Every once in a while a new reporter arrives onto the climate scene that immediately thinks they know everything there is to know about climate, and therefore doesn’t need to interview the people he/she writes about, ask questions of them, research the facts, or get any other points of view to reflect balance whatsoever, because you know, Science! Let me introduce you to Zoë Schlanger.
I direct you to the article with the words “global cooling” circled.
Now if it had been just some other blog or some minor local newspaper I would not have paid any attention to this because what was written recently by Ms. Schlanger was so laughably bad it merits ignoring it in such venues. But, unfortunately she writes for Newsweek where they’re supposed to require a modicum of “high-quality journalism”, as they advertise to their readers:
This all came about because science writer Michael Bastasch picked up my article Despite attempts to erase it globally, “the pause” still exists in pristine US surface temperature data and expanded on it at The Daily Caller with: America’s Most Advanced Climate Station Data Shows US In A 10-Year Cooling Trend.
That article on The Daily Caller got picked up by the Drudge Report where it was a top billed article for about 36 hours giving it a hugely wide circulation. And as we know when such things happen people who believe they have a duty to “crush” dissenting views, such as Matt Ridley recently pointed out about how a principal (Rob Honeycutt) of the antithetically named website “Skeptical Science” pushes such labeling:
Rob Honeycutt and his allies knew what they were doing. Delingpole points out that Honeycutt (on a different website) urged people to “send in the troops to hammer down” anything moderate or sceptical, and to “grow the team of crushers”. Those of us who have been on the end of this sort of stuff know it is exactly like what the blasphemy police do with Islamophobia. We get falsely labelled “deniers” and attacked for heresy in often the most ad-hominem way.
And labeled “deniers”, we were. Schlanger writes in her article: (bold mine)
The author quotes Anthony Watts, a former meteorologist who runs a blog dedicated to climate change denial. For the graphs on which The Daily Caller article focuses, Watts used monthly temperature data from the U.S. Climate Reference Network (USCRN) from 2005 to 2015. The USCRN is a system of temperature monitoring stations around the country, managed by the National Oceanic and Atmospheric Administration (NOAA). Watts plotted the average temperature data from those stations over 10 years on a graph and found an almost-stable trend line that indicated slight cooling. This, he writes, “clearly” shows that a “‘pause’ or ‘hiatus’ exists in this most pristine climate data,” pointing to the much-referenced argument in “skeptic” circles that there has been a global warming “pause” in recent years.
About Watts Up With That? News and commentary on puzzling things in life, nature, science, weather, climate change, technology, and recent news by Anthony Watts
Maybe she never noticed that I carry almost every press release about climate, even though I might disagree with the findings. Maybe she never noticed that with WUWT’s broad global reach, those findings often get a wide distribution.
Maybe, while she’s thinking “denier”, she also missed how I tell readers about my solar power project on my home, my electric car, and my energy and water saving upgrades to my home in her zeal to label me as “dedicated to climate change denial”. After all, I’m sure the life of a twenty-something aspiring writer living in New York City who just graduated from college two years ago must be a whirlwind of social nirvana – so apparently there’s no time left to research people you write about or ask them questions. Maybe she’s also missed my writings where I make it clear that I agree that there’s been warming observed in the last century, that CO2 has a role in some warming (as does a bunch of other drivers). I wrote then:
For the record:
I don’t “deny” climate change or global warming, it is clear to me that the Earth has warmed slightly in the last century, this is indisputable. I also believe that increasing amounts of CO2 in Earths atmosphere are a component of that warming, but that CO2 is not the only driver of climate as some would have us believe. However, what is in dispute (and being addressed by mainstream climate science) is climate sensitivity to CO2 as well as the hiatus in global warming, also known as “the pause”. Since I embrace the idea of warming and that CO2 is a factor, along with other drivers including natural variability, the label “denier” is being applied purely for the denigration value, and does not accurately reflect my position on climate.
Maybe she also missed that I lambast those who think CO2 doesn’t have any effect at all. Yes, she must have missed all those things doing that “high-quality journalism” from Newsweek.
That’s strike one.
