NOAA's odd way of presenting February temperatures

Harold Ambler writes:

NOAA's map of February temperatures across the United States got New England all wrong. It wasn't "near normal," at all, as the people of the region can well attest. Oh, and the data, too: Hartford, CT, as an example was actually 5.1 degrees below normal.

NOAA’s map of February temperatures across the United States got New England all wrong. It wasn’t “near normal,” at all, as the people of the region can well attest. Oh, and the data, too: Hartford, CT, as an example was actually 5.1 degrees below normal.

As the map above shows, NOAA seems to have struggled in creating a temperature map that accurately conveys what New Englanders recently experienced: a frigid February. Hartford was 5.1 degrees below normal for the month; Boston was 3.1 degrees below normal. Providence was 4 degrees below normal for the month. And yet all three locations fall within the “near normal” portion of NOAA’s map. What’s up with that?

How well did NOAA do representing February temperatures where you live?

Read more here:

NOAA map of February temperatures less than accurate

The climate data they don't want you to find — free, to your inbox.
Join readers who get 5–8 new articles daily — no algorithms, no shadow bans.
0 0 votes
Article Rating
162 Comments
Inline Feedbacks
View all comments
Patrick (the other one)
March 19, 2014 9:44 am

Confirming Albany, NY as a 4 degree (F) departure below the “normal” temperatures provided on Accuweather for February.

March 19, 2014 9:53 am

izen says March 19, 2014 at 8:57 am

Nothing to do with any flaw, intentional or otherwise in the graphics.

Hi Sara. How goes it at the NCDC?
Do you ever take into account how different persons with differing color perception might not be able to see or view your ‘artwork’ correctly on account of the color choices you make?
Have you ever heard of the ADA? I think your activities might just fall within the intent of that act … (and if it doesn’t prescribe in the law for that I will contact my congressman and work towards that goal.)
UPDATE – Oh, look what I just found:

Website Accessibility Under Title II of the ADA
In this chapter, you will learn how the nondiscrimination requirements of Title II of the ADA apply to state and local government websites.1 Chapter 5 answers the following questions:
Which provisions of Title II of the ADA apply to websites?
What technologies do people with disabilities use to access the Internet?
How do poorly designed websites pose barriers to people with disabilities?

http://www.ada.gov/pcatoolkit/chap5toolkit.htm
– – – – – – –
Bolding above mine.
.

March 19, 2014 10:14 am

For those few doing few design/graphics interested in seeing their work interpreted correctly (if at all in some cases!) by a group of people (estimated to be at least 10 million American males) afflicted by color perception difficulties (aka color blindness) from birth, take a moment and review this subject:
Goog search – Website guidelines for color blindness
This search returns, for instance, these sites:
Developing & Testing for Color-Blindness
http://www.epa.gov/inter508/developers/colorblind.htm
Can Color-Blind Users See Your Site?
http://msdn.microsoft.com/en-us/library/bb263953(v=vs.85).aspx
Visual Disabilities – Color-blindness
http://webaim.org/articles/visual/colorblind
.
A test – Quick check for color blindness:
. . Color vision test
. . (caveat: Refer to your eye doc/med. prof. for 100% accurate results and advice)
.

mike g
March 19, 2014 10:15 am

Stokes
Changing to new systems seems to account for just about all the global warming that has been claimed over the years.

woodNfish
March 19, 2014 10:19 am

The only thing the federal mafia knows how to do is lie. Believing anything the government says is a fool’s errand.

March 19, 2014 10:22 am

re: JohnWr says March 19, 2014 at 9:05 am
The ducks are finding it a bit cold: …
Bah!
Ducks JUST have short memories! (quoting what the ‘Winston Smiths’ from the Ministry of Trvth Climate Sci Dept. might proffer for consideration …)
.

March 19, 2014 10:44 am

maybe someone has said this but I’ll say it again. How does the temp.
know where the state bounders are? If a person wanted to make a map
with cred. the colored areas would look like blobs, ovals, funny shapes <?"?…..

Jake J
March 19, 2014 10:52 am

Curtis, interesting stuff. A couple responses.
1. Gov’t weather and climate data is scattered all over the place, and not easy to find.
2. The “divisional” anomalies in New England, anyway, simply do not match the NWS readings for the cities in those regions. Can you explain why not?
3. It would seem, if rankings are the basis for the characterizations of “much above,” “much below,” etc., that the Southwest has much less average winter temperature variation than New England.

