentier, M., Chauvel, C., and Mattielli, N., 2008, Pb–Nd isotopic constraints on sedimentary input into the Lesser Antilles arc system: Earth and Planetary Science Letters, v. 272, p. 199–211.

Labanieh, S., Chauvel, C., Germa, A., Quidelleur, X., and Lewin, E., 2010, Isotopic hyperbolas constrain sources and processes under the Lesser Antilles arc: Earth and Planetary Science Letters, v. 298, p. 35–46.

White, W.M., Dupre, B., and Vidal, P., 1985, Isotope and trace element geochemistry of sediments from the Demerara Plain region, Atlantic Ocean: Geochimica et Cosmochimica Acta, v. 49, p. 1875-1886.