Arguments
Rapid climate changes more deadly than asteroid impacts in Earth’s past – study shows.
Posted on 27 May 2014 by howardlee
The link between rapid climate changes and mass extinctions has been strengthened in a recent paper by Jourdan et al in “Geology.” The authors demonstrate that the extraordinarily huge volcanic eruptions of the Kalkarindji Large Igneous Province (LIP) in Australia were synchronous with a major extinction event in the middle of the Cambrian Period, and they also show that most of the severe extinctions in the Phanerozoic (the time since the beginning of the Cambrian Period) have coincided with rapid climate changes brought about by greenhouse gas and sulfur emissions by LIPs like Kalkarindji, the Siberian Traps, and others.
The “Cambrian Explosion” 541 million years ago was the dawn of the sort of complex animal life on Earth that we would recognize today – including creatures with shells and skeletons - which burst into the fossil record at that time. Yet this early proliferation of animal life was cut short 510.0 million years ago by the Mid-Cambrian extinction in which around 45% of genera went extinct. They did so at exactly the same time that the violent Kalkarindji eruptions let rip across an area around 1/3 the size of Australia, in a geological instant (less than 3 million years, possibly even less but this is the limit of the dating resolution). It wasn’t until 25 million years later in the Ordovician Period that life regained its radiative momentum in the Great Ordovician Biodiversification Event (GOBE).
Jourdan et al state that: “Although rapid climate changes and climate oscillations are likely to be the ultimate cause of mass extinctions, the exact trigger mechanisms arising from LIP emplacements that are responsible for these climatic shifts are less clear.” They note that the vast quantities of magma alone would result in huge releases of CO2 and SO2, and the extraordinary violence of the eruptions would have delivered these gasses to the stratosphere. In addition, sheets of magma injected underground would have baked oil-rich sediments, releasing methane and even more CO2. In fact, blobs of asphalt can be seen in the Australian lava. These attributes are all remarkably similar to the circumstances surrounding the Permian Mass Extinction, in which most complex life on earth was extinguished, as described in an earlier post here.
A Large Igneous Province (LIP) during eruption, loosely based on Svensen et al EPSL 2009, Howarth et al Lithos 2014, Elkins-Tanton GSA Spec Pub 2005, Keller et al J Geol Soc India 2011, Li et al Nature Geoscience 2014
LIP eruptions are very different from the volcanic eruptions that have occurred in the span of human history. The last LIP was the eruption of the Columbia River Basalts in the US northwest, 16 million years ago, long before the evolution of hominins from apes. Most volcanic eruptions have a temporary cooling effect from their sulfur emissions, and collectively over many millions of years their CO2 emissions keep the planet from freezing solid. LIPs, in contrast, tend to generate strong global warming through the huge quantities of greenhouse gasses they emit in a comparatively short time. The work of Jourdan et al adds to the list of extinction events that coincided with LIPs, and the severe climate fluctuations that accompanied them. Even at the end-Cretaceous, when most scientists agree a giant asteroid impact wiped out the dinosaurs, an episode of strong global warming had already triggered a major extinction event just before the impact. The authors note that the correlation between LIPs and severe extinctions is now so strong that there is a “negligible 6×10–9% probability that such correlation is due to chance alone,” which “strongly supports a causal relationship between the LIPs and severe extinctions during the Phanerozoic.”
The Mid-Cambrian extinction, occurring before colonization of land by plants and animals, also suggests that the LIP kill mechanism is unlikely to have been associated with ozone layer collapse, as was recently suggested for the Permian Mass Extinction. So oceans must be critical to the extinction mechanism (see this post), which points the finger even more clearly at rapid greenhouse-gas and sulfur aerosol-driven climate fluctuations as the cause of Earth’s largest mass extinctions.
Asteroid impacts, in contrast, have peppered our planet relatively often over the same time period, but the only Phanerozoic mass-extinction coincident with a large impact was the end-Cretaceous impact (which occurred on top of a LIP-triggered climate change as mentioned above). The popular image of asteroid impacts as the ultimate annihilator of life may be dramatic and make for exciting movies, but the hard reality is that rapid climate changes, such as humans are unleashing on the planet today, have consistently been more deadly to life on Earth.
Synchronicity between Large Igneous Province eruptions (LIPs) and large extinction events. Red denotes dated duration, pink denotes date uncertainty. See Jourdan et al for explanation of Paranà-Etendeka LIP. Redrawn from Jourdan et al, Geology 2014.
A couple of years ago there was a paper that compared modern rates of CO2 emissions/ocean acidification to geologically significant changes and mass extinction events over the last ~300 million years. It turns out that even The Great Dying happened with atmospheric CO2 increasing at rates 1-2 orders of magnitude slower than what we're causing today.
