Info

a collapse of photosynthesis, a breakdown of food chains and a drastic drop in surface temperatures with freezing conditions on continents, especially in continental interiors, and widespread and deep snowfall (Pollack et al. 1983). The dramatic change in climate might account for the relatively sudden disappearance of large parts of fauna and flora associated with large impacts. If the impact occurred in an ocean, then a longer-lasting time of much hotter conditions would follow a period of intense cold. The ocean water vaporized by the impact would increase the atmospheric moisture content considerably, and this would lead to the washing out of the tropospheric dust within a few weeks or months. After that, the remaining water and cloud in the stratosphere would generate, in the manner of a very efficient greenhouse, a large rise in global temperature, which at the surface may exceed 10°C, a temperature anomaly that would persist for some months or years, until diffusion and photochemical processes in the stratosphere return the Earth to its steady-state condition (Emiliani et al. 1981).

Recovery after a large impact would be slow. After about a decade, ecosystems would start to recuperate, but it could take more than 2.5 million years for global ecosystems to return to a 'normal' state (Table 7.4). Survival and recovery would be selective. If lack of sunlight to power photosynthesis were the primary killing agent, then detritus feeders in marine and terrestrial ecosystems would likely fare better than those dependent upon primary production (Sheehan and Hansen 1986; Sheehan and Fastovsky 1992; Archibald 1993). In the terrestrial realm, aquatic organisms did better than their fulltime terrestrial counterparts did, probably because they were generally more likely than exclusively terrestrial organisms to be members of detritus-based food chains (Sheehan and Fastovsky 1992; see also Retallack 2004). Moreover, among terrestrial animals, those able to feed in detritus-based food chains (including mammals) were more likely to survive than animals in food chains dependent on primary production (including dinosaurs). If a global pulse of infrared thermal radiation were the primary killing agent, then shelter would be the key to survival (Robertson et al.

Table 7.4 Catalogue of recovery following a large bolide impact at Chicxulub at the Cretaceous-Tertiary boundary.

Time after impact Effect

Reference

9 months Dust cloud starts clearing

10 years Very severe climatic disturbance

(mainly cooling) ends 1,000 years Vegetation begins recovering; end of 'fern spike' 1,500 years Initial recovery of deeper-water benthic ecosystems 7,000 years Recovery of some deeper-water benthic ecosystems 70,000 years Oceanic anoxia diminishes

100,000 years Final extinction of dinosaurs (possibly)

300,000 years Final extinction of ammonites (possibly)

500,000 years Large fluctuations in oceanic ecosystems start moderating 1,000,000 years Open oceanic ecosystems partly recovered 2,000,000 years Marine mollusc faunas mostly recovered 2,500,000 years Global ecosystem 'normal'

Tschudy and Tschudy (1986)

Coccioni and Galeotti (1994)

Coccioni and Galeotti (1994)

Kajiwara and Kaiho (1992) Rigby et al. (1987) Zinsmeister et al. (1989) Barrera and Keller (1994); Alcala-Herrera et al. (1992) D'Hondt et al. (1996) Hansen et al. (1993) Alcala-Herrera et al. (1992)

2004). It would have caused severe thermal stress and ignited global wildfires that would burn anything unable to find shelter: 'sheltering underground, within natural cavities, or in water was the fundamental means to survival during the first few hours of the Cenozoic. Shelter was by itself not enough to guarantee survival, but lack of shelter would have been lethal' (Robertson et al. 2004, 760). In this scenario, a heat pulse and subsequent wildfires, rather than the cessation of photosynthesis, would be the primarily killing agent, and thermal sheltering, rather than detritus feeding, would have aided survival.

