Monday, November 20, 2017

The Geology of Iguazú Falls of South America: Part II - The Paraná-Etendeka LIP, Mass Extinctions, Supercontinental Break-up, Hotspots, "Not-spots" and the Evolution of a Spectacular Waterfall

"Meditate and feel the emotion
contemplating the vibrant climax
that surrounds the eternal mist.
And do not try to describe it
Only tilt your head forehead into the abyss
that is the reflection of God."

Closing stanzas of the poem Garganta del Diablo by Alfonso Ricciutto (Argentina, 1950)

The grandeur of Iguazú Falls in central-eastern South America is vastly enhanced if one looks beyond its stunning beauty. Even though it's a recent addition on the landscape with geomorphology that is simple in construction, its evolution is the product of a long succession and complex interaction of tectonic events and geologic processes that span over a billion years. The revelation adds a depth of color far beyond the visual and enriches our understanding of the natural forces at work that shape our planet. 

Part I (here) - From the late Precambrian through the late Mesozoic, three supercontinents in succession endowed the region of Iguazú Falls with a Late Proterozoic crystalline foundation, an overlying Paleozoic sag basin and a capping veneer of Early Cretaceous volcanic rocks. 
Part II - The Paraná-Etendeka Volcanic Province has much to tell about the genesis of regional uplift, mass extinctions, mantle dynamics, supercontinental fragmentation and the creation of the South Atlantic realm that set the stage for the creation of the falls in the Pleistocene.
Part III, forthcoming, concludes with a photographic glimpse of the astoundingly rich, bidiverse and colorful rainforest that encompasses the region.

Pertinent definitions are italicized and important names are emphasized in bold. Photographs were taken on a recent visit to the falls in February 2017.

Comparisons of the world's greatest waterfalls invariably include elevations, widths, freefall, flow rates and volume. With a height of 269 feet, width of 9,500 feet and an average volume of 61,660 cfs that pours over a complex of some 275 discrete waterfalls at normal flow, Cataratas del Iguazú in Spanish) or Cataratas do Iguaçu Falls in Portuguese straddling Rio Iguazú on the border of Brazil and Argentina in south-central South America is at or near the top of the list. What's more, it's likely the most spectacular.

Pronounced eee-gwa-ZOO from the native Tupi-Guarini language, it means Great (guassu) Water (y). With an annual mean in excess of 1.5 million visitors in Argentina and Brazil, two-thirds of the falls are on the Argentinian side within 2,500 cu km sister national parks of both countries, all within a 1984-6 UNESCO World Heritage site. 

Copy the following co-ordinates into an on-line mapping program such as Google Earth and travel to the brink of Iguazú: 25°41'36.37"S, 54°26'16.33"W

South View of Rio Iguazú, the Falls and Downstream Canyon
Iguazú straddles the border of the northwestern corner of the Argentine Province of Misiones and southwestern corner of the Brazilian state of Paraná in central-southeast South America. On the Argentine side, three metallic walkways lead you to the brink at Paseo Garganta del Diablo, Circuito Superior and Circuito Inferior. The Brazilian side has an elevated promontory and overlooks.
Modified from Wikimedia Commons

The austral-sprawling supercontinent Gondwana amalgamated in the early Paleozoic from cratonic building blocks largely acquired following the Late Proterozoic disassembly of Rodinia, its massive supercontinental predecessor. In addition to closure of intervening seas and the incorporation of unifying mobile belts (ancient continent-forming, mountain-building orogenies), Gondwana's disparate landmasses came together by ~530 Ma. 

Gondwana was sutured together from two landmassesWest Gondwana (present-day South America and Africa) and East Gondwana (Antarctica, Australia, India and Madagascar). Surrounded by the Panthalassic Ocean in the South Hemisphere, it remained insular until the late Paleozoic when it amalgamated with Laurentia (the ancient cratonic core of North America) to form Pangaea, Gondwana's successor.

Final Stages of the Latest Proterozoic Fragmentation of Rodinia
The Late Proterozoic supercontinent's dissassembly was a protracted event that lasted over 100 Ma and ended ~600 Ma. Disparate rifted landmasses drifted across the globe with the two largest forming equatorial-positioned Laurentia and South Pole-sprawling, massive Gondwana. The latter began to assemble even as Rodinia began to break apart and ultimately amalgamated from East and West halves by ~540 to 530 Ma.
Modified from Paleontology

A number of Rodinia-acquired, stable and rigid Archean to Early Proterozoic cratonic fragments (old and stable continental lithosphere) - the Paranapanema block in particular - became conjugate neighbors within Gondwana. Almost from the time of assembly in the Early Cambrian, West Gondwana's margins were accosted by subducting oceanic lithosphere. It resulted in extension within the supercontinent's interior that formed a jigsaw puzzle mosaic of sag basins (aka cratonic, intracratonic, intracontinental basins). Most of the contiguous basins remained associated through a long succession of assembling and disassembling supercontinents.

Assumedly floored by Paranapanema gneissic-granitic crust, the Paraná Basin initiated formation in the Ordovician (red arrow). Minimally deformed and saucer-shaped, the intracratonic-intracontinental basin exhibited crustal thinning and subsidence that began continued through the Cretaceous (Please see post Part I here for more details). 

Crustal Building Blocks of Cambrian-Amalgamated West and East Gondwana
Along with genetically similar sag basins, the Paraná Basin (arrow) formed over an assumed Rodinia-acquired, granitic-gneissic Paranapanema crustal foundation within West Gondwana. Its association with contiguous basins was perpetuated following the break-up of Rodinia. 

Modified from Cordini et al

Six supersequences (unconformity-separated, fault-bound, multi-member sedimentary packages at ~10 million year intervals) accumulated within the Paraná Basin. The ~6,000 meter-thick stack was relatively undistorted during most of the Phanerozoic, as subsidence provided the massive accommodation space for deposition.

At first, the Paraná Basin initiated within West Gondwana as an embayment (gulf opening) of the Panthalassa Ocean but became entrapped within the supercontinent's interior, where it remained throughout its history. Basin-fill occurred not only while Gondwana was in the Southern Hemisphere, in tectonic transit across the equator and amalgamated with Pangaea in the late Paleozoic. The supersequences (below) record a host of tectono-climatic lithologies that include shallow marine, coastal and deltaic, fluvial to lacustrine and desert through glacial.