Weather Dave
March 19, 2014 11:24 am

Since we all know that the surface temperature record is ‘adjusted’ a rather simple, but tedious and informative check is the local newspapers and their archives. Usually these temperatures are from the local met office well before the data is ‘manipulated’. Yes, it’s time consuming but telling.

Jake J
March 19, 2014 11:28 am

@Weather Dave, some examples would help.

March 19, 2014 12:02 pm

re: Jake J says March 19, 2014 at 11:28 am
@Weather Dave, some examples would help.
I don’t intend to speak for WxDave, but, I thought I would ask if you had seen or are familiar with this post from 2012:
NOAA’s National Climatic Data Center caught cooling the past –
modern processed records don’t match paper records
http://wattsupwiththat.com/2012/06/06/noaas-national-climatic-data-center-caught-cooling-the-past-modern-processed-records-dont-match-paper-records/
It begins thusly:

We’ve seen examples time and again of the cooling of the past via homogenization that goes on with GISS, HadCRUT, and other temperature data sets. By cooling the data from the past, the trend/slope of the temperature for the last 100 years increases.
This time, the realization comes from an unlikely source, Dr. Jeff Masters of Weather Underground via contributor Christopher C. Burt. An excerpt of the story is below:

and then proceeds further on to point to the article upon which the post on WUWT is based:

Inconsistencies in NCDC Historical Temperature Analysis
http://www.wunderground.com/blog/weatherhistorian/comment.html?entrynum=75

Which in part says:

Ken looked at entire years of data from the 1920s and 1930s for numerous different states and found that this ‘cooling’ of the old data was fairly consistent across the board. In fact he produced some charts showing such. Here is an example for the entire year of 1934 for Arizona …

– – – – – – – – – – – –
Just some ‘food for thought’, Jake.
.

rogerknights
March 19, 2014 12:04 pm

The much above average orange has more visual impact than the record coldest blue.

March 19, 2014 12:20 pm

re: Alfred Alexander says March 19, 2014 at 10:44 am
maybe someone has said this but I’ll say it again. How does the temp. know where the state bounders are?
Alfred, the temps appear to ‘graded’ by areas called (or defined by the NCDC as) “U.S. Climate Divisions” which roughly follow state boundaries but further divide a state (presumably by local ‘climate’) as shown on the map below:
http://www.ncdc.noaa.gov/monitoring-references/maps/us-climate-divisions.php
.

rogerknights
March 19, 2014 12:22 pm

izen says:
March 19, 2014 at 8:03 am
the colder days this last winter were not exceptional compared to the coldest days ove3r the last century, in the early part there were colder winters.
it was cold compared to the last thirty years – and peoples’ memories. Perhaps you have to be over 60 to be aware that this cold winter was about average for fifty years ago, but i8s exceptionally cold compared to the last few decades.
those that think this winter was ‘excedptionally’ cold are being misled by the recent rapid warming and insufficiently long memories to noticed the climate shift.
The disparity in NOAAS catagories is because they are comparing to a much longer baseline than the short 30year one used by weathermen and local news.

Isn’t 30 years a period also used by climatologists?
If 2013 had been a hot year in the US, I suspect NOAA would have used a 30-year base period, to maximize the number of record highs (avoiding the dirty thirties).

Joseph Bastardi
March 19, 2014 1:02 pm

I posted on weatherbell.com on this matter last night and may reprise it in my patriot post blog this week ( minus my snide comments ) for the good of the general public.