Some researchers would claim that a bolide impact trigger for the end-Cretaceous extinction seems likely. However, recent work suggests that if the Cretaceous biota did succumb to bombardment, then it was to a series of impacts rather than just the one that produced the Chicxulub crater. Gerta Keller and her colleagues (2003) came to this conclusion by studying the stratigraphy and age of altered impact glass (microtektite and microkrys-tite) ejecta layers in Late Maastrichtian and Early Daman sediments of Mexico, Guatemala, Belize, and Haiti (Figure 7.5). The first impact is roughly contemporaneous with major Deccan volcanism and likely contributed to a rapid global warming of 3-4°C in intermediate waters between 65.4 and 65.2 million year ago, which coincided with decreased primary productivity and the onset of a terminal decline in planktonic foraminiferal populations. The Cretaceous-Tertiary boundary impact correlates with a major drop in primary productivity and the extinction of all tropical and subtropical planktonic foraminaferal species. The Early Daman impact, which occurred about 100,000 years after the Cretaceous-Tertiary impact, may have contributed to the delayed recovery in productivity and evolutionary diversity, as well as the demise of Cretaceous survivor species. The discovery of the two approximately 65-million-year-old craters other than the Chicxulub crater -the Silverpit crater in the North Sea (Stewart and Allen 2002) and the 24-km-wide Boltysh crater in the Ukraine (Kelley and Gurov 2002) - bolster the multiple impact hypothesis.

A final point to make is that asteroid and comet impacts may have beneficial effects on ecology and evolution (Cockell and Bland 2005). Impacts create new habitats by the redistribution of target materials and the making of new lakes within craters. Moreover, in eliminating whole groups of organisms, impacts may be instrumental in promoting the rise of new groups of organisms, including the dinosaurs and the mammals. It could well be the case that:

were it not for the boost given to evolution by environmental catastrophes, whether they result from terrestrial or cosmic causes, that life on Earth would not have advanced up the evolutionary ladder quite so rapidly. Thus we arrive at the paradox that, although catastrophic episodes in Earth history may cause mass extinctions and act to the detriment of individual species, for the biosphere as a whole they are a stimulating time.

(Huggett 1997b, 178)

Volcanism

Flood basalt volcanism, which usually occurs as sustained bouts of volcanic eruptions producing huge volumes of continental flood basalts, is a prime suspect in the mass extinction mystery. On land, changes in climate produced by protracted periods of volcanism may have resulted in environmental stress severe enough to precipitate mass extinctions. The formation of large igneous provinces undoubtedly produces huge volumes of carbon dioxide and sulphur dioxide. Take the case of the Siberian Traps, which formed at the end of

Climate warming Volcan ism MM layers

Pd (ng/g) Multiple events

Volcanic event, Pd, Ir

Reworked MM?

Climate warming Volcan ism MM layers

Pd (ng/g) Multiple events

Volcanic event, Pd, Ir

Reworked MM?

DSDP site 525A

10 12 14 16 Deccan Temperature (°C) Traps Li and Keller Courtillot (1999) (1998) C. Hoffmann et al.

Pla(l) Ir anomaly: Pd and minor Ir

Actela, Beloc, in volcanoclastic

Bochil, Coxquihui. (Haiti) and smectite-

Possibly site 1049 rich layers (Bochil)

* Ir at K/T boundary worldwide, but not in Caribbean and Central America due to incomplete sediment records.

Microtektite/Microkrystite (MM) layers

Impact event, probably

Chicxulub

Climate warming related to Deccan Volcan ism and Impact events.

DSDP site 525A

10 12 14 16 Deccan Temperature (°C) Traps Li and Keller Courtillot (1999) (1998) C. Hoffmann et al.

Pla(l) Ir anomaly: Pd and minor Ir

Actela, Beloc, in volcanoclastic

Bochil, Coxquihui. (Haiti) and smectite-

Possibly site 1049 rich layers (Bochil)

* Ir at K/T boundary worldwide, but not in Caribbean and Central America due to incomplete sediment records.

Microtektite/Microkrystite (MM) layers

Impact event, probably

Chicxulub

Climate warming related to Deccan Volcan ism and Impact events.

Figure 7.5 Multiple impact Cretaceous-Tertiary scenario based on impact glass spherule deposits and iridium anomalies in the Gulf of Mexico, Caribbean, and Central America. The oldest impact glass-spherule layer dates to 65.27 ± 0.03 million years and relates to the Chicxulub event based on glass chemistry. This impact event coincides with the global climate warming between 65.2 and 65.4 million years ago and peak intensity of Deccan volcanism. Sea-level fluctuations have repeatedly reworked younger impact glass-spherule layers in the Late Maastrichtian and Early Danian. The Cretaceous-Tertiary boundary event is generally absent in the region because of erosion and tectonic activity. A widespread iridium anomaly in the Early Danian subzone Pla(l) is tentatively identified as an Early Danian impact event at about 64.9 million years ago, and a palladium anomaly and minor iridium anomaly at the Pla(1)/Pla(2) transition may be related to a regional volcanic event. Source: Reprinted from Earth-Science Reviews, 62, G. Keller, W. Stinnesbeck, T. Adatte and Stuben, D., Multiple impacts across the Cretaceous-Tertiary boundary, 327-63, Copyright © 2003, with permission from Elsevier.

