Guest Post by Willis Eschenbach
There have been lots of articles lately discussing the retraction by the UK Sunday Times of their claims about Amazongate. Folks like George Monbiot are claiming that their point of view has been vindicated, that Amazongate is “rubbish” and that skeptics have been “skewered”. So I decided to follow the tortuous trail through the Amazon jungle, to see where the truth lies.
Figure 1. The long, twisted, rainy jungle trail leading to the facts …
First, what did the IPCC say that caused all of the furor? Here’s the quote:
Up to 40% of the Amazonian forests could react drastically to even a slight reduction in precipitation; this means that the tropical vegetation, hydrology and climate system in South America could change very rapidly to another steady state, not necessarily producing gradual changes between the current and the future situation (Rowell and Moore, 2000). It is more probable that forests will be replaced by ecosystems that have more resistance to multiple stresses caused by temperature increase, droughts and fires, such as tropical savannas. (IPCC, PDF, p. 596)
Scary stuff, climates tipping to a new steady state, 40% of the Amazon rainforest changing to savanna …
Now, this is referenced to Rowell and Moore (PDF). The first problem that arises is that this is a World Wide Fund for Nature (WWF) overview piece, and is as far from peer-reviewed science as one can imagine. The WWF says:
Up to 40% of the Brazilian forest is extremely sensitive to small reductions in the amount of rainfall. In the 1998 dry season, some 270,000 sq. km of forest became vulnerable to fire, due to completely depleted plant-available water stored in the upper five metres of soil. A further 360,000 sq. km of forest had only 250 mm of plant-available soil water left. 46
Note that already we see a difference between the citation (such as it is) and the IPCC statement. The WWF says that the forest is “extremely sensitive” to “small reductions” in rainfall. The IPCC has upped the ante, saying the forest could “react drastically” to “even a slight reduction” in rainfall. In addition, the IPCC has added an uncited claim that the South American “vegetation, hydrology and climate system” could suddenly change to a new “steady state” … be very afraid.
Now, the WWF paragraph has a citation (46). This is:
46 D. C. Nepstad, A. Veríssimo, A. Alencar, C. Nobre, E. Lima, P. Lefebvre, P. Schlesinger, C. Potter, P. Mountinho, E. Mendoza, M. Cochrane, V. Brooks, Large- scale Impoverishment of Amazonian Forests by Logging and Fire, Nature, 1999, Vol 398, 8 April, pp505
The problem is that their citation only supports the second half of the paragraph, the part that relates to the 1998 dry season. It says nothing about the extreme sensitivity of the Amazon. It says nothing about a new “steady state.” Even Dr. Lewis, who convinced the Times to issue the retraction, admits this:
The 40% claim is not actually referenced in the Rowell & Moore 2000 report (they use Nepstad to reference the specific figures in the next sentence). The Nepstad Nature paper is about the interactions of logging damage, fire, and periodic droughts, all extremely important in understanding the vulnerability of Amazon forest to drought, but is not related to the vulnerability of these forests to reductions in rainfall. I don’t see how that can be the source of Rowell’s 40% claim. Its more likely an unreferenced statement by Rowell.
And there, the trail stops. Despite Pachauri’s oft-repeated claim that the IPCC is based 100% on peer reviewed science, the IPCC has referenced a WWF document which:
1. Is not peer reviewed, and
2. Has no further citation for the claim.
So why did the Times have to retract their claim? It was the result of a letter sent to the Times by Dr. Simon Lewis, who claimed that a) he had been misquoted, and b) the IPCC claim was scientifically accurate.
From Dr. Lewis’s statement, I do believe he was misquoted. However, that does not mean that the IPCC statement was correct. Dr. Lewis defends it, saying:
The IPCC statement itself is poorly written, and bizarrely referenced, but basically correct. It is very well known that in Amazonia tropical forests exist when there is more than about 1.5 meters of rain a year, below that the system tends to ‘flip’ to savanna, so reductions in rainfall towards this threshold could lead to rapid shifts in vegetation.
