A review of state-of-the-art climate data tells a different story than what NOAA tells the public.
While media outlets scream “hottest ever” for the world in June and July (it’s summer) and opportunistic climate crusaders use those headlines to push the idea of a “climate crisis” the reality is for USA is that so far most of 2019 has been below normal, temperature-wise.
Little known data from the state of the art U.S. Climate Reference Network (which never seems to make it into NOAA’s monthly “state of the climate” reports) show that for the past nine months, six of them were below normal, shown in bold below.
| 201810 | -0.18°F |
| 201811 | -2.56°F |
| 201812 | 2.39°F |
| 201901 | 0.63°F |
| 201902 | -3.15°F |
| 201903 | -2.81°F |
| 201904 | 1.55°F |
| 201905 | -1.13°F |
| 201906 | -0.14°F |
Above: Table 1, U.S. average temperature anomaly from October 2018 to June 2019. Full data file here
Note the below average value for June, 2019 at -0.14°F
The data, taken directly from NOAA’s national climate data page, shows not only that much of 2019 was below average, but that the US Temperature average is actually cooler now for 2019 than we were in 2005, when the dataset started.
The U.S. Climate Reference Network (USCRN) was established to give the most accurate temperature readings compared to the old Cooperative Observer Network (COOP) which suffers from urban encroachment, siting problems, and a multitude of human induced inhomgeneities such as station moves, incomplete data, closed stations, and runway condition stations at airports that were never designed to report climate data.
Readers here know of my work to highlight these problems, and as a result, there was a 2011 report by the GAO about the problems with the old COOP network. They investigated a subset of the larger COOP network (The USHCN) and said:
According to GAO’s survey of weather forecast offices, about 42 percent of the active stations in 2010 did not meet one or more of the siting standards. With regard to management requirements, GAO found that the weather forecast offices had generally but not always met the requirements to conduct annual station inspections and to update station records. NOAA officials told GAO that it is important to annually visit stations and keep records up to date, including siting conditions, so that NOAA and other users of the data know the conditions under which they were recorded. NOAA officials identified a variety of challenges that contribute to some stations not adhering to siting standards and management requirements, including the use of temperature measuring equipment that is connected by a cable to an indoor readout device— which can require installing equipment closer to buildings than specified in the siting standards.
NOAA does not centrally track whether USHCN stations adhere to siting standards and the requirement to update station records, and it does not have an agency wide policy regarding stations that do not meet its siting standards. Performance management guidelines call for using performance information to assess program results. NOAA’s information systems, however, are not designed to centrally track whether stations in the USHCN meet its siting standards or the requirement to update station records. Without centrally available information, NOAA cannot easily measure the performance of the USHCN in meeting siting standards and management requirements.
Source: GAO-11-800 August 31, 2011, Climate Monitoring: NOAA Can Improve Management of the U.S. Historical Climatology Network https://www.gao.gov/highlights/d11800high.pdf
NOAA’s response? Rather than fix it, they closed reporting the USHCN subset of COOP stations in 2012. They now say this on their climate data page:
National USHCN monthly temperature updates have been discontinued. The official CONUS temperature record is now based upon nClimDiv. USHCN data for January 1895 to August 2014 will remain available for historical comparison.
Source: https://www.ncdc.noaa.gov/temp-and-precip/national-temperature-index/time-series?datasets%5B%5D=uscrn&datasets%5B%5D=cmbushcn¶meter=anom-tavg&time_scale=p12&begyear=2004&endyear=2019&month=12
Yet, while the USHCN was closed, and the data from it are no longer reported in the monthly and yearly NOAA climate reports, the problems identified in the USHCN persist in the larger COOP Network, of which several thousand stations remain:
All NOAA did was treat the visible symptoms we identified (by excising them), while leaving the larger disease uncured, and continuing to use the majority of data in it, data with the same sort of problems and inhomgeneities discovered in the USHCN subset. The USHCN was 1218 station out of over 8700 COOP stations, and that remaining data is used to calculate the U.S. Climate Divisional Dataset used to report “official” temperature averages today. Basically all they did was sweep the problem under the rug, and report that they have algorithms to “fix” bad data.
In any other branch of science, in the stock market, or in criminal forensics, “bad” data would be thrown out as unreliable.