the Permian. The estimated volume of erupted basaltic lava is about 1.6-2.5 million km3 (Renne and Basu 1991). If spread evenly over the Earth's surface this material would form a layer 3 m thick. The eruptions would have produced about 1013 t of carbon dioxide (Wignall 2001), which after a brief cooling phase resulting from the emission of sulphur and formation of sulphur aerosols in the stratosphere, would have caused global warming. The warming would have had two important results. First, it would have promoted anoxic conditions in oceans because less oxygen can dissolve in warm water and because the pole-to-equator temperature gradient would probably weaken, so reducing the oceanic circulation (Hallam and Wignall 1997, 141). Second, the warming might have triggered the dissociation of gas hydrates, which would cause even more warming (p. 124).

Early work on the connection between large igneous provinces and mass extinctions focussed on the Late Cretaceous event. The aim was to account for the more or less gradual increase in extinction rate for many groups of organisms, followed by a catastrophe lasting a few tens of thousands of years or less. Peter R. Vogt (1972) recognized the proximity of the Deccan Traps to the Cretaceous-Tertiary boundary. Likewise, Dewey M. McLean (1981) noted that the Late Cretaceous mass extinction coincided with one of the greatest outpourings of flood basalt in geological history and hypothesized that the outgassing of carbon dioxide associated with the lava created an atmospheric greenhouse in which the heat was high enough to render the dinosaurs infertile. He thought that the bout of protracted volcanism had caused an extraordinary, global hiatus that gives the illusion of sudden extinction, and that it gave rise to the geochemical signatures misinterpreted by some as evidence of the impact of an extraterrestrial body. Later researchers embellished McLean's ideas, arguing for a scenario of environmental deterioration caused by increased volcanism over an extended period as a possible explanation of the pattern of Late Cretaceous extinctions (Gledhill 1985; McLean 1985; Officer and Drake 1985; Officer et al. 1987). Large injections of sulphates to the atmosphere would be potentially disastrous, producing prodigious volumes of acid rain, reducing the alkalinity of the surface ocean, cooling the atmosphere, and depleting the ozone layer (Stothers et al. 1986). Injections of ash from contemporary explosive volcanoes would enhance the cooling of the atmosphere. The very end of the Cretaceous period saw a paroxysm of intense volcanicity. The iridium peak marks this intense episode. The source of the volcanic dust and gases would have been the flood basalts that poured over large parts of India, and possibly the North American Tertiary Igneous Province that appears to have been active at the same time as the Deccan Province (Courtillot and Cisowski 1987). Charles Officer and his co-workers (1987) identified the Late Cretaceous paroxysms of volcanism as the chief culprit of plankton extinction and ecological catastrophe among terrestrial plants. They envisaged a relatively gradual deterioration of the environment putting many species under stress, and then a short period of rapid deterioration associated with intense volcanic outbursts that, for many species, were the coup degrace.

Some researchers still favour the flood-basalt volcanism scenario for some mass extinctions, but the debate rumbles on. Paul Wignall (2001), comparing the timing of mass extinctions with the age of large igneous provinces, found seven out of 11 major flood basalt provinces coincide with some form of extinction episode, though in just five cases was there a close correspondence. The four best correlations are for consecutive mid-Phanerozoic events: the end-Guadalupian extinction with the Emeishan flood basalts, the end-Permian extinction with the Siberian Traps, the end-Triassic extinction with central Atlantic volcanism, and the early Toarcian extinction with the Karoo Traps. Oddly, in these four cases, the onset of eruptions slightly postdates the main phase of extinctions. Wignall opined that the link between large igneous province formation and extinctions remains enigmatic: the volume of extrusives and the speed of province formation are unrelated to extinction intensity, and the violence of eruptions appears unimportant. He did find that six out of 11 provinces coincide with episodes of global warming and marine anoxia or dysoxia, this association being suggestive of a connection between volcanic carbon dioxide emissions and global climate. In contrast, he noted little if any geological evidence for cooling associated with continental flood basalt eruptions, which implies that sulphur dioxide emissions have little long-term impact on global climates. The emission of volcanic carbon dioxide is by itself insufficient to account for the large carbon isotope excursions associated with some extinction events and intervals of flood basalt eruption. This was borne out by