Indeed, some leading models of future climate change impacts show a die-off of more than 40% Amazon forests, due to projected decreases in rainfall. The most extreme die-back model predicted that a new type of drought should begin to impact Amazonia, and in 2005 it happened for the first time: a drought associated with Atlantic, not Pacific sea-surface temperatures. The effect on the forest was massive tree mortality, and the remaining Amazon forests changed from absorbing nearly 2 billion tonnes of CO2 from the atmosphere a year, to being a massive source of over 3 billion tonnes.
The Amazon drought impacts paper was written by myself and colleagues in Science (attached). Here is the press release explaining the sensitivity: http://www.leeds.ac.uk/news/article/36/amazon_carbon_sink_threatened_by_drought
Now, there’s a couple of things to note about this claim. First, other than a paper by Dr. Lewis himself about Amazon carbon sinks, there are no citations. The paper about carbon sinks is interesting, but it does not show anything about a “flip” to savannah, and doesn’t mention the 40% claim.
Second, he does not present any evidence that the 40% statement is correct. Instead, he says that climate models show that the statement is correct … Now, climate model results are interesting, but they are not evidence of anything but the assumptions of the programmers of the models.
And in fact, the 40% claim is called into question by another paper by the same Nepstad cited by the WWF document. It says:
During the severe drought of 2001, PAW10m [plant-available soil water to 10 metres depth] fell to below 25% of PAWmax in 31% of the region’s forests and fell below 50% PAWmax in half of the forests.
Now, if the Amazon were so sensitive, if it “could react drastically” to even a “slight reduction” in rainfall, certainly such a large reduction would make a big difference … but that didn’t happen. There was no “flip” to savannah mentioned in the paper.
Third, Dr. Lewis seems to want us to think that some fraction of the rainforest becoming savannah is supportive of the IPCC claim that:
… the tropical vegetation, hydrology and climate system in South America could change very rapidly to another steady state …
That’s just misdirection. Dr. Lewis does not provide any evidence in support of the alarmist claim that the South American climate is in danger of a rapid change to some other steady state. Which is no surprise to me, since I know of no historical evidence of such a rapid large-scale change in the tropical climate to a much dryer state.
And finally, even Dr. Lewis recognizes that there is no scientific certainty about this question, saying:
This is not to say this there isn’t much uncertaintly as to exactly how vulnerable how much of the Amazon is to moving to a savanna system.
Well … yeah. Given that uncertainly, his claim that the IPCC statement is “basically correct” is unsupportable. “Much uncertainty” means that we cannot make scary statements like the IPCC has done, and we certainly can’t say that they are “basically correct”. All we can say is that they are uncertain.
Before going on to look at some actual data, lets review the story so far:
1. The IPCC made a claim that “Up to 40% of the Amazonian forests could react drastically to even a slight reduction in precipitation”, and that the South American climate could change rapidly to a new steady state.
2. This was referenced to a WWF review paper which was not peer reviewed.
3. The WWF paper had no citation for that claim.
4. Dr. Lewis says the claims are correct. However, like the IPCC, he does not provide a citation for his claim that the 40% statement is correct. He points us to a 2009 paper, of which he is a co-author. It doesn’t contain any support for the 40% claim. He refers to a few climate models, but shows no evidence.
5. Dr. Lewis says that there is “much uncertainty” about the question.
6. Dr. Lewis does not provide any evidence to support the idea that the South American climate is likely to change rapidly to a new steady state.
Now, having reviewed the story so far, lets think about this a bit dispassionately. First, is it theoretically possible for the Amazon to “flip” from rainforest to savannah?
Certainly it can. If the Amazon rainfall went to a tenth of the current value, it would all be savannah. So how much would a “slight reduction” affect the Amazon rainforest?
