It misses a most important point: when people talk about a "global warming hiatus" from 1998 to 20XX, they are always full of shit.
1998 was an abnormally warm year. When people say things like "The last N years were N of the N+1 warmest on record", 1998 is always that +1. The trend-line for global warming only just now started to creep above 1998; it was like 2 or 3 standard deviations above the average.
If you start your trend lines at 1998, they always look flat, because you are starting at an outlier. If you take a trend line over a different period - for instance, 1990 to 2010 - the "global warming hiatus" goes away and the data looks like a normal noisy data series conforming reasonably to a linear trend.
If you see someone talking about climate data over a period of time starting at 1998, they are full of shit. They are using one of the most obvious and well-known lies-told-with-statistics you'll ever come across.
Ars Technica explains this quite well in a piece [1] published yesterday. The paper's original dataset was replaced with an updated one that was run through the same algorithms that NOAA had previously used. The updated dataset better reflects findings that other scientific bodies had released that counter the claim of a warming 'slowdown', partly because they used a richer array of sources. At the end of the day, the trend is not that different, and the trend is still one of warming. Ho-hum.
What is with the weird title there? It is written as if we should already know about the existence of such a whistleblower, or maybe it was written up before the name was known.
One possibility is he tried to whistleblow to the Washington post, but they sent it to the whistleblow targets, allowing them to develop talking points:
"He submitted an earlier, shorter version of this essay to the Washington Post, in response to the 13 December article (climate scientists frantically copying data). The WaPo rejected his op-ed, so he decided to publish at Climate Etc."
https://judithcurry.com/2017/02/04/climate-scientists-versus...
It was published a day after the Mail article was, and that so it's not unreasonable to assume that readers are aware of its content - ie, that such a whistleblower exists.
Why is his name so important to someone who would have either not heard of this, or just read an article prominently featuring the guy? It is definitely odd.
>[This] latest article is noteworthy in that it appears to reveal the supposed “whistleblower” who has been cited by the US House Science Committee in its ongoing clash with climate scientists at the National Oceanic and Atmospheric Administration (NOAA).
The Mail made Bates a subject of import, and another outlet is responding. What is odd about this, exactly?
Well, I found this. Now it looks like that part of the Ars article is "fake news":
"Yesterday, Bates said he was contacted by the Science Committee for the first time only after the story broke. He said he has not communicated with anyone there before and was not a whistleblower for the committee previously but that he expected to be invited to Washington to testify at a future hearing."
That part of the Ars article is restating a claim made in the Daily Mail article. Also, Lamar has been making the claim for awhile [1] that he has whistleblowers in the NOAA, which is where the link comes from.
>"That part of the Ars article is restating a claim made in the Daily Mail article."
Wow, this is why I ignore the news. What a clusterfuck. However, I cannot find the part you are talking about in the Daily Mail article after a search. Can you quote it?
Thanks, I didn't read that article (only the headline... I skipped to the blog post to hear from the "horses mouth") and was unaware there was an ongoing mystery whistleblower. It is all news to me.
I really don't understand the complaint in the article.
> He claimed to have developed a way to raise sea-temperature readings that had been collected by buoys: He would adjust them by using higher temperature readings of sea water collected by ships.
So actually it's due to ships being used before, that the data needs to be corrected. And they're using comparison between the buoys and the ships running at the same time. This is the same story as the sensors which were running in the cities and needed to be moved into the remote countryside in the US to account for city heat. I get that he doesn't like that the bouys readings were raised rather than ships, but relatively, why would it matter?
The linked article says:
> Now, some of those same authors have produced the pending, revised new version of the sea dataset – ERSSTv5. A draft of a document that explains the methods used to generate version 5, and which has been seen by this newspaper, indicates the new version will reverse the flaws in version 4, changing the buoy adjustments and including some satellite data and measurements from a special high-tech floating buoy network known as Argo.
After looking it up quickly, it seems that one step really was raising the buoy values by a single constant, but this was only a tiny part of a much more involved process.
"The SSTs from ships or buoys were accepted (rejected)
under a QC criterion that observed SSTs differ from the first-guess
SST from ERSST.v3b by less (more) than 4 times standard
deviation (STD) of SST (Smith and Reynolds 2003).
The ship and buoy SSTs that have passed QC were
then converted into SSTAs by subtracting the SST climatology
(1971–2000) at their in situ locations in
monthly resolution. The ship SSTA was adjusted based
on theNMATcomparators; buoy SSTA was adjusted by
a mean difference of 0.12 C between ship and buoy
observations (section 5). The ship and buoy SSTAs were
merged and bin-averaged into monthly ‘‘superobservations’’
on a 28 3 28 grid. The number of superobservations
was defined here as the count of 28 3 28 grid
boxes with valid data. The averaging of ship and buoy
SSTAs within each 28 3 28 grid box was based on their
proportions to the total number of observations. The
number of buoy observations was multiplied by a factor
of 6.8, which was determined by the ratio of randomerror
variances of ship and buoy observations (Reynolds and
Smith 1994), suggesting that buoy observations exhibit
much lower random variance than ship observations.
The SSTAs of superobservations were further
decomposed into low- and high-frequency components.
