The Breakup of Pangea: The North America - Africa Connection

Roy W. Schlische
Department of Geological Sciences
Rutgers University
610 Taylor Road, Piscataway, NJ 08854-8066 USA

By the end of Permian time, ~250 million years ago (Ma), [Figure>>] all of the major continents were assembled into a giant landmass called Pangea (see Figure at right). During this time, eastern North America was situated next to western Africa. Specifically, Nova Scotia, Canada, was adjacent to Morocco, and New Jersey was adjacent to the Western Sahara [Figure>>]. By Late Triassic time (~235 Ma) and possibly earlier, this landmass began to break apart, and the continents began to drift toward their present-day positions over the last 200 million years (m.y.). As these landmasses drifted away from one another, the Atlantic Ocean gradually widened [Figure>>]. How do we know that Pangea [Figure>>] existed? More specifically, how do we know that New Jersey and the Western Sahara lay next to one another?

The first piece of evidence comes from the floor of the Atlantic Ocean. Running roughly in a north-northeasterly direction through the approximate center of the Atlantic Ocean is a huge undersea mountain range known as the Mid-Atlantic ridge [Figure>>]. Much of the rocks exposed on the seabed are an igneous rock called basalt. Farther away from the ridge, the basalt is draped by a veneer of sediments that have settled out of the water column. With increasing distance from the ridge, the thickness of sediment increases [Figure>>]. More importantly, the age of the sediment just above the basalt also increases with increasing distance from the ridge. The basalt underlying the sediment is marginally older than the sediment. Therefore, the basalt is youngest at the ridge itself and gets older with increasing distance from the ridge [Figure>>]. Furthermore, the oldest oceanic rocks are found adjacent to the offshore areas of eastern North America and northwestern Africa.

The Mid-Atlantic ridge is the site of undersea volcanic eruptions and earthquakes [Figure>>]. Here, the crust is being pulled apart. As it does so, molten rock (magma) rises to the surface, erupts or intrudes older rocks, and cools to form new crust. That's why the youngest rocks are found on the ridge itself. As this process is repeated many times, the crust gradually drifts away from the ridge. Because the oldest crust is found adjacent to eastern North America and northwestern Africa, this crust must have formed at a ridge when the Atlantic Ocean was very narrow and then drifted to its current position during the last 175-200 m.y.

These age relationships allow us to reconstruct the positions of the continents at various times in the past. If we were to take a very large knife and cut out all oceanic crust that is 0 to 10 m.y. old on either side of the ridge and then squeeze the rest of the oceanic crust along with the attached continents together, we would know approximately what the continents looked like at 10 Ma [Figure>>]. The problem with this approach is that we may not know exactly how to squeeze the pieces together after removing the cut-out oceanic crust. Fortunately, the oceanic crust contains features called fracture zones [Figure>>]. As crust forms and drifts away from a mid-ocean ridge, it always moves parallel to the trend of the fracture zones, much like a train always moves parallel to railroad tracks. Therefore, when we move the pieces back together, we do so in a direction parallel to the fracture zones [Cartoon figure]. By successively repeating this procedure for successively older segments of the oceanic crust (e.g., 10-20 Ma, 20-30 Ma, etc.), we can eventually restore the continents to their position at the time just before continental breakup [Figure>>].

There are, however, some uncertainties with this approach. First, the age of the oldest oceanic crust is not as well known as other parts of the oceanic crust. Second, because the oldest oceanic crust is covered with a thick blanket of sediments, the trends of fracture zones are not well defined [Figure>>]. To refine our reconstruction for this critical time period, we need to turn our attention from the ocean floor to the continents themselves.

