JLI Final Forecasts for 2014

Guest Post By Walter Dnes

The NOAA(NCDC) January data set update was delayed. It came in during the afternoon/evening of March 6th. With all the January data being in, now is the time for the January Leading Indicator “JLI” algorithm forecast to “put up or shut up”, and make forecasts for 2014. As described here and here, the JLI algorithm is not a “real forecast” per se, but rather a “zero skill baseline” that a “real forecast” has to beat in order to show skill. The only excuse I’ll use for missing the forecasts is a Pinatubo-scale event, i.e. a major volcano (or meteorite/comet impact) that kicks up a significant amount of particulates/sulfates/etc. into the stratosphere.

First, the raw data. Because some of the data sets adjust their past data every month, the algorithm would produce slightly different results each month for the quantitative forecasts. In close cases, even the qualitative forecasts can change. In order to allow reproduction of the results, the January 2014 data sets, as downloaded in February 2014, are attached here, along with the spreadsheet used for the calculations.

The Qualitative Forecasts

  • HadCRUT v3

    The January 2014 HadCRUT3 monthly anomaly was 0.472 versus 0.392 in January 2013. The 2014 annual mean anomaly is forecast to be warmer than the 2013 annual mean of 0.459.

  • HadCRUT v4

    The January 2014 HadCRUT4 monthly anomaly was 0.506 versus 0.450 in January 2013. The 2014 annual mean anomaly is forecast to be warmer than the 2013 annual mean of 0.488.

  • GISS

    The January 2014 GISS monthly anomaly was 0.70 versus 0.63 in January 2013. The 2014 annual mean anomaly is forecast to be warmer than the 2013 annual mean of 0.603.

  • UAH v5.6

    The January 2014 UAH5.6 monthly anomaly was 0.291 versus 0.497 in January 2013. The 2014 annual mean anomaly is forecast be cooler than the 2013 annual mean of 0.236.

  • RSS

    The January 2014 RSS monthly anomaly was 0.262 versus 0.439 in January 2013. The 2014 annual mean anomaly is forecast to be cooler than the 2013 annual mean of 0.218.

  • NOAA (NCDC)

    The January 2014 NOAA (NCDC) monthly anomaly was 0.6480 versus 0.5491 in January 2013. The 2014 annual mean anomaly is forecast to be warmer than the 2013 annual mean of 0.625.

The Quantitative Forecasts

Due to the noisiness of the data it is possible for the qualitative forecast to indicate a warmer value than the previous year, while the quantitative forecast indicates a cooler value (or vice versa). This type of mixed signal occurs for 2014 in the land data sets, where the qualitative forecast is for warmer than the previous year, but quantitative forecast is for a cooler year.

Tab “jan_and_avg_2” of the spreadsheet has some statistics in the block P1:V4, comparing the January anomalies with the annual anomalies. These include slope() and intercept(). Once we have the January anomaly, we can apply the old “y = mx + b” linear equation to get a quantitative prediction for the year.

  • HadCRUT v3 * The slope in cell Q3 is 0.81614. The intercept in cell R4 is 0.02345. The Jan 2014 anomaly is +0.472. Applying the standard “y = mx + b” equation, we get a predicted 2014 annual anomaly of +0.409 with an unknown error margin.
  • HadCRUT v4 * The slope in cell R3 is 0.77609. The intercept in cell R4 is 0.02637. The Jan 2014 anomaly is +0.506. Applying the standard “y = mx + b” equation, we get a predicted 2014 annual anomaly of +0.419 with an unknown error margin.
  • GISS * The slope in cell S3 is 0.81358. The intercept in cell S4 is 0.03062. The Jan 2014 anomaly is +0.70. Applying the standard “y = mx + b” equation, we get a predicted 2014 annual anomaly of +0.600 with an unknown error margin.
  • UAH v5.6 * The slope in cell T3 is 0.64062. The intercept in cell T4 is 0.01732. The Jan 2014 anomaly is +0.291. Applying the standard “y = mx + b” equation, we get a predicted 2014 annual anomaly of 0.64062 * 0.291 + 0.01732 = 0.204 with an unknown error margin.
  • RSS * The slope in cell U3 is 0.64755 and the intercept in cell U4 is 0.03456. The Jan 2014 anomaly is +0.262 The predicted 2014 annual anomaly is 0.64755 * 0.262 + 0.03456 = 0.204 with an unknown error margin.
  • NOAA (NCDC) * The slope in cell V3 is 0.84179 and the intercept in cell V4 is 0.04571. The Jan 2014 anomaly is +0.648 The predicted 2014 annual anomaly is 0.84179 * 0.648 + 0.04571 = 0.591 with an unknown error margin.

In weather forecasting, one generally goes with the model consensus, or at least the majority opinion. The JLI …

  • qualitative forecast indicates 4 (surface) data sets warmer and 2 (satellite) data sets cooler
  • quantitative forecast indicates all 6 data sets cooler

The “cooler” runs outnumber the “warmer” runs 8 to 4. So I’ll go with a somewhat cooler year overall.

The Met Office 2014 Prediction

19 December 2013 — The global average temperature in 2014 is expected to be between 0.43 C and 0.71 C above the long-term (1961-1990) average of 14.0 C, with a central estimate of 0.57 C, according to the Met Office annual global temperature forecast.

Their forecast is based on the average of HadCRUT4, GISS, and NOAA(NCDC) anomalies. Using the numbers from the JLI quantitative anomaly forecasts, the JLI equivalent forecast is…

( 0.419 + 0.600 + .591 ) / 3 = 0.537

I acknowledge that I have an additional 2 months of data available compared to what UK Met Office had when they made their forecast.

0 0 votes
Article Rating

Discover more from Watts Up With That?

Subscribe to get the latest posts sent to your email.

68 Comments
Inline Feedbacks
View all comments
March 7, 2014 9:25 pm

To be clear, the serious work on the 2020-2030 global cooling forecast came from Paleoclimatologist Tim Patterson of Carleton University.
I was writing an article for the Calgary Herald and phoned Tim and said: “Tim, you and I both believe climate change is natural and cyclical, correct?” Tim immediately agreed. So I said “OK, when is it going to get colder?” He then said, with a pause of just a few seconds, “2020 to 2030”. I asked why, and he explained that he based his answer on his research into the Gleissberg Cycle, which is about 90 years long. I asked Tim if the ~60 year PDO cycle might be a better fit, but he preferred the Gleissberg.
If the PDO governs, then global cooling has probably already begun, but it will take a few more years to be sure.
I am increasingly convinced that CO2 is utterly irrelevant as a driver of global temperature. Wait ten years and this will be the new conventional wisdom in climate science. Some people will say they knew it all along… 🙂
Regards to all, Allan
.

Editor
March 7, 2014 10:11 pm

Allan M.R. MacRae says:
> March 7, 2014 at 4:52 pm
>
> Serious Question – HELP please – I need this.
>
> Does ANYONE out there have a strong predictive
> track record, say 3 or 6 or 9 or 12 months in the
> future, for North American winter temperatures?
You can get general forecasts from commercial firms if you’re willing to pay. I can’t personally vouch for anybody’s accuracy.

Richards in Vancouver
March 7, 2014 10:14 pm

Latitude says:
March 7, 2014 at 3:54 pm
“Steven Mosher says:
March 7, 2014 at 1:24 pm
What is known
===
whew, what a relief
At least we know we don’t know squat………”
I respectfully disagree with you, Latitude. I think we DO know squat.

daddylonglegs
March 7, 2014 10:19 pm

We are about to get a La Nina, not el Nino. This will pull temperatures down.

David L
March 8, 2014 12:33 am

“Due to the noisiness of the data it is possible for the qualitative forecast to indicate a warmer value than the previous year, while the quantitative forecast indicates a cooler value (or vice versa). This type of mixed signal occurs for 2014 in the land data sets, where the qualitative forecast is for warmer than the previous year, but quantitative forecast is for a cooler year.”
…………..
In other words,the average slope of their linear temperature rend continues to be zero. When noise is included, 2014 could either be above or below 2013.

David L
March 8, 2014 12:47 am

Steven Mosher:
“What is known
1. An El Nino event occurs about once every three to seven years,
2 As the ocean builds up heat in the western Pacific Ocean, some of the heat goes into the atmosphere through evaporation.
3.Right now the El Nino Southern Oscillation is in its “neutral” phase — neither warm nor cool
4. If we get an El nino, temperatures are likely to warm.
5. models make forecasts”
………….
Difference in opinion about what’s “known” may be around certainty. Your statement above is filled with “about, some, if, likely”.
In grad school I knew two profs (a philosopher and and physicist) who were arguing over the certainty of science. The physicist said “if you drop a pencil I can tell you the speed it falls, the distance, the kinetic energy, etc.”. The philosopher said “can your science tell me if I’ll drop the pencil”.

Solomon Green
March 8, 2014 1:57 am

” The Met Office 2014 Prediction
19 December 2013 — The global average temperature in 2014 is expected to be between 0.43 C and 0.71 C above the long-term (1961-1990) average of 14.0 C, with a central estimate of 0.57 C, according to the Met Office annual global temperature forecast.”
Can anyone explain why the Met Office refers its prediction on a long term (1961-1990) average? Why not (1951-1980) or (1971-1990) or the most recent (1981-2010)? Is it because the anomaly may look higher when using 1961-1990 as the base? Or is it because they might find it difficlult to exclude UAH and RSS from their averages if they selected the later periods?

March 8, 2014 2:47 am

RE Seasonal Weather Forecasting in North America
http://www.cpc.ncep.noaa.gov/products/predictions/long_range/tools/briefing/seas_veri.grid.php
NOAA predicted in November 2013 that this winter (December January February) would be warmer than usual in northeastern and south-central USA. It was actually much colder than usual in these regions.
NOAA’s recent Temperature Forecast Heidke Skill Scores are terrible (near-zero), but still seem inflated given their utter failure.
Would it be unfair to suggest that NOAA can compete head-to-head with the UK Met office for “worst seasonal weather forecaster”?
Is any organization good at this? Names?

RichardLH
March 8, 2014 3:15 am

I predict a weak El Nino towards the end of this year based on this graph/predictor
http://climatedatablog.files.wordpress.com/2014/02/uah-tropics.png

March 8, 2014 7:13 am

Just the facts
Here is what you asked
“Please enlighten us as to “what is known”. Do you think that “we may see a mini global warming phenomenon”?…”
And before
“Thanks for the ENSO 101 there, but if you try some reading comprehension, I never said “we know nothing”, I wrote that, “In summary, we have no idea what’s going to happen”, and per this comment;”
So let me answer your first question again
Of course we have an Idea of what is going to happen. It’s pretty simple. some people have an Idea that El nino may happen. It’s pretty clear what that means. I have an idea that the sun may go down today. The fact that I am uncertain about it doesnt make it any less of an idea.
##############
Next question was
Have you considered entering the legal field? Between you and Bill Clinton you could spend your twilight years parsing out the meaning of the words “may” and “is”…
Parsing out the meaning of the words “May” and “Is” would be Philosophy son not law. I find it ironic with your name “Just the facts” that you dont pay more attention to facts which are about what ‘is’ and ideas which can be about what “may” exist.
funny actually.
################
There is no “probable knowledge”, we are talking about forecasting here, and as I wrote several years ago;”
Of course there is probable knowledge. Further citing yourself to make an argument is really funny.

March 8, 2014 7:46 am

Uk.
Why are we building windmills?
It would depend. Which “we” are you referring too?
Regardless of which however you are asking for an explanation of motivations. These explanations
Are always probable not certain. So you might have
An idea and you might point to statements
About motivations but in the end you just have probable
Knowledge. Another word for that is belief. Its funny when
Skeptics are not skeptical.

RichardLH
March 8, 2014 7:57 am

Steve: I am unclear as to what you think will actually happen.
As far as I can tell all you have said is that there ‘may’ be a El Nino.
As you correctly observe, the sun ‘may’ come up tomorrow.
Care to give any actual prediction of how large or small the upcoming El Nino might be? And when it might happen? Just to get off the fence.

RichardLH
March 8, 2014 8:18 am

“They should be using 1981–2010 as our base period “in order to comply with a recommended World Meteorological Organization (WMO) Policy, which suggests using the latest decade for the 30-year average.””
You can align all the data sets in their overlap period of 1979-today and thus remove the base period sampling problem and get them all on the same page.
You can go from
http://climatedatablog.files.wordpress.com/2014/02/hadcrut-giss-rss-and-uah-global-annual-anomalies-with-gaussian-annual-and-15-year-low-pass-filters-from-sources1.png
to
http://climatedatablog.files.wordpress.com/2014/02/hadcrut-giss-rss-and-uah-global-annual-anomalies-aligned-1979-2013-with-gaussian-low-pass-and-savitzky-golay-15-year-filters1.png

March 8, 2014 4:44 pm

Apart from Nino4 things are running in the cool side currently:
http://www.bom.gov.au/climate/enso/indices.shtml
The QBO should move into the easterly phase around July, which is more conducive for La Nina:
http://www.esrl.noaa.gov/psd/data/correlation/qbo.data
I’m looking at a trend for a more positive AO July through Oct which again would indicate a La Nina bias like last summer.

goldminor
March 9, 2014 3:09 am

I see a La Nina coming that will last around 3 years, breaking back to El Nino in 2016/17. It should be close to a -2 event in scale. The end of this year might see a short small El Nino.

Solomon Green
March 9, 2014 4:33 am

My thanks to Just The Facts for answering my questions. Which leaves me with only one more question. How can Climate Scientists really believe in their fudged data?
Thanks, also for your reponses to Steven Mosher. It would be interesting to learn just what percentage of medium term (3-12) months have been accurate. For the UK many of us have noticed that the Met Office’s predictions have a very high degree of correlation with the outcomes. Unfortunately it is negative correlation.