I’ve been working on two different essays for school this past week. One is about the Grapes of Wrath. Bleccch. The other, however, was much more enjoyable. Compare three journalism texts on the same topic, see how they differ, and why.
I chose to cover the publication of the Kaufman et al study, with articles from the Winnipeg Free Press, New York Times, and Sydney Morning Herald. As my essay ended up being very relevant to the purpose of this blog, I thought I’d publish it here. Enjoy.
On September 4th, 2009, one of the foremost scientific journals in the world published a significant study regarding global warming. Scientists have known for years that the Arctic has been warming at an unusual rate, but, by putting the recent trend in historical perspective, the Kaufman et al study was able to show just how unusual. Nearly every major newspaper in the world covered the publication of the study, some more accurately than others. In this document, three articles, from Canada, the United States, and Australia, are examined.
This issue is not about the warming of the Arctic but, rather, a specific study regarding the warming of the Arctic. Therefore, it is fairly easy to gain an objective basis of the issue, simply by reading the study. The conclusions of the study can then be compared to the newspaper articles to see how well they summarize its findings.
The study in question, “Recent Warming Reverses Long-Term Arctic Cooling”, was published in Science magazine, which is one of the most reputable scientific journals in circulation. The study was in the field of paleoclimatology, which uses lake sediment, glacier ice, and tree rings to reconstruct past climates. Notably, this study was the first of its kind to reconstruct 2000 years of Arctic climate at a high resolution (temperature measurements which were only years or decades apart). Previous studies of this resolution only measured the Arctic climate of the past 400 years.
The results were striking. As expected from calculations of the Earth’s cyclical changes in orbit, the Arctic was cooling for the first 1900 years. The Holocene thermal maximum (warmest part of the interglacial) was between 6 000 and 10 000 years ago, and the Earth has been slowly moving toward another ice age ever since. However, in the past century, anthropogenic greenhouse gases caused this trend to sharply reverse. The rapid warming, at a rate of 1.4 C/century (compared to the previous cooling of -0.022 C/century), caused the Arctic to become even warmer than it had been at the beginning of the graph. The period of 1999-2008 was, therefore, the warmest decade in the past 2000 years.
It is important to note that the focus of the study was on the previous cooling trend, not the recent warming trend. As observational data is present for the past 100 years of Arctic temperatures, reconstructions were unnecessary. The recent data simply had to be inserted at the end of the process. Therefore, the rapid warming was only mentioned at the very end of the study, occupying less than one page out of four. The recent warming trend, especially in the Arctic, is already common knowledge in the scientific community. It is unsurprising that it took such a minor role in a paper which was, after all, about reconstructing climate from past centuries.
However, “Arctic warmer than in last 2000 years, new study finds”, printed in the Winnipeg Free Press, did not properly articulate this focus. The author, Bob Weber, called the study “groundbreaking”, which is very true, as it provides the first detailed reconstruction of Arctic climate going back farther than 400 years. However, Weber then devoted nearly the entire article to the recent spike in temperatures, implying that it was this recent anthropogenic trend, rather than the detailed reconstruction, which was so groundbreaking. For example, in the introductory paragraph, the study is referred to as providing “real-world evidence to back mathematical climate models that suggest greenhouse gases are behind global warming”. The authors of the study did compare the reconstruction to model simulations, but primarily in the context of the previous cooling, to see how well their data correlated with orbital calculations. They cared much less about how the observational warming data matched up with computer models, as that has been done countless times in other studies. Also, the models do not merely “suggest” that greenhouse gases are behind global warming. “Suggest” is likely too weak a word, especially for members of the general public, who are not well-phrased in the tentative language of science. In fact, there is not a single climate model which has been able to explain the recent warming without taking into account human activities. If there was, you can be very sure that everyone in the climate blogosphere would have heard about it far too many times from angry commenters.
A similar error is made when Weber writes, “The study says the fact that no other major variables changed about that time – there were no large volcanic eruptions, for example – suggests there could be only one culprit for the warming Arctic: carbon dioxide emissions that began increasing rapidly during the Industrial Revolution.” This error is somewhat more grievous than the last, as the study never made such a claim. The lack of an alternative explanation for the recent warming certainly is well known in the scientific community, but it is not covered in this particular study. Saying that it is implies that the study was focused on recent trends in temperature, rather than paleoclimatic reconstructions. Therefore, since the study is “groundbreaking”, the idea of an anthropogenic influence on climate is portrayed as groundbreaking as well – when, in actuality, scientists first began to examine the issue over a century ago. Svante Arrhenius’ claim, in 1896, that industrial emissions of carbon dioxide could one day warm the planet was groundbreaking. This study making such a claim was not.