NW to SE Schematic Cross-section through the Uplifted Paraná Basin
Cratonic basins within central West Gondwana, subsiding subsequent to collisional events that occured on the periphery of the supercontinent, subsided as sag basins. It provided accommodation space in the case of the Paraná Basin for six supersequences from the Ordovician through the Early Cretaceous, when it acquired flood basalts of the Serra Geral Formation.

From Mineropar and Milani and Zalan, 1998

The Paraná's sixth and final supersequence - the Serra Geral Formation (brown stratum) - occurred in the Early Cretaceous. Lavas emanated from fissures fed by a vast system of dikes and sills and blanketed the basin's uppermost sedimentary strata. It's one of the planet's greatest manifestations of continental flood basalts

It not only ended the basin's protracted fill-history but heralded a period of large scale global tectonic change that continues to this day - the opening of the Atlantic Ocean and dispersal of the world's continents. And most relevant to this post, the process(es) that resulted in the extrusion of the flood basalts uplifted the Paraná Basin and directed its nascent fluvial system continentward. But of course, I'm getting ahead of myself.

Isla San Martin within the Channel of Rio Iguazú below the Falls
Although heavily vegetated on its uppermost surface, the island beautifully displays columnar basalts of the Serra Geral Formation. Zodiac tours reach the island by negotiating the rapids of Rio Iguazú Inferior and allow visitors to ascend stairs and view Iguazú Falls its Throat of the Devil from below.

Serra Geral lavas not only flooded the surface of the Paraná Basin in South America but its basinal counterpart in the Etendeka region of western Africa with the Awahab and Tafelberg Formations. Emplacement occurred before West Gondwana rifted apart and the South Atlantic had begun to open. The two-country region is referred to as the Paraná-Etendeka Volcanic Province (purple blob) and includes a number of enigmatic but genetically related volcanic features on the intervening Atlantic seafloor. 

Incipient Break-up of Pangaea (~120 Ma) and the Paraná-Etendeka LIP
In the Lower Cretaceous, Pangaea had already begun to fragment apart with the opening and subsequent closure of the earlier Paleo- and later Neo-Tethys oceanic realms. The event geometrically and kinematically correlates with opening of the Atlantic Ocean that followed, which occurred in overlapping, discontinuous, rate-variable stages beginning with the Central Atlantic and progressing to the South and North regions. 

It's one of the world's largest preserved and well-exposed Large Igneous Provinces (LIP). Earth history has been punctuated with a number of these flood basalt episodes of high volume, short-lived magmatism that don't fit into the current tectonic plate paradigm (Keep reading!). With a present-day volume in excess of 1,500,000 cu km, the timing and duration of Paraná-Etendeka magmatism has been a matter of debate with the main pulse at ~132 to 134 Ma and lasting from ~1 to 17 Ma, which is a relatively short geological interval at either extreme. With the South Atlantic opening between South America and Africa shortly afterward at ~120 Ma, the timing has fueled a number of genetic hypotheses regarding LIPs, rifting, seafloor spreading and even mantle architecture.

Rifting divided the volcanic province with the Paraná Basin receiving 95% of magmatism in Brazil, Argentina, Paraguay and Uruguay and the remainder in Namibia and Angola. The rift-separated provinces consist of nearly coeval and lithological and geochemical counterparts. Volcanic rocks, however, are formed of heterogenous lava packages with distinct architectural, morphological and geochemical differences. What might this imply?

Distribution of Serra Geral Igneous Rocks and Main Tectonic Structures
 The continents of South America and Africa are shown in an early rift, early-drift reconstruction with the South Atlantic initiating opening. The bulk of the Paraná-Etendeka LIP is located within the Paraná Basin. Its pre-volcanic sediments are blanketed by volcanic rocks of the Serra Geral Formation. Geosyncronous, syntectonic and nearly geochemically equivocal magmatic counterparts are within the formerly juxtaposed Etendeka region of African Namibia and Angola. Notice rhyolitic rocks (pink Acidic Units) along both continental margins and the two regions based on TiO2 basaltic geochemistires (blue and green PEIP). 
See text below for explanation.
From Rosetti et al

Compositionally, they're basalts and basaltic andesites (97.5%) with minor quantities of acidic rhyolites (2.5%). Typical of many LIPs where silicic components are often present, Paraná-Etendeka's rhyolitic rocks (pink map regions above) are located near southeastern South America's and southwestern Africa's continental margins from its pre-rift locale. 

Together, they cover the bimodal compositional spectrum (red ellipse below) from mafic ("basic"/basaltic) to felsic ("acidic"/granitic) igneous rocks. The association is not unique to the Paraná-Etendeka Province, which is an extensional, mafic-dominated, continental rift regime. Bimodality is also recognized in a number of tectonic settings such as collisional felsic-dominated subduction zones. And, there's more!

In two distinct regions within the Paraná-Etendeka Volcanic Province, tholeiitic basalts (the most common volcanic rock and produced at mid-ocean ridges and continental rifts) exhibit two chemically similar subtypes that overlap temporally and spatially: a low-TiO2 lava field in the south and high-TiO2 in the north (blue and green regions above). The northern phase appears younger than the southern and Etendeka, which is also stratigraphically demonstrable. What is the genetic implication of the subtypes and their distribution?

Bowen's Igneous Rock Reaction Series
Discovered by geochemist Norman Bowen in the 1920s and 30s, the series depicts the production of igneous rocks and their mineralogical components during crystallization (cooling). The process has profound implications for the fundamental physics of the Earth and the evolution of igneous rocks. Paraná-Etendeka volcanic rocks are bimodal in their chemistry and distribution across the province's landscape.

There are a number of explanations for the origin of LIPs, the wide range of magma chemical compositions and source of magmatism such as plumes, hotspots, mantle or crustal sources and decompression melting. Regardless of the diverse theories, continental volcanic margins exhibit thinning and extension that led to break-up, seafloor spreading, the formation of oceanic crust and passive margins on the continents that of the newly formed Atlantic realm. 

• What mantle process(es) explains Serra Geral geochemistries, distribution, timing and duration of emplacement?  
• Does bimodality suggest a single volcanic source? 
• Did LIP emplacement trigger rifting and South Atlantic opening? On a grander scale, is the process responsible for the break-up of Pangaea and opening of the North and Central Atlantic?
• What are some of the mantle models for Paraná-Etendeka magmatism? Was magmatism attributable to decompression (upwelling magma-melting due to less pressure), conductive heating or the ascent of an asthenospheric plume that triggered lithospheric mantle melting?
• Does partial melting (a portion of solid rock melts and forms minerals different from the solid) of coevally emplaced basaltic rocks explain the presence of silicic rocks? 
• Both basaltic and alkaline igneous complexes are asymmetrically distributed across the LIP, while silicic magmatism is more symmetrical at the site of continental break-up. Does it suggest a link to lithospheric extension?
• What does the V-shaped orientation of seafloor magmatism imply about the process that formed them? If the South American and Africa plates are diverging, why aren't the ridges aligned linearly? 
• Does the LIP's south to north age progression of magmatism reflect the direction of rifting?