Joseph Bastardi
March 19, 2014 1:10 pm

“it was cold compared to the last thirty years – and peoples’ memories. Perhaps you have to be over 60 to be aware that this cold winter was about average for fifty years ago, ”
that is nonsense. The winters of the early 60s were cold, but not this cold as you can see if you simply put them against the recent 30 year means
winters 1960-1961 through 64-65
http://www.esrl.noaa.gov/psd/cgi-bin/data/composites/comp.pl?var=Air+Temperature&level=Surface&mon1=11&mon2=1&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&ipos%5B1%5D=1961&ipos%5B2%5D=1965&ineg%5B1%5D=&ineg%5B2%5D=&timefile0=&tstype=0&timefile1=&value=&typeval=1&compval=1&lag=0&labelc=Color&labels=Shaded&type=2&scale=100&labelcon=1&switch=0&cint=.5&lowr=-10&highr=10&proj=USA&xlat1=0&xlat2=90&xlon1=0&xlon2=360&custproj=Northern+Hemisphere+Polar+Stereographic&level1=1000mb&level2=10mb&Submit=Create+Plot
this year was colder than those, and next year will be also with the modiki enso after the prolonged cold enso ( similar to 09-10)
http://www.esrl.noaa.gov/psd/cgi-bin/data/composites/comp.pl?var=Air+Temperature&level=Surface&mon1=11&mon2=1&iy=2014&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&iy=&ipos%5B1%5D=&ipos%5B2%5D=&ineg%5B1%5D=&ineg%5B2%5D=&timefile0=&tstype=0&timefile1=&value=&typeval=1&compval=1&lag=0&labelc=Color&labels=Shaded&type=2&scale=100&labelcon=1&switch=0&cint=.5&lowr=-10&highr=10&proj=USA&xlat1=0&xlat2=90&xlon1=0&xlon2=360&custproj=Northern+Hemisphere+Polar+Stereographic&level1=1000mb&level2=10mb&Submit=Create+Plot
since the pdo flip in 07 winters are growing colder . The winters of the early 60s and indeed through the 70s were during the cold PDO, WHICH IS OUR POINT, THAT OUR WINTERS WILL TURN COLDER. WAIT TILL THE AMO FLIPS. and they are cold in tandem… LIKE THE 60S AND 70S

JP
March 19, 2014 1:46 pm

Bastardi,
If am not mistaken, the Winter of 1961-62 (Northern Hemisphere) was the coldest winter on a global scale for the 20th Century. And I think the winters of 46-47 and 41-42 were also in the top 10 for the Century.
My personal memory can still recall the bitter winters of 75-76 and 77-78. The winter of 75-76 was especially cold, as the frigid temps arrived in late November. However, I can never recall this can of cold lasting past mid Feb. Normally, even after a bitter late Dec through mid Jan, a 2-3 week thaw sets in through most of the country.
Finally, I also remember the drought/heatwaves of the mid to late 1970s. The summer of 77 was especially nasty. I was spending it with my sister in Boston, and I can still remember the heat induced black-outs up and down the East Coast.

March 19, 2014 1:49 pm

By coming up with a new system and/or putting the numbers into a new system it will be more difficult for us “Joe-six-packs” to use thing like TheWayBackMachine (http://archive.org/web/web.php) to check and see how past numbers have been changed.

JP
March 19, 2014 1:50 pm

Gieger,
But, NOAA and NASA use means from the late 20th Century. What Tamino admits to is a moving of goal posts. Many years ago I advocated using a 100 year mean instead of the 30 year mean that is so popular. You cannot have it both ways. Well, actually, the Alarmists do.

March 19, 2014 1:55 pm

Up where I live (right in that good old white “near average”) we ran 5-20° below normal for all but a few days in February (3, I think). The furnace has been running nearly continually since late October, & even the older people round here are agog at thier heating bills (well, there’s also that whole “Under my plan, energy prices will necessarily skyrocket.” thing that a certain light-bringer once promised, but the actual physical amount of fuelstuffs consumed is quite a lot above average this Winter).
But the NOAA can surely tell much better from their $1800 executive chairs in Washington DC than us yobs out in it every day.

March 19, 2014 2:47 pm

Theo Goodwin
March 19, 2014 at 7:48 am
When you are criticizing someone, technical details matter. NOAA has a defined methodology that they use in constructing these maps, and have used several years. Are you suggesting that they should have changed their methodology just because you think the US had a cold winter? If not, if their defined and pre-existing methodology classifies a temperature as “Near normal” or “Near average” (although the term “near median” would be better), then that is how they should portray it on the map.
Turning to your specific example, south West Virginia is split among three divisions, South Western, Central, and Southern. Central extends south almost to the south border of West Virginia, but also north to the Northern Border. They had rankings of 35, 43, and 51 respectively. Had the central division been three rankings lower, it would have been classified as “below normal”, while the southern division was below median but definitely near to it.
Switching to anomalies with respect to 1901-2000 (the anomaly interval used on the map), the divisions had anomalies of, respectively, -2.0 F, -1.4 F, and -0.4 F. That certainly seems consistent with the classification. The anomaly with respect to the current “climate normal” (1981-2010), ie, the 30 year period used for baselines as per WMO recommendations, are -3.1 F, -2.8 F, and -1.6 F.
For the southern division, that compares to anomalies compared to the current climate normal of plus 7.0 F in 1927 (rank 120), and -11.8 in 1895 (rank 1), making it far closer to the warmest February than to the coolest. As recently as 1978 the February average has been -11.6 F (rank 2), and as recently as 2007, it has been -8.0 F (rank 13), so you don’t even have the XKCD excuse.