These continents contain a series of rift basins, which are depressions bounded by faults [Figure>>]. A fault is a fracture along which the two sides of rock have slid past one another; the slippage sometimes produces earthquakes [Figure>>]. As Pangea started to break apart, the crust was stretched and thinned (much like a stiff wad of chewing gum), and it fractured in places to produce faults. The specific type of fault associated with a rift basin is called a normal fault. Along a normal fault, one block moves up and the opposite block moves down. The downward moving block forms a depression that can fill with sediments [Figure>>]. Some of the depressions (rift basins) that formed during the breakup of Pangea were quite large, and some contained large lakes, much like the large lakes (e.g., Tanganyika, Malawi) in east Africa [Figure>>], which also are contained within rift basins. Notable rift basins associated with the breakup of Pangea include the Newark basin (located in Pennsylvania, New Jersey, and New York); the Fundy basin (New Brunswick and Nova Scotia, Canada); and the Argana basin (Morocco) [Figure>>].

The layers of rock, called strata, that fill these rift basins provide valuable clues about what the conditions were like when the layers were laid down. For example, the Newark basin contains lots of dark gray and black strata [Figure>>] that accumulated in deep lakes in a semi-tropical setting. The Fundy and Argana basin have strata that contain or once contained evaporites (like salt) that suggest the evaporation of very shallow lakes [Figure>>]. The strata also contain tilted sublayers (e.g., Red Head Member and Tadrat Ouadou Member) identical to what is found inside modern sand dunes [Figure>>]. The dunes and evaporites indicate that the strata in the Fundy and Argana basin formed in an arid setting. All three basins contain one or more lava flow units [Figure>>]. There are three lava flow units in the Newark basin [Figure>>], and one each in the Fundy and Argana basins [Figure>>]. The chemical composition of these flow units in all basins is very similar, suggesting that they were all fed by magma from the same source region in the Earth's interior or possibly even the same magma body[Figure>>]. This would be virtually impossible with the continents in their current position but is much more likely if Africa and North America were adjacent to one another. Radiometric dating of the lava flow units in the Newark and Fundy rift basins yields the same age, ~200 Ma, for all lava flow units.

The lava flow units in the Fundy and Argana rift basin and the oldest flow unit in the Newark basin are stratigraphically located a few meters above a distinctive gray to black shale unit [Figure>>]. The assemblage of pollen species [Figure>>] changes dramatically across this unit, with many species becoming extinct. The sedimentary layers immediately above the boundary are dominated by fern pollen [Figure>>]; ferns are typically the first plants to colonize an area after a catastrophe. Other fossil data indicate that many other plant and animal species became extinct at the time interval marked by the distinctive gray to black shale unit [Figure>>]. This unit marks the boundary between the Triassic (below the boundary) and Jurassic periods. The sheer number of plant and animal species that died off across the Triassic-Jurassic boundary qualifies it as a mass-extinction event [Figure>>]. [A mass extinction also occurred at the Cretaceous/Tertiary boundary. This mass extinction was caused by an asteroid impact and caused the extinction of the dinosaurs, among numerous other plants and animals. The pollen assemblages above the Cretaceous/Tertiary boundary are also dominated by ferns.] Because the Newark, Fundy and Argana basins (among other basins) all contain the distinctive gray to black shale always located a few meters below the basalt, we know that all these basins recorded the mass extinction event and then, very shortly thereafter, the eruption of the basalts.

As noted above, the stratigraphy of the Newark, Fundy, and Argana basins in the vicinity of the Triassic-Jurassic boundary [Figure>>] is very similar. In fact, large parts of the entire section of rocks in the Fundy and Argana basins are quite similar). This suggests two hypotheses: 1) the rocks units were deposited in a single rift basin which then got split apart when Africa and North America separated; or 2) the Fundy and Argana basins were separate basins, but the same conditions affected both basins and, therefore, their rock units are quite similar. Hypothesis 1) is not favored because both basins are bounded by separate faults [Figure>>], which is consistent with Hypothesis 2. As discussed previously, both basins contain sand-dune deposits [Figure>>] and evaporites, which suggest that both basins were located in an arid climatic region.