What's Down There?

Although not without controversy, the consensus is that the Paraná-Etendeka LIP was fed by the Tristan mantle plume. Its name is derived from the Tristan da Cunha volcanic island group and adjacent Gough Islands. They are located some 500 miles east of the South Mid-Atlantic Ridge spreading center, the divergent boundary between the North American and Eurasian plates) where the plume is currently thought to reside. They are all associated with the opening of the Atlantic and represent a single genetic province some 9,000 km broad. 

The Paraná and Etendeka LIP includes two V-shaped, aseismic (without seafloor spreading or earthquakes except at the hotspot) submarine volcanic chains on opposite sides of the Mid-Atlantic Ridge. They are the NW-oriented Rio Grande Rise on the west and NE-oriented Walvis Ridge to the east, respectively on the South American and African Plates. The assumption is they all arose from a single source - a deep mantle plume.

Simplified Structural Map of the South Atlantic Ocean and the Paraná-Etendeka Volcanic Province
Formerly juxtaposed, the Paraná (P) and Etendeka (E) Volcanic Provinces lie across the opening Atlantic realm. The Tristan hotspot (T) is thought to initially lie beneath the Mid-Atlantic Ridge and is responsible for the LIP and Rio Grande Rise and Walvis Ridge. The Paraná-Etendeka LIP is also referred to as the South Atlantic Igneous Province, which is named in accordance with the North and Central Provinces and their similarly named LIPs.

The majority of volcanoes are located at plate boundaries - mid-ocean ridges and subduction zones. About 5% are within plates such as the Hawaiian Islands. The magmatism is classified as a hotspotstationary, time-progressive regions of local surface volcanism. Some are coincident with or near mid-ocean ridges - such as Iceland and the Paraná-Etendeka LIP - and are thought by many to be produced by mantle plumes. 

Hotspots and the Plume hypothesis have become incorporated into Tuzo Wilson's almost universally accepted 1960s Theory of Plate Tectonics (a rigid outer lithospheric layer glides over a plastic, partially molten asthenospheric layer). It provides a mechanism for continent assembly and disassembly and volcanism both at plate boundaries, while convective upwellings of the Plume hypothesis account for volcanism elsewhere such as at the Emperor-Hawaii ridge within the Pacific plate. Since its inception by Morgan in 1971, the 20 initially identified have ballooned to about 122, although the record is 5,200.

Plumes are deep-seated (sub-lithospheric), fixed (spatially persistent in location), diapirs (narrow vertical upwelling columns of unusually hot (but not molten due to extreme pressure), mantle-derived material that is heated by the Earth's core. Plumes are thought to originate from the core-mantle boundary and buoyantly ascend to the asthenosphere-lithosphere boundary where they produce a high rate of hotspot volcanism at the surface. 

Model of Plume Head and Hotspot Volcanism and Extension
Left, A modeled plume head with a narrow tail at the top of its ascent at the base of the lithosphere where it is predicted to flatten. Right, In this simplified and vertically exaggerated diagram, the head is drawn into the developing spreading center where melting forms a flood basalt province. Its arrival leads to uplift and extension, continental break-up, and eventually, the formation of a new ocean basin - the Plume hypothesis.
Modified from Ian Campbell

In concert with the descent of subducting lithosphere, plumes drive plate tectonics, explain flood basalts on land and sea, produce unidirectional time-progressive volcanic chains and even break up supercontinents forming new oceans! It's a grand picture that explains the genesis of the Hawaiian Island system and was actually inspired by it. 

The concept makes great sense (but not to everyone). Why is there no evidence of plume-induced precursory uplift, flood basalts and a hotspot conduit beneath the volcanically active Big Island? How can the chain's "bend" and time-progression possibly occur given slow plate movement? Why do geochemical signatures indicate a shallow mantle source? Why are there are no reliable seismic tomographic images, only computer models?

Convection Currents in the Mantle Drive Plate Tectonics
Melted magma is returned to the surface via upwelling plumes and redelivered during subduction. The convective process drives plate tectonics (and more) and is a manifestation of our cooling planet.
Modified from "Hotspots and Mantle Plumes" from Earth's Dynamic Systems, Chapter 22

At the top of its ascent, a plume head is thought to flatten into a 2,000 to 2,500 km disk and enter a continent's spreading center where eruptive products generate a hotspot. Beneath the continent, magma becomes molten due to adiabatic decompression (decreased pressure due to ascent without heat transfer, raising temperatures or adding flux like water). Basaltic magma then penetrates the crust and is influenced by it. As it melts, it fractionally crystallizes (Bowen's process whereby magma differentiates into a progression of igneous rock types as it cools)

Initially, it is thought that plumes actively force a continent apart and more recently, thermally and chemically erode the base of the lithosphere and promote a melt that exacerbates lithospheric weakening. The process is thought to account for Paraná-Etendeka bimodality and geochemistry of rocks of lower mantle origin versus basalts of mid-ocean ridges that are asthenospheric in origin (higher mantle) that develop by partial melting (the chemical differentiation of crustal rocks). 

The Plume Hypothesis
This scenario illustrates the release of sulfur-rich volatiles (extinction discussion below) but also nicely shows a mantle plume ascending into the lithospheric mantle beneath a continent. This model does not entirely account for the wide-range of trace-element signatures in basalts and, to some, indicate that the melts are derived from both convecting AND lithospheric mantle.
Modified from Guex et al

The Tristan plume head is thought to cause domal uplift followed by crustal extension and flood basalts at the plume tail. The formation of seafloor ridges on a time-progressive track, continental break-up of West Gondwana and the opening of the South Atlantic ensues. Paraná-Etendeka basalts erupted just ~5 Ma before seafloor spreading started. The timing of voluminous magmatism and the LIP preceding the opening of the South Atlantic as well as the geochemistries seem to fit, but the concept is not without opposition. 

