Worst
Drought in 1,000 Years Could Begin in Eight Years
By
Bruce
Melton, Truthout
21
February, 2013
What
will desertification look like? Great Sands National Monument, south
central Colorado. (Photo: Bruce Melton)
Beginning
in just eight years, we could see permanent climate conditions across
the North American Southwest that are comparable to the worst
megadrought in 1,000 years. (1)
The
latest research from the Lamont-Doherty Earth Observatory at Columbia
University published in December 2012 has some truly astounding news.
The megadroughts referred to in the paper published in Nature Climate
Change happened around about 900 to 1300 AD and are so extreme that
they have no modern counterpart for comparison (these megadroughts
will be referred to in the following as the "12th century
megadrought"). The research was funded by the National Oceanic
and Atmospheric Administration (NOAA).
We
have been warned for decades that we would be facing a megadrought if
we did not do something about climate pollution. We did not, and now
according to the projections of a new study, that is just what the
future may hold. And remember, projected conditions similar to the
worst megadrought in 1,000 years would be the baseline conditions.
Dry periods, which we normally refer to as drought times today, would
be superimposed on
top of
the megadrought extremeness.
The
Lamont-Doherty research not only includes one of the four new climate
scenarios, but also uses the new high-resolution climate models that
provide more detail and accuracy. Both will be found in the
forthcoming 2013 Intergovernmental Panel on Climate Change report
(IPCC 2013). The authors tell us about the new climate scenario:
The
RCP85 scenario involves stronger anthropogenic radiative forcing
[than the old IPCC scenario] and was chosen to reflect the present
lack of any international action to limit CO2 emissions.
Let
me interpret. "RCP85" is one of four new scenarios the IPCC
has requested the research community to prepare. The four scenarios
were chosen from existing international research literature in 2007
at a meeting in the Netherlands consisting of 130 international
stakeholders.
The
old scenarios (40 of them) were based on a complex "storyline"
involving the way our global society changed over time, what type of
and how much energy we used, when and how fast we changed our land
from forest to fields, how quickly population grew or did not grow,
and different population growth rates in different regions. The new
scenarios represent the concentrations of greenhouse gas pollutants
in the atmosphere and the amount of warming they create instead of
the vastly varying emissions of the old scenario storylines. (The
"RCP" in the scenario's name means Representative
Concentration Pathways.) The new scenarios do not represent any one
future snapshot of the way our society evolves. Different
evolutionary paths could result in the same greenhouse gas
concentrations. The new scenarios are the simple end result of
greenhouse gases emitted to our atmosphere by any number of societal
evolution pathways. It's a simpler process, and it updates the old
scenarios prepared in 2000 with current greenhouse gas data. It also
reflects a revelation in research that because we have failed to act
on climate change, the old worst-case scenario was optimistically
good. (2)
The
new models have more grid squares (higher resolution) in that they
can "see" a smaller piece of the earth compared to the old
models. The old models took forever to run on supercomputers, and so
do the new ones, but we can see smaller areas and smaller scale
climate processes are better represented now. The new models also
include volcanoes, changes in the sun's strength and more complex
interactions between clouds and pollutants like nitrous oxides and
sulfur dioxides (both manmade and natural), and their results agree
better with observations of our past climate.(3)
The
results of the new scenarios and most current modeling (as compared
to the old scenarios and models) are that warming is greater, drying
in dry areas is greater and increasing wetness in wet areas increases
further. (4)
And
just to finish up translating that quote, as I have been doing for
four paragraphs now, let's take "Anthropogenic radiative
forcing." "Anthropogenic" is human-made; and
"radiative forcing" refers to the stronger-than-normal
greenhouse warming we will experience because of climate pollution.
Of equal importance in the quote is the acknowledgement, in a
peer-reviewed research publication, that there has been a lack of
action in the international community on reducing CO2. This type of
statement is something we are seeing more of these warming days.
The
results of the new research are critically deserving of an alarmist
tone. That we could slip into profound continuous drought so soon is
certainly a surprise to most of us, to say the least. The typical
consensus opinion of unrestrained climate pollution impacts by the
year 2100 only tells us that permanent drought will come to many
parts of the world and, basically, that dry areas could become drier.
