January 2017 Projected Temperature Anomalies from NCEP/NCAR Data

Guest Post By Walter Dnes

In continuation of my Temperature Anomaly projections, the following are my February projections, as well as last month’s projections for January, to see how well they fared.

Data Set Projected Actual Delta
HadCRUT4 2017/01 +0.736
HadCRUT4 2017/02 +0.817 (incomplete data)
GISS 2017/01 +0.95 +0.92 -0.03
GISS 2017/02 +1.02
UAHv6 2017/01 +0.462 +0.300 -0.162
UAHv6 2017/02 +0.544
RSS 2017/01 +0.530 +0.409 -0.121
RSS 2017/02 +0.606
NCEI 2017/01 +0.9011 +0.8817 -0.0194
NCEI 2017/02 +0.9849

The Data Sources

The latest data can be obtained from the following sources

Miscellaneous Notes

A review of the past year of data indicates that a slightly better correlation between NCEP/NCAR and 4 of the 5 data sets can be achieved by sliding the monthly window back one day; e.g. for February’s projection, use January 31st through February 27th NCEP/NCAR data. This is probably due to the fact that NCEP/NCAR uses an arbitrary 0000 UTC day boundary, while in real life, daily temperatures follow local standard time around the globe. Another advantage of sliding back one day is that NCEP/NCAR data runs 2 days behind. This means that February 26th data is used as a guesstimate for only the 27th, rather than the 27th and 28th. Effective this month, my projections will switch to a window trailing one day behind the calendar month. E.g. for February, I’ll be using January 31st through February 26th data, and assuming that February 27th’s anomaly is equal to February 26th’s anomaly.

At the time of posting 4 of the 5 monthly data sets were available through January 2017. HadCRUT4 is available through December 2016. The NCEP/NCAR re-analysis data runs 2 days behind real-time. Therefore, real daily data through February 26th is used, and the 27th is assumed to have the same anomaly as the 26th.

The projections are derived from the previous 12 months of NCEP/NCAR anomalies compared to the same months’ anomalies for each of the 5 data sets. For each of the 5 data sets, the slope() value (“m”) and the intercept() value (“b”) are calculated. Using the current month’s NCEP/NCAR anomaly as “x”, the numbers are plugged into the high-school linear equation “y = mx + b”, and “y” is the answer for the specific data set. The entire globe’s data is used for HadCRUT, GISS, and NCEI. For RSS and UAH, subsets of global data are used, to match the latitude coverage provided by the satellites.

February’s NCEP/NCAR anomaly has risen to the highest level since April 2016. The relatively warm January has made any potential long pause more difficult to achieve. Even the “short pause” isn’t that great. My post for January had the “pause” in the daily NCEP/NCAR data going back to July 7th 2015 https://wattsupwiththat.files.wordpress.com/2017/01/daily.png but this month it has retreated to July 11th 2015.

The graph immediately below is a plot of recent NCEP/NCAR daily anomalies, versus 1994-2013 base, similar to Nick Stokes’ web page. The second graph is a monthly version, going back to 1997. The trendlines are as follows…

  • Black – The longest line with a negative slope in the daily graph goes back to early July, 2015, as noted in the graph legend. On the monthly graph, it’s August 2015. This is near the start of the El Nino, and nothing to write home about. Reaching back to 2005 or earlier would be a good start.
  • Green – This is the trendline from a local minimum in the slope around late 2004, early 2005. To even BEGIN to work on a “pause back to 2005”, the anomaly has to drop below the green line.
  • Pink – This is the trendline from a local minimum in the slope from mid-2001. Again, the anomaly needs to drop below this line to start working back to a pause to that date.
  • Red – The trendline back to a local minimum in the slope from late 1997. Again, the anomaly needs to drop below this line to start working back to a pause to that date.

NCEP/NCAR Daily Anomalies:


NCEP/NCAR Monthly Anomalies:


0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
February 28, 2017 7:01 pm

OOPS! Title should read…
“February 2017 Projected Temperature Anomalies from NCEP/NCAR Data”

Reply to  Walter Dnes
February 28, 2017 7:05 pm

What’s a month between friends, all is forgiven.

Reply to  Walter Dnes
February 28, 2017 8:28 pm

I’m trying to get a good estimate of the monthly numbers for the 5 datasets, before they’re officially released. UAH usually comes out a day or 2 after month-end, and RSS in a week or so.GISS and NCEI usually come out in approximately 3 weeks, and HadCRUT comes out after 4 or 5 weeks, so it’s interesting to have an idea of what’s happining, right away.

