본문 바로가기
읽기 · 학술 · 공개 라이선스 원문

Theory and classification of mass extinction causation

Algeo, Shen · 2023

공개 학술지 National Science Review 에 실린 총설 논문입니다(2023년). CC BY 4.0 으로 공개된 글이고, 그림 · 표 · 참고 문헌 표시를 뺀 본문을 실었습니다.

출처 Algeo & Shen (2023), National Science Review · 라이선스 CC BY 4.0 · 원문

장 5개 올림 이 장 2795낱말
← 앞 장

이 장은 원문만 있습니다. 문장별 해석은 앞 장부터 차례로 붙입니다.

5장 · Theory and classification of mass extinction causation — Biocrises with probable tectono-oceanic triggers

Tectono-oceanic processes, e.g. orogenic uplift, opening and closing of oceanic gateways, and latitudinal redistribution of continents and oceans, are capable of triggering large-scale climate changes in the Earth system, as shown by many events through Earth history. This mechanism is not considered to have been the ultimate cause of any of the Big Five mass extinctions, but it may have caused or contributed to several second-order biocrises (Fig. 4). Given that the underlying endogenic forces precipitating such changes operate at timescales of millions to tens or hundreds of millions of years at exceedingly slow rates, it might be expected that biotic communities adjust to the accompanying environmental changes through migration or gradual evolutionary processes. However, tectono-oceanic processes can trigger moderately rapid environmental changes by the crossing of a key tipping point or climatic threshold (e.g. one that triggers a rapid response, such as the onset of continental icesheet growth; cf) or by the opening of continental pathways or oceanic gateways that result in faunal migrations or exchanges. For all such events, a key question is whether such changes can proceed swiftly enough to create sufficient ecosystem stress to precipitate a biocrisis.

The Late Mississippian mass extinction, which occurred during the Serpukhovian Stage at ∼330–324 Ma, resulted in large losses of diversity among brachiopods and reef fauna, although this biocrisis was at least partly due to low origination rates. It coincided with a major expansion of continental ice mass in Gondwana, a large mid-Carboniferous eustatic fall that produced a deep unconformity in North America and elsewhere, and strong climatic cooling, as recorded in O-isotope records. These features, accompanied by PCIEs, mark the Late Mississippian biocrisis as a carbon-burial event (Fig. 3). These observations suggest that global climatic cooling was the main cause of extinction. The trigger for this event remains uncertain, but the lack of any obvious candidate other than tectonism suggests that climatic cooling driven by uplift of the Appalachian-Hercynian orogens reached a critical threshold around the end of the Mississippian (Fig. 4).

The ∼90-Ma end-Cenomanian mass extinction, which severely reduced diversity among rudistid bivalves, benthic foraminifera and ammonoids, was probably related to the opening of an oceanic gateway between the paleo-Southern Ocean and the Neotethys, in response to India's initial northward movement away from the other Gondwanan continents. This event was associated with an unusual combination of global warming and higher marine carbonate δ¹³C values. Based on the former observation as well as widespread development of anoxia, an LIP trigger was proposed, although the source of magmatism has been variably cited as the Columbian-Caribbean Province, the High Arctic LIP, the Madagascar LIP or the Ontong-Java Plateau. Although the Cenomanian–Turonian Boundary (CTB) interval was associated with peak oceanic spreading rates, the weakness of Hg enrichments throughout this interval argues against an LIP trigger, and positive C-isotope anomalies (Fig. 2F) reveal it to have been a carbon-burial event rather than a carbon-release event (as is typical of LIPs). Changes in Nd-isotopic compositions around the CTB across the Atlantic to Tethyan regions provide evidence of major changes in global-oceanic circulation, consistent with the opening of an ocean gateway triggering this event (Fig. 4).

The ∼34-Ma end-Eocene biocrisis, which resulted in modest extinctions among forams, mollusks and mammals, was linked to rapid climatic cooling and initial growth of the Antarctic icesheet. Although the proximate cause of the biocrisis was doubtlessly climate change, it is unclear whether the ultimate cause was a bolide impact or merely a non-linear, threshold climate response to a long-term, tectonically driven decline of atmospheric CO2. Several potential impactors have been identified, with the 34.86-Ma Chesapeake Bay impact being a better fit temporally than the 35.7-Ma Popigai Crater; the former event was large enough to produce the widespread North American tektite strewn field (i.e. an ejecta layer). It is also possible that the end-Eocene biocrisis was a response to a combination of long-term CO2 fall and short-term impact-related cooling (see below).

