January temperature in Norway

From the eKlima database [L44] I have extracted temperature data for January from 108 stations scattered throughout mainland Norway. Most of the stations have been in continuous operation from 1960 until today. For each year from 1960 to 2019, I have calculated the average for all the stations. The result can be seen in the figure below:

The most striking thing about this figure is the large variations from year to year. Throughout the period, a temperature rise of a few degrees is also visible, but this trend drowns in the annual variations.

The figure illustrates how it is possible to lie with statistics. For example, if we only look at the period 1989 to 2019, we see a declining trend. We can also use the figure to compare the two 30-year periods 1960-1989 and 1990-2019. If we choose to count the number of years where the January temperature has been below -6 °C, we get the following result:

Period 1960-1989: 9 years below -6 °C
Period 1990-2019: 1 year below -6 °C

which shows increasing temperature over the entire period.

From a relatively large amount of data, we have, in two different ways, limited the data so that we get two conflicting conclusions. In the first case, we selected a sub-period which most clearly showed a declining trend. In the second case, we reduced the amount of data by concentrating on the number of observations below a certain limit, and we selected this limit carefully so that the result clearly showed an increasing temperature. If we had chosen a different limit, the result would not have been so clear (although all choices had shown the same tendency).

It is doubtful whether these data can be used to detect any trend at all. But it has been an interesting exercise in how the eKlima database may be used and how the downloaded data can be analysed.

Latest update: 2021-07-21