The paper is here (paywalled), and there's a good write-up summarizing it here.
In 2013, the eminent palaeontologist, Steven Stanley was bestowed the Penrose Medal, the top honour awarded by the Geological Society of America. Prof Stanley has long been a proponent of the idea of punctuated evolution, as opposed to the classical, more gradual evolutionary pressures first suggested by Darwin and Wallace.
Stanley's proposed mechanism for these evolutionary spurts is by extinctions (mass and otherwise) either freeing up niches or increasing the competion therein. His suggested mechanism for these extinction events? You guessed it - climate change. Interestingly, Stanley feels that climate change in either direction can be equally efficacious as an extinction mechanism.
Being a thicko engineer, I never covered any of this stuff during either of my degrees. I only have a vague grasp through a general interest in science and wanting to learn more about our planet's history. (And hence getting some insight into its future.)
The OP mentions that "most scientists agree a giant asteroid impact wiped out the dinosaurs". This is obviously out of my field, but I had discussed the K/T (I'll never get used to calling it the K/Pg) on many occasions with one of my neighbours - sadly no longer with us. The gentleman in question however, was a Fellow of the Geological Society of London and was keen to disabuse me of the idea that this was a settled issue.
According to Myles, there were (are?) two diametrically opposed camps in the debate. (Sound familar?) There are those who are convinced by the arguments put forward by the Alvarez/Alvarez/Asaro/Michel camp relating to the Chicxulub bolide as the Deus ex Machina. However, as indeed mentioned in the OP, others look towards the massive basaltic eruptions that spewed forth from the Deccan Traps.
By amazing coincidence, my current bedtime reading is a book called "The Sixth Extinction", by Richard Leakey and Roger Lewin. They also view climate change as a very likely vector for the many of the extinctions the planet has experienced - and continues to experience to this day.
I'm curious what the differential is between pre and post extinction events, if there's a threshold beyond which massive die-off's are (or appear to be) set in motion. Maybe as a percentage of current atmospheric levels, or an absolute change? I've seen there are links to how fast the changes occur, but how much isn't clear (at least not to me). Would love to see an article about CO2 levels at different times, and what they were before and after these events. I imagine such a discussion would have to take into account feedbacks and absorption saturations, et cetera... I can hear the deniers already, "CO2 is already almost saturated! Adding more won't do anything!"
On the volcanic diagram, in the white box at the lower left, it says "fossil fuels". Shouldn't that read fossil oil instead? It became a fuel from our modern persapective.
[JH] Nope. Natural gas is a fossil fuel. It is not fossil oil.
climatelurker - what is important is the rate of change. A rapid change over 3C may have more effect than a very slow change over say 6C. The initial cooling by aerosols and then rapid warming by GHG are probably both important. I think the most useful numbers would be rate of change in forcing (W/m2/century) which is much more informative than % concentration. However, for paleoevents, it is very difficult to get high precision both dating and forcing. CO2 is only indirectly measured by change in a proxy for pH which in itself must have a large time-constant for ocean mixing. Estimating the forcing from aerosols on a volcanic eruption that has never been observed is even more complex. We struggle to measure aerosols even today.
@ climatelurker
As scaddenp rightly states, rate of change is absolutely critical in terms of extinction potential. In the case of both bolides and mega-volcanism, there is a double whammy. The initial spectacular aerosol cooling is likely to be pretty short-lived, as these particles will not have a long atmospheric residence time. On the otherhand, so-called greenhouse gases have this habit of lingering on, and on, and on...
One of the fingerprints of massive CO2 build up is the presence of cap carbonate deposits, as discussed in this piece from the Snowball Earth site.
Also, the following from the BBC describing some of the dramatic events on the planet's timeline might be of more than passing interest. Scroll down to the bottom of that screen and there are links to various extinction theories (incl climate change) as well as the varying habitats that would have been extant at different times.
Incidentally, I've just noticed that one of the scientists appearing in the BBC video clips (Paul Hoffman, from the Earth and Planetary Sciences Dept at Harvard) is also involved with the Snowball Earth project.
Villabolo - you are correct for the Kalkarindji LIP - the authors mention oil specifically. In the Permian Siberian Traps LIP it looks like it was mainly coal. In this generalized figure I used "fossil fuels" as a collective noun and to make clear the parallels between the past and today's climate change. In all cases the heat from the magma baked the organic-rich sediments converting them to methane and CO2 (with a host of other nasties mixed in).
Climatelurker - It''s clear that some LIP events have been associated with severe extinctions and others have had more mild extinctions. The Emeishan LIP that occurred just before the Siberian Traps generated global warming and an extinction event much less severe than the Siberian Traps did.