Meanwhile, perfectly good data gets ignored in favor of “fixed” bad data. NOAA says this about the new state-of-the-art Climate Reference Network network:
The vision of the USCRN program is to maintain a sustainable high-quality climate observation network that 50 years from now can with the highest degree of confidence answer the question: How has the climate of the Nation changed over the past 50 years?
These stations were designed with climate science in mind. Three independent measurements of temperature and precipitation are made at each station, insuring continuity of record and maintenance of well-calibrated and highly accurate observations. The stations are placed in pristine environments expected to be free of development for many decades.
Source: https://www.ncdc.noaa.gov/data-access/land-based-station-data/land-based-datasets/us-climate-reference-network-uscrn
The data from the rest of the world, as reported by NOAA’s Global Historical Climatology Network is largely composed of the same type of stations, with an equal to or even greater (due to lack of consistent quality control) set of data problems.
Unfortunately, NOAA doesn’t seem to think the data from this state-of-the-art US Climate Reference Network is worth reporting to the public. A scan of the last 5 years of yearly and monthly “State of the Climate” reports has not a single mention of this high quality data, preferring to cite the data from the old COOP network instead, now repackaged as the U.S. Climate Divisional Dataset.
In fact, for the June 2019 State of the Climate Report, NOAA is claiming that the US was 0.2F above average in June, when in fact the US Climate Reference Network reported the June data as below average at -0.14°F
The June contiguous U.S. temperature was 68.7°F, 0.2°F above the 20th century average, ranking in the middle third of the 125-year record.
Source: https://www.ncdc.noaa.gov/sotc/national/201906
So what is the correct US Temperature for June?
In May, 2019, there was also a disparity. USCRN reported the national average temperature as below average at -1.13°F
While NOAA’s “official” climate report said it was only about half that, reading warmer by 0.43°F:
For May, the average contiguous U.S. temperature was 59.5°F, 0.7°F below the 20thcentury average and ranked in the bottom third of the 125-year record.
Source: https://www.ncdc.noaa.gov/sotc/national/201905
It seems NOAA can’t even agree on reporting what the actual temperature of the United States is on a monthly basis, using their own old and unreliable data, while neglecting to report the best data they have. Why?
That’s a travesty of government incompetence worth investigating.
One wonders if screaming headlines about “hottest ever” this month would even exist if the world had a global version of the U.S. Climate Reference Network where the data was quality controlled, and measurements taken far away from the human induced heat of urbanization.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I care not about record heat or cold where I am not living. I care about trends. Records will always be set because we have so little data (100 years of data is minuscule). It is the trend that matters, not today’s temperature.
It is likely that there is never a “flat trend” of any significant length, although using an average might produce one (such as the top portion of a curve where low temperatures on both ends are mixed with high temperatures in the middle). I believe nature is always changing it’s mind (i.e. lot’s of cycles that affect temperature).
Every city will have it’s own UHI effect, and as long as the city grows, the UHI of that city will change. So it’s pointless to try to measure a natural trend inside a city, or even to compare trends between cities. There is just too much noise to be useful.
Use well situated stations that are kept in good repair. Keep good records. And use the raw data from such stations to prevent human bias from creeping in. This will measure the natural trend.
From the article: “One wonders if screaming headlines about “hottest ever” this month would even exist if the world had a global version of the U.S. Climate Reference Network where the data was quality controlled, and measurements taken far away from the human induced heat of urbanization.”
As commiebob said above: “How about UAH” the satellite temperature data. If NOAA used UAH as the temperature chart of record, instead of the bastardized Hockey Stick they use, there wouldn’t be numerous “Hottest Year Evah!” claims from NOAA.
The reason for that is because the UAH chart shows 1998 as the second-warmest year between 1998 and 2016, so all those NOAA claims about the year 2012, or 2013, or 2014, or 2015 are all incorrect if going by the UAH chart. The only year that is warmer than 1998 subsequent to 1998, is the year 2016, which was 0.1C warmer than 1998, a statistical tie.
NOAA bastardized 1998 and turned it from the most significant, El Nino, warm year since the 1930’s, and reduced it to insignificance in order for them to be able to make these “hotter and hotter” claims. They did this bastadization of 1998 right in front of all of our eyes. I think the pause might have been one impetus for them. They somehow needed to show that things were continuing to warm, so they did what they did earlier and cooled the past to make the present look warmer. Devious little climate scientists, aren’t they.