However, changing the focus of the study was not the only error in the Free Press article. Weber made an incorrect choice of words which would have made anyone well-phrased in climatology cringe. When discussing how the study used lake sediment, glacier ice, and tree rings to reconstruct past climate, Weber stated that “all three methods are well-accepted ways of estimating weather.” It is really quite incredible that a science journalist could have gotten away with such an error, as the difference between weather and climate is covered in the first chapter of nearly every book about climate change. In reality, estimating weather with proxy reconstructions is virtually impossible, unless paleoclimatology were to advance to a point where proxy temperature measurements could be determined on a weekly scale. All that can be done – and all that really matters for our purposes – is to estimate climate, the standard conditions of temperature, precipitation, and variability in an area. Climate refers to a long period of time (classically 30 years). It is not affected by conditions of day-to-day weather. As it is, Weber’s assumption that the two words are interchangeable only serves to reinforce public confusion on their differences.
The article concluded with a statement about climate models, claiming that they “are often used by scientists to predict the effects of global warming and are just as often criticized by skeptics as mere speculation”. It is difficult to understand who Weber is referring to as “skeptics”. Does he mean the mobs of angry laypeople, omnipresent on YouTube, amateur blogging sites, and letters to the editor sections, nearly all of whom have little to no scientific background? If so, should we really trust them on the matter of how reliable climate models are? Do their shouts really count as criticism if they don’t know what they’re talking about? Alternatively, perhaps Weber is referring to actual scientists, who understand the physics, mathematics, and statistics of climate models, yet feel that they are little better than random guesses. We have no solid figure as to how many of these scientists exist worldwide, but minimal research shows that most, if not all, of them also hold the belief that humans are not affecting the climate. Such scientists make up roughly 3% of the climatological community (Doran and Zimmerman, 2009). They perhaps make up roughly a dozen worldwide. Giving them space in an article about a credible peer-reviewed paper reinforces yet another public misconception – that scientists are fairly equally split on whether or not humans are the cause of global warming.
Finally, the article was published on page A2 of the Free Press. Any regular reader of this newspaper will know that such a page is the most trivial in the issue. Page A2 is almost entirely covered by ads and indexes. There is a tiny space for an article, which is usually about someone’s lost parrot, or how a digital image of a squirrel is taking over the Internet. It is surprising that such a “groundbreaking” study was printed on this page, instead of in the “Science” or “World” section. What does it say about our mainstream media when an article of such scientific importance is squeezed into the easiest possible place to overlook? What does it say about our society?
Luckily, “Global Warming Could Forestall Ice Age”, printed in the New York Times, did not contain these errors. It was printed in the “Environment” section. Its focus was on the Earth’s glaciation cycles rather than the recent trends in temperature. It did not mention skeptics. The only similar mistake made was a mischaracterization of what was and was not stated in the study. Andrew Revkin, the author, writes that “summer temperatures in the Arctic region would be expected to cool for at least 4000 more years, given the growing distance between the Sun and the North Pole during the summer in the Northern Hemisphere, the study says.” The study did not say such a thing – it was common knowledge long before the study was written, since Milutin Milankovitch first calculated the Earth’s orbital cycles in 1941. Similarly, Revkin states that “the next [ice age], according to recent research, could be 20 000 or 30 000 years off discounting any influence by humans.” Again, this research is not recent. Portraying it as so implies that it is still somewhat shaky and unproven, which could lead the public to wonder if scientists really understand ice ages. Some individuals would be all too eager to claim that the Earth was simply approaching a new thermal maximum, which was causing the warming, not greenhouse gases.