That some LIPs have more severe effects on life than others is a puzzle. It probably relates to eruption rate, presence/absence of organics in the sediments at the LIP location, and other random things. As the authors of one paper noted: "...larger eruptions of flood basalts which clearly had no effect on global ecosystems (e.g. Tarim). The reason why some LIPs are contemporaneous with mass extinctions could be related to random conditions which cannot easily be predicted such as: parental magma composition, geographic location, composition of country rock, or vulnerability of species."
billthefrog - I think most workers in the field accept the end cretaceous was a "double whammy" - LIP-generated global warming triggering a significant extinction, with the impact dealing a final blow. It's interesting to note that the pendulum is definitely swinging back to emphasize the role of the Deccan Traps LIP as originally highlighted by Gerta Keller.
@billthefrog - Thanks for letting us know about the interesting Snowball Earth web site. So far as I can tell, though, the last news postings at that site were in 2006. Is there a reason why the site hasn't been kept up? Are there any other sites dealing with the Snowball Earth hypothesis that might provide more current information?
@ Joel
Oops! I discovered the Snowball Earth site a few years back, but hadn't noticed that it had gone inactive. Mea culpa
NOAA has an interesting palaeoclimatology portal that you might find useful.
Examples of teaching resources covering this material can be found here and here.
A far more detailed description of the first (?) of these events way back in the Palaeoproterozoic Era, the Huronian glaciation, was given in the Proceedings of the National Academy of Sciences.
Hope that helps.
Joel and Bill - There have actually been 2 "snowball" episodes, and a number of major ice ages on the planet, long before the most recent Pleistocene ice age.
About 2.9 billion years ago there was an ice age when microbes evolved methanogenesis, which reduced the CO2 levels in the atmosphere. This was early in Earth's life a time when the sun was less bright than now.
The first snowball Earth (actually more of a slushball because some areas were not frozen) coincided with the rise of atmospheric oxygen around 2.5 billion years ago. The Earth froze because there had been a lot of methane in the atmosphere, which oxidized on contact with the newly-minted oxygen, removing a greenhouse gas blanket and plunging the planet into a deep freeze in 3 separate ice ages spread over 200 million years.
The second snowball Earth (the famous one - but even this was more of a slushball) occurred around 700 million years ago. This was in response to the first greening of land surfaces, which boosted oxygen levels that had been declining since the Great Oxidation Event 2.5 billion years ago. Boosting oxygen once again removed methane that had recovered somewhat as oxygen declined, but more importantly it boosted rock weathering rates which drew down CO2 from the atmosphere. The episode involved at least 3 separate global ice ages, 720-700 million years ago, 650-630 million years ago, and briefly 580 million years ago.
There was another brief ice age around 517 million years ago known as the MICE event in response to the Cambrian Explosion proliferation of life (and probably enhanced biological pump).
There was another major ice age in the Ordovician in response to the rise of vascular plants and the enhanced weathering and carbon drawdown they triggered.
The ice ages in the Carboniferous-Permian were probably a combination of so much carbon being locked away as coal (CO2 levels down to about 220ppm) and the coincidence of high continental areas at the South Pole. There's some suggestion that an asteroid impact may have tipped the balance to ice age at one point also.
I like following the Skeptical Science discussions but I'm only a lay person with regard to climate science. I was wondering whether there were any studies that related continental drift and the break up of the supercontinents and shifting of the continental plates around the planet to the excessive volcanic activity that led to these mass extinctions.
I know that the IPCC do have a climate prediction outcome for business as usual, but has anyone done a maximum emissions outcome if all the known reserves of fossil fuels are consumed. Also, it seems to me that the only analyses regarding emissions is based on current consumption, where great comfort is taken by deniers that there will be 100 to 200 years of coal left to burn. But, this assumes that the developing nations will remain low emitters per capita. What would happen if everyone in the world were emitting per capita the same as the high emitter per capita countries like Australia or the US? Unlike the CO2 doubling debate, perhaps there also needs to be a worst case maximum emission scenario debate, because this would get to the heart of the equity debate. My simplistic analysis of such a scenario indicates that there would only be about 40/50 years of coal left if we consumed at a maximum emissions rate and we would put the CO2 levels back to the time of the dinosaurs in a very short time. I wonder what the predictions regarding the climate and mass extinctions would be in such a scenario.
mancan @18, a worst case emissions scenario would not just consume proven reserves (on which your figure is based) but a substantial proportion of the total resource base (TRB). That is, it will access coal not currently commercially viable either due to increased energy prices, or due to improved technology (such as coal seam gassification). Unsurprisingly, estimates of the TRB are less certain than those for proven reserves. At the outside, determined exploitation of the fossil fuel TRB would lift atmospheric CO2 levels above 5000 ppmv (based on the IEA 2011 estimate). At the low end, determined exploitation of the TRB will lift atmospheric CO2 levels to about 1500 ppmv (based on WEC 2010 figures). The former will raise temperatures to levels at which multicellular life may be impossible - and certainly impossible for large, warm blooded vertebrates such as humans. The later will lift temperatures to levels at which periodic tropical heat waves will result in 100% mortality for large, warm blooded vertebrates. That may be devestating, even apocolyptic for the economy, but should not represent an extinction threat for humans (many of whom live outside the tropics). Intermediate estimates will reproduce conditions very similar to those during the end Permian extinction, making the tropics seasonally uninhabitable for large, warm-blooded invertebrates.
