Scandinavia

RUTI Scandinavia

Fig1. This writing shows the temperature data from the above areas with a combined area nearly 2 mio kvm2, almost the same size as Greenland. Working with Scandinavian temperature series is very rewarding due to the fact, that we have the Nordklim data series in addition to GHCN. This has a lot of consequences, first of all great data availability, but also we see many temperature series from GHCN and even Hadcrut without large adjustments. Nordklim data ends at latest in 2002 – has sadly not been updated for long time – and therefore the coupling Nordklim-GHCN can give us the complete data series.

Nordklim is also marvellous because we don’t constantly have to deal with interrupted time series, periods taken out of time series etc. Nothing left out in Nordklim, not one single data series in Denmark, Sweden, Norway, Finland, Iceland, Greenland and more has been interrupted. Fantastic.

Nordklim is created as a corporation between all Scandinavian national meteorological institutes.

 
Bravo! Click to go to Nordklim homepage.

In most cases, there is hardly any difference between GHCN, Hadcrut and Nordklim:
  

Fig2. Scandinavia is covered well, perhaps except one country: Denmark – see later. Another critique from me is that Nordklim has far more southern Swedish stations placed in the Baltic on islands than within the main land, the large bulk of Sweden.

And more: In the station overview of Nordklim online today there are more stations listed than actually available to download right now (!?)
Stations thus apparently now missing are:
Finland: Huittinen Hattula Heinola Virolahti Lavia Virrat Orivesi Vaasa Ylistaro Aehtaeri Maaninka Joensuu Kestilä Yli-Ii Pudasjärvi  (all rural except Vaasa?)
Norway: Skjaak Reinli Nedstrand Vetti Oerskog Lien I Selbu Bodoe Kraakmo (All rural except Bodoe?)
Sweden: Lund Havraryd Krokshult Haavelund Vaenersborg Linköping Aalberga Vaesteraas Lisjoe Malung Graengesberg Soesjoe Sidsjoe Junsele Leipikvattnet Loevaanger Tjaamotis Kiruna (mostly rural)


1) DENMARK
 

Fig3. For Denmark, certain coastal areas – especially the west coast of Jutland – may not appear to be best choice for representing temperatures for the bulk of Denmark, especially the warm periods with SW to W winds.   (DMI Pictures taken from Swedish fine article: http://helahavetstormar.nu/urban-varmeo-runt-kobnhamn/)

The weakness for the west Jutland coastal stations might be that in cold periods they equals average Danish temperatures well while in warm periods they don’t show the heat well.

(More on Coastal temperature stations, see RUTI Coastal temperature stations)

The Danish temperature stations used by GHCN and Nordklim:
         

Fig4. From GHCN we have 4 rural temperature stations shown in blue. From Nordklim we have 2 rural stations shown in green. These 6 stations are relevant, but we also need stations from the inland bulk of Denmark. Red stars: We also have 1950-2010 from the Urban station Alborg in North Jutland and data 1880-2010 from large urban Copenhagen area is available as the only GHCN station in Denmark. 
 
Fig5. Taken from: http://www.dmi.dk/dmi/tr99-13.pdf
There are plenty of rural inland meteorological stations in Denmark, but not online available data it seems?

In addition, GHCN data for all stations except the large Urban Copenhagen temperature series are cut with the effect that we cannot compare today’s temperatures with the warm 1925-45 period. In fact, it is mostly years after 1970 that are missing in data. As if the Danish temperature data suddenly got too poor and useless after 1970?

With these comments in mind, here are my so far best attempt to create a Danish temperature series:
 
Fig6. Green stars: Nordklim / Blue stars GHCN. As you can see, not a single rural or near rural station representing the bulk of Denmark: The inland of Jutland, Fyn and Zealand.
To improve quality I exchanged Copenhagen data 1880-1990 with Swedish Falsterbo data, rural,
Falsterbo:
 
Results:
  

Fig8 A s mentioned: The weakness for the west Jutland coastal stations might be that in cold periods they equals average Danish temperatures while in warm periods they don’t show the heat well.
However, this issue is not present in the smaller Danish waters like Kattegat.