Ken Caldeira and Michael R. Rampino (1990), who calculated that the total global warming from carbon dioxide release during the Deccan eruptions, even if they were concentrated in a short and sharp 10,000-year pulse, would be about 0.8°C, too low an increase to cause mass extinctions. A companion mechanism, such as the dissociation of gas hydrates, may be necessary to trigger cataclysmic global environmental changes owing to runaway greenhouses. There is some evidence that this might have happened in the end-Permian and latest Palaeocene events (see below).

Combined impacts and volcanism

Many researchers attribute the three largest mass extinctions over the last 300 million years, which occurred at the very end of the Permian, Triassic, and Cretaceous periods, to massive continental volcanism or bolide impacts, or to a combination of the two (a sort of geological double-whammy). Interestingly, large bolide impacts and massive flood-basalt volcanism occur more frequently than do mass extinctions. It seems reasonable to conclude therefore, that, in isolation, neither of these events will trigger the biggest mass extinctions. Indeed, it is questionable whether either event could lead to the collapse of ecosystems world-wide. However, if large bolide impacts and massive flood-basalt volcanism events were to coincide, then an extinction of truly massive proportions might result. Three such combined cosmic and volcanic events seem to have occurred during the Phanerozoic - the end Cretaceous extinction, the end Triassic extinction, and the end Permian extinction (Figure 7.6). A statistical study by Rosalind V. White and Andrew D. Saunders (2005) showed that large bolide impacts and massive flood-basalt volcanism may coincide, and that the probability of doing so is about one combined event per hundred million years.

Keller and her colleagues (2003) found evidence for the combined action of impacts and volcanism in the Gulf of Mexico, Caribbean, and Central America around the Cretaceous-Tertiary boundary (p. 121). A problem may arise in resolving the relative importance of impacts and volcanism in mass extinctions. Keller (2005), working on DSDP Site 216 on Ninetyeast Ridge in the Indian Ocean, showed that effects of volcanism and impacts on biota are virtually the same. During the late Maastrichtian, Ninetyeast Ridge passed over a mantle plume, which led to volcanic eruptions, islands building to sea-level, and catastrophic regional environmental conditions for planktonic and benthic foraminifera. The biotic effects of this mantle plume volcanism were severe: benthic and planktonic species were dwarfed, species richness plummeted to six to 10 species, species diversity fell by 90 per cent, all ecological specialists disappeared, most ecological general-ists in surface waters disappeared, and blooms of Guembelitria (a disaster opportunist) alternating with low oxygen-tolerant species dominated the ecosystem. Keller (2005) noted that these faunal characteristics are nearly identical to those of the Cretaceous-Tertiary boundary mass extinction, except that the fauna recovered after Site 216 passed beyond the influence of mantle plume volcanism about 500,000 years before the end of the Cretaceous. Her conclusion was that impacts and volcanism cause similar environmental catastrophes, which calls for a review of current impact and mass extinction theories.

Methane hydrates

Methane hydrates occur under the sea-floor and in permafrost. Should some mechanisms open these huge reservoirs of methane, then global environmental changes may ensue. Methane is a powerful greenhouse gas and reacts with oxygen to form carbon dioxide.

