Concern
Over Catastrophic Methane Release — Overburden, Plumes, Eruptions,
and Large Ocean Craters
(Yamal Crater as seen from the air. Image source: The Siberian Times.)
9
March, 2015
Depending
on who you listen to, it’s the end of the world, or it isn’t. A
loud and lively debate that springs up in the media every time a new
sign of potential methane instability or apparent increasing emission
from methane stores is reported by Arctic observational science.
On
one side of this debate are those declaring the apocalypse is nigh
due to, what they think, is an inevitable catastrophic methane
release driven by an unprecedentedly rapid human warming of the
Arctic. A release large enough to wipe out global human civilization.
These doomsayers are fueled by a number of scientists (usually Arctic
observational specialists) who continue to express concern — due to
an increasing number of troubling, if not yet catastrophic, rumblings
coming from the Arctic carbon store. The Arctic is warming faster
than it ever has, they accurately note. And this very rapid rate of
warming is putting unprecedented and dangerous stresses on carbon
stores, including methane, that have lain dormant for many millions
of years. The risk of catastrophic release, therefore, is high enough
to sound the alarm.
On
the other side are a number of mainstream news outlets backed up by a
group of established scientists. This group claims that there’s
generally no reason to worry about a methane apocalypse. The methane
releases so far are relatively small (on the global scale) and there
are all sorts of reasons why future releases will be moderate, slow
in coming, and non-catastrophic. The methane store most pointed
toward by methane catastrophists — a frozen water methane known as
hydrate — tends to self-regulate release, in most cases, acting as
a kind of pressure valve that would tend to moderate emission rates
and prevent instances of catastrophic eruption.
A
third group appears to have somewhat sidestepped an otherwise
polarized discourse. Outlets like ThinkProgress and others have
continued to quietly report observations without drawing conclusions,
one way or the other, on the issue of near term methane apocalypse.
They point, instead, to what are, admittedly, some rather odd and
scary methane rumblings going on near the pole. Among this ‘middle
ground’ group are a survey of about 100 researchers who’ve
identified a likely carbon release (including both methane and CO2)
from the Arctic equaling between 10 and 35 percent of the human
emission by the end of this Century. It is a ‘middle ground’ that
is troubling enough. For 10-35 percent of the human carbon emission
coming from the Arctic is a massive release in the range of 1 to 3.5
gigatons of carbon (with a fraction as volatile methane). If such an
emission does materialize, it will equal (on the low end) or exceed
the annual rate of environmental carbon release last seen during the
PETM — a hothouse extinction 55 million years ago that turned the
oceans into killers and forced life on land to shrink in size and
burrow to avoid the awful heat and stifling atmosphere of that age.
Regardless
of where you stand in this discourse, the Arctic itself continues to
provide cause for both debate and appropriate concern.
Methane
Overburden
(Barrow surface methane observations by NOAA ESRL show methane readings that range about 60 ppb above the global average. Note the 50 ppb increase over the past decade coincident with numerous ‘outlier’ spikes [green cross hatches] from local sources. Image source: NOAA ESRL.)
Perhaps
the most obvious sign that there’s something not quite right going
on in the Arctic is a large overburden of both methane and CO2 in the
region. Looking at NOAA’s ESRL site, we find that methane levels at
Barrow, Alaska (one of just a handful of Arctic sensor stations in
the ESRL network) are in the range of 1910 parts per billion. By
comparison, NOAA’s Mauna Loa Station, on the edge of the tropics
and well away from the polar overburden, records about 1850 parts per
billion (ppb).
At
current rates of atmospheric methane increase, it will take about 9
years for Mauna Loa to catch up to where Barrow is now. But by that
time Barrow may be pushing 1970 ppb or more. In addition, all Arctic
stations record numerous anomalous spikes in methane from local
sources. The ESRL site lists these spikes as outliers. But, for all
the ESRL reporting stations, the Arctic stations are the ones that
host by far the most numerous such outliers. The local methane
sources, therefore, appear to be quite active in the Arctic.
(Global distribution of methane averaged over 2011 by NASA/AIRS. Note the very high concentrations in the Arctic region. For this map, the highest concentrations occur in the Yedoma region of Russia, a region of multiplying methane emitting tundra melt and Thermokarst lakes [see below]. Image source: NASA/AIRS.)
Perhaps
the most reliable way to sample the Arctic methane overburden is to
get a full view of it through satellite sensors. The above NASA image
taken in 2011 shows a massive methane overburden in the upper
latitudes that slowly diffuses southward. Note the highest
concentrations in this image are near the permafrost zones in Yedoma
in northeastern Russia.
NOAA
also provides its METOP array which frequently finds methane
concentrations at above 2400 parts per billion at the 10,000 to
20,000 foot level in broad blankets over the Arctic region —
especially in the months of September through November and then again
in January. Again, these measures are the highest in any region of
the globe and they occur directly over the Arctic.