The UAH chart puts the lie to the “hotter and hotter” meme. “Hotter and Hotter” is NOAA-generated propaganda.
Here’s the UAH chart. See how many “hottest years evah!” you can find between 1998 and 2016. The answer is none. None of those years are as warm as 1998, according to UAH. The only time NOAA could claim a “hottest year evah!” was 2016, and that by 0.1C over 1998. The current global temperatures, three years after 2016, are currently about 0.4C cooler than 2016. You can see that on the UAH chart, too.
http://www.drroyspencer.com/wp-content/uploads/UAH_LT_1979_thru_June_2019_v6.jpg
There’s a lot of fraud going on at NOAA and NASA Climate. On government time and using taxpayer money.
Basing arguments on NOAA and NASA climate data is a meaningless exercise. How can anyone have any confidence in their numbers after watching what they have been doing to manipulate the temperature record for all these years? I certainly have zero confidence in anything they put out. They have an agenda, and it’s not science, it’s politics. They are pushing a storyline.
I’m not very good with databases but it is very easy to adjust data sets, curves, trends etc. to fit reality as you see it, whatever the numbers are it will become fact after while when no one remembers (or cares) where it came from in the first place. QA? Who QA’s data even if they know and or care about it’s relevance? I’ve paid a well meaning lab (who wanted more business) a few grand to do something as simple as determine specific heat of a simple water based solution and they screwed it up, my simple guess based on a gross heat balance was closer to actual fact. So, even well meaning QA is problematic, not too mention the rest of the humans gathering the data. Gets only exponentially worse when everyone starts looking at it, Heisenberg raises his ugly head and the greediest common denominator holds sway (more taxes anyone?). The only way to know the future is to survive long enough to experience it…but…you must have a high tolerance for stupidity along the way.
Anthony, good article. Measurement station problems were what drew me to WUWT in the first place, and like you, I believe are still a major soft point for any climate discussions.
Your article could benefit from a flow chart to show the relationships of the various datasets.
Not a mathematician here, but– aren’t we missing a real basic consideration? If some reported datum is +/- some value different than the mean, isn’t the real question one of whether or not that datum is within 2 SD of the mean? Is a home run into the upper deck really better than one into the first row of seats? “Normal” is normal.
The USHCN was 1218 stations out of over 8700 COOP stations, and that remaining data is used to calculate the U.S. Climate Divisional Dataset used to report “official” temperature averages today.
…not a single mention of this high quality data, preferring to cite the data from the old COOP network instead, now repackaged as the U.S. Climate Divisional Dataset.
Which leaves me a bit confused. Is the U.S. Climate Divisional Dataset just the old USHCN data under a new name (and reorganized into geographical areas)? We at least knew where to get the locations and metadata of the USHCN stations. Who has a link to such information for the Divisional Dataset?
For whatever it’s worth, probably very little, I agree with Ananymoose’s comment above:
It would be nice if the surfacestations.org database would get out of maintenance mode.
If the Divisional Dataset is using the same stations, then the work that has been done here is still relevant.
Last summer I visited the Turnbull NWR near Cheney, WA., home of one of the three CRN sites in Washington State. Asked a NWR staff person if I could walk over to the array and take some photos, and they were not concerned. (Possibly because they knew nothing about it.) Interesting setup, certainly sited better than the AWOS surrounded by pavement at Spokane Intl Airport 17 miles north. We are all familiar with UHI effect, but I predict complaints soon about the “cool bias” at CRN locations. Yesterday (7/29/19), Turnbull CRN site had a low of 37F. Spokane International Airport’s low was 58F.
Now that’s a HUGE UHI differential
Bryan A: It is indeed huge. Not hard to see why, looking at the airport station:
http://www.desk-net.net/Spokane_AWOS-b.JPG
Windsong: I’d be interested in seeing your picture of the CRN station
Juan, I scrolled back in my phone to September and found just two. One is very similar to the one featured on noaa.gov for Turnbull NWR (TRNW1). There is a 14 mb pdf on the page linked below that has photos of all CRN locations. I also saw that tomorrow is that site’s “birthday,” in service August 1, 2007.
https://www.ncdc.noaa.gov/crn/photos.html
Thanks for the link. I’ll be driving up to Spokane from Rufus, Oregon on Saturday, and I might zip over and take a look at the CRN station, just for the fun of it.