Note that I circled the phrase “global cooling” in the graphic above showing Schlanger’s title on Newsweek. I’ll point out that the phrase “global cooling” does not appear anywhere in my article on WUWT, nor in the Bastasch article on The Daily Caller. It is important to note that the use of the phrase “global cooling” is a construct known as a “straw man argument“, and it only appears in Ms. Schlanger’s article so she can easily knock it down. Both my article on WUWT and Basatsch’s article on The Daily Caller focused on pointing out that the temperature record (and slight cooling trend) was only about the contiguous United States.
Neither article suggested “global cooling” was observed, nor was there an extrapolation of the U.S. temperature trend to the world. We both pointed out it was a curiosity in a warming world, and that the difference looks to be attributable to the high quality measurement environment and state of the art technology (requiring no adjustments) of the US Climate Reference Network (USCRN).
That’s strike two.
Then there’s the claim of “cherry picking”, the most laughable Schlanger failure of all:
First, in 2005, the USCRN was far from complete. As of January 1, 2005, only 69 of its 114 temperature monitoring stations, or just 60 percent of the ultimate total, had been installed, according to NOAA’s Howard Diamond, who is the program manager of USCRN. The last and 114th station wasn’t installed until September 2008, which means that comparing the data from 2005 to 2008 with data after 2008 produces a severely lopsided analysis. This is especially important because of the geographic nature of temperature monitoring: Since only stations in certain areas of the U.S. were up and running before 2008, there is a lot of information missing from the averages of those early years.
If Watts had chosen to exclude the data from before the USCRN was complete and start his analysis on, say, January 1, 2009, to the present, he’d actually see “a slightly increasing trend of temperature anomaly data in the contiguous U.S.,” according to Diamond, as shown in the graph below. “So the same upward trends in temperature data we have seen have been and continue to be the case.” In other words, the U.S. is still getting warmer.
Then she presents a graph from the NCDC plotter tool citing “as shown in the graph below” combining three data sets, but that graph has no trend line.
From the Newsweek article:
Note the caption saying “A graph showing the trends in temperature anomalies monitored by USCRN from 2009 to 2015”. There is no trend line plotted, just the data.
Tell me Ms. Schlanger, how does one determine a trend in a graph without actually plotting a trend line? I find this hilarious that she didn’t take this most basic step, especially since I gave all the tools, including a link to a free trial program, DPlot, in my WUWT article to allow complete replication of my findings. This should have been simple enough for anyone to do, even a journalist.
And the claim of “cherry picking” is doubly hilarious because I plotted the entire available USCRN dataset as published by NOAA, making no choices of any kind on start dates or data omission, while she advocates:
If Watts had chosen to exclude the data from before the USCRN was complete and start his analysis on, say, January 1, 2009, to the present…
The hubris on display here by Ms. Schlanger is breathtaking, especially since it was predicted at WUWT in comments:
That’s strike three.
At The Daily Caller, Mike Bastasch notes in an update to his story, essentially the same issue with her report:
Update: An article published in Newsweek claims TheDCNF misused temperature data “to fabricate an argument that opposes the scientific consensus on climate change.” Newsweek basically argues that TheDCNF used incomplete temperature data to give a “lopsided” analysis of the U.S. temperature record over the last decade. Newsweek then quotes an environmental lawyer who talks about how skeptics like to “cherry-pick” data to disprove global warming. Ironically, Newsweek’s article makes the very error it accuses TheDCNF of making by further limiting the scope of the USCRN temperature record.
TheDCNF’s article uses the full data set presented by NOAA on its online database for the USCRN. Newsweek says it’s “lopsided,” but it was data NOAA saw fit to plot that data on their website even though the USCRN wasn’t fully completed until Sept. 2008. If this data is so lopsided, why does NOAA even bother plotting it? If it’s not reliable, why display it in conjunction with post-2008 data without some sort of disclaimer?
Newsweek also notes that starting USCRN data in Jan., 2009 shows a slight warming trend. That seems true, but here Newsweek is simply doing some “cherry-picking” of their own. For example, the USCRN temperature trend beginning in Oct. 2008 (the month after the final station was added to the network) shows virtually no trend in the temperature, while starting in Nov. 2008 shows a slight cooling trend. Same thing with Feb. 2009. What TheDCNF did was show the entire temperature record for the USCRN instead of “cherry-picking” a start date
Yes, please tell us Ms. Schlanger, why NOAA/NCDC sees fit to include this USCRN data from 2005, for public use, alongside other data sets on the same NOAA/NCDC website? I’d really like to hear that explanation.