February 28, 2017 7:42 pm

I know how Learch felt. It’s like opening the door to some unpleasant thing. Ugggg. Ever month this **** comes out. What have they done Now?

February 28, 2017 7:51 pm

Tropical Lower Tropospheric (LT) temperature as measured by UAH in Jan2017 is ‘way down.

Usually Global LT temperature follows Tropical LT temperature by about one month. Wait and see if there is a big drop in Global LT temperature in Feb2017.

Reply to  Allan M.R. MacRae
February 28, 2017 8:57 pm

I think LT will rise, but not as much as Walter expects. The surface estimates look reasonable, though.

Reply to  Nick Stokes
March 2, 2017 6:23 am

UAH global 0.35 for February. Still following the NCEP/NCAR data in terms of consecutive warmer/cooler months. UAH diverging quite strongly now from the monthly ENSO3.4 + ~4 month lag formula described above.

Reply to  Allan M.R. MacRae
March 1, 2017 3:06 am

For clarity, my above comment refers to UAH Lower Tropospheric (LT) global temperature, which “should” drop to a low of ~0.0C in February 2017 (based on a Nino3.4 Index Anomaly of -0.73C in Oct2016, and then should increase thereafter for several months.

My equation is, which is fairly good, but not perfect (see plot below) is::
UAHLTcalc Global (Anom. in degC, ~four months later) = 0.20*Nino3.4IndexAnom + 0.15


Here is the plot of my equation; note the cooling impact of major volcanoes:

I have no opinion on Surface Temperatures – too many “adjustments”.

Reply to  Allan M.R. MacRae
March 1, 2017 6:08 am


Your formula suggests there should have been a month-on-month decline in UAH starting March 2016 onwards, matching the continuous decline in ENSO3.4 starting November 2015. In fact there have been several month-on-month rises in UAH since March 2016. Particularly since June 2016 your formula and UAH diverge quite widely.

Since Jan 2016 UAH has been much more closely correlated with the NCEP/NCAR data discussed above. In each month since Jan 2015 UAH has risen and fallen in tandem with the month-on-month rises and falls in the NCEP/NCAR data; albeit with smaller degrees of magnitude in more recent months.

For this reason it seems more likely to me that Feb 2017 will see a rise on Jan 2017 in UAH, because Feb 2017 is the most anomalously warm month in the NCEP/NCAR index since April 2016. Maybe somewhere in the high 0.3 or low 0.4 areas. We’ll find out in the next day or so.

tony mcleod
Reply to  Allan M.R. MacRae
March 1, 2017 4:37 pm

Alan, what is the difference between the evidence of the: “cooling impact of major volcanoes”, versus the evidence the warming trend is the impact of AGH?

Reply to  Allan M.R. MacRae
March 2, 2017 12:14 pm

I’ve got a similar model that was predicting 0.032 instead of your 0.0. Actual is 0.35. Not good. Record model divergence. 3 of last 4 months have diverged massively. It may be time to transition to explaining the Step soon and why globe doesn’t fully cool after El Nino’s.

Reply to  Allan M.R. MacRae
March 3, 2017 4:45 am

Well, my model clearly did not work in Feb2017 as is usually does – see the chart above.

Note I wrote “My equation is, which is fairly good, but not perfect” above.

I am curious to see what happens in March – up or down?

Tony, you asked:
“Alan, what is the difference between the evidence of the: “cooling impact of major volcanoes”, versus the evidence the warming trend is the impact of AGH?”

Presume AGH – Anthropogenic Global Heating

The evidence is in my plotted above graph – note how the actual temperature is significantly below the model temperature, and slowly recovers after the two major volcanoes.

February 28, 2017 8:56 pm

Above the “pause” and the trend line, eh? How hot would it have to get to convincingly show that the world is warming? On the chart shown the range is > 1 degree in 20 years. Another degree in 20 more years won’t matter? 2 in 40 more? 5 in 100?

Reply to  Afterthought
February 28, 2017 10:08 pm

What chart shows “> 1 degree in 20 years”? The one showing trend lines much less? And please consider that the beginning of 1997 was somewhat on the cool side.

Here is HadCRUT4 cherrypicked for start date (1974) to show rapid warming afterwards:

The cheerypicked-for-maximizing linear trend that includes the recent El Nino spike is about .795 degree in 42.75 years, or about .186 degree/decade. A higher trend can be found starting with 1974 and ending with 2004.75 (a late but arguable start date of “The Pause” / “The Slowdown”), at .207 degree/decade.

