A 70-80 year peridiocity identified from tree ring temperatures AD 550 – 1980 in Northern Scandinavia

Posted: January 28, 2014 by tallbloke in Celestial Mechanics, Cycles, Natural Variation, Ocean dynamics, solar system dynamics, Tides, trees

Here’s a nice confirmation of Harald Yndestad’s work on Long term lunar tides in the north Atlantic. He found a 74 yr cycle.

A 70-80 year peridiocity identified from tree ring temperatures AD 550 – 1980 in Northern Scandinavia

  • a Finnish Meteorological Institute, P.O. Box 503, FI – 00101 Helsinki, FINLAND
  • b Department of Geosciences and Geography, P.O. Box 64, FI – 00014 University of Helsinki FINLAND


The classical Maximum Density data of 65 Torneträsk trees from years 441-1980 AD are studied in millennial, centennial and volcanic scales. The millennial scale is analyzed applying a specific filtering method. In that scale, the climate is cool after 1200-1400 AD. This more or less steady period is suggested to be due to volcanic episodes, which reduced the northward heat transport in the North Atlantic. The century scale variation, on the other hand, is suggested to be due to internal oscillations in sea surface temperature (SST) and to be connected to variations in the Arctic sea ice. Specifically, these oscillations have caused an additional warming and cooling trend in Northern Fennoscandian temperatures before and after 1930’s, respectively.

Variations in the temperature estimates are explained by the results for different temporal scales. All of them show local impacts leading to differences when compared with hemispheric estimates. The long-term estimate of the temperature as derived from the present Torneträsk data is found to be biased. The source of that is unknown.

1. Introduction

Millennial temperature variations at one location can naturally differ from the hemispheric or polar ones. In Esper et al. (2012a) it was observed that different tree ring estimates of the temperature for the Fennoscandian area differ even from each other. This was not expected because the area is limited and the data samples used were partly overlapping. The initial aim of this paper was to study estimates which differed most. The basic data selected for that aim were the historical Torneträsk MXD (Maximum Density) measurements of 65 trees from years 441-1980 AD as given in Schweingruber (1988). However, the work led us to analyse climatic variations and so extended the initial aim.

The hemispheric temperatures at the interval 1200-2000 AD have recently been simulated in Mann et al. (2012). Simple energy-balanced and general circulation models were applied. The simulation included solar, volcanic, aerosol and greenhouse forcing. When compared with the tree ring temperatures (D’Arrigo et al., 2006), the agreement was, by taking into account model errors and estimation errors in the temperatures, excellent. Especially the warming since 1800 was well captured.

Kaufman et al. (2009) derived temperatures of polar areas from proxy data. The main features in their results are like those in the case studied by Mann et al. (2012), especially the recent warming is well evident. Both studies seem to describe a similar global development. There is one exception in Kaufman et al. (2009). From their long data they could determine a systematic decreasing trend of -0.22 °C in 1000 years. Such a decrease could be expected because at northern latitudes Milankovitch or orbital cooling should be observable.

The Fennoscandian area serves as a good source for tree-ring data. The time series are continuous, long and well cross-dated. They have been actively studied in several papers. Nevertheless the results pose questions. No unique description of the temperatures is found. Esper et al. (2012b) mention “that choosing one Scandinavian record instead of another can alter constructed temperatures by 1.5-3 °C during Medieval times”.


  1. Stephen Richards says:

    Trees are not thermometers and never can be. Their growth needs are too complex and too varied.

  2. tallbloke says:

    Stephen, agreed, but as always timings are more discernable than magnitudes.