The Plume hypothesis is challenged by a number of anti-plume alternatives. One is edge-driven by changing plate boundaries on continental margins with slab-pull that stretches the lithosphere and initiates rifting. Another contends that the Plume hypothesis has more difficulties than certainties and relies on ad "hoc variants" and contrived modifications for any given situation. Proof is over 70 colorful plume types such as singles, clusters, superplumes, cactoplumes, fossil plumes, stealth plumes, dying plumes, finger plumes, baby plumes, pulsing plumes and even spaghetti plumes. 

The Plate hypothesis contends that the location of volcanism and plumes are "inconsistent with many first-order observations" (Foulger) such as a lack of evidence for geophysical indicators (mantle temperatures, time-progressive volcanic tracks or seismic anomalies in the lower mantle). They believe hotspots aren't there or are even needed!

"The cactoplume – the ultimate fix for any surface phenomenon"
From by platist Erik Lundin

Rather than being contrived as an adjunct to the tectonic plate hypothesis, the Plate hypothesis is fully in keeping with it and simply postulates that "anomalous volcanism" results from lithospheric extension. Thus, melts are permitted to passively ascend from the asthenosphere. It is the conceptual opposite of the Plume hypothesis in stating that volcanism is a shallow, near-surface process not a deep-seated plumal one and that it is a consequence of lithospheric processes rather than an active driver of them. 

In other words, like a rift valley that permits melts to ascend from shallow depths, intraplate surface magmatism is explained as a passive response to the stretching of lithospheric plates.

Platists disavow that the Tristan plume severed the West Gondwana continent and opened the South Atlantic. Instead, they believe the northward-propagating mid-ocean ridge crossed a preexisting zone of weakness - a major transtensional intracontinental structural discontinuity with prior rifting history - the Paraná-Chacos basin shear zone between the Paraná and Chacos Basins (See Part I). Some interpret it as a failed rift arm of a triple plate junction where magmatism was focused. The assumption is in keeping with the Plate hypothesis that operates under the geologically established premise that ancient collision zones may re-activate and transform into regions of extension with associated magmatism.

Non-plumists believe that a plume head was not a driving force in the opening of the South Atlantic or even existed during its opening. In addition, anomalous mantle melting occurred only locally. In fact, its purported size is surprisingly small in comparison to the oft-cited diameters of plume heads. What's more, its location is poorly constrained with some placing its location near the Paraná flood basalts and others at the African plate. 

Schematic of Non-plume Induced Break-Up Model for the South Atlantic
A, Hotspot forms low-degree melts that vent to the surface at the hotspot; B, Changing plate boundary forces stretch the lithosphere and initiate rifting as high-degree. Adabiatic decompression melting at thinned areas generates a melt manifested as flood basalts on the surface and the formation of a LIP. The onset of seafloor spreading is characterized by by melt extraction that builds volcanic margins. C, Additional plate movement over the hotspot forms the two aseismic ridges and seaward dipping basaltic reflectors.
From T. Fromm et al

They believe the Paraná-Etendeka LIP in the South Atlantic is one of many examples of plumeless magmatism. A similar scenario explains the evolution of Iceland in the North Atlantic, where it is almost universally assumed to be underlain by an ascending hot plume from deep within the mantle. Instead, platists believe that Iceland's presence is explained as a natural consequence of relatively shallow processes related to tectonics. 

Conceptual Schematic of the Icelandic Plume
From Geovolcanix

Iceland lies where the Mid-Atlantic Ridge crosses subducting crust of the Caledonian suture, a structural discontinuity of the disintegrating Pangaean supercontinent. It formed ~400 Ma by closure of the Iapetus Ocean when Baltica (Greenland and Scandinavia) collided with northern Laurentia (in concert with Avalonia's collision with southern Laurentia). Its re-activation permits the ascent of magma from the shallow, upper mantle and results in the formation of the island in the Mid-Atlantic at sea level. There's no need for plumes, if hotspots don't exist.

The Plume-less Emergence of Iceland
Confirmed by geochemical analyses of generated melts and studies of tectonic structure, Iceland may have formed along a transverse branch of an ancient suture zone (dashed line) where it crossed the North Atlantic mid-ocean ridge (red line). The Caledonian collision zone is associated with the closure of the Iapetus Ocean at 400 Ma by convergence of Laurentia, Avalonia and Baltica. A similar process may explain LIPs elsewhere such as the Paraná-Etendeka that is attributed to an ascending mantle plume. 
Modified from Foulger et al

Getting back to the Paraná-Etendeka's enigmatic V-shaped seafloor ridges, plumists attribute their geometry to northerly components of three diverging oceanic plates (called an RRR junction) that formed over the Tristan plume. Platists, on the other hand, believe the ridges formed via the accommodation of stress in the lithosphere due to rifting. Release may have occurred along several transform fault segments or failed rift arms or from remelting of detached continental lithospheric mantle. The orientation developed as the South American and African plates diverged, while following a relative track to the NW and NE along arms.

Vector Diagram of South American and African Absolute Plate Motions
If one knows the relative motions of the plates to the mid-ocean ridge, then absolute motions can be deduced. It accounts for the geometry of the Paraná-Etendeka seafloor ridges.

In addition to mantle dynamics and genesis, the Paraná-Etendeka Province finds itself in the midst of another controversy, one involving an enigmatic extinction-volcanism association. 

The crux is a near-perfect temporal association between mass extinction events, catastrophic global climate change and at least a half-dozen LIP eruptions of massive basalts in the Phanerozoic with rapid magma extrusion rates. No other phenomenon indicates such a high correlation, even bolide impacts. The most compelling and dramatic are the end-Permian Siberian Traps and Deccan Traps in India. Although coincidence doesn't prove causality, the frequency of a volcanism-extinction connection is compelling.

Typical of the many continental LIPs in the mid-Phanerozoic, the Paraná-Etendeka's vast surface outpourings emplaced over a short geological time frame, although the duration is only variably constrained. Considering the magnitude of the eruption, which is thought comparable to other LIPs associated with mass extinctions, massive volumes of atmospheric gases were likely released with a strong potential for adverse environmental impact. Yet, the LIP has fostered little attention, since it appears to have formed when extinction magnitudes were low. What's up?

Worldwide Distribution of LIPs and Associated Volcanic Basins
Besides the Paraná-Etendeka LIP (encircled), other volcanism-extinction examples include the Permian Emeishan Traps, the end-Permian-Triassic Siberian Traps (the most severe known extinction), the end-Triassic Central Atlantic Magmatic Province (precursor to Central Atlantic opening) and the Cretaceous-Paleogene Deccan Traps (possibly bolide related). The three LIPs associated with the opening of the Atlantic - NAIP (North Atlantic Igneous Province), CAMP and Paraná-Etendeka - are investigated for their association with mass extinction events. 