To investigate this, we can look at the amounts of rainfall around the Amazon. Figure 2 compares the vegetation and the rainfall:
Figure 2. Vegetation map of central South America. The Amazon rainforest is dark green. Violet rectangle show area of measured rainfall shown below in Fig. 3. Red lines show rainfall in millimetres per year.
There are several things we can see from this map. First, rainfall is not the only thing that is limiting the Amazon rainforest. There are areas with less than 1600 mm which are rainforest, and areas with more than 1600 mm which are not rainforest.
Second, at the left edge of the rainforest, we have the Andes mountains. In these areas, the Amazon is limited by elevation rather than by rainfall.
Now, suppose that the rainfall drops by 10%. I’d call that a “slight reduction” in rainfall. Will that affect 40% of the rainforest? No way. If we were to shrink all of the red lines by 10%, we’d only get about a 20% reduction in area … but there are large areas which are not rainfall limited in that sense. So a 10% reduction in rainfall might, and I emphasize might, give us a maximum of a 20% reduction in rainforest area. To get to 40% rainforest loss, we’d need a large reduction in rainfall, not a slight reduction.
But who is claiming that there will be a large reduction in Amazon rainfall? That is a model prediction, and not even one that appears in all of the models. Dr. Lewis says:
Indeed, some leading models of future climate change impacts show a die-off of more than 40% Amazon forests, due to projected decreases in rainfall.
This, of course, also means that some leading models do not show a die-off. Even the models don’t all agree with the IPCC claim.
However, all of this, all of the claims and counterclaims, and the models, and Dr. Lewis’s letter, and the cited scientific documents, all run aground on one ugly fact:
The data shows no change in Amazon rainfall in a century of measurements.
Figure 3 shows three different ground-based observational datasets, along with the recent Tropical Rainfall Measuring Mission (TRMM) satellite data.
Figure 3. Four Amazon rainfall datasets, covering the rectangular area shown in violet in Fig. 2 (2.5°N–12.5°S, 72.5°W–50°W). Note the generally good agreement between the four datasets (including the TRMM satellite data)
The main feature of this dataset is its stability. Note the lack of any trend over the last century, and the lack of any large excursions in the rainfall. It stays between two and two and a half metres per year. There are no really wet years, and no really dry years. 95% of the years are within ± 10% of the average rainfall. There are individual dry years, but no prolonged periods of drought.
So while Dr. Lewis says (correctly) that rainforest can change to savannah, he is not correct that 40% of the Amazon is at risk from a “slight reduction” in rainfall. More to the point, there is no evidence to indicate that we are headed for a reduction in Amazon rainfall, “slight” or otherwise. That is a fantasy based on climate models.
The reality is that despite the globe warming by half a degree or so over the last century, there has been no change in the Amazon rainfall. As usual, the IPCC is taking the most alarmist position possible … and Dr. Lewis is doing all he can to claim that the IPCC alarmism is actually good science.
Unfortunately for both the IPCC and Dr. Lewis, here at the end of a long, twisted, and rainy jungle trail, we find that the facts inconveniently disagree with their claims.
[UPDATE] Credit where credit is due. I love writing here because I always learn something. The Amazongate story was originally broken by Richard North, whose blog is EUReferendum. Give it a look, lots of good stuff.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Their was an error in my earlier post, the Cox die back theory is not the most absurd of the Amazon theories the dubious honor goes to none other that Dan Nepstad. In a paper he produced for the WWF in 2007 with the conservative title “Amazon’s Vicious Cycles” Nepstad claims that 55% of the Brazilian forests could be destroyed within 20 years, note the nice qualifier “could”. This document was the basis of the REDD submission presented at the Bali conference in December 2007 by the Woods Hole Research Center, authored by none other than rather busy Dan Nepstad , who seems to be alternately employed by the WHRC, WWF and the Moore Foundation( a major contributor to the WWF). The REDD scheme’s intention is to preserve the worlds forests by using a market based emissions trading scheme. The WHRC , is aided by Goldman Sachs, a foundation sponsor of the WHRC. It is great when an independent scientific organization, despite the numerous environmental activists on their board, and a philanthropic financial institution combine their resources for the benefit of the world.