A much more serious and troubling error appears later in the article. Revkin writes that “the ability to artificially warm the climate, particularly the Arctic, could be seen as a boon as the planet’s shifting orientation to the Sun enters a phase that could initiate the next ice age.” Anthropogenic greenhouse gases certainly could offset the next ice age. However, Revkin phrases this idea in a way which suggests that it is a good thing. Multiple misconceptions are present in this chain of logic. Firstly, Revkin assumes that warming is better than cooling. This is an easy assumption to make when you live at high latitudes, such as in the United States. However, from a broader, but still anthropocentric, worldview, it is the amount of change in climate which matters more than the direction it moves in. For example, warming might have some local benefits in North America and Russia, but it also causes subtropical drying, making agriculture in areas such as North Africa, Mexico, and Australia less and less viable. Cooling, in the other hand, allows the boreal forest to expand southward into richer soils, but it also makes the climates of high latitudes less conducive to human settlement. Both warming and cooling would have impacts on ocean and wind currents. Alternatively, perhaps Revkin believes that warming is more conducive to life on Earth. This assumption is incorrect, even at high latitudes – just look at the giant mastadons and other mammals which were present throughout North America during previous ice ages. Warming also causes desertification in some areas of the world, leading to water stresses that can threaten species. It seems obvious that the best situation for any species, including humans, is for the global climate to remain stable. Therefore, the magnitude of any change matters much more than whether it is warming or cooling.
So, let us look at the magnitude and rate of change in the transition to an ice age compared to the observed and expected warming. As the study states, the Arctic was previously cooling at a rate of -0.022 C/century. This rate remains relatively steady as the Earth moves into a glacial state. However, the present warming is occurring at a rate of 1.4 C/century – 60 times faster, which is expected to accelerate dramatically as it progresses. Obviously, the current change in climate is much more dramatic than a descent into an ice age, and will cause much more stress for people as well as other life on Earth. Anthropogenic global warming is in no way preferable to an ice age.
Critically examining these two North American articles makes one lose faith in popular science journalism. However, when we look at the opposite corner of the world, Australia, a much more accurate article emerges. “Cooling trend of 1900 years reversed in decades” was published online through the Sydney Morning Herald. It is significantly shorter than the other articles, but contains none of the glaring scientific errors. In fact, the study is perfectly summarized in the first sentence of the article, which reads, “The Arctic was cooling for 1900 years because of a natural change in the Earth’s orbit until greenhouse gas accumulation from the use of fossil fuels reversed the trend in recent decades, scientists say.” In a remarkably concise fashion, Renee Schoof, the author, is able to convey the purpose, focus, and conclusions of the survey. The article goes on to explain how the Arctic would still be cooling if it weren’t for human activities, how global warming is expected to affect the Arctic first and foremost, and how warming in that area has serious implications for sea level rise and the stability of methane hydrates. Although the article was short, it lacked the editorial slant of the previous two pieces. One wonders if the Australian society is more accepting of the idea of anthropogenic climate change than North America is, as its media, at least in this instance, does a remarkably better job reporting it.
What does this analysis tell us about the reporting of science, particularly climate change, in North America? As we can see, it’s much worse here than it is in other parts of the world, as the two North American articles contained glaring errors, misrepresentation, and editorial slants which only served to perpetrate popular myths about climate change. The Australian article, in contrast, perfectly summarized the study. Are the Australians better informed, and therefore demand accurate journalism on climate change? Or are their media simply more responsible? Conversely, are North Americans blind to scientific situations which might have policy implications, and thus welcome journalism that downplays the status of the science? Or are the North American media corporations influenced by individuals and think tanks who wish the public to remain confused about whether or not climate change is a serious threat? Perhaps the two sides of the situation, the public and the media, feed off each other. The media influences what the public thinks, and the public expects the media to confirm their preconceived opinions. Once this feedback loop begins, it is difficult to stop. Hopefully it is not impossible.
People can also take the different proxies and recreate the paper in a spreadsheet. They have provided the data here:
Is A2 a separate section? I would think the second page of a newspaper would be pretty significant.
[Not in this paper. A1 and A3 are very important, but A2 is the index and the ads, and a teeny tiny space for a (usually trivial) article. -Kate]
Journalism seems to greatly benefit from assuming that new research is “groundbreaking”, even though it only supports what scientists have known for years. For example, when that 47-million-year old apelike fossil was discovered, it was reported in dozens of newspapers as “the missing link”. Not “a missing link”, but THE missing link. As though we didn’t have any evidence of the connection between humans and apes, and now we suddenly do. Right.
If a newspaper gets to report on stuff as though it is new and groundbreaking, it makes the newspaper look good. Newspapers like to do that.