A final example of what was certainly a tectonically driven biocrisis was the demise of various megafauna, especially marsupials in South America, during the ‘Great American Interchange’, which was triggered by the formation of the Panamanian land bridge at 2.8 Ma. This event led to nearly 25% of native South American mammalian genera going extinct, and by the end of the period of exchange more than half of the genera in South America were of North American origin (n.b., the effects of this event on the North American mammalian fauna were much smaller). This biocrisis was clearly of tectonic origin and carbon-neutral in its carbon-cycle effects (Fig. 4).

Other issues related to mass extinction causation

Discerning causation from biotic patterns

An undervalued approach to discerning proximate causation and its possible implications for ultimate extinction mechanisms is examination of the ecological patterns of mass extinction. Whereas extinction intensity tells us about the severity of a biocrisis, it provides no information concerning the types of ecological changes or the degree of ecosystem restructuring accompanying a mass extinction. The idea that biological selectivity and ecological patterns during an extinction event could be information-rich was proposed in the 1980s. Subsequently, the hypothesis that morphological and ecological features promoting survival during mass extinctions differ from those favoring success during non-crisis intervals was formulated. Ecological selectivity may be operating during the LQME, with extinction intensity predicted to be highest in high-latitude regions, but total species losses to be greater in the tropics due to the much higher biodiversity baseline.

Some general biotic patterns are evident in all biocrises. Taxon extinction probability increases with body size, due in part to larger animals being higher in the food chain and thus more susceptible to disruptions anywhere at lower trophic levels, and in part to smaller population sizes. One example of this is the disappearance during the EPME of all 25(+) species of gorgonopsids, the apex terrestrial predator of the Late Permian, whereas tetrapod clades at lower trophic levels (e.g. cynodonts, dicynodonts, procolonophonids and temnospondyls) survived in part. Other known factors influencing survival probability include geographic range and, for higher taxonomic levels, lower-order taxon richness (i.e. genera containing more species are less likely to go extinct).

For the purpose of discerning ultimate causation, biotic patterns that are not shared by all mass extinctions are more useful than the general patterns discussed above. Based on an analysis of extinction selectivity during the EPME, it has been inferred that hypercapnia (high CO2) exerted a strong influence on marine invertebrate faunas, with carbonate-secreting organisms being especially susceptible as well as those with a high sensitivity of respiratory pigments and/or a limited ability to adapt metabolically to high or fluctuating CO2 levels. Secondary environmental factors contributing to mortality included hyperwarming, anoxia and sulfide toxicity, with trophic stresses becoming acute at higher levels in the food web as ecosystem collapse proceeded. An analysis of extinction patterns among fish during the EPME concluded that broader tolerance to environmental variation was an important factor in survival, although factors such as a more active physiology and larger body size also conferred survival advantage. A key point is that the responses of individual taxa as well as entire ecosystems were driven by a confluence of environmental stresses that operated simultaneously, in some cases in mutually reinforcing ways.

Multiple causation of mass extinctions

Many studies have inferred multiple proximate causes for mass extinctions. Indeed, given that global catastrophes cause changes throughout the Earth-surface system, concurrent changes in temperature, ocean redox conditions, acidity and other environmental variables must be the norm. In most cases, it is difficult or impossible to tease apart the effects of multiple concurrent climato-environmental changes and determine their relative importance for an extinction event. What can be stated with assurance is that redox and temperature changes are regarded as having been particularly important in many Phanerozoic biocrises.