All that may change with the discovery of Early Cretaceous Valanginian age (~137 to 132 Ma) black shales in pelagic (bottom) sediments of the Tethysan oceanic realm (that opened in eastern Pangaea before break-up). They reveal major positive δ13C perturbations of the global carbon cycle, the earliest of the Cretaceous system, and a warmer climate. The shales suggest a biocalcification crisis of nannoplanktons that best tolerate low nutrient conditions.

Volcanic activity coincident with the Paraná-Etendeka LIP, prior to the break-up of West Gondwana, may have played a major role in this environmental change by increasing adversely tolerated nutrient conditions, ocean acidification or both via outpouring of bio-limiting atmospheric gases from the mid-ocean spreading center. The crisis may be tied to a global episode of anoxic (oxygen-poor) ocean deposition called the Weissert Event.

Mass Extinction Intensity Associated with Volcanism
The strong temporal association between LIPs and mass extinctions suggests causality. Three of the largest crises are Siberian and Deccan Traps and the CAMP. The graph suggests an association for the Paraná-Etendeka does not exist, although recent data possibly suggests otherwise. The magnitude and duration of genera affected in extinctions is represented by the gray field. The band size is related to continental LIPs or oceanic plateaus and represents estimated volume of lavas plotted as a function of their respective age. Continental LIPs are color coded as a function of age of the underlying lithosphere. Modified from Guex at al

If no extinction occurred, it may be due to emplacement over a shorter timescale or lower volatile concentrations or less deleterious ones, fluorine and chlorine in particular that degrade the UV-shielding ozone layer and cause ecosystem stress. It may even be due to the age and composition of the underlying mantle. The Paraná-Etendeka erupted through thermally eroded Proterozoic lithosphere versus the Central Atlantic Magmatic LIP and Deccan Traps, both with high extinction correlations that emplaced through Archean cratonic lithosphere and releasing atmospheric sulfur (above graph).

Tonemapped Photo of Salto Bosetti on the Eastern Side of Iguazú Falls 
Bosetti Falls is one of some 275 individual falls that comprise Iguazú Falls. Who would suspect, that while standing at the brink as its waters cascade off tiered basalts of the Paraná Volcanic Province that a mass extinction of life was in the making some 130 million years ago?

Thus far, the Paraná intracratonic sag basin formed in West Gondwana in the Ordovician first as an embayment and later became landlocked within the continent's interior. The subduction of Panthalassa oceanic lithosphere along the continent's southwest margin resulted in intracontinental extension and basin subsidence that provided accommodation space within the Paraná Basin for a thick stack of sedimentary supersequences

The emplacement of Serra Geral continental flood basalts in the Early Cretaceous, whether from hotspot activity or alternative genetic origin, terminated the basin's protracted depositional history and gave rise to the Paraná-Etendeka Volcanic Province, one of the largest in the world. The event was a precursor to rifting that fragmented West Gondwana into South America and Africa and the opened the Atlantic. 

Marginal oceanic slab compression not only induced intracontinental extension within the Paraná Basin but caused uplift at various intervals in the Paleozoic. The most significant is thought to have occurred prior to rifting, when the basin was also tilted down to the west and segmented into a succession of three "compartimentos" or planalto (plateaus). 

Três Planaltos Paranaenses
Beginning on the east, the geomorphological units of Paraná include: the Atlantic Littoral Plain (coastal shoreline); the Serra do Mar (coastal range that borders the crystalline plateau on the east); First Paranaense Planalto (Precambrian crystalline terrain); Second (Paleozoic terrain); and, Third (Cretaceous basaltic plateau). 
Modified from
The three plateaus of Paraná are separated by escarpments that progressively expose younger strata of the basin's billion-year history. The First Planalto is formed by igneous and metamorphic rocks with Serra do Mar its eastern boundary. The Second constitutes outcrops of uppermost Paleozoic supersequences sediments that is overlapped by Serra Geral volcanics of the Third. Quaternary sediments occur in all regions especially river valleys. A third subbasin exists in the northwest with

West to East Schematic Cross-section of the 
Modified from

Although poorly understood despite its long history and highly controversial, the river system assuredly began to develop concurrent with the rifting process. Megadomal rift-arches developed centrifugal drainage patterns that remain partially evident on the landscape, but the history of the modern system didn't begin until the complete break-up of Gondwana. 

Initially, as South America and Africa broke apart, a single, large, longitudinally elongated basin - the Afro-Brazilian Depression - captured drainages and gave rise to an interconnected lake system. The two continents became separated at the end of the middle Cretaceous between 98 and 93 Ma. Following Gondwana break-up, erosive denudation of the uplifting Brazilian crystalline shield occurred as grabens (subsiding rift-blocks) formed offshore sedimentary basins (such as the Santos and Spirito in southeastern Brazil) while isostatic uplift may have occurred in the Sierra do Mar or possibly also related to Mid-Ocean Ridge "push" and far-field Andean orogenic slab "pull. 

Uplifted rift-blocks resulted in isolated endorheic drainages (closed basins without outlet) that extended continentward under arid to semi-arid conditions (evidenced by a vast sand sea across southeastern Brazil that directly underlies Serra Geral basalts) probably were the first evidence of a fluvial system.  

Granitoid Massifs of the Serra do Mar Coastal Range Rise Above the South Atlantic
The otherwise heavily deforested Serra do Mar complex of mountains is protected by a system of national and state parks, ecological systems and biological reserves in easternmost Brazil. With a Precambrian  crystalline metamorphic basement derived from supercontinent Rodinia, granitoid igneous rocks derived from the Brasiliano-Pan-African orogeny during Gondwana assembly and basaltic magmatism from the arrival of a plume head in the Late Cretaceous-Paleogene, the composite range runs some 1,500 parallel to the Atlantic Ocean between the coast and eastern uplifted Paraná Basin.

Concomitant with rifting and uplift into the Neogene, onshore basins became separated from coastal drainage systems on the newly-formed, "Atlantic-style" passive margin by the uplifted, fault-scarped, coast-paralleling, 1,500 mile-long Serra do Mar hydrologic divide. Its eastern escarpment (above) initiated E-W erosive retreat concomitant with deformation that further delineated the evolution of drainage patterns in addition to the capture of surrounding headwaters from adjacent hydrographic systems. The current drainage system developed between the Miocene and Pliocene, concomitant with the initiation of the modern Paraná system.