I do agree, though, essays are enjoyable as long as you have a good teacher. And what is a blog but a bunch of mini-essays? (Well, some blogs aren’t, but I don’t read those blogs.)
[The problem with essays vs blogs is that essays have to have structure and good grammar. -Kate]
One (minor) point I would defend Revkin on is his characterizing as “recent” the results that a new glaciation might be @ 30K-40K years off . IIRC the relevant paper is from about five years ago. Admittedly what counts as recent for Revkin or me may not for you. :)
A point of criticism I would add is that if he was going to mention the glaciation offsetting effect at all he should have noted that anthropogenic influence has already far exceeded the level needed for that.
“The Australian article, in contrast, perfectly summarized the study. Are the Australians better informed, and therefore demand accurate journalism on climate change? Or are their media simply more responsible?”
But the article actually originates in North America! It’s by Renee Schoof of McClatchy. The headline is unique to the Sydney Morning Herald, and there are some other textual changes, but if you search on the first few sentences you will find that this story was carried in North American media too.
Australian newspapers are all owned either by the Fairfax media group or by Rupert Murdoch’s News Ltd. The Fairfax papers are a little more left-leaning and the SMH is one of theirs. The Australian, the main national paper, is a Murdoch paper, and it’s notorious for providing a platform to deniers, delayers, and skeptics. The Australian climate activist Anna Keenan, a perennial optimist, says she nonetheless draws heart from the fact that the paper has a permanent section on climate now. The issue is certainly extremely prominent in Australian political life (example), mostly because of the drought. We have a whole ministry of climate change, created after the change of government at the most recent election, and if the opposition can’t assemble a coherent position on the issue, they’ll probably lose the next election as well.
So it’s true that climate politics is a big part of the Australian scene, but don’t think this means that we’re all on the same page down here! The government has a Copenhagen-compliant emissions-trading policy that is not yet law, the ex-government (who still hold the balance of power in the upper house) are torn between wanting to weaken that policy and wanting to block it outright, and most climate activists reject the policies of both centrist parties and support the Greens, who are 350.org-style radicals with a few Senate seats.
[Aaah….I missed the fact that it was written by an American. However, the fact that the paper decides to publish that article, over something more “controversial”, shows that it has more scientific responsibility.]
It got no mention at all in the rest of the Australian press that I could see. Global Warming is not reported in some of the media unless it cannot be avoided. (ie they would look foolish)
[Yeah, it’s pretty similar here in Canada. There was a fair amount about climate change as well as environmental issues for the past few years….right up until the economic crisis began. Since then it’s been off the radar. -Kate]
Oops, one of the local WA papers did an article but I can’t find it.
PS Good essay
Nice essay. My only critique would be that it’s not really fair to characterize an entire country’s media coverage on the subject based on one article. Revkin always does a bad job reporting on climate science, but there are of course a number of American journalists who do much better.
Not to defend the American media, which I think on the whole does a terrible job covering climate science. But it’s still a little rough to generalize based on a single article.
[The assignment was limited to three news pieces, so I was sort of forced to broadly speculate based on a likely unrepresentative sample. Eventually (ie when I actually have time….) I might look at a wider sample of journalism sorted by geography and see how it compares. -Kate]
“summer temperatures in the Arctic region would be expected to cool for at least 4000 more years, given the growing distance between the Sun and the North Pole during the summer in the Northern Hemisphere, the study says.”
Did Revkin really say that?
I think he’s confusing distance with angles.
[citations needed – Kaufman et al criticism]
I gave the citation for the Kaufman et al criticism. The data from Kaufman et al and the paper Tiljander et al. They are at odds with each other. Tiljander is the original author of Proxy #20.
[citations needed – if you’re going to criticize a paper, find a peer-reviewed source that criticized it and cite that. Piecing together data from different sources and creating your own conclusion is still armchair analysis. There are a lot of other places you can do that – Tamino, for example, as he’ll be able to assess your argument. I can’t. -Kate]
S2, this is actually a correct way to describe what is happening, though perhaps not the way you are used to. Successive Earth-Sun distances at boreal summer solstice are longer and longer. And yes, it’s angles doing it ;-)
LOL, Tamino deleted my post asking him if head an opinion on the upside-down use of Tiljander.
You say this site is about evaluating credibility, and I pointed out why I find this paper you brought up to be not credible.