Dr.
Leonid Yurgonov uses the AIRS/AQUA satellite sensor to provide a
record of Arctic methane overburden. One that is clearly visible
here:
In
the above image we see methane measurements at the 18,000 foot
altitude above the Arctic and upper latitudes. The progression is
from January of 2009 (furthest left) to January of 2013 (furthest
right). Orange coloration represents methane readings in the range of
1850 to 1950 parts per billion. Deep red coloration is in the range
of 2000 parts per billion. Note the rapid shift from blues and
yellows (1700-1800 ppb) to oranges and reds (1850-2000 ppb) during
the five years from 2009 to 2013.
So
not only does the AIRS sensor show overburden, but it also finds
rapid methane build-up over the period measured.
These
combined measures alone provide more than enough evidence of a
methane overburden in the far northern region together with a rate of
buildup that maintains the overburden and leads the global methane
measure.
Steady
Increase So Far
But
even if we do have both a buildup of methane in the polar region
together with what looks like an ominous overburden, we should be
quick to point out that the rate of increase, especially on the
global scale, has been mostly steady so far.
Under
any catastrophic methane release scenario, we would expect Arctic
methane to rapidly jump higher, dragging the global measure along
with it. In general, we’d expect almost all sensors to pick up the
signal of an exponentially ramping curve. And we don’t see that as
yet.
Though
there has been a bit of an uptick in global and Arctic methane
increase rates during recent years, they have maintained about a 4-7
ppb annual increase since ending a decade-long pause from 1995 to
2005.
It
is worth noting, however, that the global methane measure increasing
at an exponential rate would be a trailing measure indicator —
occurring only in the wake of any catastrophic or large-scale
release. So, as a predictor, the global methane measure isn’t very
useful.
Thermokarst
Lakes
Which
brings us to the key question — what are the leading indicators of
major methane releases or of catastrophic releases of the kind some
have feared?
Since
we have never directly observed one, and since large-scale or
catastrophic releases are merely theoretical at this time, we can
only point toward evidence of past large scale releases, and an
ongoing, but apparently growing, smaller scale release happening now.
The
first such related observation may well have come in the form of an
increasing methane emission from Thermokarst Lakes. Thermokarst Lakes
form when sections of permafrost thaw and collapse, creating a
depression. In wet regions, water soon pools within these hollows.
Organic material at the bottom of the pool is provided by thawing
permafrost. In the anaerobic lake bottom environment, methane is
generated as the organic material is broken down.
Over
recent years, this increasingly widespread Thermokarst thaw and
formation has resulted in a number of Arctic ‘fire lakes’ popping
up — lakes whose methane emission is so great that bubble
concentrations are high enough to burn. During winter, these bubbles
are trapped beneath ice and when released, create an explosive
mixture.
(Methane production in a thermokarst lake. Image source: The Royal Society.)
From
the 1970s through the mid 2000s, it is estimated that some regions of
the Arctic experienced as much as a 58 percent increase in methane
release due to Thermokarst Lake formation alone. An important measure
since a number of studies found that Thermokarst Lake formation was
one of the primary drivers of methane release from the Arctic at the
end of the last ice age.
But
as a catastrophic release driver, Thermokarst Lake formation is
relatively mild, even if it is capable of pushing Arctic methane
release levels higher. As such, the next indicator — a discovery of
large methane releases from the ocean floor in the Arctic — was
somewhat more concerning.
Oceanic
Plumes
For
as of 2011 an expedition to the East Siberian Arctic Shelf (ESAS)
found massive plumes of methane as large as 1 kilometer across
emitting from the shallow sea bed region off Northeastern Siberia.
The researchers, Shakhova and Similetov, seemed very concerned that
this might be a sign of a potential impending large scale release on
the order of 1 to possibly 50 gigatons. The methane stores for the
ESAS alone were massive — in the range of hundreds of gigatons. So
even a fractionally small release from this source could be
devastating. For reference, a 1 gigaton release would more than
double the annual methane release from all global human and natural
sources. A 5 gigaton release, on its own, would be enough to more
than double atmospheric methane concentrations. And since methane
traps heat more than 20 times as efficiently as CO2 over a century
time-scale, such a release would result in far more rapid warming
than previously predicted by scientific bodies such as the IPCC. A
very rapid rate of warming that would be extraordinarily difficult
for human civilizations to adapt to.
Of
course this announcement set off amazing controversy. We couldn’t
be certain what the source of this methane was, some said. Was it
submerged permafrost methane? Was it hydrate? Was it free gas
methane? And how could we be certain that this release hasn’t been
ongoing for some time?