The news that we could be experiencing permanent drought on the scale
of megadrought proportions - beginning in only eight years - should
be considered a global threat of the highest order.
So
why, once again, is there no alarm? The prepublication press release
for this paper came out on December 23 and while it did get picked up
by a few sources, the only major outlet was Agence France Press. All
of the coverage referenced the 10 percent reduction in streamflow
that this work's modeling projects for the near future. This
seemingly small number appears to have limited journalists' interest
in the results of the research as a whole. No one who wrote about the
paper seems to have recognized the megadrought reference.
Implications
for increased evaporation and the seasonality of the projected
drying, both of which the authors say were important, were only
regurgitated from the press release with little additional thought.
It is the megadrought reference, however, that gives us a true
understanding of exactly how worried we should be.
The
NOAA Climate Variability and Predictability Program's press release
at Columbia University gives us an authoritative view of this
research:
Long
tree ring records allow estimations of past variations in Colorado
River flow and suggest a 15% reduction of flow during the 12th
Century megadrought. Therefore the new paper concludes ...
Anthropogenic
climate change is projected to lead to a potential reduction of
Colorado River flow comparable to the most severe, but temporary,
long-term decreases in flow recorded. (5)
This
"most severe, but temporary, long-term decrease in flow
recorded" is the concept we need to understand. This is the
megadrought reference. A 10 percent reduction beginning 2021 to 2040
is extreme enough for these researchers to compare the average
conditions projected for the very near future to the 12th Century
megadrought. This single message is critical and it was missed by
popular reporting. Just to be sure I am clear: this quote "temporary,
but long-term decreases in flow" here refers to these 75- to 200
year-long megadroughts, the last one occurring about 1,000 years ago
or in the 12th Century. These droughts were temporary, like the
droughts of today, but in the near
future, conditions comparable to these droughts will be the average
climate condition. Dry periods that we know as drought today will
be on
top of
megadrought dryness.
In
2004, Edward Cook and a team of researchers published a paper in
Science that describes this 12th century megadrought. These
researchers were from Lamont-Doherty, as well as NOAA, the National
Climatic Data Center and the Universities of Arizona and Arkansas.
The
12th century megadrought was a part of a series across western North
America over a 400 year period. According to the authors, it "dwarfs"
the ongoing western drought we are currently experiencing. (6)
This ongoing western drought also differs in a surprising way from
the Dust Bowl, as we can begin to see in the graphic below and as I
will explain further:
Current
western drought pales in comparison to prehistoric megadrought.
Conspicuously absent from this record is the Dust Bowl. This data was
averaged using a 60-year period. This “averaging” smooths out the
record of the Dust Bowl because it was short and situated more in the
Great Plain instead of across the entire U.S. West.
This
work by Cook and colleagues was compiled from tree rings found in
some really odd places, like ancient Anasazi cliff-dwelling timbers
and tree stumps beneath the waters of lakes in the Great Basin that
were alive during the megadroughts and were subsequently submerged as
the lakes rose afterwards. Clues were also found in sediment deposits
in North Dakota, evidence of drifting sands in Nebraska and lichen
residue in Texas. These megadroughts were powerful enough to lower
the level of the lakes in the Great Basin (including Great Salt Lake)
by 200 feet and change parts of the Great Plains into a sea of
shifting sand.
What
the above graph shows us is the great difference between the
megadroughts and our current western drought. The century scale
length of the megadrought periods is profound relative to drought our
society has experienced. Even more striking is the relationship
between the current western drought and the Dust Bowl. In the graph
above, the Dust Bowl does not appear to be represented and there are
a couple of reasons for this.
The
Dust Bowl was a little more centered in the Plains. Cook looked at
all of the United States west of the Mississippi River. But more
importantly, the graphic above uses a 60-year smoothing. What this
means is that the graphic shows us the 60-year average of drought
conditions. Because the Dust Bowl lasted only about 10 years, its
extremeness is diminished in the averaging process for the smoothing.