The issue of whether any mass extinctions had multiple ultimate causes is intriguing and as-yet unanswered. In general, invoking multiple causes for a single mass extinction appears to violate Occam's Razor, which favors the simplest possible explanation for a given set of observations. However, there are some examples in Earth history of biocrises with possible multiple causation—the most likely being cases where one factor has produced long-term environmental or ecosystem stresses that have amplified the effects of a second, unrelated factor. In this regard, the biocrisis that has received the most attention to date is the ECME, which coincided with both a bolide impact (the Chicxulub crater in Mexico) and an LIP (the Deccan Traps in India). Recent high-resolution dating has demonstrated that flood basalt activity in India spanned a ∼750-kyr interval bracketing the ECME, whereas the age of the Chicxulub crater exactly matches that of the biocrisis. Thus, if the Deccan Traps eruptions played a role in the ECME, it was probably to generate widespread environmental stresses that made the effects of the Chicxulub impact more severe than they might otherwise have been, although the possibility of an eruptive stage exactly coinciding with the ECME requires further investigation. The hypothesis that overlapping bolide–LIP events yield systematically higher extinction rates than either mechanism alone is inherently plausible, although the number of such events is too small for a robust statistical evaluation.

The question of multiple causation has arisen also in the context of the LDME. The land-plant–weathering hypothesis offered two scenarios: a ‘strong version’ in which both the long-term trends throughout the Devonian (i.e. climatic cooling, atmospheric CO2 drawdown) and the short-term biocrises of the end-Frasnian, end-Devonian and other horizons were due to land-plant bioevolutionary events, and a ‘weak version’ in which land-plant evolution merely created a backdrop of environmental stresses on which other events (e.g. bolide impacts or LIP eruptions) then operated episodically to produce extinction rate maxima. Although both possibilities remain under consideration, it should be noted that the end-Frasnian and end-Devonian biocrises were associated with strong short-term cooling and PCIEs, i.e. hallmarks of productivity-driven biocrises inconsistent with the temperature-driven processes that are typically associated with bolide impacts and LIP eruptions (see above). The potential productivity drivers of these short-term events are uncertain, although the Late Devonian coincided with the spread of archaeopterid forests and seed plants. The latter development was particularly important because it permitted vascular plants to move out of moist lowland habitats and invade the drier upland regions that make up large parts of continental interiors, making it arguably the single most important evolutionary innovation in terms of areal expansion of land-plant habitat. Thus, the ‘strong version’ of the land-plant–weathering hypothesis appears more likely at present, although this matter certainly warrants further investigation.

One objection to the land-plant–weathering hypothesis as a trigger of short-term biocrises relates to rates. Specifically, it has been argued that early land plants spread slowly across the globe over a time span of tens of millions of years and therefore could not have caused short-term biocrises such as the Kellwasser (end-Frasnian) and Hangenberg (end-Devonian) events, the estimated duration of which are between 50 kyr and a few hundred kyr. This view reflects a misunderstanding of the nature of biotic expansions, which are generally not slow and continuous but episodic or ‘punctuated’ (note that the same debate regarding evolutionary tempo took place 50 years ago). This point is illustrated by the history of humans: although their initial evolution in Africa was comparatively slow (>1 Myr), the subsequent spread of modern man (Homo sapiens) across the globe was much more rapid, i.e. into southwest Asia and southern Europe by 100 ka, into Australasia by ∼50 ka, into the Americas by ∼15–13 ka, across the Indo-Pacific at ∼3–1 ka, and into all remaining areas globally during the Age of Discovery (0.5–0 ka) (Fig. 5). Although punctuated, the spread of modern humans has been rapid at geological timescales because of their high level of mobility—in the ancient world, jumps of plant and animal species across oceans may have occurred more sporadically in time. It also illustrates a general pattern that, although the evolution of a new taxon or clade may be protracted on its landmass of origin, subsequent expansions in range (i.e. across newly colonized landmasses) are rapid events. This latter point is well-illustrated by the rates of spread of modern invasive species across newly invaded continents, a process that typically requires 10¹–10³ years, i.e. a timescale that is effectively instantaneous from a geological perspective (Fig. 5).