If
such a methane release was building up to a catastrophic event, what
mechanism would be the cause? In other words, how might gigatons of
methane suddenly blow up from the sea bed?
(Lower troposphere methane concentrations over the Kara, Laptev, and East Siberian Seas during September-November of 2009-2012 shows overburden in active oceanic release zones. Image source: Dr. Leonid Yurganov).
This
point is worth a bit of further exploration. The issue is that the
most unstable form of methane when warmed is the methane hydrate
store mentioned above. Methane hydrate is a frozen combination of gas
methane and water. It crystallizes into a kind of fire ice under high
pressure and in low temperature environments. It typically forms
about 200-600 feet below the sea bed as methane bubbling up from
warmer regions below contacts seawater, high pressure and cold. If
the layer is warmed under human heat forcing, the hydrate thaws
releasing its gas. The gas now becomes stored in pockets under high
pressure. The gas below pushes against the sea bed above and some of
it bubbles out (and these releases are found in the large plumes
along the ESAS and elsewhere). But most of it, so far, has remained
entombed.
What,
then, could cause the large stores of entombed gas releasing from
destabilizing hydrate, to break through hundreds of feet of seabed —
hitting first ocean water and then atmosphere?
Over
the past four years conjecture over this issue has raged on. Swelling
at points when Shakhova and Similetov would make a new announcement
and then ebbing as a wave of reassurances would rush in from
scientific critics and mainstream media.
By
summer of 2014 a discovery of new, large-scale plumes in the Laptev
Sea by the SWERUS C3 expedition set off another wave of media
speculation and controversy. But as the dust settled it became clear
that the Laptev sea floor had been added to the list of methane hot
spots in the Arctic, following in the footsteps of the ESAS region as
an area to watch for potential increasing rates of release.
Tundra
Blowholes
In
nature, gasses under high and increasing pressure often find pathways
for escape. Typically, the escape is gradual — we see this in
volcanic regions in the release of magma gasses through cracks in the
earth and through vent pathways. And sometimes the escape is far more
violent — with hot volcanic gasses blowing away even hill or
mountainsides in spontaneous eruption, or bubbling out, en mass,
through volcanic lakes to spill toxic plumes over a countryside.
The
gas source in question for Arctic methane release — hydrate — is
very large. Even at the low end, it is estimated that hundreds of
gigatons of the stuff lay buried beneath frozen tundra ground or in
ocean stores beneath the seabed. A gigaton is one billion tons. A
billion tons of frozen hydrate would cover roughly one cubic
kilometer. One cubic kilometer of a flammable gas under high
pressure.
And
in the Arctic, hundreds of billions of tons lay under rapidly warming
permafrost both on land and in the submerged seabed.
(Graphic of permafrost and gas hydrate methane by Carolyn Ruppel. Note that 75 percent of the ESAS sea floor is in the range of 50 meters in depth or shallower and that buried hydrate deposits can be found in the range of 200-300 feet. Image source: Methane Hydrates and Contemporary Climate Change.)
As
of 2011, some scientists were warning that we were seeing a slow
release from some of this submerged hydrate store in the ESAS. By
2014, the potential slow release had expanded into the Laptev Sea.
But
that year, 2014, also saw something else. A potential catastrophic
release of methane. For in the frozen region of Yamal, Russia the
ground near a remote Siberian village began to shake and bulge. Soon
after, according to eyewitness accounts, the area began to smoke.
Then, with a flash and a thunderous boom, the ground erupted.
When
the smoke cleared, a massive crater was found where only flat, frozen
tundra was there before. A giant plug of frozen earth had been
ejected violently. And all that remained was an ominous gray-black
crater.
(Yamal Crater as seen from the air. Image source: The Siberian Times.)
Researchers
investigating the crater found 10 percent atmospheric methane
concentrations at its base.
Overall,
it was estimated that about 11 tons of TNT equivalent explosive force
was enough to remove this 100+ foot wide and 220 foot deep plug from
the Earth. Exploding and burning methane in the range of about 10
tons would have been enough to generate the crater. Gas under high
pressure in the hundred + ton range may have been able to explosively
excavate this hole.
As
a result, the amount of methane in question for this single event was
relatively small, especially when one considers the hundreds of
billions of tons in the still frozen store.
It
appeared that the rapidly warming Yamal territory and a broad region
of nearby Northwestern Siberia may be seeing tundra warming extending
deep enough to begin to destabilize pockets of relic hydrate. The
hydrate in some of these pockets was beginning to thaw and
catastrophically erupt to the surface.
By
early 2015 a total of seven primary craters and scores of secondary
craters of this kind had been discovered throughout this section of
Siberia. Local Russian authorities were very concerned — moving
seismographs into the area to monitor ground stability in a region
that includes one of their largest natural gas developments.