To see the relationship between the Dust Bowl and the current western
drought more easily, see the graphic below, which does not include
60-year smoothing:
This
image shows the details of drought area across the western U.S.
without averaging. The authors tell us the following about the
current Western Drought, “Its 4-year duration appears to be unusual
over the past 104 years.”
The
current western drought, though it has not yet lasted as long as the
Dust Bowl and its worst peak has not been quite as high, shows more
continuous red uninterrupted by wet years. This continuous stress is
much worse on an ecosystem than an extreme drought punctuated with a
year or two of wetness here and there, like we saw in the Dust Bowl.
Cook and his team also observed that the four-year duration of the
current drought "appears to be unusual over the past 104 years."
This "four-year duration" is based on their data, which
ended in 2003. The drought did ease somewhat in 2005 and then again
in 2009 and 2010, according
to archived maps found on National Drought Mitigation Center's
Drought Monitor.
Otherwise, it has been in existence across large parts of most of the
American West since Cook's work was published in 2003. (7)
The
Dust Bowl was also a singular event strongly enhanced by agricultural
practices. A series of wet years in the early 20th century lulled us
into thinking that the Great Plains region was an agricultural
nirvana. Sodbusters arrived by the tens of thousands and ripped the
moisture-sustaining prairie grasses from the land with little thought
to long-term consequences. When dry times returned, little of the
natural prairie grasses remained to conserve moisture. This led to
more evaporation than normal - aggravating the drought. With the
increased dryness came increased winds, themselves induced by the
drought. The winds made evaporation even higher; the Dust Bowl ensued
and sands began to move.
Today,
modern agricultural practices diminish the wholesale drying
experienced during the Dust Bowl, but extreme drought persists in the
West. This is because the average temperature in the West has warmed
at nearly twice the global average according to the Rocky Mountain
Climate Change Organization. (8)
There
are a few more things I need to mention to drive home the
significance of the 10 percent reduction in stream flows. One is
timing. The modeling shows that spring, summer and fall see a greater
reduction in flows in most places than in winter. In many places,
more mountain precipitation is now falling as rain in the winter, and
this will increase. More runoff in winter means less snowpack, less
water slowly percolating down into the aquifers, lower aquifer levels
and a longer evaporation season as the snowpack disappears early. (9)
Compounding
the increased length of the evaporation season, a little more warmth
means a lot more evaporation. It is not a one-to-one relationship.
The impacts are compounded in one more way. In the high country where
most of the West gets its water, a little warming, and its
corresponding evaporation, takes water that should slowly melt and
feed aquifers or run off into reservoirs and evaporates it directly
into the sky.
Another
confusing aspect of this work is that numerous places in the press
releases, and in the findings themselves, tell us and show us in
graphic form that not all seasons in all areas experience drying and
increased evaporation. The Columbia University press release tells us
that, "The Colorado headwaters are expected to see more
precipitation on average," and the NOAA Climate Variability and
Predictability Program press release at Columbia tells us that,
"Despite the fact that precipitation might increase in some
regions and seasons (e.g. winter in northern California)." The
most telling example of this climate confusion comes at the bottom of
the Columbia University press release. This statement by Mingfang
Ting, one of the paper's authors and a specialist in precipitation
extremes, tells us: "For Texas, the models predict that
precipitation will decrease and evaporation rates will also go down
in spring and summer, but only because "there is no moisture to
evaporate." (10)
Climate
scientists have been pulling this alarm for 20 years. It is real -
the building is on fire. To pull the fourth alarm on this one:
Truthout wrote Professor Seager and asked him to confirm the
assumption that natural drought cycles would add to or be on top of
the projected megadrought drying. He confirmed, adding, "For the
next one to three decades, results are not greatly different across
the [different scenarios] because so much of what will happen is
already in the pipeline, so to speak."
What
does this mean? It means that even the best-case scenario that the
IPCC is now considering results in an outcome that is the same, or
"not greatly different," from the worst-case scenario of
the new IPCC scenario family, for the next 10 to 30 years.
Things
will get far worse if we do not do something about climate pollution
as fast as we can. But there is good news. Reality tells us that,
contrary to what the voices of denial and delay are saying, the
solutions to climate pollution will be no more expensive than the
cost of clean drinking water across the planet every day. (11)
References:
1.