… And there, the trail stops […] other than a paper by Dr. Lewis himself about Amazon carbon sinks, there are no citations […]
Really? Lewis refers us to his colleague Daniel Nepstads article on the issue both online and as an Appendix to his PCC Complaint. This is amply referenced, here’s the relevant extract:
” Our 1999 article (Nepstad et al. 1999) estimated that 630,000 km2 of forests were severely drought stressed in 1998, as Rowell and Moore correctly state, but this forest area is only 15% of the total area of forest in the Brazilian Amazon. In another article published in Nature, in 1994, we used less conservative assumptions to estimate that approximately half of the forests of the Amazon depleted large portions of their available soil moisture during seasonal or episodic drought (Nepstad et al. 1994). After the Rowell and Moore report was released in 2000, and prior to the publication of the IPCC AR4, new evidence of the full extent of severe drought in the Amazon was available. In 2004, we estimated that half of the forest area of the Amazon Basin had either fallen below, or was very close to, the critical level of soil moisture below which trees begin to die in 1998. This estimate incorporated new rainfall data and results from an experimental reduction of rainfall in an Amazon forest that we had conducted with funding from the US National Science Foundation (Nepstad et al. 2004). Field evidence of the soil moisture critical threshold is presented in Nepstad et al. 2007. ”
Which leads Nepstad to state:
In sum, the IPCC statement on the Amazon was correct. The report that is cited in support of the IPCC statement (Rowell and Moore 2000) omitted some citations in support of the 40% value statement.
Hope this helps.
If there is any conversion to savanna, it is likely to come from local logging and burn-clearing for agriculture. Because the heavy rain leaches nutrients from the soil, most Amazon nutrients are held in the biomass. Since logging removes that biomass immediately and slash and burn ag removes it over a longer time, you don’t get the regrowth, but that isn’t a rainfall problem.
With reduction in rainfall, you’d first go through all the forest stages that populate non-rainforest tropical areas before you got to open savanna, and that might take centuries. Soil nutrients will continue to wash down slowly from the Andes.
Reduced rainfall would be beneficial to agriculture in that respect, leaving imported potash and locally produced nitrates available in the soil longer. Gotta look on the bright side.
I have to say, I am becoming increasingly un-willing to even discuss MODELS as actual climate. I wish articles would state at the beginning that the predictions in them are based on models. Then it would be clear that we are discussing models and not actual events.
I’m sure models can be useful tools, but until they can actually predict the past or the future I’m afraid they are just academic.
Any honest climate scientist’s interview would conclude with “but we just don’t know yet”.
I’m beginning to feel that the people creating “models” are using Edsels as benchmarks. It’s time for the modeling police to require “real world possibilities” and not be filled with worst case scenarios and adjustments for “public attitude adjustment”(propaganda). The models look like they are all “rube goldberg” scenarios of run away reactions.
That’s amazing – even their own models contradict the claim…..
Any idea exactly how much rainfall reduction would be needed to cause a 40% die-off? I’m guessing quite a bit.
As any good alarmist would say: “Don’t confuse me with the facts, my mind is made up.”
Is there any IPCCish claim, which can withstand exposure to real data?
Temperature: http://climexp.knmi.nl/data/icrutem3_285-310E_2.5–12.5N_na.png
Precipitation: http://climexp.knmi.nl/data/iprca_285-310E_2.5–12.5N_na.png
One gotta love the KNMI Climate explorer.
Note, that IPCC claim contains twice “could” and once “probable”.
“The great tragedy of science is the slaying of a beautiful hypothesis by an ugly fact.” – Thomas Henry Huxley (1825-95)
As I sat here reading this article an advert for the WWF was broadcast on the television. They were asking for money to save the Jaguar because “20% of the Amazon rainforest has already disappeared”. This time they blamed logging etc (not climate) for the present decrease and future decreases. Obviously they don’t have enough confidence in the IPCC version to claim another 40% will disappear due to climate change unless you give them money.