The low-frequency component was constructed by applying
a 268 3 268 spatial running mean using monthly
superobservations where the sampling ratio is larger
than 3% (five superobservations). An annual mean
SSTA was then defined with a minimum requirement of
two months of valid data. The annual mean SSTA fields
were screened and the missing SSTAs were filled by
searching the neighboring SSTAs within 108 in longitude,
68 in latitude, and 3-yr in time. The search areas
were tested using ranges of 158–208 in longitude, 58–108
in latitude, and 2–5 yr..."
>"I get that he doesn't like that the bouys readings were raised rather than ships, but relatively, why would it matter?"
I highly doubt they just raised all the buoy data by a single constant. The value probably varied over time and the "reverse" operation would not have exactly the same effects on any trends. We would have to look closer at the paper to see what was actually done though.
I think, as usual, the media has obscured rather than clarified what is going on here. If you read his own words, you will see John Bates is accusing NOAA (in particular, Tom Karl) of p-hacking. He claims they are changing the significance cutoff, choosing which data/processing to include based on statistical significance of a trend, and that this process is not reproducible.
"I questioned another co-author about why they choose to use a 90% confidence threshold for evaluating the statistical significance of surface temperature trends, instead of the standard for significance of 95% — he also expressed reluctance and did not defend the decision.
[...]
Tom Karl constantly had his ‘thumb on the scale’—in the documentation, scientific choices, and release of datasets
[...]
Tom Karl liked the maturity matrix so much, he modified the matrix categories so that he could claim a number of NCEI products were “Examples of “Gold” standard NCEI Products (Data Set Maturity Matrix Model Level 6).”
[...]
Since the version GHCNM3.X never went through any ORR, the resulting dataset was also never archived, and it is virtually impossible to replicate the result in K15.
[...]
The withholding of the operational version of this important update came in the middle of a major ENSO event, thereby depriving the public of an important source of updated information, apparently for the sole purpose of Mr. Karl using the data in his paper before making the data available to the public."
Personally, I know such statements about statistically significant deviations from a strawman nil-null hypothesis are always meaningless. However, perhaps this will open up a few more people's eyes to the awful state of affairs NHST has brought us.
lol. Difficult to take the original Mail article seriously. A look at the temperature rise graph [1] tells you a lot. Bizarrely they've started the y-axis at a value lower than the '97 baseline meaning the slope of a line through the data looks shallower than it is.
Secondly their interpretation of the graph is messed up. The Met Office data tracks the NOAA data, but looks to be lower by a constant. This means you'd expect that the slope of a regression line through each of the two data sets would be very similar.
Why did they choose 1997 as a baseline (why does anyone choose the baselines they do)? '1997' doesn't appear in the article. '2000' appears in the article. Oh look, the temperature was lower in 2000 meaning the slope would appear greater if the graph matched the discussion. [I don't know who is pulling a fast one here]
The National Review isn't a science magazine. When it says things like "And in the most Obama-esque move" and references The Cato Institute, you have to take what they are saying with a grain of salt.
The main thing that caught my eye in this article was that it using dailymail.co.uk as it's source. It's rather common to see an article originate on dailymail, then get repeated by a hundred other news-like sites. They all use that first article as their only source, and no follow up.
I started reading this, and was a little sceptical by the first paragraph, but when I saw Obama-esque and the Daily Mail references I had to check out their homepage. Definitely a bad source.
It's funny how David Rose continues to get baselines wrong despite it being explained to him hundreds of times. Even more strange is how his 'inadvertent' baseline errors always end up supporting his latest wacky conspiracy theory…
What [Rose] fails to mention is that the new NOAA results have been validated by independent data from satellites, buoys and Argo floats and that many other independent groups, including Berkeley Earth and the UK’s Met Office Hadley Centre, get effectively the same results.
It's sad some scientists go to far in manipulating data to fit their model, but that doesn't mean every scientist is crooked.
Anecdotal fallacy – using a personal experience or examples to extrapolate without a statistically significant number of cases that could form scientifically compelling evidence.
There's a wide variety of views among HN members. Reading a variety of sources, including those you may not agree with, is a good way to understand issues from different perspectives.
If you believe a submission is inappropriate for HN, flag it and move on.
That sort of depends on believing that all publications are of equal quality, or that there is no universal measure of quality in journalism and science.
While there are a range of perfectly good publications from across the political spectrum (well, probably not right across it but the fringes rarely produce well researched pieces), it is not true to say that any publication / PoV is of equal validity / newsworthiness as any other. The Daily Mail (in the UK) is a good example of a publication with |editorial value| < ε.
Sure, there's crap journalism all over the spectrum, some of it barely worthy of the name. The National Review is hardly the Daily Mail. While you may not agree with the opinions of its commentators, it's far from a tabloid. Over the years it's consistently valued and published high quality commentary and political opinion.
It misses a most important point: when people talk about a "global warming hiatus" from 1998 to 20XX, they are always full of shit.
1998 was an abnormally warm year. When people say things like "The last N years were N of the N+1 warmest on record", 1998 is always that +1. The trend-line for global warming only just now started to creep above 1998; it was like 2 or 3 standard deviations above the average.
If you start your trend lines at 1998, they always look flat, because you are starting at an outlier. If you take a trend line over a different period - for instance, 1990 to 2010 - the "global warming hiatus" goes away and the data looks like a normal noisy data series conforming reasonably to a linear trend.
If you see someone talking about climate data over a period of time starting at 1998, they are full of shit. They are using one of the most obvious and well-known lies-told-with-statistics you'll ever come across.