Permanent climate conditions across the North American Southwest
that are comparable to the worst megadrought in 1,000 years ...
Seager et al., Projections of declining surface water availability
for the southwestern United States, Nature Climate Change, December
2012, page 5, last paragraph.
Earth
Institute press release:
http://blogs.ei.columbia.edu/2012/12/23/smaller-colorado-river-projected-for-coming-decades-study-says/
2.
Scenarios: Old scenarios (SRES) are emissions based... IPCC
Special report on Emissions Scenarios, see
Foreword.
New
Scenarios (RCP) are based on radiative forcing, or warming caused by
greenhouse gases... Intergovernmental
Panel On Climate Change, Representative Concentration Pathways
(RCPs), first paragraph.
Creation
of New Scenarios... Moss et al., Towards New Scenarios for Analysis
of Emissions, Climate Change, Impacts, and Response Strategies,
Intergovernmental Panel on Climate Change, Geneva, 132 pp, 2008,
first paragraph, page i.
http://www.aimes.ucar.edu/docs/IPCC.meetingreport.final.pdf
3.
New Climate Models... Knutti and Sadlacek, Robustness
and uncertainties in the new CMIP5 climate model projections,
Nature Climate Change, October 28, 2012, last paragraph, left column,
page 1.
Old
models, future CO2 concentration... United Nations Environmental
Program, GRID ARENDAL, Past
and Future CO2 Concentrations.
(accessed 021313)
4.
The results are that warming is greater, drying in dry areas is
greater and increasing wetness in wet areas increases further...
Knutti and Sadlacek, Robustness
and uncertainties in the new CMIP5 climate model projections,
Nature Climate Change, October 28, 2012. This statement is reflected
in Figures 1, 2 and 3 on pages 2 and 3.
5.
Long
tree ring records allow estimations of past variations in Colorado
River flow
... comparable to the most severe, but temporary, long-term decreases
in flow recorded ... NOAA Discussion of the paper, last two
paragraphs, accessed January 30, 2013.
6.
Dwarfs the current western drought ... Cook et al., Long Term
Aridity Changes in the Western United States, Science, November 2004,
page 1017, top of page right column.
7.
Unusual
in the last 104 years ... Ibid. page 1017, second paragraph, right
column. Drought conditions since 2004: North
American Drought Monitor.
8.
The
Rockies have seen nearly twice the average global warming... Hotter
and Drier: The West's Changed Climate,
Rocky Mountain Climate Organization, February, 2008, page iv. The
American West experienced 70 percent more warming than the average
for the rest of the world.
9.
More mountain precipitation is falling as rain in the winter ...
Ibid., page v.
10.
There is no moisture to evaporate... State of the Planet, Water
Matters,
Smaller Colorado River Projected for Coming Decades, Study Says,
sixth paragraph, accessed on January 31, 2013.
The
Colorado headwaters are expected to see more precipitation on
average... Columbia
University Press Release, Sixth paragraph.
Despite
the fact that precipitation might increase in some regions and
seasons (e.g. winter in northern California)... NOAA
Climate Variability and Predictability Program press release,
forth paragraph.
11.
The solutions to climate pollution will cost 1 percent of global
GDP per year ... The annual cost, for about the last 100 years -
every year - for providing our global society with clean drinking
water has been about 1 percent of global GDP annually, at $500
billion. Alley, Richard. Earth:
The Operators' Manual,
WW Norton, 2011.
Other
things that cost one about $500 billion per year:
-Annual US
military budget has averaged $500 billion
since about 1980, not counting wars.
-Advertising
at $500 billion per year
globally. $492 billion in 2011. Forecast to $629 by
2015.
-Agriculture damages from normal weather at $500 billion per
year in the US alone. In the US alone we see $485 billion normal
weather damages to agriculture every year. Lazo et al., US
economic sensitivity to weather events,
American Meteorological Society, June 2011. Press
release.
-US
health care costs
in 2009 were five times the cost of global clean drinking water at
$2.5 trillion per year.
No comments:
Post a Comment
Note: only a member of this blog may post a comment.