Periodicity in the mass extinction record

One area of mass extinction research that is closely linked to the bolide impact model is the issue of periodicity (or ‘quasi-periodicity’, as argued in some studies). The hypothesis of a ∼26-Myr periodicity in Phanerozoic biocrises was first promulgated by Raup and Sepkoski, with variations on this theme invoking other periodic intervals. A key assumption of any periodic mechanism is that all extinctions must have had the same ultimate cause. Most such hypotheses have invoked astronomical controls, e.g. the orbit of an unidentified companion star (‘Nemesis’) through the Oort Cloud, vertical motions of the Solar System across the galactic plane, or rotation of the Milky Way Galaxy (see review in), although it also has been suggested—rather improbably—that non-astronomical intrinsic biological mechanisms could yield extinction periodicity. However, quite apart from the statistical problems of such analyses (which are addressed in), the fact that Phanerozoic mass extinctions have been triggered by multiple, fundamentally different and unrelated mechanisms effectively precludes any form of periodicity in their distribution through time. Although ‘mass extinction periodicity’ is still a popular topic in the literature (e.g., Fig. 1), it is well past time to discard the idea that there is any temporal regularity to the extinction record.

Long-term changes in mass extinction causation

This survey of mass extinction causation reveals a feature of potential interest, i.e. that over the course of Earth history there has been a transition from dominantly carbon-burial to dominantly carbon-release events (Fig. 4). This is indisputably the case for the Big Five mass extinctions, among which those associated with carbon burial (i.e. LOME and LDME) preceded those associated with carbon release (i.e. EPME, ETME and ECME). Even though some exceptions to this pattern can be found among the second-order biocrises, it is nonetheless noteworthy. This pattern is perhaps to be understood in the context of the ‘Earth as fuel cell’ theory of, in which the Earth's biosystems have reduced carbon and effected its net transfer from the atmosphere-hydrosphere to the lithosphere over several billion years, charging the planetary fuel cell, after which the crustal reservoir of reduced carbon has been repeatedly tapped by various processes that have released a portion of the stored energy. These carbon-tapping processes include bolide impacts, LIP eruptions, and now anthropogenic use of fossil fuels. In this regard, the shift from dominantly carbon-burial biocrises to carbon-release biocrises toward the end of the Paleozoic Era represents the development of a highly charged planetary fuel cell, a condition that may have been associated with ‘peak organic carbon burial’.

FINAL THOUGHTS

Scientific progress has benefitted enormously from application of the principle of multiple working hypotheses. However, the literature on mass extinction causation is strewn with poorly tested hypotheses that have been uncritically adopted for many biocrises. Some of the hypothesized triggers for mass extinction events have lived on well past their sell-by dates, despite the fact that overwhelming evidence points in other directions. One example of this is the Late Quaternary extinction of megafauna in the Americas. As noted in, ‘the overkill hypothesis, at least in general terms, already has been ‘proven’ as thoroughly as any historical hypothesis can be. All of the key evidence was available years ago, and all of it firmly refutes competing, ecologically oriented hypotheses. The event's timing, rapidity, selectivity and geographic pattern all make good sense according to the anthropogenic model, and no sense at all otherwise.’ It is a form of false equivalence that all extinction mechanisms must be given equal weight in a scientific study, much as we see in the media today regarding political issues, that all sides must be given equal time in the limelight. More nuanced judgments are needed, and failed ideas should be jettisoned. In this regard, the ‘one size fits all’ approach to the analysis of mass extinction causation should be permanently laid to rest. It is long since time to acknowledge that extinctions in Earth history have multiple ultimate causes. Apropos of this, the perspicacious words of the American author F. Scott Fitzgerald are called to mind: ‘The test of a first-rate intelligence is the ability to hold two opposing ideas in mind at the same time and still retain the ability to function’. In the present instance, the intellectual task is less onerous—what is needed is not the ability to reconcile two opposing ideas but, rather, to recognize that two or more causation mechanisms for mass extinctions can simultaneously be valid.

Conflict of interest statement. None declared.

이 글은 위 출처의 원문을 그 라이선스(CC BY 4.0)에 따라 옮겼습니다. 참고문헌 목록과 인용 번호, 그림 · 표 · 각주 · 수식 · 부록은 빼고 절별로 나누어 실었으며, 그 밖의 문장은 원문 그대로입니다. 원 저작자와 발행 기관은 이 화면을 만들거나 검토하지 않았고, 이곳의 해석과 낱말 풀이는 공식 번역이 아닙니다.