It’s in the WWF own financial interest to promote alarmist forecasts. The more alarmist the claims, the more money they can fool the public into giving. But any claim they make has to get past the Advertising Standards Authority and I doubt this one would.
Thanks Willis.
You are a great investigator for us.
Thank you for taking us by the hand and leading us through the arguments step by step. All very logical and reasonable.
I now wait a step by step refutation of all your points by Dr Lewis or one of his cohort, with the same clarity.
And wait…….
And wait….
As usual Willis puts in the missing link, logic, well done Willis, and all this work without a grant? but then grants don’t pay for logical results.
Rainfall
positive feedback
clearance fire smoke -low rainfall – fires/smoke -low rainfall
[http://] irina.eas.gatech.edu/EAS_spring2008/Andreae2004.pdf
Greenhouse on ENSO
[http://] iprc.soest.hawaii.edu/users/axel/GL11677W01.pdf
Drought and Fire in the Greenhouse
Ecological and Climatic Tipping Points of the World’s Largest
Tropical Rainforest, and Practical Preventive Measures
WWF! but An independent scientific review of the content of this report was conducted by Prof. Yadvinder Malhi, Professor of Ecosystem Science at Oxford University.
[http://] [www] .whrc.org/policy/pdf/cop13/Amazon-Vicious-Cycles.pdf#search=”Nepstad et al amazon”
Tipping point.
This is a very silly statement , made without backup “Now, if the Amazon were so sensitive, if it “could react drastically” to even a “slight reduction” in rainfall, certainly such a large reduction would make a big difference … but that didn’t happen. There was no “flip” to savannah mentioned in the paper”
It is obvios to even me that the tipping point of ecology is not a rainforest-to-no-forest-in-a-year-situation. Ecological tipping points take time to be evident but once the tipping point is passed it may be difficult to turn back. All trees do not die in a year. They may, for example, become less resistant to pest attack and then die over a number of years etc.
Good analysis, Willis. You have destroyed the IPCC claim in two ways. Firstly, by following the paper trail which leads to a dead end of oft vague and unsupported assertions, and secondly, by taking a practical, common sense appraisal of the actual rainfall data, to show that a small reduction cannot possibly lead to a changeover to savannah.
To my mind however, the saddest thing about this episode with “The Times” is that they appear totally unable to defend themselves from the ludicrous charges of Lewis. Have they no scientific journalists who can defend their position, as you have done? It is as if a national newspaper, on uncovering a major political scandal, suddenly published a retraction because some minister wrote a letter saying that the scandal had not in fact occured.
Amazongate Mk II
How to hide the decline of IPCC credibility !!
.
Nicely dealt with.
Now what if the Amazon WAS savannah? Professor Philip Stott wrote this a few years ago: Tropical Rain Forests – Exposing the Myths http://www.probiotech.fsnet.co.uk/trf.html
“At the end of the Last Ice Age, only some 12,000 to 18,000 years ago, the tropics which are today occupied by these so-called ‘ancient cathedrals’ were seasonal savanna grasslands, both cooler and much drier than now.”
It is based on an earlier paper called Jungles of the Mind, http://www2.csusm.edu/spanish/undergradcenter/jungles_of_the_mind.htm
Interesting stuff.
The definitive account of Amazongate
http://eureferendum.blogspot.com/search/label/Amazongate
Nice one, Willis.
It’s curious how people with an apparent political point to make don’t seem to think that evidence matters.
WFF pfft. WWF has become the largest get rich quick scheme factory in the world.
If they can’t, or don’t, set up a daughter organization, or company, to charge other organizations and companies if they don’t like it.