Independent evidence from paleomagnetism supports the hypothesis that the Fundy and Argana basins lay adjacent to one another in Triassic/Jurassic time and that both basins lay in an arid climatic belt. Paleomagnetism is based on the principle that some rocks--when they cool from a melt or are deposited--retain an impression of the orientation of the Earth's magnetic field. Most of us are familiar with the effect of the magnetic field on a compass: the field causes the compass needle (which is attached to small bar magnetic) to point north. What most of us are not aware of is that Earth's magnetic field also wants to cause the compass needle to point toward or away from the Earth. The angle between the inclined compass needle and a line that is perpendicular to the Earth's surface is called magnetic inclination. Magnetic inclination varies systematically with latitude [Figure>>]. At the Earth's magnetic pole, the inclination is 90° (a compass needle would point straight down); at the Earth's magnetic equator, the inclination is 0° (a compass needle is parallel to the Earth's surface). Everywhere else between the equator and the pole, the inclination is greater than 0° but less than 90°, with the inclination progressively increasing with increasing distance from the equator. In rock samples containing a fossil magnetism, the inclination can be measured, and this can be used to determine the latitude at which the rock formed or was deposited.

Paleomagnetism of rocks from around the Triassic-Jurassic boundary (~200 Ma) in the Fundy basin indicate that the Fundy basin was located at a latitude of ~15° north of the equator; the Argana basin was located at ~13.5° north of the equator [Figure>>]. Thus, both basins were situated at a very similar latitude. This latitude falls within the arid climatic belt, which is consistent with the rocks which accumulated in these two basins. This similar latitude supports the reconstruction of Pangea [Figure>>]. Interestingly, the present-day latitude of the Fundy basin is ~45° north of the equator. Thus, in the last 200 m.y., the Fundy basin has drifted northward by ~30° of latitude, which is equivalent to about 3330 km. The average rate of movement was 1.67 cm/yr. We must point out that paleomagnetism only deals with latitudes and northward or southward movements. Thus, 3330 km only reflects the northward component of the movement of North America; the eastward or westward component of movement is unknown.

In contrast to the ~15° latitude of the Fundy basin, the Newark basin was situated at a latitude of ~10° at 200 Ma. The present-day latitude of the Newark basin is ~40°. Therefore, the Newark basin has also drifted northward by 30° or 3330 km. Because both the Newark and Fundy rift basins have drifted northwards by the same amount, we know that both basins have always been part of the same continent, North America, which drifted northward as a coherent block. At 200 Ma, the Argana basin was located at ~13.5° latitude. Its present-day latitude is ~33.5° (the present-day latitude of the Argana basin shown in the figure is given in "North American coordinates"). Thus, the Argana basin (and Africa) have drifted northward by ~20° of latitude or 2220 km, more than 1100 km less than North America. This is excellent independent evidence that Africa and North America were once joined but have gone their separate ways.

As we noted earlier, the Newark basin contains some very deep-water lake deposits (Lockatong Formation) [Figure>>] whereas the Fundy and Argana basins contain evaporites and sand-dune deposits. The Lockatong Formation (the gray shaded interval in the Figure>>) was deposited (~223 Ma to ~218 Ma) when the Newark basin was located at a latitude of 5°-7°. This would put the Newark basin in the tropical climatic belt. Rainfall is high in the tropics, and this would favor the formation of deep lakes. In contrast, rocks of the same age in the Argana basin were deposited at a latitude of ~10°, near the edge of the arid climatic belt. Consequently, rocks that are ~220 Ma in the Newark basin are quite different from rocks of the same age in the Argana basin, mainly because the two sets of rocks accumulated in different climatic belts. To find rocks in Africa that are similar to the deep-water lake deposits of the Lockatong Formation, we would have to look in rift basins in the Western Sahara, which lay adjacent to New Jersey's Newark basin ~220 Ma [Figure>>]. Unfortunately, there are no rift basins exposed in the Western Sahara, but there are rift basins underneath the marine waters of offshore Western Sahara.