[You’re mixing up credibility and coherence. Here you’re assessing coherence, based on your own analysis, or the analysis of other individuals/professionals. If a rebuttal comes out in a journal (Kaufman et al is still fairly new so I’m not surprised there isn’t anything yet) then you can cite that for your criticism. But when you start piecing together multiple sources and creating your own conclusion, it’s pretty easy to misinterpret somebody’s work.
I will not allow nonsensical claims to be published on my site, and I don’t have the time or the education to investigate all claims based on coherence. That’s why the comment policy has a solid set of rules about how the basic conclusion of your scientific claims must be supported by peer-reviewed literature. -Kate]
[citations needed – “lake varves in the Tiljander proxy should be interpreted as being at their warmest iaround the 1100s”. The Tiljander paper says exactly the opposite. You’re not really backing up your arguments with citations when the citations come to a different conclusion than you do. If you have complaints about Tiljander et al, it’s most effective for you to follow standard scientific practice and write a letter to BOREAS. That’s your best shot at fixing what you see as problems in the peer-reviewed literature. Choosing instead to only talk about it in blogs has the aura of spreading rumours, whether or not that is your intent. -Kate]
[citations needed. I still don’t understand where you’re getting the “warmest in the 1100s” thing. Even if it was warm in the 1000s and 1200s doesn’t necessarily mean it should also be warm in the 1100s. If it does, give us a citation for it. -Kate]
The 1100s had a cold spell that is confusing things.
Let’s try this.
According to Tiljander et al
1) It was warmer in 1020s than in 1120s
2) It was warmer in 1220s than in 1120s
3) It was warmer in the 1000s(century) and 1200s than in times like the 1400s, 1500s, and 1600s
Fig 5 at http://www.climateaudit.org/pdf/others/Tiljanderetal.pdf
According to Kaufman et al
The Tiljander proxy at Lake Kottajarvi says
1) It was cooler in the 1020s decade than in 1120s
2) It was cooler in the 1220s decade than in 1120s
3) It was cooler in the 1000s and 1200s than in times like the 1400s, 1500s, and 1600s.
For points 1 and 2) from the Kaufman data the value in the 1120s is 1.04, while in the 1020s and 1220s it is -1.82.
Here is a chart of TIljander proxy data listed by Kaufman et al
Here is the overall average of proxies
[Okay, that makes more sense now. The data set differs. However, the implications are not necessarily heinous. It may be something as simple as revising the Tiljander data for error. Robert Grumbine gives a nice explanation. If you’re unsure as to whether the data was used correctly, email Kaufman or send a letter to Science. That’s what the peer-review process is set up for. -Kate]
The warmest in the 1100s, I said around the 1100s realizing the 1000s and 1200s were the warmest periods, and didn’t notice how much dropoff there was in the 1100s in between. I think this is for only 30 years from 1115-1145.
I don’t think the paper lists the coldest period as being in the 1100s, but yes I should have noticed the significant drop. I should have just said 1000s and 1200s.
Yea, the data sets are different. That’s why I have trouble giving credibility to the paper. The datasets appear to be the same , but upside down(just flip the chart I linked). The Kaufman paper gives no description of any error corrections.
Robert Grumbine’s explanation is not helpful, as I’m looking at the single proxy, and he is criticizing a WattsUpWithThat misrepresentation of criticism of a different paper, and is talking about upside down hockey sticks. I accept that the results of the other paper are not upside down. So most of Robert’s explanation, while correct, doesn’t reflect on this paper. Also, evaluating the results without this proxy is intersting, but not what I am evaluating here.
Again I am assessing credibility. The use upside-down, makes it difficult to give credibility. When having it pointed out and not correcting, or giving an explanation, also impacts on credibility. Having to rely on random bloggers guessing about why the paper is valid, and the author made no mistakes, doesn’t lead me to assign more credibility to an author.
[If you’re not sure how to interpret the difference, email the author and ask. He’s the best person to help you. -Kate]
Yes he is, and how should I interpret the lack of a response?
Instant gain in credibility. Kaufman has responded with:
>That was an error on my part. I have submitted a correction to SCIENCE.
[Awesome. That’s good to know. How do journals like Science deal with corrections? Do they republish the article? -Kate]
I don’t think they republish, they just publish the correction. The author says the correction is pending review.