Another interesting plot: http://woodfortrees.org/graph/hadcrut4gl/from:2002.09/to:2017/trend/plot/hadcrut4gl/from:1974/to:2002.09/trend/plot/hadcrut4gl/from:1974/to:2017

February 28, 2017 9:25 pm

I expect the satellite-projected lower troposphere measures to continue their recent months shortfall from the projections by Walter Dnes, fair chance even get another month of the shortfall increasing. Reason why: Snow and ice coverage anomaly is getting increasingly negative, and I think more so when weighted by insolation, especially considering that sunlight is increasing in the areas in and near North America that have less snow and ice coverage than normal. As in, I expect increased positive anomaly of surface heating in areas that I think have upward mobility of the lapse rate from the surface to a couple kilometers aloft.

February 28, 2017 9:25 pm

I find the first figure confusing. I have no idea what its point is. I find the second figure wrong because it shows warming where cooling exists. This one is easy to explain because UAH satellite records show a cooling of 0.1 degrees Celsius in the decade from 2002 to 22012 which your figure 2 shows as warming. Clearly their temperature data source shows non-existent warming there. There are several points that clearly show exaggerated warming throughout. To start with, the bgeginning hump of the twenty-first century is shown as being 0.28 degrees Celsius higher than the super El Nino of 1998 is. In actual fact the superr El Nino towers over that hump by o.45 degrees Celsius. This much of an error in a wel-known temperature region is simply inexplicable for professional climate scientists. Mine are satellite figures which I advise you to use also in place of their falsified mishmash. It gets even more unbelievable when we look at the 22016 El Nino in their graph. It is shown as being 0.63 degrees higher than the 1998 super El Nino is. Bear in mind here that a majority of El Ninos have a total height about half a degree to 0.6 degrees Celsius. That temperature chart of theirs must be considered a complete fake. And it does not even show the beginning of tyhis fakery because it is left off the chart on purpose. It all started with the hiatus of the eighties and nineties. Never heard if t? That is because their fakery goes that far back ,being parallel to James Hansen’s time at GiSS. NOAA, GISS, and the Met Office all got together and changed the eighties and nineties temperatures into a non-existent warming. But I I had already used the original data in my figure 15 in the book “What Warming?” and protested. I was totally ignored so I put a note about it into the preface of my book. In an 18 year stretch of the eighties and nineties they raised the temperature by 0.18 degrees Celsius which amounts to 1 degree Celsius per century. More That in less than two decades, when the claimed warming for the entire twentieth century is o.8 degrees. It would not surprise me if their success with the eighties and nineties did not go to their heads and make them want to continue that warming for the rest of the twenty-first century. This would explain the excess heat in the period that follows, including the extreme height of that El Nino in 2016.However, long-term exaggerated warming we are dealing with makes it likely that the warming was simply increased to get rid of that cooling. The steady rise which their twenty-first century shows is illusory and will need much explication from those fake warming advocates. Since you need to use satellite data to see the cooling their temperature scale must be considered compromised. I advice that this temperature source should not be used to derive any scientific conclusions about global temperature.,Right now, the cooling has been temporarily blocked by the 2016 El Nino, a temporary source of warming. Once that is over we will see global temperature drop to the values we have not seen since the twentieth century. That conclusion itself follows simply from extrapolating the cooling section linearly. And, oh, I forgot that these three temperature wizards had their file ends polished by a common computer device. It left high spikes in the exact same places in all three data-sets. You can see one of them sitting right on top of the super El Nino of 1998 in the second figure above, on its left shoulder. There are around ten or more of these around and you can pick them all out by comparing their temperature curve with satellite temperature curves.

Reply to  Arno Arrak (@ArnoArrak)
February 28, 2017 10:15 pm

Some of the surface warming being greater than warming of the satellite-measured lower troposphere is true. Have a look at Figure 7 of http://www.drroyspencer.com/2015/04/version-6-0-of-the-uah-temperature-dataset-released-new-lt-trend-0-11-cdecade/

One significant reason I see for this is reduced snow and ice coverage causing more surface warming in places where there is upward mobility in the lower tropospheric lapse rate.

Javert Chip
Reply to  Arno Arrak (@ArnoArrak)
March 1, 2017 8:47 am

Somebody needs to show Armo how the paragraphs key on his computer.