For practical purposes, I do not think these scenarios are relevant in that they would require sustained, high tech exploitation of fossil fuel resources well into the next century - something that may be made economically impossible by global warming, and certainly socially impossible. I think the upper end of realistic scenarios is well represented by the IPCC's RCP 8.5 scenario out to 2100:
Beyond 2100, we will need to reduce CO2 emissons to zero to avoid the very dangerous tail of the RCP 8.5 scenario.
billthefrog @6, while rate of change is the main driver of extinction, the Earth can, and has in the past, reached temperatures where the absolute temperature value drives extinction. The obvious examples are the two slushball Earth episodes discussed by howardlee @12. Equivalent episodes today would extinguish all, or nearly all complex multicellular life on Earth, regardless of the rate of onset.
Of more concern are episodes like the end Permian mass extinction where tropical temperatures rose sufficiently to be seasonally uninhabitable. Global temperatures of about 10 C above the preindustrial will make large sections of the tropics uninhabitable to large warm blooded vertebrates based on absolute values of wet bulb temperature. There is evidence that the absolute limits are to be found for most complex multicellular life at temperatures about that level.
Humans would have to be terminally foolish to raise temperatures so much that absolute limits on temperature become a major factor in extinctions. Currently, however, terminally foolish best describes global policies on geen house gases.
Thanks Howard,
What you say pretty well reinforces what I understood from the Climate Change MOOC I recently did at Exeter University.
The reason I put a question mark on the Palaeoproterozoic snowball/slushball event having primacy was simply cause I don't know enough about the earlier Archean Eon. Could the Faint Young Sun of those far off days have been instrumental in an even earlier snow/slush ball? Do we have enough preserved crust from that long ago to tell? I know that some zircon crystals have been radio-dated at around ~4.3 billion years, but whether there is sufficient preserved crustal material - I just don't know.
Moving to more recent times and the Pleistocene glaciation. If my understanding is correct (cue a huge guffaw from the wife at this point), temperature wise we've been on a downward slope since the heady days of the Eocene. The Indian Plate went sailing full tilt into the Eurasian Plate starting about 65 MYA and causing (initially) a major carbon release into the atmosphere. (Chicxulub, the Deccan Traps and the start of the Himalayan orogeny - a lot of stuff going on about then.)
About 50 MYA, the Himalayas and the Tibetan Plateau had popped up, exposing lots of nice fresh silicate surfaces just ripe for weathering - with the concommitant CO2 drawdown. (I don't really know how much credence is given to the postulated Azolla Event and its resulting major carbon sequestration which is, by some, thought to have happened at about the same time - to within a million years or so.)
Obviously, the continents were still dancing around, getting their rocks off, so to speak. (Stealing a line from the great Terry Pratchet.) However, I understood that it was the sealing off of the Panama Ithmus that was the final piece of the jigsaw that allowed the reformation of massive polar ice sheets.
Have I got that just about right?
mancan18 - you might also like to have quick look at this guide to the RCPs. Economics would prevent burning everything but working out atmospheric concentrations also require taking into account how the natural systems draw-down (or otherwise) excess CO2. A typical pseudo-skeptic will argue that environment will continue to draw down 50% of emissions, ignoring that as temperatures rise, the oceans eventually emit CO2, rather than absorb it.
You can play with really extreme scenarios but the RCPs are a far better guide to what is realistic (and frightning enough at that).
Billthefrog
Another important event around 30-35 million years ago was the separation of Sth America and Antarctica, allowing the formation of the Antarctic circum-polar current, tending to isolate Antarctica and allowing it to colmore
@Tom & Glenn - Thank you
@ Graham RE: the RCP guide- Wow
Thanks Howard for another innteresting article about extinction.
I want to understand this sentence:
I'm not sure, if you refer to http://dx.doi.org/10.1016/j.palaeo.2012.06.029 as the source of that sentence because I don't have free access to it. So I want to confirm with you wheather this number is or is not a typo. How can we be so sure about the causes of events so deep in the paleo (60-500Ma)? What do you mean exactly by "due to chance alone" and how do you exclude an event so defined with such inbelievably high certainty?
Thanks Tom
Thanks scaddenp
Chriskoz @20 - That quote is direct from Jourdan et al's paper. The supplementary data can be found at ftp://rock.geosociety.org/pub/reposit/2014/2014190.pdf. see p 15 for randomness calculation.
They admit it is a "crude estimate" but go on to say: "Nevertheless, this series of
calculations suffice to demonstrate that the mass extinctions – LIP association cannot be due to chance with a probability of 6x10-9 % that the synchronicity between LIPs and mass extinctions is random."
Tom Curtis @ 15 - I concur. Thanks!
Many of these extinction events are associated with a Carbon Isotope Excursion (CIE)- a surge in C12 enriched carbon into the active carbon cycle. This C12 enriched (C13 depleted) surge of carbon is best explained by the dissociation of several trillion tons of methane hydrates, many scientists think.
The carbon isotope excursions, the oceanic anoxia, and the low level runaway climate change can all be tied together into a general theory of most mass extinction events, triggered by these flood basalt erruptions and subsequent release of methane from the hydrates.
So, now this middle Cambrian extinction joins this list, it appears. The paper is behind a pay wall, though. Does anyone know if there is a carbon isotope excursion associated with this extinction, and if so does anyone have a link to a paper claiming a coincident CIE associated with this extinction event?
A classic paper on the End Permian mass extinction and the probable role of methane release in that extinction is here:
How to kill (almost) all life - the end-Permian extinction event
" The extinction
model involves global warming by 6 degrees C and huge input
of light carbon into the ocean-atmosphere system
from the eruptions, but especially from gas hydrates,
leading to an ever-worsening positive-feedback loop,
the ‘runaway greenhouse’."
Billthefrog @16 wow, there's a lot in there! I'll do my best...
The Faint Young Sun Paradox is still just that - a paradox. We know from sediments at the time that there was liquid water and normal sedimentary processes, so the Earth was not frozen solid. The work of William Moore shows that very early Earth was essentially in Large Igneous Province mode all the time. There was a lot of CO2 in the atmosphere as plate tectonics handn't started yet (subduction started ~3.2 billion years ago) so the world was basically covered in volcanic islands, but lacked the large mountain ranges and surface area for weathering to draw down CO2 so fast. Yet experiments on fossil raindrops suggest the atmosphere was not so dense, so alternative atmospheric gas mixes have been inferred. We also know the oceans had about 26% more water in them, so the Earth's albedo was likely much lower. By the time of the ice ages @2.9, & 2.5Ga, subduction had started and continents had grown, and oceans had reduced somewhat, but until oxygen arrived the atmosphere was methane-rich. There's much more than I can fit here and it is still an area of ongoing research, and there has even been a suggestion that the sun back then might have been 5% larger (effectively negating the faint young sun paradox).
Regarding the slow decline of CO2 and temperatures since the Eocene hyperthermals, see this post. This decline has been correlated with a reduction in subduction zone length by Prof Zeebe and others. One critical event was the marooning of Antarctica by continental drift and the establishment of the Antarctic Circumpolar Current 33.5-30 million years ago, but increased weathering (Hymalayas, Andes) had a role. I reccomend THE book on the subject: "Earth's Climate Past and Future" by Bill Ruddiman. Hope this goes some way to answering your post.
Leland @24 from the Jourdan et al paper: "the Stage 4–5 transition is associated with a global sea-level rise and negative δ13C and positive δ34S excursions recorded in stratigraphic sections worldwide (e.g., Montañez et al., 2000; Hough et al., 2006)." [stage 4-5 is the same time as Kalkarindji.]
Yes we see a pattern of such events. Here's a list grabbed from a couple of papers - note that the dating of some of the events is better than others. The coincidence of LIP and Mass Extinction/Climate event is strongest where the latest high-precision dating has been applied (Permian, Triassic, Mid-Cambrian).
LIP event /extinction or climate event:
Columba River 17ma (Mid Miocene Climate Optimum)
Yemen/Afar 31ma (none?)
North Atlantic 62/56ma ?PETM/Hyperthermals?
Deccan Traps 66ma (Cretaceous extinction precursor)
Sierra Leone 70ma (?)
Caribbean 90ma (Cenomanian/Turonian Anoxic Event);
Madagascar 90Ma (ditto)
Hess Rise 100ma (?)
SE Africa/Maud/Georgia 100ma (?)
Kerguelen 120ma (?Aptian)
Ontong Java 122ma (Aptian Anoxic Event);
High Arctic LIP 130ma
Parana-Etendeka 132ma
Shatsky Rise 145ma
Karoo-Ferrar-Dronning Maud Land 183ma (Toarcian OAE)
Central Atlantic 201 (Triassic Mass Extinction)
Angayucham 210ma (?)