Numerous contemporary rivers headwater within the west flank of the Brazilian Highlands as does Rio Iguazú that originates within Serra da Baitaca State Park. It is directed to the west by basinal tilt, many faults and old and deep episodically active NW-SE, E-W and NE-SW structural lineaments re-activated during Gondwana break-up. 

Patterns of topography and drainages, block faulting and river capture provide suggestive evidence of more recent uplift such as from neotectonic (Neogene) compression from far-field stresses from the Andean orogeny in the Miocene, all of which likely contributing to the reorganization of the fluvial basin to the morphology we see today.

The Paraná River System and Iguazú Tributary
With an asymmetric drainage basin of about 1,081,000 square miles, the Paraná River in south-central South America is second in length to the Amazon river and runs through Brazil, Paraguay and Argentina for some 3,030 miles. Downstream, it merges first with the Paraguay River and further on with the Uruguay River to form the estuary-delta Rio de la Plata that empties into the Atlantic Ocean. Upstream, its eastern, left-hand tributaries headwater essentially from the west flank of the coastal Serra do Mar mountain range. 
Modified from

It completes a ~1,320 mile, cuesta-escarpment and waterfall-punctuated journey as the river and its floodplain meanders cratonward across sedimentary and volcanic rocks of the Paraná Basin's three plateaus. After cascading off Serra Geral basalts of the Third Plateau at Iguazú Falls, the river joins mainstem Rio Paraná as a left-hand tributary at the Triple Frontier, the confluence of Uruguay, Brazil and Argentina. The system's collective waters reach the Atlantic at the estuary-delta Rio de la Plata between the cities of Montevideo, Uruguay and Buenos Aires, Argentina. 

Iguazú Falls is the largest waterfall of the Paraná River Basin, although prior to 1982, it was rivaled by Sete Quesdas. The series of seven waterfalls, also on the Serra Geral Formation, was inundated by the impoundment of the locally controversial and largest in the world Itaipo Reservoir and Hydroelectric Dam upstream on the Paraná River in 1982. 

To be discussed, notice that following a dramatic bend in the serpentine river channel immediately above Iguazú Falls (below), the otherwise broad river curiously undertakes a near reversal of direction, while below the falls, it constricts into a narrow gorge that strikes linearly away from a U-shaped chasm called the Devils Throat (tiny white spray). 

The Region of Iguazú Falls
Below the falls, the confluence of the Paraná River and Iguazú tributary form the border between three countries, while upstream the latter separates two. The region around Iguazú Falls has clearly succumbed to deforestation, agriculture and human settlement (computer-enhanced). Sharply delineated, Argentina and Brazil have notably set aside park land, whereas, Paraguay has permitted development.
Modified from 2001 NASA Earth Observatory image

Where does all that water go? Downriver, Rio Paraná is joined by Rios Paraguay and then Rio Uruguay before emptying into the Atlantic Ocean at Rio de la Plata between Buenos Aires, Argentina and Montevideo, Uruguay.  

The Mainstem Paraná River and Wide Floodplain below the Confluence with Iguazú River

From the waterfront of Buenos Aires, Rio de la Plata looks more like a huge, placid lake than a river. Geographically, the 290 km-long estuary (where tides enter the mouth of the river) is either a gulf of the Atlantic or the world's widest river (256 km). Historically, the "River of Silver" was named for the abundance of the mineral assumed to be in the region, which is plentiful but only since it was brought downriver from far inland Bolivia, which is less accessible by the Paraná's Paraguay tributary than the Amazon River.

Summer Photo of Rio de la Plata from Space
The Paraná River empties into Rio de la Plata and the southern Atlantic Ocean. The current forms a massive delta and carries a long plume of iron-stained sediments out to sea. Also delta-like in shape is the city of Buenos Aires, visible on the southern side of the river as is smaller Montevideo across the river on Uruguay's southern side. A marshy floodplain and braided stream system accompanies the river through rich farmlands. It's the largest wetland corridor in the world and extends well beyond Iguazú Falls in the north.
Modified from NASA's Visible Earth Catalog

Directly above the falls, Rio Iguazú is known as Iguazú Superior, where it flows over a number of small steps carved into basaltic bedrock and skirts a slew of small vegetated islands set precariously in the channel. In the midst of initiating another clockwise meander, the river channel dramatically shallows and widens to 1,500 m, as if doubling back on itself.

Iguazú Superior
The mid-channel islands above the falls provide a sanctuary for birds such as toucans and parrots and exotic tropical butterflies. In the near distance, the Throat of the Devil sends a plume of mist skyward that is easily mistaken for a low cloud as one approaches on one of three metallic catwalks. 

Paseo Graganta del Diablo, the metallic catwalk above the falls that is reached by taking a short train ride from the visitor center, leads you to the chasm from island to island. With every approaching step the roar of the falls grows louder. Colorful birds and butterflies are everywhere.

Frothy, brilliant white and churning violently, about half the river's flow plunges off the volcanic plateau in a huge 2.7 km arc from Argentina to Brazil, while sending a cloud of spray skyward that's visible from space. At the brink, a large portion of Iguazú Superior converges into an enormous mist-shrouded, thunderous funnel that's 230 feet high. Called Garganta del Diablo in Spanish or Throat of the Devil. Curiously, it aligns with the strike of the river channel and gorge downstream from the falls, which is best seen from the air or on map view. 

Staring into the Throat of the Devil
Shouting to be heard over the roar, the spray is welcomed relief in the 90 degree heat.

Below the falls, Iguazú Inferior tumples into a broad, shallow plungebool that is littered with large blocks of displaced basalt. The channel has completed a hair-pin reversal of direction as it converts to a linear, narrow (~80-90 m) and deep gorge (~70 m) that is somewhat steeper along the north, right bank (below). The channel-chasm-gorge morphology begs the question, what structural aspects caused the river to double back on itself and contribute to the distinctive upfalls and downsfalls morphology? 

Iguazú Inferior
Oblivious to the turbulence of the falls it has just experienced, Iguazú Inferior calmly flows away in a linear and relatively narrow gorge. It will soon join parent Rio Paraná at the three country confluence.