Fish farmers bad — unless they pay. Loggers bad — unless they pay. Oil companies bad — unless they pay. Heavy industry bad — unless they pay. Et Cetera. Unless you pay for WWF certificate of approval or what ever everyone is bad. Other “climate organization” are of course approved, somewhat, to tax everyone as well if it’s a “good” scheme.
Small point, Willis: it was the Sunday Times which retracted, not the (daily) Times. Different newspapers, different editors, different staff – though same proprietor (R Murdoch).
Baa Humbug says:
June 27, 2010 at 3:08 am
Baa Humbug,
Here are some other things I find strange. It is claimed 40% of the Amazon is “extremely sensitive” to small reductions in rainfall.
Yet…
1998 saw, not a small reduction in rainfall, but a DROUGHT.
That was 12 years ago.
Certainly, if 40% of the Amazon is “extremely sensitive” to a “small reduction in rainfall”, wouldn’t any of the Amazon react to a DROUGHT?
I would expect if all the above were true, this drought affected region would be well on its way to savannah hood. Yet, nothing is mentioned in follow up reports.
I did catch some episodes of River Monster. It seems there’s still enough water in the Amazon to grow some mighty big and fearsome fish.
Now there’s a guy with a good job. He gets to go fishing and paid for it to boot.
Almost as good as a programmer for a GCM. Paid to program and you don’t need to produce anything verifiable.
Mike MCMillan @ur momisugly 3.34am
The dangers from logging have diminished in recent years. Since its peak in 2004 logging has reduced been by 64% due to a government crackdown on illegal logging, the establishment of protection zones that cover more than 30% of Brazil’s forests and act as a buffer zone between cleared land and the core rain forests, limiting the clearance of privately owned land to 20% and the growing lack of land suitable to pasture or agriculture. What the alarmists never mention is that pasture land soon degrades and that 32% of cleared land has been revegitated either by plantation timber or natural regrowth. They also fail to mention that the FAO consistently over estimated logging until recent years when satellite coverage provided a more accurate means of assessment.
Amazon = $
Lots of attractive graphs, including a hockey stick, from Nepstad.
http://news.mongabay.com/2008/0124-nepstad.html
Mr Eschenbach, if you had actually followed the debate instead of trying to reinvent the wheel, you would be aware that the WWF has asserted that the problem stems from a reference accidentally omitted from its report, as in the claim that “WWF acknowledges that a reference to Fire in the Amazon as the source of the 40% claim outlined above was mistakenly omitted during the editing process of the Global Review of Forest Fires report.” It goes on to say:
WWF’s source for this statement is Fire in the Amazon, a 1999 overview of Amazon fire issues from the respected Instituto de Pesquisa Ambiental da Amazônia (IPAM – Amazon Environmental Research Institute). The source quotation from Fire in the Amazon reads “Probably 30 to 40% of the forests of the Brazilian Amazon are sensitive to small reductions in the amount of rainfall.”
Intriguingly, the WWF then goes on to say that: “Our report does NOT say that 40% of the Amazon forest is at risk from climate change.”
The argument has thus already progressed to a discussion over the provenance of “Fire in the Amazon”. It turns out that this is edited (in fact, largely written) by Daniel Nepstad, and is not peer-reviewed. Thus, we have the IPCC citing an assertion in a WWF document which is “accidentally” unreferenced but, when the reference is identified, it relates to a non-peer-reviewed publication.
It then gets better, for “Fire in the Amazon” refers to the Brazilian rainforest (less then half the total) while the IPCC asserts that the 40 percent mentioned applies to the Amazon basin as a whole. Thus, even if “Fire in the Amazon” could be relied upon as a legitimate source (which it cannot), it still does not support the IPCC assertion.
All this and much more can be read about on EU Referendum, including this comprehensive analysis which pre-dates yours by three months and which, had you noted it, would have significantly improved your exposition.
I suppose this is a long-winded way of saying that you are not the only toiler in the vinyard. Some of us have been at this specific theme a lot longer than you and a little bit of acknowledgement would not only not go amiss, but might actually benefit WUWT readers.