As we discussed above, the basaltic rocks in the Newark, Fundy, and Argana basins are very similar. Although the Newark and Fundy basins were separated by ~4.5° of latitude and were in different climatic belts, basalt--which is an igneous rock--is not affected by climate, and thus should not be different in different locations, assuming that the basalts came from the same source region or magma body. Because the chemistry of the basalts from the different basins is so uniform [Figure>>], we know that the basalts came from the same source region or magma body, which must have been at least 4.5° of latitude in length (500 km). Because similar basalts are found in the Culpeper rift basin in Virginia, the source region or magma body must have been at least 720 km long (i.e., the northward distance between Virginia and Nova Scotia, Canada).

The eruptions of basalt occurred on a massive scale. We know from the geological record that the basalts in the various rift basins are always located a few meters above the Triassic-Jurassic boundary [Figure>>]. Therefore, the eruptions occurred AFTER the extinction event and could not have caused them. There are, however, very large bodies of basaltic rock buried beneath very thick sediments (and ocean water) along most of the eastern continental margin of North America [Figure>>]. These basaltic rocks have not been dated precisely, and therefore it is possible that some or all of them are older than the Triassic-Jurassic boundary and the mass-extinction event. We all know that eruptions can be lethal, mainly due to the explosive nature of some volcanic eruptions). However, the highly fluid basaltic lava generally erupts non-explosively. Nonetheless, the secondary effects of these massive outpourings of lava can be lethal. For example, the 1783 Laki eruption in Iceland [Figure>>] poured lava out of giant cracks in the Earth that were about 30 km long [Figure>>]. These eruptions of basalt also emitted large quantities of volcanic gasses [Figure>>] including sulfur dioxide (which, when mixed with water, produces sulfuric acid), fluorine, and chlorine. These noxious compounds resulted in the death of ~50% of the livestock in Iceland in 1783. Consequently, ~20% of the Icelandic population starved to death. One can only imagine what the massive effects would be a giant basaltic eruption in a region that is at 100's of kilometers long, as was the case in eastern North America 200 m.y. ago [Figure>>]. On the other hand, the very large bodies of basalt in the offshore region of eastern North America could be younger than the Triassic-Jurassic boundary. In that case, the mass extinction must have been caused by some other mechanism, perhaps an asteroid impact [Figure>>], as was the case at the Cretaceous-Tertiary boundary. No impact crater of the correct age (~200 Ma) has yet been discovered. However, because ~70% of the Earth is covered by oceans, an oceanic impact site would be difficult to find. Furthermore, the vast majority of oceanic crust 200 m.y. old would have been subducted at oceanic trenches where one plate dives below another. An iridium anomaly at the Triassic-Jurassic boundary section in the Newark basin [Figure>>] could have been produced by an asteroid impact. More research is needed to determine the cause or causes of the mass extinction at the Triassic-Jurassic boundary.

In summary, we have discussed various lines of evidence, from both the ocean floor and the continents themselves, that suggest that North America and Africa lay next to one another around 200 Ma. In the intervening time, these two originally joined continents broke apart and drifted away from one another, as the Atlantic Ocean opened and widened. The Fundy rift basin (Nova Scotia, Canada) and the Argana rift basin (Morocco) were originally adjacent to one another, and contain similar types of rocks deposited in an arid setting. Rocks in the Newark basin accumulated in a more tropical setting. All three basins contain lava flows derived from a very large source region or magma chamber in the Earth's interior. The lava poured out ~200 Ma, just after a mass extinction event that wiped out many plant and animal species. This period in Earth's history was certainly an interesting time, and we are fortunate to have an excellent geologic record of it preserved in New Jersey's Newark rift basin.

Introductory Geology Lab Website>>
Rutgers Geology Home Page>>