Figure 7.6 Extinction rates of marine genera versus time, with eruption ages of continental flood-basalt provinces and three huge bolide impacts shown. Three of the severest extinctions, the end-Permian, the end-Triassic, and the end-Cretaceous, correspond with the eruption of the Siberian Traps, Central Atlantic Magmatic Province, and Deccan Traps, respectively. Evidence of impact (*) has also been reported at these times (Alvarez et al. 1980; Becker et al. 2001; Olsen et al. 2002; Basu etal. 2003). The end-Cretaceous crater is about 180 km in diameter; the size and existence of craters associated with the end-Permian and end-Triassic events are unconfirmed. The end-Guadalupian extinction, which occurred around 259 million years ago, coincided with eruption of the Emeishan Traps (Zhou et al. 2002), but no evidence for impact has been noted for this boundary. Oceanic plateaus may also have had profound environmental consequences (e.g. Kerr 1998), and selected oceanic plateaus are therefore included on this figure, but as text only, because the preservational bias of the geological record towards younger examples would otherwise render the diagram misleading. Source: Reprinted from Lithos, 79, R V. White and A. D. Saunders, Volcanism, impact and mass extinctions: incredible or credible coincidences, 299-316, Copyright © 2005, with permission from Elsevier.

Age (million years)

Figure 7.6 Extinction rates of marine genera versus time, with eruption ages of continental flood-basalt provinces and three huge bolide impacts shown. Three of the severest extinctions, the end-Permian, the end-Triassic, and the end-Cretaceous, correspond with the eruption of the Siberian Traps, Central Atlantic Magmatic Province, and Deccan Traps, respectively. Evidence of impact (*) has also been reported at these times (Alvarez et al. 1980; Becker et al. 2001; Olsen et al. 2002; Basu etal. 2003). The end-Cretaceous crater is about 180 km in diameter; the size and existence of craters associated with the end-Permian and end-Triassic events are unconfirmed. The end-Guadalupian extinction, which occurred around 259 million years ago, coincided with eruption of the Emeishan Traps (Zhou et al. 2002), but no evidence for impact has been noted for this boundary. Oceanic plateaus may also have had profound environmental consequences (e.g. Kerr 1998), and selected oceanic plateaus are therefore included on this figure, but as text only, because the preservational bias of the geological record towards younger examples would otherwise render the diagram misleading. Source: Reprinted from Lithos, 79, R V. White and A. D. Saunders, Volcanism, impact and mass extinctions: incredible or credible coincidences, 299-316, Copyright © 2005, with permission from Elsevier.

Methane hydrates are a large reservoir of very light carbon (carbon isotope ratios in the range -60 to -65 per mille) in the atmosphere-ocean system (Dickens et al. 1995, 1997). In theory, once released, the methane would lead to ocean warming and the dropping of the oceanic thermocline, perhaps with continued dissociation and a 'runaway greenhouse' effect (Dickens et al. 1995). Such a process may account for the end Palaeocene carbon isotope excursion, which coincides with the warming event known as the Late Palaeocene thermal maximum (Dickens et al. 1995). However, the degree of warming due directly to methane release was probably modest, with a global surface temperature rise of about 2°C, although changes in the oceanic thermohaline circulation may have generated an added temperature rise (Dickens et al. 1997; see also Berner 2002).

Methane release might explain other isotopic excursions in the stratigraphic record, including the end-Permian (Krull and Retallack 2000; Wignall 2001; Berner 2002; Retallack et al. 2003), the Aptian (Jahren et al. 2001), and the Toarcian (Hesselbo et al. 2000). For instance, there is evidence that such a process aided the end Permian extinctions (Retallack et al. 2003). Massive releases of frozen methane hydrates would have reduced atmospheric oxygen levels, making it difficult for terrestrial vertebrates to breath, and would have raised the atmospheric carbon dioxide content, causing a dramatic warming of climate. Sharp changes in carbon isotope ratios may reflect such a change. Interestingly, Lystrosaurus survived the extinction event, arguably because it was adapted to living in burrows, which have low oxygen levels and high carbon dioxide levels. Several features of Lystrosaurus show this adaptation - a barrel chest, thick ribs, enlarged lungs, a muscular diaphragm, and short internal nostrils. Therefore, while most Permian animals died of asphyxiation, Lystrosaurus survived and spread rapidly, accounting for 90 per cent of post-extinction fauna in some areas. Coal swamps and coral reefs, which disappeared for millions of years after the mass extinctions, are sensitive to low oxygen levels and may have succumbed to oxygen depletion, too. Against this view, it seems that carbon dioxide released from the Siberian Traps volcanism could have caused the carbon-isotope excursion observed across the Permian-Triassic boundary without calling on methane release (Grard et al. 2005).