Siberian Traps 252ma (Permian Mass Extinction)
Emeishan traps 260ma (end Guadaloupian extinction)
Tarim 280ma (none?)
Skagerrak- Barguzin–Vitim - Carboniferous Rainforest Collapse (Moscovian and Kasimovian stages);
Viluy - End Tournasian;
Pripyat–Dniepr–Donets - End Famennian–end Frasnian;
Kola/Kontogero - End Frasnian;
Altay–Sayan - End Silurian (?);
Ogcheon S Korea - End Ordovician?;
Central Asian intraplate magmatism - End Late Cambrian;
Kalkarindji - End Early Cambrian;
Volyn - End Ediacaran;
(From Kravchinsky 2012 & Bryan and Ferrari 2013) ma= million years ago.
mancan18 @13 - Regarding the link between continental breakup, volcanic activity and these LIP & extinction episodes:
LIPs do seem, tentatively, to be concentrated at times when continents are breaking up. But they are caused by mantle plumes that extend from the core-mantle boundary, and not directly with the regular continental drift process.
It's important to emphasise that these LIP monsters totally dwarf any volcanic eruption ever witnessed by humans. In recent decades the total combined annual volcanic greenhouse gas emissions, including undersea volcanics and mid-ocean ridges, are equivalent to the annual emissions of a single state like Ohio or Michigan.
There is an association between long-term CO2 levels, climate, and the total length of subduction zones at any given point in geological time. But the climate is complex, and other factors (rock weathering rates related to mountain building, ocean currents related to continent configuration, life innovations, asteroid impacts...) all interact and play a role in global climate.
Thanks to billthefrog @ 11 and howardlee @12 for their helpful, informative responses to my comment about the Snowball Earth web site @ 10. And thanks to everyone @1-27 for a wealth of fascinating information about paleoclimatology!
Thanks howardlee
howardlee @26-
Wow, that's a great list, thanks. Best I've ever seen. :)
So there are isotope excursions at a coincident time and in the correct direction to fit the flood basalt erruption / methane and CO2 release / oceanic acidification and anoxia general theory of most mass extinction events.
It appears now that there is a subclass of methane hydrates that might be dramatically less stable than most methane hydrates are - high salt "triple point" hydrates.
DYNAMICS OF SHALLOW MARINE GAS HYDRATE
AND FREE GAS SYSTEMS- Xaoli Liu (Thesis)
"We show that the hydrate system at South Hydrate Ridge is already
everywhere at the three-phase boundary, and therefore it is highly sensitive to changes in ambient conditions, offering a mechanism for rapid release of methane from gas hydrate deposits." [page 2]
Other scientists including Peter Flemmings are starting to write about these high salt hydrates. Models of the high salt hydrates tend to confirm Xaoli Liu's predictions.
These hydrates appear to be at the triple point of the hydrate/methane gas/sea water system. This would result in extreme temperature sensitivity and gas phase transport of methane within the high salt region. These high salt methane hydrate deposits, in combination with albedo change from loss of sea ice and permafrost rotting, could act as a bridge between mild CO2 based warming and massive hydrate dissociation.
The high salt concentrations in these sediments would normally be diluted out by diffusion into surrounding sea water and low salt sediments. The salt is generated by the well known "purification by crystallization" process in which the methane hydrates tend to exclude salt from their crystal structure when they crystallize out, leaving the salt behind in the sediments. So, a continuous flow of methane gas from deeper in the deposit is necessary to sustain these high salt triple point hydrates, before the salt diffuses away.
The need for a continuous flow of methane tends to put the high salt deposits in pinnacles at the top of the hydrate formations - in deposits at the top of the hydrate stability zone. So the relative shallowness of these deposits may make them more vulnerable to temperature changes.
These high salt deposits are in principle detectable by seismic mapping - there is a blank looking "wipeout zone" created by free diffusion of gas beneath a methane producing pinnacle.
So, just by looking again at existing sonic mapping data, we could in principle discover just how common these high salt methane hydrate deposits are, and determine how likely they are to push the climate system past tipping points into low level (?) runaway global warming.
Leland @30 - I know that conventionally volcanic CO2 is not considered sufficiently isotopically light to explain the strong -ve Carbon isotope excursions. So hydrates are commonly pointed to as a likely source of light carbon. Others point to the geothermal baking of organic-rich sediments releasing high volumes of light carbon. Still others point to crustal melting and mixing with plume magma generating the volumes of light carbon. We also know that the geochemistry of these deepest-mantle-derived magmas is a little different from more shallow-melted magma.
Personally, I'm slightly skeptical of the hydrate idea as the initial driver of the ancient global warming events because the rapid, strong warming points to the initial atmospheric buildup of carbon (CO2/methane) faster than surface ocean/biosphere can buffer the changes (ie centuries to a few millennia at most). IPCC AR5 points to uncertain but expected slow (millennial-scale) release of hydrates, then slow subsequent release of methane from the sea to the atmosphere (see p 531 of the report - warning large pdf). Also recent observations suggest that microbial metabolism would mitigate ocean methane release somewhat.
The "tripple point" shallow hydrates you refer to may indeed be more responsive, and affect the same shallow ocean reservoir that would be saturated by volcanic CO2 and geothermal methane/CO2.
Given the magnitude of the temperature changes observed for these ancient events and the difficulty in resolving dates under about 2,000 years, it seems likely that a cascade of knock-on effects unfolded after the initial perturbation of the system. Hydrates and permafrost methane release would - in my personal view - be triggered as a response to the initial volcanic/geothermal carbon-shock global warming. I recall somewhere related to the Permian extinction that there are multiple isotopic kicks in a very expanded geological exposure which supports the idea of various carbon reservoirs being released in sucession, but I can't find the paper right now. There was also a good paper suggesting that the Eocene hyperthermals after the PETM were driven by permafrost melting at orbital pacing (ie a long tail feedback, long after the initial carbon shock of the PETM - which I am personally convinced was triggered by the North Atlantic LIP).
The suggestion that such huge and long-lasting feedbacks could be triggered by an initial carbon-shock of sufficient rate and magnitude (at rates similar to or slower than today) is truly scary to me.
Hi howardlee, thanks for the reply and information.
Yes, the triple point hydrates may be a minor player in the whole scenario- nobody knows. Each extinction event may be different, but follow the same general theory. The rate of onset and the magnitude of the triggering event necessary for massive hydrate dissociation might depend on the total hydrate inventory, and whether there is a large region of methane hydrates near the poles or subsea permafrost that is shallow enough to be abruptly triggered. Our own East Siberian Arctic shelf seems particularly vulnerable. And hydrate inventories could be high, since we are coming out of a series of ice ages with low bottom water temperatures, although water pressure also factors in, and ocean levels are low. So, we could have more total hydrates than average, and in fact our "methane capacitor" could be fully charged and ready to be triggered.
The good news, if there is any, about the high salt triple point hydrates is that these pinnicles will have their salt concentrations diluted as the hydrates dissociate into methane and fresh water. The high salt hydrates might be important in the early phase of the extinction events, adding mostly to ocean acidification in the early stages, with little methane escaping into the atmosphere. But, along with melting permafrost, and other carbon sinks turning into carbon sources, they might act as a bridge to the destabilization of the greater hydrate mass.
Later on, as the extinction scenario unfolds, the amount of methane released could start to overwhelm the local oxidation capacity of the ocean. Basin scale anoxia and acidification will likely occur, if the modeling done by the IMPACTS group at the national labs is any indication of real events. IT should be noted, however, that the modeling done by Reagan et al does not take into account the high salt hydrates:
Basin Scale Assessment of Gas Hydrate Dissociation in Response to Climate Change
According to this modeling, the northern Pacific will be more severely affected than the well ventilated Arctic Ocean. The sea of Okhotsk, that most Americans have never heard of, could create a plume of anoxic deep water stretching all the way across to Alaska, down the west coast of North America, all the way to Baja California, wandering out into the Pacific, then continuing down the coast of South America in diluted form.
It scares me too, howardlee. It sure would be nice to be wrong about this stuff, wouldn't it?
But, I don't believe it. The whole theory hangs together too well, and explains too much geological information. The theory of flood basalt triggered methane catastrophes has too much predictive ability, too much explanatory ability, too much unifying ability, and it too consistent to be basically wrong, I think.
On the other hand, the high salt methane hydrates could be quite significant and could be a major piece of the puzzle of how these past mass extinction events occurred. The high salt hydrates could also be crucial to predicting how our manmade extinction event will proceed, and what the eventual outcome will be.
From an earlier thread on Skeptical Science:
Such high salt methane deposits may be fairly common, according to authors including Maria Torres and Miriam Kastner:
OCCURRENCE OF HIGH SALINITY FLUIDS ASSOCIATED WITH MASSIVE NEAR-SEAFLOOR GAS HYDRATE DEPOSITS
The existence of these deposits may be the answer to the disconnect between the geological evidence of past methane catastrophes and our current lack of understanding of how these mass extinction events occurred.
One way that the high salt hydrates could be important, though, is that as the local methane oxidation capacity of the oceans is exceeded, basin scale acidification and anoxia are predicted to occur, by the IMPACTS group modeling of Reagan et al.
And, as basin scale anoxia and acidification occurs, more methane is transported to the atmosphere, adding to warming from direct and indirect atmospheric chemistry effects of methane, according to the modeling of Isaksen, et al.
In this scenario, the high salt hydrates could lead to local methane emission hotspots, like for example the Hydrate Ridge and Cascadia Margin area of the Pacific Northwest, off the coast of Washington and Oregon. These areas are known to contain high salt hydrates, and have plumes of methane and acidified water from anerobic oxidation of methane a kilometer or two wide extending into the bottom water:
Gas hydrate destabilization: enhanced dewatering, benthic material turnover and large methane plumes at the Cascadia convergent margin
The authors of this 1999 paper did not know about the high salt hydrates, apparently, and proposed a different mechanism for methane release. But they did document the anoxic and acidified plumes around Hydrate Ridge, and calculated that the oxygen demand from the hydrates is a thousand times or more the demand from an equivalent area of sea bottom.
Hi Howardlee-
I had a crazy idea one time, and I dont think it's very likely, but I thought I would share it, and maybe other people can tell me why it's wrong. I do hope the idea is wrong. I don't really expect a reply on this old thread.
The idea was that significant polar melting events might make flood basalt erruptions worse, by transfering mass from the poles to the equatorial regions (as the ice caps melt and the oceans rise) slowing the rotation rate of the earth by conservation of angular momentum.
The idea is that as the ice caps melt and the oceans suddenly rise, changes in isostatic balance enhance volcanic activity. Also, mass is transferred from the poles to the equatorial regions, slowing the rotation of the crust of the earth, by conservation of angular momentum (this effect is often seen in ice skaters, extending their arms to slow their rates of rotation while they are spinning). The crust of the earth wants to rotate slower, but the massive core of the earth wants to remain rotating at the original rate. This sets off massive stresses in the crust of the earth, changes the path of tectonic plates and can set off rifting events, such as the opening of the North Atlantic associated with the PETM, according to this idea.
So, it's just an idea, and hopefully it's wrong. Becasue if it's right, our current round of icecap melting could set off rifting events and trigger massive flood basalt erruptions, similar to those associated with past mass extinctions.
Here's a paper on a similar idea, involving mass transfers from the oceans to the poles following major impact events, and a possible link from such impacts to geomagnetic reversals:
Muller - Geomagnetic Reversals Driven by Abrupt Sea Level Changes
What set off this round of speculation is the observation that flood basalt erruptions sometimes seem to be dated slightly later than mass extinction and rapid warming events, rather than slightly before them.
What we know from all the comments is that the Earth will have episodes of warming and cooling, all by itself, regardless of mankind. Some of these events will be rapid or slow, depending on the cause. The rapid change's are associated with mass extinctions. There is evidence that man's emmissions are causing a rapid change. But I'm not certain that the Earth isn't warming all by itself regardless of man, or that man's activities causing warming isn't preventing the next ice age. It's all so speculative.
Both the "asteroid" theory and the "Decan traps" theory have three very glaring problems. The first is the classic scientific blunder: correlation is not causation. There is no evidence that rapid increases in CO2 harms life or that the temperature rose. They only know that the volcanos erupted (or maybe an asteroid hit the earth) and then they speculate about the possibilities of what might have happened.
Secondly, the extintion of the dianasaurs and the Decan traps aren't even necessarily correlated. The extintion of the dinasaurs isn't well defined and some believe it happened over millions of years. The inaccuracy of dating the extinction of dinasaurs means that even the correlation is speculation.
Third, over the course of evolution, CO2 has dropped from seven thousand ppm to 180 ppm. A catastrophic rise in CO2 is not consistent with what was happening, which is a decrease in CO2.
The only reason people even believe a far fetched story like "an asteroid hit the earth" or a volcanic eruption is that nobody has proposed anything more plausible.
So here's a proposal: There are two things that support all life on earth: CO2 and sunlight. There is only one of those components that we know for sure has decreased by orders of magnitude over an extended period of time. Hint: it isn't sunlight.
CO2 is the ingredient of life. The catestrophic events in history are likely caused by organisms (plants or plankton) that evolved and depleted CO2 to a new low, killing everyting that lived off the plants that required higher CO2 concentrations. Adding CO2 to our atmosphere is likely to be the best thing we could do for our planet.
I predict that in 200 years from now people will look back on the global warming issue with bewilderment. How could restoring the chemical basis of life (CO2) be a bad thing. Our progeny will think we are as unsophisticated as the society that encarcerated Galileo for saying the earth revolves around the sun.
So was the extinction due to the initial cold caused by the mongo massive volcanic exhaust solids including carbon from burned fossil hydrocarbons blotting out of the sun, or the proposed latter GHG induced warming?