Given that CRN is great and the other stations are crap
we have the following test.
Compare the great to the crap
Should see a difference.
nope
https://www.ncdc.noaa.gov/temp-and-precip/national-temperature-index/time-series?datasets%5B%5D=uscrn&datasets%5B%5D=climdiv¶meter=anom-tavg&time_scale=p12&begyear=2004&endyear=2019&month=12
nope
https://www.ncdc.noaa.gov/temp-and-precip/national-temperature-index/time-series?datasets%5B%5D=uscrn&datasets%5B%5D=climdiv¶meter=anom-tavg&time_scale=p12&begyear=2004&endyear=2019&month=7
Funny thing. I just finish a roject comparing all CRN to all the urban neighbors within 40km.
answer. Nothing to see.
Yes, and the start of the USCRN record happens to coincide roughly with “the Pause” that Warmists spent a decade or more denying existed, until the data became too overwhelming to ignore.
If the advent of USCRN is now keeping the serial adjusters honest, that’s a step in the right direction.
Stephen,
I think there is something to see, but I am not sure I can explain it.
Your plots are too busy to see what is going on in the comparison. Try this one instead:-
https://www.ncdc.noaa.gov/temp-and-precip/national-temperature-index/time-series?datasets%5B%5D=uscrn&datasets%5B%5D=climdiv¶meter=anom-tavg&time_scale=12mo&begyear=2004&endyear=2019&month=6
These are in effect 12-month trailing averages to June posted for each year from 2006 onwards. The excellent match of the early years 2006 to 2013 suggest that there is no systematic problem with choice of monthly baseline for anomaly calculation. (This can be readily confirmed by running out each of the plots for one month averages of the individual calendar months.)
However, the graph shows the appearance after 2014 of a significant (systematic?) difference amounting to about 0.2 deg F in the annual averages – with ClimDiv < CRN. This seems like the wrong way round to support Anthony's narrative, but I am puzzled that it is there at all.
Exactly! I posted this problem below, before reading yours.
It serves to illustrate that in climate pseudo-science ‘bad’ data are in fact the good data and ‘good, reliable’ data are the rotten ones. You just call them, the good data, ‘discordant’ as Mann did with the thermometer record and leave it out.
Three different ‘official climate science’ temperature anomalies published, showing differing negative and positive anomaly values, unequivocally demonstrate there is no ‘settled science’ to be had from the fraudulent climate change industry. And the usual climate alarmist trolls are offering flaccid equivocations to evade the blunt and unarguable failure of their much ballyhooed fraud. Pathetic – all the way ’round.
Enjoy the beautiful and highly variable summer weather, folks! Winter, like the next phase of continent spanning glaciers, is certainly coming….
“The data, taken directly from NOAA’s national climate data page, shows not only that much of 2019 was below average, but that the US Temperature average is actually cooler now for 2019 than we were in 2005, when the dataset started”
Well yes, but so does the ClimDiv data set. In fact, the USCRN data set warmed up by about 0.2°F vs. the ClimDiv data set. So the USCRN shows even more warming (or less cooling) since 2005.
Anthony– Can you work with Dr Spencer to obtain the monthly CONUSA satellite temperature data for 2019 to compare it with the anomalies from USCRN?
https://www.bbc.co.uk/news/science-environment-49167797
I’m not sure what your point is but…………..
https://notalotofpeopleknowthat.wordpress.com/2019/07/31/uks-ten-warmest-years-what-the-bbc-forgot-to-tell-you/#more-40629
All the bits the BBC/MO left out.
somewhat off topic:
Oscillations of the baseline of solar
magnetic feld and solar irradiance on a millennial timescale (V. V. Zharkova, S. J. shepherd, S. I. Zharkov & E. Popova) Received: 11 January 2019, Accepted: 4 June 2019 Published: 26 06 2019
Whats about peer review ??
Worth to discuss ?
I live by the Atlantic Ocean. My car’s reading of the outside temperature declined by 8 degrees in the 10 minutes it took me to drive from a point several miles inland.