Climatic change

Climatic change was for a long time deemed the principle cause of extinctions (e.g. Simpson 1953). While recent decades have seen the watering down of the link between climate and extinction, there is still much evidence that points to climatic change as potential disruptor of ecological stability (Stanley 1984a, 1984b). Indeed, the global changes induced by bolide impacts and volcanism act primarily through climatic change. Nonetheless, many other factors cause climate to change, including continental drift.

The rearrangement of landmasses may causes changes of climate large enough to precipitate bouts of extinction. If a continent should drift from one climatic zone to another, extinctions are likely: the northwards drift of India probably led to the demise of several groups of land plants (Knoll 1984). Plate motions affecting the Gondwanan landmass and south-western Laurasia at the close of the Triassic period brought South America and southern Africa into low latitudes and produced increasing aridity on those continents. The drier climate in turn brought about floral changes: new plants evolved the better fitted to the arid conditions. This floral evolution had repercussions higher up the food chain: the mammallike reptiles and rhynchosaurs became extinct because they were unable to feed on the lowland bush vegetation that had previously supported them (Tucker and Benton 1982).

Steven M. Stanley believed that periods of global cooling, caused by the encroachment of continents on one or other of the poles, have been a prominent cause of marine biotic crises. He felt that climatic cool periods have had a far greater effect on the marine biosphere than have reductions in sea-floor area associated with global regressions of the sea. The oldest biotic crises identified by Stanley were in the Palaeozoic era - in late Ordovician, late Devonian, and late Permian times (Stanley 1988a, 1988b; see also Copper 1994). Each of these biotic crises was long-drawn-out. In each of them, tropical marine biotas, including stenothermal calcareous algae, declined greatly, and reef communities were decimated. As the Ordovician and Permian crises wore on, so warm-adapted taxa were displaced towards the equator, and as the Devonian crisis got under way, so tropical taxa died out in New York State while cold-adapted hyalosponges expanded. In the aftermath of each marine crisis, biotas became cosmopolitan, little or no reef growth took place, and the deposition of carbonates diminished. Stanley (1988a, 1988b) makes much of the coincidence between these crises and the occurrence of glacial episodes, apparently triggered in each case by the proximity of a major continent to one of the Earth's poles. To him, this coincidence and the similar pattern of taxial change in each crisis - preferential loss of tropical taxa, replacement at low latitudes of warm-adapted by cold-adapted forms, and an aftermath in which cosmopolitan faunas prevailed, limestone production was diminished, and reef production was for a long while suppressed - implicates climatic deterioration, probably resulting from plate movement, as the primary agent involved in these Palaeozoic extinction events. A similar sequence of events may account for biotic crises in late Eocene times and at the boundary of the Pliocene and Pleistocene epochs, too, have been linked with the expansion of glaciers and ice sheets (Stanley 1984a, 1986).

Marine regression and transgression

Thomas Chrowder Chamberlin (1909) noted an apparent relationship between regressions and mass extinctions in the marine realm. Raymond C. Moore (1954) explored this relationship in Palaeozoic rocks of North America. However, Norman D. Newell was the first to promulgate an unambiguous hypothesis connecting regression with mass extinctions during the Phanerozoic. Newell firmly believed that regressions and transgressions of the sea were potent forces of mass extinction. He reasoned that the present relief of the continents is much greater, and the land surface more uneven, than has been usual through geological history. Less relief would mean that relatively small epeirogenic movements or changes of sea level could produce enormous geographical and climatic changes. A sea-level rise (or fall) of just a few metres would have sufficed to cause the initiation of mass extinctions (Newell 1962, 1963). Newell was unclear as to the length of time involved in these revolutionary changes, but where the land surface was low and very flat, the migration of the strandline might have been rapid enough to have a cataclysmic effect. Such swift changes of sea-level and resulting mass extinctions are, according to Newell, recorded in the stratigraphical column: many transgressions and regressions have affected much of the world in short spans of time. Later researchers have also stressed the important role of sea-level change in some mass extinctions (e.g. Hallam 1984a; Wiedemann 1986; Wignall and Hallam 1993; Bardet 1994; Sandberg et al. 2002).

Anoxia and hypoxia

Marine transgressions commonly lead to the widespread development of hypoxic to anoxic ocean water and the formation of black shales, though the mode of formation is unclear. Not all transgressions produce black shales, but only the globally distributed black shales correlate with mass extinctions (Hallam and Wignall 1997, 251; see also Wignall and Twitchett 2002).