Javert Chip
Reply to  Javert Chip
March 1, 2017 9:09 am

…and while they’re at it, show me how to write a cogent 12-word sentence.

Reply to  Javert Chip
March 1, 2017 1:41 pm

Your computer has a paragraph key? Is it the key between the Enter and the Shift keys?

Reply to  Javert Chip
March 1, 2017 7:25 pm

If you are an English teacher, Javert, you should mind your spelling. My name is Arno, not Armo. Sorry about those missing paragraphs, Sir. I shall try to improve myself next century.

Reply to  Javert Chip
March 1, 2017 7:43 pm

I call that ‘the carriage return.’ Still.

Reply to  Arno Arrak (@ArnoArrak)
March 2, 2017 12:16 am

Arno Arrak

“…UAH satellite records show a cooling of 0.1 degrees Celsius in the decade from 2002 to 22012 which your figure 2 shows as warming…”

That’s because the chart shows estimated surface data data from 1997 data up to 2017. You’re referring to satellite lower troposphere data (UAH) over a shorter time span within that period.

Even in the UAH data there will be different trends if you chose different start and end dates. For instance, as you point out there was cooling in UAH if you start in 2001 and end in 2012. However, if you start UAH in 1997 (the year the above NCEP chart starts) and run it to the latest update, the cooling trend between 2001 and 2012 is consumed in the overall warming trend: http://www.woodfortrees.org/graph/uah6/from:1997/plot/uah6/from:2001/to:2012/trend/plot/uah6/from:1997/trend

March 1, 2017 5:43 am

Joe D’Aleo at WeatherBell put up a graph of the past three solar cycles and when the El Niño’s and La Niña’s occurred. Seems to be very regular, having them occur at the same time during each cycle. Based on the last two cycles, expect an El Niño coming up followed by a La Niña at the bottom of this cycle.

Looking at the “global temperatures” the current anomalous high LT temps are at the poles and tropical areas have cooled. I always thought the poles are where the warm air and water goes to die, so as the heat dissipates into outer space the patterns flip and we go colder.

The other thing I find interesting is ice free water also gives up the heat faster than water covered with ice. Cycle up, cycle down.

Reply to  rbabcock
March 1, 2017 8:08 pm

Yes, high anomalies are heat on its way to space by T⁴, yet we’re using them as a sign of planetary heat accumulation, the exact opposite. This is like estimating the population of a city by counting cars on outward bound freeways.

March 1, 2017 10:01 am

Any data on this Reuters/ Climate Research Program claim?

March 1, 2017 2:39 pm

Next time Di Caprio goes viral for saving the planet…

Leonardo DiCaprio Flew an Eyebrow Artist 7,500 Miles for the Oscars
Elle Videos•February 28, 2017
DiCaprio, as well as several other A-list stars, employ eyebrow artist, Sharon Lee-Hamilton, to get them ready for the big night. Hamilton, however, lives in Austrailia, meaning celebs fly her 7,500 just to get their eyebrows done. Read more on ELLE.com!

Caligula Jones
Reply to  TomRude
March 2, 2017 9:38 am

C’mon! He plants TREES, people. TREES, I tell you, to compensate for this.

I get the idea he got involved with climate change when someone mentioned that they use models…

richard verney
March 2, 2017 5:39 am

For those that are interested, UAH Feb data is now in. It is form +0.3degC in January, to 0.35degC in February.

The ENSO metre is now in positive so the ENSO region is now warming. I do not know whether that will continue, and there appears always to be some lag before this is picked up by the satellite. This might have some bearing on future temperatures in the coming months.

richard verney
March 2, 2017 5:43 am


May I suggest that in future publications, you set out your delta for the preceding 6 months. We can then get a feeling as to whether you are continuously under or over estimating, and the approx margin.

March 2, 2017 7:58 am

Isn’t it time to adjust the predictions? UAH anomali for February comes in almost 0,2 C below the prediction.

Reply to  OleKlemsdal
March 2, 2017 12:50 pm

Since I use the previous 12 months to establish the trendline. the numbers should eventually line up, unless UAH anomaly is not a linear function of the NCEP/NCAR anomaly.

Reply to  Walter Dnes
March 6, 2017 10:27 pm

Thanks for the explanation! And no, I don’t thinkUAH anmaly is a linear function of the NCEP/NCAR anomaly.

March 2, 2017 12:55 pm

HadCRUT4 for January came in today at +0.741, versus my projection of +0.736.

%d bloggers like this: