Latest News (hidethedecline)

Original Temperatures: Introduction

Posted by Frank Lansner (frank) on 24th December, 2013
Latest News (hidethedecline) >>


In Denmark, first of April 1872, most meteorological tasks where collected in the new organization: “Meteorologisk Institut” – later “DMI”. 2 of the 3 main purposes are
1) Collect observations
2) Spread these to the public
Tax money spent on this institution has a main purpose to collect and spread out observations. This goal is probably not that different from the goals of all other tax funded meteorological institutions, but are the meteorological institutions actually doing their very best to spread out temperature observations from the stations as they are paid to do? Showing the public (the taxpayers) the truth, the whole truth and nothing but the truth when it comes to temperature observations from the stations?
Primary goals:
1) Restore and present original temperature data trends for the public.
2) Evaluate: Does original temperature data tell us something about the effect of Greenhouse gases in the Earth’s atmosphere?
3) Evaluate temperature data used in climate science, that is, shed light on the adjustments made to original temperature data and pinpoint if the temperature records used for climate science represent reality as well as possible.

The ORIGINAL TEMPERATURES project is mostly based on temperature data from the following sources:

National Meteorological yearbooks
National Statistical yearbooks
Data from National Archives
The World Weather Records compilation
Data successfully required from National Meteorological Institutes
NOAA´s data rescue files – mostly scanned meteorological yearbooks
  (- For some reason no data has been added to this compilation after 2003 although plenty is missing).
Books and articles available on libraries
Smaller data bases like NACD and Nordklim
Online databases like Tutiempo and ECA&D

Fig 1

The bulk of data (possibly all) presented by the ORIGINAL TEMPERATURES project has already been made public at some point by the relevant institutions like national MI´s and thus ORI-TEMPS (short for ORIGINAL TEMPERATURES) is simply presenting what was already made public. Thus, I don’t think I need to seek asylum in Russia just yet for this “leak” of temperature data to the public.

The earlier “RUTI” writings were based mainly on GHCN v2 raw.
“RUTI” is an abbreviation for “Rural Unadjusted Temperature Index” indicating that such data are the preferred temperature data for these investigations. The concept of ORI-TEMPS brings us a large step closer to this kind of high quality temperature data simply because the number of datasets to select from is exploding when expanding the sources for temperature data is done in ORI-TEMPS.

Tutiempo vs. ECA&D (for Europe):
Sometimes the online data presented by Tutiempo ( is not identical to the data you can find using ECA (
In the FAQ section ECA&D write:

“the older series may have changed, because of improved data quality control or data archaeology by the data providing institutions.”

That is, ECA&D tends to show adjusted recent versions of data.

For Tutiempo, this is not quite the case. I asked the Tutiempo team if their data are original or adjusted versions and I got this answer:
“The data shown are the original provided by the station.
Nobody notify us if they change.”
And they explain to me that they use data from the World Meteorological Organization.

We cannot know for sure if Tutiempo has always been provided with original unadjusted temperature data to begin with, but it does seem likely that Tutiempo versions are rarely – if ever – updated.
Generally this suggests that Tutiempo versions of temperature data are normally older than the ECA&D versions.

WWR – World Weather Records
- climatic data from stations all over the world published since 1927.

WWR-NOAA: As far as I can see, only a fraction of the data is available online from NOAA, check out for example 1961-90 / South America:

WWR-NCAR: Another source for WWR data, NCAR:
I tried to download WWR data, but got the message that I needed to get a user first. After some time I had a user that also was approved, but here the result of trying to download with my new user:
Fig 2

“NCAR-Only Access”.

In good old Denmark instead I found several parts of the WWR climate data in The Royal Danish Library in Copenhagen, so WWR data appears anyway here and there in the writings.


Estimating areas of similar temperature trend.


1) An area of multiple similar temperature trends increase the reliability of data   
2) An area of multiple similar temperature trends is the best tool to fast and easy reveal faulty datasets, outliers , sudden breaks in data  and faulty adjustment.
3) Stitching datasets or filling in data from other series should be done using other datasets from the same area of similar temperature trends. (Some data sources like GHCN V2 often provides only parts of data series, and thus to get the entire temperature trend for an area you might need to stitch data sets.)
4) Make sure that a temperature trend related to a smaller area is not used to represent a much larger area. (Not rarely for example, coastal stations are used in climate science to represent much larger areas than just the coast. )
5) Finally, identifying areas of similar trends turns out to yield an overwhelming challenge to the claim that CO2 is responsible for a strong recent warming. We will get back to this.

How to estimate the areas of similar temperature trend?

To a large extend, areas of similar temperature trends appear to be areas with similar exposure to winds from the ocean. For example, far most ocean coasts show a warmer temperature trend the nearby non-coastal areas during the 20´ieth century, results from 37 study cases:

Fig 3

Taken from (

“OAS” = “Ocean Air Shelters”
The non-coastal areas must be divided further by estimating roughly which areas are “OAS” (Ocean Air Shelters) and which are “OAA” (Ocean Air Affected):
Fig 4
Actual temperature trends from stations can then be used to define the areas more precisely. The difference of temperature trend between OAS and OAA is generally so large (0,5-2 K 1900 – 2010) that estimation of OAS vs. OAA areas is mostly rather straight forward and easy to work with.
Temperature data from the OAS areas (blue) have colder temperature trends and are thus more controversial than temperature data from the other areas. Temperature data from the OAS areas are often adjusted or not used in climate science.
On the illustration above I have used red colour for Urban and Coastal areas, yellow for OAA and Blue for OAS.

OAS areas often show little or no warming trend 1930-2010. This is a challenge to the role of the greenhouse gas CO2:
-> How come areas with no ”help” from ocean air temperature trends cannot show strong warming trend 1930-2010?
(CO2 is present in all areas, obviously.)

How much of the land is then OAS and how much is OAA?

It depends on the geography:

Fig 5

10% of the area is OAS, 90% OAA. Ex: Northern Germany, Poland, West and Central Australia. Temperature trends of such OAA areas are slightly warm as is the overall area.

Fig 6

90% of the area is OAS, 10% OAA. Ex: US Midwest, Hungarian Valley, Central South America, SE Australia. Large areas with none or little warm trend 1925-2010. Such OAA mountain peaks (like the Alps and some peaks of the Rocky Mountains) show strong warming trends but represent just a small area.

Fig 7
EX: The Alps, Turkey, Himalaya. In areas with plenty high peaks closely packed together mostly just the highest tops will be directly affected by ocean air and OAS areas dominate. That is, even some mountain tops can be in shelter of ocean air. If icy tops are affected this might affect lower areas indirectly or lead to positive feedbacks etc. but for now this is out of scope.

Colder temperature trends predicted in most of the area of the Alps are exactly what we find when examining all existing datasets from the Alps. A Web site like “HISTALPS” strongly ignore cold trended stations in the Alps and strongly warm adjust the few valley stations used. See ORI-TEMPS THE ALPS.

Narrow valleys, UHI and rivers: Urban heat will always be an issue for urban temperature stations, but the UHI problem is amplified in narrow valleys where the heat cannot escape easily. If possible always chose the station connected to the most rural area, especially when dealing with deeper valleys.
Also the effect of lakes and rivers (downstream from OAA areas) will have its temperature trend carried downhill and amplified in narrow valleys.

As complicated as it gets: Germany
Here’s how the German map looks when divided into such areas:
Fig 8

Dominant wind directions from oceans areas are SW to NW. The elevation of Germany rises gradually the longer you get away from the coast. Also, areas with hills and mountains appear in numerous places and orientations. None the less it was easy to find the areas of similar temperature trends using the approach of OAS and OAA mentioned. Germany also has numerous areas (OAS) with no or hardly any temperature rise 1925-2010.

Last changed: 24th December, 2013 at 16:16:49



None Found

Add Comment