From the Article “Hansson D, Omstedt A (2008) Modelling the Baltic Sea ocean climate on centennial time scale; temperature and sea ice. Climate Dynamics, 30, 763-778, doi:10.1007/s00382-007-0321-2.”
We have an estimate of temperatures for the central Danish waters “Kattegat” and “Skagerrak” based on measurements from ferries:
 

Fig9. These data are pretty close to the Danish results so far, perhaps a 0,2-0,3 K colder trend over 100.

2) SOUTHERN SWEDEN

 

Fig10. Green stars: Nordklim / Blue stars GHCN.

Here are some of the public available GHCN data from Southern Sweden:
  
Fig11. GHCN do deliver some useful data from Swedish town Jonkoping but otherwise its mostly the Nordklim database we can use for Southern Swedish temperatures. When nothing is written, data is from Nordklim:

 

Fig12. The Southern Swedish temperature data matches the Kattegat-Skagerrak temperatures, slightly colder trend than the Danish mostly coastal data.

 
Fig13. Green stars: Nordklim / Blue stars GHCN.

3) CENTRAL AND NORTHERN SWEDEN

Stations with both blue and green star are Nordklim data until around year 2000, and then GHCN until 2010.
 

 
Fig14 . Fine agreement for example in Haparanda, but for example in Karesuando  (Not that far from Haparanda) GHCN data suggests year 2000 to be a full Kelvin warmer than Nordklim data. No other station in the area has such a warm year 2000 peak:

 
Fig15.

Temperature trends:
 

Fig16. We see a slightly colder trend in central and Northern Sweden than in Southern Sweden and a significantly colder trend than for Denmark. None of the Swedish temperature series from Nordklim shows a peak in year 2000 as GHCN suggested for Karesuando.

Notice how homogene data are for such a large area. These data must “homogenized” perfectly? And yet it is hard to claim much trace of unprecedented heat recently?

4) SOUTHERN NORWAY
 

Fig17. A very strong peak around 1989-1991 (See RUTI adjustment of the French and Austrian peak) and we don’t see recent temperatures significant warmer than previous warm peak.

5) CENTRAL NORWAY 

Fig18. Rather similar to the above series for central Norway.


 
Fig19. Finland And Karel (RU)

6) NORTHERN FINLAND-KAREL(RU)

In general, the Northern Russian temperature series (and Siberian) are rather intact from GHCN, and thus we can include the Karel area data too.

 

Fig20. Same picture as for most Scandinavia, allthough the warm peaks of the 1930´ies appears stronger than in recent years.

7) SOUTHERN FINLAND-KAREL(RU)

 

Fig21. In Southern Finland-Karel, temperatures of recent decades are similar to the warm peak before 1950..

 

CONCLUSION

For the entire Scandinavian land area - including the Russian Karel part – there is no warming measured by thermometers. The area has approximatel same size as Greenland. Any extraordinary warming claims for the Scandinavian therefore would have to be based on massive changes in data, even though practically all thermometers tell a very similar story indeed. It’s a tough job to claim that all Scandinavian thermometers not only has errors, but errors that result in very similar “error” trends. No, the conclusion is: No extraordinar warming in Scandinavia at this point.

OK, for Denmark .based on coastal ocean-wind sensitive temperature stations from 15% of the country, we did see a slight warming in recent years  compared to to the pre 1950 peak. I would very much like to see long temperature trends from the more central parts of Jutland, Sealand or Funen.

Perhaps the area of Scandinavia with most cooling trend was area 6, the most northern area of Scandinavia. Was not the Arctic supposed to warm extremely fast?

See also: “RUTI Greenland, Iceland and Spitsbergen” And “RUTI Siberia”.

COMING UP:
- A graph of the 7 Scandinavian areas for compare.