A
large upheaval of this kind in the wrong place would easily rupture a
pipeline or destroy sections of a gas production operation. But the
deeper irony was that continued gas production in this region was
contributing to a problem that may well be making the ground far, far
less stable and setting up the risk for even larger-scale eruptions.
For
the Yamal crater wasn’t important due to the relative size of its
methane release — the release was very small in the global context.
A mere drop in an ocean of greenhouse gasses being emitted now by
humans. It was important due to two other, and perhaps more stark,
reasons.
The
first was the very violent nature of its release — an eruption
similar to that of a volcano — represented a severe geophysical
upheaval that was all too likely triggered by a rapid human warming
of the tundra. This kind of release, as the Russians in the region
were quick to realize, represented a danger to both inhabitants and
to infrastructure.
But
the second reason is, perhaps, more important. It is the fact that
the Yamal crater may well be evidence of the kind of mechanism for
catastrophic methane release some of the more conservative scientists
have been demanding. It’s possible, then, that the Yamal crater is
in microcosm, what a truly catastrophic methane release might look
like on the much larger scale. And the critical question to ask here
is — could there be a connection between the methane blowholes we
are now observing in the Arctic and a number of mysterious and
gigantic craters discovered on the sea bed around the world?
Giant
Craters on the Seabed
In
2013, marine geophysicist Dr Bryan Davy from GNS Science found what
may be the world’s largest gas eruption craters on the seafloor
about 310 miles east of Christchurch, New Zealand.
The
craters, which the researchers called ‘pockmarks,’ formed in an
active gas zone along the ocean bottom. They measured from 250 meters
to 7 miles in diameter and about 300 feet deep. With the largest
crater able to encompass all of lower Manhattan.
(Giant craters off Christchurch New Zealand are thought to have formed due to large gas eruptions during previous episodes of sea bed warming. Could human warming set be setting off something similar for the Arctic? Image source: Mysterious Giant Crater Like Structure Found Near New Zealand.)
The
craters are thought to have formed during ice ages when sea levels
lowered off New Zealand causing the sea bed to warm and gas hydrate
to thaw. Eventually, the gas is thought to have erupted into the
surrounding water with a portion bubbling up into the atmosphere.
GEOMAR
seismic records indicated active gas pockets beneath the crater
zones. Dr Joerg Bialas, a GEOMAR scientist noted:
Gas
release from the larger pockmarks may have been sudden and possibly
even violent, with a massive volume being expelled into the ocean and
atmosphere within hours or days.
The
300 foot depth of the craters touched the hydrate stability zone even
as their large size indicated that massive pockets of the gas lifted
away large sections of sea bed suddenly and violently. It’s the
kind of rapid destabilized gas release that may well represent a
worst-case Arctic warming scenario.
Cause
for Appropriate Concern
So
the question must be asked — is the Yamal crater physical
validation of a catastrophic methane hydrate release mechanism that
has circulated, as theory, through the geophysical sciences for
decades? One that involves large eruptions that displace massive
sections of earth and seabed during a violent release process. Are
the Siberian methane blowholes smaller examples of what can happen on
a much greater scale? And does the methane overburden in the Arctic,
the documented increasing Thermokarst Lake release, the sea bed
methane release in the Laptev and ESAS, and the new formation of
methane blow holes in Yamal in the context of a rapidly warming
Arctic tundra and sea bed (seeing unprecedented rates of warming)
represent a growing risk for this kind of release?
Under
even a ‘moderate’ 1 to 3.5 gigaton Arctic carbon release rate by
end century given by the survey of 100 Arctic scientists, there will
likely be more than enough potential freed methane to include large
scale catastrophic releases similar to the kind seen off New Zealand
and elsewhere (250 meter to 7 mile wide cratering events).
In
this context, the issue is not one of ‘apocalypse now’ or
‘apocalypse not.’ That framing is all wrong. This issue is one of
how much or how little geophysical upheaval and related methane
release we will see — and how soon. One of how rapidly humans can
stop making the situation even worse, by drawing down their own
catastrophic emission rates as rapidly as possible.
There
is, therefore, more than enough cause for appropriate concern and
continued monitoring of what appears to be an ongoing destabilization
of Arctic carbon stores — large enough to represent a variety of
hazards both terrestrial and atmospheric.
Links:
Scientific
hat tips to Dr. Leonid Yerganov, Dr. Gavin Schmidt, Dr. David Archer,
Dr. Igor Semiletov, Dr. Natalia Shakhova, Dr. Carolyn Ruppel, Dr.
Jason Box, Dr. Peter Wadhams, Dr. Bryan Davy, Dr Joerg Bialas, SWERUS
C3, GEOMAR and The Russian Center of Arctic Exploration
No comments:
Post a Comment
Note: only a member of this blog may post a comment.