Oxygen shortage also affects land organisms. Raymond B. Huey and Peter D. Ward (2005), noting that background extinction rates and ecosystem turnover for terrestrial vertebrates were elevated for much of the Late Permian and well before the mass extinction, suggested a period of sustained environmental degradation before the final catastrophe. To account for this pattern, they invoke a combination of reduced atmospheric oxygen and climatic warming inducing hypoxic stress (Figure 7.7). The effect of less oxygen in the atmosphere is to compress altitudinal zones (Figure 7.8). Simulations indicated that the degree

Figure 7.7 (a) Oxygen and carbon levels over time and relative to present-day levels. Global hypoxia would have occurred in the Late Permian and Triassic because of falling oxygen combined with rising temperatures. The dashed vertical line indicates the mass extinction at the Permian-Triassic boundary. (b) Present-day altitude with partial pressure of inspired oxygen equivalent to that at sea-level in the Phanerozoic. Thus, at sea-level, the Triassic oxygen minimum would be equivalent to oxygen levels found at about 5 km today. Source: Reprinted with permission from R. B. Huey and P. D. Ward (2005) Hypoxia, global warming, and terrestrial Late Permian extinctions. Science 308, 398-401. Copyright © 2005 AAAS.

Age (million years)

Figure 7.7 (a) Oxygen and carbon levels over time and relative to present-day levels. Global hypoxia would have occurred in the Late Permian and Triassic because of falling oxygen combined with rising temperatures. The dashed vertical line indicates the mass extinction at the Permian-Triassic boundary. (b) Present-day altitude with partial pressure of inspired oxygen equivalent to that at sea-level in the Phanerozoic. Thus, at sea-level, the Triassic oxygen minimum would be equivalent to oxygen levels found at about 5 km today. Source: Reprinted with permission from R. B. Huey and P. D. Ward (2005) Hypoxia, global warming, and terrestrial Late Permian extinctions. Science 308, 398-401. Copyright © 2005 AAAS.

Age (million years)

Figure 7.8 Predicted maximum altitude over time for hypothetical species having graded tolerances (2-8 km) to hypoxia. From the Late Permian through the Jurassic, however, partial pressure of inspired oxygen was sufficiently low that ranges would have been compressed to near sea-level and some species would have gone extinct. Source: Reprinted with permission from R B. Huey and P. D. Ward (2005) Hypoxia, global warming, and terrestrial Late Permian extinctions. Science 308, 398-401. Copyright © 2005 AAAS.

Age (million years)

Figure 7.8 Predicted maximum altitude over time for hypothetical species having graded tolerances (2-8 km) to hypoxia. From the Late Permian through the Jurassic, however, partial pressure of inspired oxygen was sufficiently low that ranges would have been compressed to near sea-level and some species would have gone extinct. Source: Reprinted with permission from R B. Huey and P. D. Ward (2005) Hypoxia, global warming, and terrestrial Late Permian extinctions. Science 308, 398-401. Copyright © 2005 AAAS.

of altitudinal compression was high and severely reduced the altitudinal range of all hypothetical species modelled. If the terminal Permian oxygen level were about 16 per cent, the partial pressure of inspired oxygen would have been the same as that found today at 2.7 km elevation. None but the most hypoxia-tolerant species would have survived. In addition, the altitudinal compression would have forced extinctions by reducing habitat diversity, fragmenting and isolating populations (even a low mountain would present a formidable obstacle against dispersal), whilst rising sea levels would have reduced the area of suitable habitat, perhaps causing additional extinctions through the species-area effect.

Diseases

Several researchers stress the potential role of diseases as drivers of mass extinctions. Lethal pathogens carried by the dogs, rats, and other animals associated with migrating humans may have caused the Pleistocene mass extinctions (MacPhee and Marx 1997). Similarly, it is possible that the terminal Cretaceous extinction event might have resulted from changes of palaeogeography, in which land connections created by falling sea-levels allowed massive migrations from one landmass to another, leading to biotic stress in the form of predation and disease (Bakker 1986, 443).

Was this article helpful?

0 0

Post a comment