About 20,000 years ago, the regional climate changed for the last time from cold and dry to hot and rainy, But, the continental rainfall regime in southeastern and southern Brazil is a result of moisture from the Amazon to the north. Interacting with cold masses from the south accounts for the Paraná Basin's high rainfall. Of the major tributaries of the Parana River - the Paranaíba, Grande, Tietê, Paranpanema and Iguazú- the latter's importance to the system is based on its drainage basin of 62,000 sq km, length of 1,320 km and an annual mean volume of 1,746 cu m/s. Thus, it has great capacity to generate exceptional discharge (a maximum of 12,799 cu m/s), the lifeblood of the falls. 

Although subtropical, unlike many other South American rivers where annual temperature variations are relatively limited, the volume, color and content of Iguazú's waters vary considerably with season. During summer rains from October to March, the reverse of the North Hemisphere, the river swells within its channel and becomes laden with silt and clay from basaltic red soils stained with oxidized iron and aluminum acquired from basaltic mafic minerals - largely pyroxene, feldspar, hornblende, mica and magnetite.

Wide, Calm, Turbid and Meandering, a Swollen Rio Iguazú Superior Approaches the Falls in the Wet Season

The region's soils, called latosols that typically develop in tropical rainforests, form from the same minerals. They endow the region with its fertile, deeply colored "red earth " known as "tierra roja" in Argentina and "terra roxa" in Brazil and promote its agricultural bounty beginning with coffee in colonial times.

In the dry, winter season from April to July, Rio Iguazú transports a relatively low quantity of suspended sediment, running clear or slightly greenish and unhurried within a moderately wide, basalt-floored, shallow channel across the gently downwest-sloping Paraná Plateau. During these times to the joy of millions that visit it, water copiously spills off the Paraná plateau at ~1,500 cu ft/s. It's a jaw-dropping spectacle for all the senses that visitors can't get enough of!

During an extreme drought in May and June of 1978, the falls actually dried up completely for 28 days due to nonexistent flow. In contrast, 2014 rains in the Argentine and Brazilian regions of Misiones and Paraná reached historic levels that resulted in a flow rate of 46,300 cubic meters per second - 33 times the usual flow rate. The previous record of 36,000 was reached in 1992. Both times, officials closed the catwalks for safety as the discharge completely obliterated the falls within a single wall of murky-brown water.

Iguazú Falls During Extreme Drought and Overflow of Biblical Proportions
Modified from

Iguazú's construction is surprisingly simplistic in light of the billion year-plus processes that contributed to its formation. Essentially, three uppermost layers of Serra Geral lava give rise to a two-step staircase effect. At normal rates of flow, the system consists of some 275 individual waterfalls, although the number fluctuates from 150 to 300. About half the river funnels into Garganta del Diablo that lies on strike with Iguazú Canyon below the falls. The remainder spills over a curvilinear front that lies perpendicular at the apex of the channel's sweeping, clockwise turn. The falls of San Martin, Adam and Eva, Penoni, and Bergano are the largest of the individual waterfalls along the front.

The explanation for the linear geometry of the gorge and chasm is a NNW to NW intraformational, of which there are a number locally that strike N 80° and N 170°. Large-scale structural controls also exist from re-activated NW and WNW lineaments and epeirogenic (unwarping) since the Pliocene-Pleistocene that mobilizes large-scale blocks.

South-Southwest View of Iguazú Falls
The massive steps of Iguazú Falls are clearly visible as is the sweeping change of direction the river makes on its approach to the falls. Below the falls, a narrow gorge strikes linearly away from the mist-enshrouded plunge basin of Garganta di Diablo. Elliptical-shaped Isla di San Martin stands midriver below the falls. With the border essentially lying mid-river, the red roofed hotel is in Brazil, while everything else is Argentina. From Wikimedia Commons

As one might expect, waterfalls form slower in erosion-resistant, homogenous, stratified bedrock-channeled rivers such as the igneous rocks of the plateau. Characteristically, trapps (or trapps, Swedish for "stairway" in mafic rock) form that are controlled by bed thickness, degree of consolidation, bedding plane discontinuities, joints and fractures.

Schematic Profile of Iguazú Falls
Surfaced with vesicular basalt, two traps control the falls' staircase morphology.
From Latrubesse

From the Argentine side, the lava flows can be seen laterally in the system of prismatic, horizontally fractured, vertical columnar jointing in the vegetated walls of Isla de San Martin, the blocky island below the falls. It displays the lower two of the plateau's superficial-most three basalt flows against a background of the falls' three flows.

Movie goers will recognize the island and waterfall from The Mission filmed in 1986. 
A species of Black Vultures calls the island home.

One of the most conspicuous and yet not completely understood features occurs along downcutting bedrock-floored channels such as found on the volcanic plateau. Knickpoints, rapids and waterfalls form at an abrupt break or sharp change in slope in longitudinal channel profiles that dip upstream. In conjunction with bedrock erosion (degradation), the knickpoint and waterfall retreat upstream (headward migration). Upstream advancement occurs as blocks of bedrock are hydraulically plucked (quarried) from the lip of the falls by material travelling as bedload (versus suspended load).

Channel incision typically occurs as a flowing body of water finds its way to the lowest point to which it can flow (base level) on the way to the sea. The process is facilitated by climatic change, tectonism, bed lithology, stream kinetic energy and entrained load. Quarrying is the most rapid means of eroding a bedrock channel and is facilitated by hydraulic wedging (where clasts forcefully ratchet apart fractures and joints).

Undercutting of the waterfall cliff (headwall) from rocks that scour and abrade the riverbed at the plungepool (deep depression in the streambed below the falls) causes collapse of the knickpoint lip. It occurs more readily when erosionally susceptible rock (footrock) underlies the strata at the lip (caprock). 

As the waterfall retreats, a gorge (80-90 m wide and 70-80 m deep) typically develops in the channel downstream, linear in the case of Iguazú as it follows a fault along the edge of the volcanic plateau. The entire process of waterfall evolution occurs in the youthful stage of river development, which typically occurs in a river system's upper reaches where gradients and channel slopes are greater (oversteepened), water supply is more abundant and flow velocities are faster. 

Pothole Formation and an Undercutting Waterfall with Plungepool
Retreat is often viewed largely as a function of pothole consolidation and the erosion and toppling of susceptible, loosely or weakly consolidated, or jointed or fractured stratal blocks at the brink and the consolidation of plunge pools below waterfalls that undermine or undercut the headwall and cause collapse.
Modified images from and

Cylindrical holes in the riverbed above and below the falls form when repetitive current eddies cause sediment and pebbles to abrade the bed. The action progressively exposes pothole walls by keyholing (from below) and incising (from above). Their consolidation promotes vertical and lateral channel incision, headward migration and knickpoint advancement. River polish and fluting commonly occurs on resistant bedrock surfaces.