March 19, 2014 3:15 pm

Jake J says:
March 19, 2014 at 10:52 am
1) If you are going to write a blog on how unbelievable the NOAA rankings map is, you have an obligation to find that data. It isn’t that hard. You go to NOAA’s home page, click on “Climate Information”, then “Climate of the US”, then “Climate at a Glance”. By default, that brings you to the time series page, which shows the data as both a graph and a table. You can then press on maping to see the data for a single month or year laid out on a map of the contiguous USA. It took me less than 15 minutes to find, and I was not even writing a blog.
2) The NWS figures use a 1981-2010 baseline. In the case of the Massachusetts coastal division, that increases the anomaly figures by 2.7 F relative to the 1901-2000 baseline used on the Climate at a Glance map. That is, in the Massachusetts coastal division, February temperatures from 1981-2010 averaged 2.7 F greater than they did over the period 1901-2000. That makes the temperatures seem much colder to us, with our recent perspective than they actually are compared to the centenial average, let alone the 1895-2014 interval used for the rankings. For comparison, the anomalies are:
1981-2010: -2.3 F
1901-2000: 0.4 F
1895-2014: 0.2 F
I don’t think anybody would be inclined to argue that a temperature 0.2 F greater than the mean is not “near average”. Our perspective is distorted, however, from recent (1981-2010) warm temperatures.
3) The North East probably does have more climate variability than the South West, in that climate variability, on average, increases as you move away from the equator towards the poles. The NOAA map does not show that, however. Rather it shows that the South West was warm this year while the North East was cool relative to recent experience. The US is a big place, and does not have the same weather everywhere.

March 19, 2014 3:22 pm

Some people have complained about the relative obscurity of the light blue “below normal” areas on the map. That obscurity is likely a function of the monitor they use. On my monitor, colours near that colour tend to be washed out to white when viewed directly, but are very clear when viewed at an indirect angle. I do not know how wide spread that feature is. If it is, NOAA should not use that colour in graphics. However, it is an oddity of my current monitor in my experience, so not something you would automatically think of. (I did not notice it until I discovered error margins often printed in that colour, and became puzzled as to which error margins people were referring to.) If you are experiencing that problem, I certainly recommend contacting NOAA to let them know about the problem – but I doubt anything can be read into their not having noticed so obscure an issue.

March 19, 2014 3:49 pm

Tom Curtis says March 19, 2014 at 3:22 pm
Some people have complained about the relative obscurity of the light blue “below normal” areas on the map. That obscurity is likely a function of the monitor they use.

Are you color blind like (statistically) more than a few of us are? Few ppl take any pains to assure their graphics are viewable by this small percentage of color-vision impaired. Your response comes off as a little more then arrogant on this point, BTW.
Of course, you didn’t see this posting made earlier on the subject of considering allowances for the color-vision impaired:
http://wattsupwiththat.com/2014/03/18/noaas-odd-way-of-presenting-feb-temperatures/#comment-1593902
Anybody competent creating material on a government website should already be informed on this subject, especially some like Sara Veasey whose specialty and responsibility in the department is shown on the official org chart as ‘Graphics’. Why are we paying these specialists if they fail repeatedly in their positions and responsibilities? (preceding sentence was rhetorical)
This is aside from the ‘artful’ choice of colors chosen by anyone at NCDC to deceive/mislead any member of the public who may casually view these graphics.
.

March 19, 2014 5:03 pm

jim, I took the colour blindness test you linked to. The result: “normal answers: 31/31”. So, I would hazard a guess that just as in my youth, I am not colour blind.
The feature that I described is a real fact about my monitor. The colour shown depends on the angle of viewing, which results in light blue colours becoming washed out, or disappearing when viewed directly rather than obliquely. Consequently, the light blue “below normal” areas on the map are very faint, or invisible (depending on the version of the map) when I look at the monitor directly. That is not a result of colour blindness, and it is arrogant of you to assume so.
It is even more arrogant of you to assume that the choice of colour was “artfull” and done with intent to “deceive/mislead”. It is also downright obnoxious.
As to my own supposed arrogance. I described a problem with my monitor, and suggested it might be more widespread but specifically disavowed knowledge of how wide spread the problem is. I’m happy to concede that for some people the problem may be colour blindness, and that NOAA should provide versions of their maps easily readable by colour blind people. Of course, they have done so in the map that actually lists rankings by division.