And of course, she doesn’t show the graph she’s criticizing in her Newsweek article so that people can make a comparison. Here it is:
That’s strike four.
So since she couldn’t do it herself, let’s help Ms. Schlanger do the trend line. I’ve taken her advice “If Watts had chosen to exclude the data from before the USCRN was complete and start his analysis on, say, January 1, 2009, to the present…” and cherry picked that set of data, and used the DPlot program I provided to plot it.
Of course, Ms. Schlanger’s demand to plot the data in the way she says will show a warming trend is entirely replicable, using the data provided by NOAA and the DPlot program, yet she didn’t do it to demonstrate her own skills while criticizing others.
That’s strike five.
Ms. Schlanger stated in her Newsweek article:
“So the same upward trends in temperature data we have seen have been and continue to be the case.” In other words, the U.S. is still getting warmer.
Well, maybe not. You see if you use the other tool that NOAA/NCDC provides, the “climate at a glance” tool, using the Climate Division Data instead of the pristine USCRN data, you also get a cooling trend for the period set in the controls, shown below:
That’s strike six.
Granted, this is annual, not monthly like I originally plotted, but NOAA doesn’t allow a month by month plot of data, though they do allow specific months. So, if I “cherry picked” a dataset, why would a cooling trend for the Contiguous United States also show up in another data set?
We can go back to the plotting tool provided by NOAA/NCDC and look at that same Climate Division data on a month by month basis:
Here is the data provided by NOAA/NCDC, so that you can replicate these graphs yourself.
USClimDiv-time-series 2005-2015 (Excel worksheet)
The trend line will tell us whether there is a warming, a pause, or a cooling in the last decade in the Contiguous United States using the other dataset, from the U.S. Climate Division:
So, it seems no matter what data set you use, The “pristine” U.S. Climate Reference Network or the U.S. Climate Divisions based on the entire “non-pristine” Cooperative Observer Network consisting of thousands of stations, there does seem to be a “pause” or even a slight cooling in the Contiguous U.S. Surface Temperature Record for the past decade.
That’s strike seven. But, I saved the best for last:
But, what about the claims that I used the wrong start point, i.e. “cherry picking”? As Steve McIntyre famously says: “you have to watch the pea under the thimble“.
Compare these two excerpts from Ms. Schlanger’s Newsweek article:
“…according to NOAA’s Howard Diamond, who is the program manager of USCRN. The last and 114th station wasn’t installed until September 2008, which means that comparing the data from 2005 to 2008 with data after 2008 produces a severely lopsided analysis.”
“If Watts had chosen to exclude the data from before the USCRN was complete and start his analysis on, say, January 1, 2009, to the present…”
Wait, what? The dates don’t match. If the USCRN was complete in September 2008 (as they assert) why would we wait until January 2009 to plot the data? I plotted that too:Why Yes, there is a slight warming trend as Mr. Diamond and Ms. Schlanger assert. Both climate networks show this.
But, so what? Isn’t this use of a partial dataset cherry picking too? You’d think Ms Schlanger would have suggested starting in October 2008 when the USCRN was commissioned, (see NOAA October 2008 press release here) not when the last station was installed.
Now, all this is splitting hairs, but it still leaves the question of why if NOAA’s collective science team thinks the USCRN data was not spatially representative of the contiguous United States in 2005, the date they allow for the start of data plotting on their own web page, and they cut off earlier dates– why did they allow the public to use it?
Why did NOAA’s Mr. Diamond have to assert that we should not have used it, and why did Ms. Schlanger suggest I instead start in January 2009? I await their explanation but I don’t think either Ms. Schlanger and Mr. Diamond will answer that. In the meantime,readers can draw their own conclusions.
In any event, since the USCRN data from January 2005 and the US Climate Division data both essentially match, their argument about start dates for plotting trends in USCRN (due to it supposedly not being spatially representative of the contiguous USA until 2008) is moot:
It will be very interesting to see how the USCRN data plays out, because right now, just like the satellite temperature record, it shows no warming trend over the past decade or for that matter, several years longer:
So to sum up what we learned about Ms. Schlanger’s article, borrowing from Ms. Schlanger’s subtitle for her story:
Don’t believe everything you see in graph form.
Based on her lack of research skills, I’d modify that to say:
Don’t believe everything you read in Newsweek.