Pothole-riddled, Surface Polished and Fluted Bedrock Exposed above the Falls
Many of the potholes, some with trapped bedload, have consolidated, while others have begun to keyhole, where erosion through the wall creates an opening in the downstream face leaving behind an arch of bedrock. The keyholes provide a path to flush or evacuate erosive material and enable vertical and lateral incision of the channel floor. The reddish-gray color is due to weathering of iron-rich minerals.

The pothole remnant in the gorge's wall below the falls has been progressively exposed by channel quarrying to the extent that it reveals its structure laterally. Abandoned potholes and plungepools provide evidence of upstream retreat and when linear, imply the direction of flow. Notice the stratigraphy of the flows in the transected wall. Characteristically, basalt forms a columnar jointing pattern upon cooling, which, when supplemented by horizontal fractures, facilitates channel excavation and waterfall formation. 

Fully incised Pothole

It's currently thought that, at least since the Pleistocene, Iguazú Falls has been advancing upstream from the confluence of Rios Iguazú and Paraná by headward erosion and knickpoint retreat some 21 km to its present-day location at Garganta del Diablo. The rate is estimated to be 1.4 to 2.1 cm/yr over the last 1.5 to 2 million years.

Many waterfalls of the Paraná River's tributaries have succumbed to the needs of the region's growing population and have been eliminated by hydroelectric projects and dams. It leaves Iguazú Falls as a poignant last example of the greatest falls of the river system. Fortunately, this precious resource remains protected within the two national parks and UNESCO World Heritage site. 

In this post, I've attempted to show there is often far more to a landscape or landform than meets the eye. And, that by looking deeper - both in time and space - the revelation adds a profound dimension of richness and enhances our understanding of the natural forces at work that shape our planet.

I am extremely grateful to Edgardo M. Latrubesse, PhD of the University of Texas at Austin and Professor Eduardo Salamuni, PhD of the Federal University of Parana State in Brazil, who contributed extremely helpful information regarding the evolution of the Paraná Basin and geomorphology of Iguazú Falls. Dr. Salamuni's personal communications were of great value in formulating many of the ideas found in this post. Special thanks are also in order for his addition of my blog to his Facebook page (here).

An Alternative Model for Iceland and the North Atlantic Igneous Province by G.R. Foulger et al, Journal of Volcanology and Geothermal Research, In press 2017.
Climatic Events During the Late Pleistocene and Holocene in the Upper Parana River: Correlation with NE Argentina and South-Central Brazil by Jose C. Stevaux, Quaternary International 72, 2000.
Contrasting Stress Fields on Correlating Margins of the South Atlantic by Eric Salomon et al, Gondwana Research, 28, 2015.
F,Cl,andS Concentrations in Olivine-hosted Melt Inclusions from Mafic Dikes in NW Namibia and Implications for the Environmental Iimpact of the Paraná–Etendeka Large Igneous Province by Linda Marks et al, Earth and Planetary Science Letters, 392, 2014. 
Foz do Iguaçú: Geomorphological Context of the Iguaçú Falls by Marga Eliz Pontelli and Julio Cesar Paisani, Landscapes and Landforms of Brazil, Chapter 31, 2015.
Iceland is Fertile: The Geochemistry of Icelandic Lavas Indicates Extensive Melting of Subducted Iapetus Crust in the Caledonian Suture by G. R. Foulger et al from
Iguazu Falls: A History of Differential Fluvial Incision by José C. Stevaux and Edgardo M. Latrubesse, Geomorphological Landscapes of the World, Chapter 11, 2010.
Large Igneous Provinces by Richard E. Ernst, Cambridge University Press, 2014.
Morphostructure of the Serra Do Mar, Paraná State, Brazil by Edenilson Roberto do Nascimento et al, Journal of Maps, 12, 2016.
• On the Causes of Mass Extinctions by David P.G. Bond and Stephen E. Grasby, Palaeogeography, Palaeoclimatology and Palaeoecology, 478, 2017.
Origin of the Paraná-Tristan-Etendeka Igneous Province by G.R. Foulger, available on-line.
Origin of the South Atlantic Igneous Province by G.R. Foulger, Journal of Volcanology and Geothermal Research, 2017.
Paraná Flood Basalts: Rapid Extrusion Hypothesis Confirmed by nNw 40Ar/39Ar Results by David S. Thiede and Paulo M. Vasconcelos, Geology, 38, 2010.
Petrogenesis of Early Cretaceous Silicic Volcanism in SE Uruguay: The Role of Mantle and Crustal Sources by Michele Lustrino et al, Geochemical Journal, 44, 2010.
Petrology and Geochemistry of Early Cretaceous Bimodal Continental Flood Volcanism of the NW Etendeka, Namibia. Part 1 and 2: Characteristics and Petrogenesis of the High-Ti Latite and High-Ti and Low-Ti Voluminous Quartz Latite Eruptives by A. Ewart et al, Journal of Petrology, Vol. 45, 2004.
Review of the Aareal Extent and the Volume of the Serra Geral Formation, Paraná Basin, South America by Heinrich Theodor Frank et al, Pesquisas em Geociências, 36, 2009.
Tectonic History and the Biogeography of the Freshwater Fishes from the Coastal Drainages of Eastern Brazil by Alexandre Cunha Ribeiro, Neotrop. Ichthyol. 4, 2006.
Thermotectonic and Fault Dynamic Analysis of Precambrian Basement and Tectonic Constraints with the Parana Basin by L.F.B. Ribeiroa et al, Radiation Measurements, 39, 2005.
The Link Between Large Igneous Province Eruptions and Mass Extinctions by Paul Wignall, Elements, Vol. 1, 2005.
Thermal Erosion of Cratonic Lithosphere as a Potential Trigger for Mass-extinction by Jean Guex et al, Scientific Reports, 2016.
• Timescales and Mechanisms of Plume–lithosphere Iinteractions: 40Ar/39Ar Geochronology and Geochemistry of Alkaline Igneous Rocks from the Paraná–Etendeka Large Igneous Province by S.A. Gibson et al, Earth and Planetary Science Letters, 251, 2006.
Zircon U–Pb Geochronology from the Paraná Bimodal Volcanic Province Support a Brief Eruptive Cycle at ~135 Ma by Viter Pinto, Chemical Geology, Vol.1, 2011.