Good to know I am not the only one who notices these kinds of discrepancies.
I think it might also be worth discussing topics similar to ones brought forth by research papers like these: https://www.tucson.ars.ag.gov/unit/publications/PDFfiles/2538.pdf
The paper finds that rainfall has intensified very slightly over the past 59 years and of course the explanation is climate change (in their defense anthropogenic/man-made climate change wasn’t actually mentioned in the paper).
While their data shows an increase in rainfall it’s barely significant and certainly not anything to be alarmed about in my opinion especially living in the region. Page 3 figure 1c and 1d kind of sum up the reality of the situation. The dataset starts in 1961 with around 10mm of precipitation over the 11 year average and by 2017 there is around 30mm of rainfall over an 11 year average but the introduction and conclusions that climate change and temperature are directly linked to more precipitation ignores the 30mm rainfall average in the mid 1980’s when temperatures were apparently below the anomaly.
I lived in Phoenix at the turn of the millennium and apparently at this site some 80 miles southeast of Phoenix they were recording below average rainfall… But not in Phoenix as I recall!
As mentioned Figure 1d shows that annual and average temperatures were well below the temperature anomaly in the mid 1960’s through the mid 1970’s and yet rainfall in this region was much higher than normal during the same period of time which kind of obliterates their assumptions that higher average temperatures lead to more rain. On top of that total precipitation in 1966 comes up just about 2 inches shy of the 2015 record on the graph. In fact just by looking at those graphs it appears that higher annual rainfall totals only recently coincide with higher annual temperatures since around the year 2010. Even in the early 80’s there was a huge spike in total precipitation but the temperature was still clearly below the temperature anomaly according to the data and the averages bare that out as well.
Germane to this discussion is R. Geiger’s book “The Climate Near the Ground” It is available in the internet. Look for climatenearthegr032657mpb.pdf
The book is a superb text about climate measurements above and below the ground. The illustrations are informative and imaginative. The prose is exquisite and wonderfully crafted. I dug it up after seeing Dr Jim Steele recommend it.
If I was king, I’d make the climate whoppers read it out loud to each other. And then make them repeat the measurements. And then discuss the measurements. In coherent English.
DL link at archive.org</a<
Hear in Australia the Golden rule about temp stations is they must be in a Stephenson screen and no reforestation or deforestation can occur within 200m of the site. So what did the BOM do here in Melbourne? Moved the site to skyscraper city and right opposite a river. Jeez you couldn’t even make this stuff up. Maybe they thought one would cancel out the other.
With the media constantly chirping about Heat! Hot! Hottest EVAH!, don’t most people think to themselves, realize the weather changes quite a bit, and go back to looking for the baseball scores?
These endless debates about temperature data contribute what to World Peace and Prosperity?
Medieval Warm Period, Little Ice Age, The Pause,
This is the Information Age. Trained engineers who understand the rules of Data realize this is nothing more than Sturm and Drang.
Meaningless. You should all just stop, try to remember the rules of Significant Digits from your undergrad education. Except for Mosher who clearly has never studied this.
If is is hotter than it used to be, most people would notice this. It is not….
Of course: when processing temperature time series, I’m a simple layman compared with persons working within institutions like NOAA or… BEST!
Nevertheless I made a test based as usual on my preferred source, the GHCN daily data set (because it is really raw):
– extract, out of the entire data set, all stations located within CONUS (i.e. US minus Alaska and Hawai);
– extract, out of the CONUS data subset, all stations flagged ‘CRN’;
– compare the two subsets’ series within their common period (I started with 2004), with 2009-2018 as reference period for anomaly computation.
Here is the result:
https://drive.google.com/file/d/1zg9M-GZwNoIBln404Ay0voAL8V4PmSdK/view
This is a comparison of (in the mean for 2004-2019)
– 130 CRN stations located within 70 cells of a 2.5 ° grid
with
– 6500 CONUS stations located within 170 grid cells.
The linear estimates for 2004-2019, in °C / decade
– CRN: 0.36 ± 0.16
– CONUS: 0.27 ± 0.15
The reason why the CRN set average shows a higher trend than all CONUS stations you easily can see when looking at their respective running means: CRN was cooler in the early phase, and warmer in the later phase of the series.
The greatest CONUS difference above CRN was +0.68 °C, below CRN -0.62 °C.
The average difference was no more than +0.02 °C!
Please don’t wonder when comparing this graph with those shown by Steven Mosher, like this one:
https://tinyurl.com/yxwm2w5s
because station sets, baseline and anomaly computation differ.
For example, nothing similar to e.g. Pairwise Homogenisation algorithm was used here; all stations which did not provide for sufficient baseline data were automatically rejected.
*
I keep wondering about the similarity between the two time series, despite the station set difference!
That reminds me words written by Steven Mosher years ago below a WUWT guest post, something like “The planet is oversampled”. He was really right!
How else could we explain that about 100 stations give results so similar to those provided by over 6000? The primary condition of course is to have a well balanced spatial distribution over the country you want to observe.
*
To be honest, I don’t understand why Anthony gets raging about differences
– between a minuscule set of CRN stations and a huge set of anonymous stations, no matter where they come from;
– arbitrarily picked out of such a small time period like Oct 2018 – Jun 2019!
Regards
J.-P. D.
Source: ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/daily/
I was wondering if you plotted the absolute temperature trends for the USCRN stations. When doing that on the NOAA website, the cooling trend is pervasive from 2005-2019 whereas using the older data from 1979-2004, the warming trend is in line with the MSM.
Apparently the summers are warming slower in an overall US warming trend (a moderating effect on summer highs and winter lows), probably mediated in some respect by moisture content. This suggests that perhaps land use changes started in the 1930s are more important to US outcomes than average temperatures (e.g. 1936 was not an unusual year by average), or that globally warmer air just carries more moisture (e.g. the conditions at the Last Glacial Maximum are known to have been extremely arid/dusty by today’s standards), or some combination of the two.
Sadly, all of the comparisons made here of USCRN data, which begin only in 2005, with much-longer time series overlook an essential point: the major discrepancies between pristine and UHI-corrupted locations developed in the the course of the 20th century. All the major changes in urbanization, transportation and mechanization took place during that period. That the discrepancies in common-base anomalies prove to be minor so far the 21st century cannot be taken as valid indication of similar agreement throughout earlier history.
1sky1
“… overlook an essential point: the major discrepancies between pristine and UHI-corrupted locations developed in the the course of the 20th century.”
First of all, your elegant comment is here out of topic: the discussion is here restricted to CRN vs. ‘crap’.
But should you be able to do that: please communicate a list of 100 of your ‘pristine’ CONUS stations having been active during the last century!
They should be present in the GHCN daily record, so the same job can be done again, exactly like this one with 46 USHCN stations considered ‘pristine’ around 2011/12 by ‘surfacestations.org’:
https://drive.google.com/file/d/1nfrk9qylSAyYA1yNilPXrC6E3Sul5rVA/view
https://drive.google.com/file/d/1L5oQowWa2O7n2GRMvRRjQyMt5Oe8Iqpf/view
As you easily can see, your UHI-based point does not fit very well to real observation.
You seem to have problems with logic. My essential point is that 21st century data comparisons cannot be projected backwards. And what others may consider “pristine” is not the basis I use for vetting century-long records. Having invested considerable effort in that vetting process, I’m not inclined to gratify the desire of an uncomprehending troll to eat the fruit of that work. Without further ado, here’s the discrepancy between UHI-corrupted and relatively uncorrupted CONUS records over the 20th century:
http://s1188.photobucket.com/user/skygram/media/Publication1.jpg.html?o=0
Note that how discrepancy flattens going into the 21st century.
Just to the east of St. Paul MN. I’ve turned on the air conditioning for 4 hours this year.
This week, morning temperatures are around 58 F. With good insulation a couple of fans cool the house at night and the house is okay through the day.
This is a cool summer in MN. If any NASA map show MN hot though August 1, they are lying and you should never again believe NASA for anything. If they lie at this, even their aeronautical data is suspect and therefore worthless. In Latin: Falsus in uno, falsus in omnibus.
Where do you go with things when the number cruncher s don’t tell the truth ?
NASA has no credibility in the weather fabrication industry .
Stick to space or …don’t bother if you are willing to politicize climate .
Trump still has a lot of cleaning out to do .
My utility company reported all but one anywhere from 4-11 degrees cooler than for the same time last year.