Do European tree-ring analyzes indicate a recent anomalous hydrologic climate?

From Dr. Judith Curry’s Climate, etc

by Frank Bosse and Nic Lewis

Not really.

A recent article (MB Freund et al. 2023, MBF23 later) in “Earth and Environmental Nature Communication” investigates the variability of summer droughts since 1600. It uses the method of “isotope analysis” stabilizer C13/O18″ for expansion”Normalized precipitation-evaporation index (SPEI) from 1950 to present back to 1600.

The paper describes and uses a multiproxy network over large areas of Europe (see Figure 1 of MBF23) to reconstruct the history of summer drought over a longer historical period. It found interesting results about the dependence of those events on volcanism and solar forces. It is a document worth reading and we are interested in whether the main title makes sense and likewise this statement in the Summary:

“We show that the recent European summer drought (2015–2018) was highly unusual in the context of centuries…”

Thanks to the authors for using SPEI regeneration every year data are available, so we can do the calculations to check these claims.

The first unambiguous “confirmation” of the paper’s title appears in Figure 3a in MBH23:

Figure 1: Copy of Figure 3a of MBH23. European annual average SPEI data are shown in blue/red, the output of the low-pass filter is shown in black.

The black line in this figure shows the effect of applying a 13 year low pass, so it is related to the recent past. Indeed, after 2010, the 13-year-old Chebyshev filter showed a “significant” decrease in precipitation readings much lower than at any other time during the 1600-2018 reconstruction period. However, on closer observation, one also sees periods of aridity, before 1950 the beginning of the classical SPEI dataset is marked with “SPEI”, or before 1880 marked in grey. bold in Figure 1, and the low-pass filter does not react the way it did after 2010.

The reason for this behavior is quite simple: All smoothing filters have difficulty with the beginning and the end of the filtered dataset. They estimate the output because there is no predecessor/inheritance in the raw data. To test the effect of this property, we used the same data with the similar filter (Loess) and compared with Figure 1 but stopped the filter in 1949:

Figure 2: Figure 1, but with the smoothed SPEI Index ending in 1949.

If the paper was written in 1950 it would find an “unusual recent hydroclimate”, in 2023 it would find the same for recent conditions due to filter failure. The head after 1600 is also unusually wet at the filter output for the same reason.

The final drop in Figure.3a of the MBH23 is bogus, it’s an artifact of a used filter.

A simple running averaging filter, although it has no output in the early years, is not generated, which more fairly smooths out the 1600-2018 oscillations:

Figure 3: Summer SPEI data (black) filtered with the running mean (red). The previous minimum of this filter is shown as a red dashed line. It is evident that there are minimums in the years 1870 and 1680 as well as at the end of the period 1600-2018.

Figure 3 gives results contrary to the headlines of MBF23: by 2018 (the last data point in the MBF23 set), it indicates that the recent European summer climate was NOT unusual, the index SPEI is within the natural variability range.

To prove that the statement in the Summary (“2015-2018 very unusual”) is not true, we took a closer look at the data and calculated the average of those 4 years over the entire time period. .

It turns out that for many time periods, the 4-year average SPEI data is more negative than for 2015-2018, with this average being -0.273:

Since 1900 there have been four such periods, all of which took place in the years leading up to 1950: 1947-1950; 1946-1949; 1945-1948; 1944-1947. The period before 1950 (not strongly influenced by human coercion) was indeed marked by very dry summers, not mentioned in a word in MBF23.

Before 1900 there were several periods:

1892-1895; 1760-1763; 1759-1762; 1738-1741; 1688-1691.

“European Summer Drought 2015-2018” is NOT so unusual in the context of centuries”, as falsely stated in the abstract.

To reinforce this point, we also looked at whether longer-term averages were “very anomalous”.

It turns out that the 5-year average produces 10 periods over the period 1600-1950, a period influenced mainly by natural variability, with more negative SPEI values ​​than the most recent period. as of 2018; 10 year average for 9 such periods prior to 1951. And the 3-year average produces no fewer than 57 periods prior to 1951 with more negative SPEI values ​​than the most recent period.

In addition, we looked at the change in annual data after 1950 (the “native SPEI” period) and prior to this year, the “network-based European hydroclimate” reconstruction period. grid of isotope-stabilized tree rings of the oxygen-carbon ratio” in MBF23. We calculated the standard deviation (sigma) for 21 years of annual data (Figure 4):

Figure 4: Change of SPEI data year on year. The averages before 1950 and after this year are marked with a dotted line. Note the jump.

The lower time variation of the reconstruction has raised some doubts, such as whether the reconstruction of SPEI 1600…1950 is useful for 1:1 comparison of the newer original SPEI data with historical reconstruction data prior to 1950. It appears that the reconstruction process, even if otherwise valid, significantly underestimates natural variability. This is a common problem with proxy-based reconstruction. It results in an exaggerated level of volatility in the instrumental SPEI period after 1950 relative to the natural variation, so that normal fluctuations can appear anomalous.


MBF23 is a valuable paper dealing with the description of the variation of summer droughts in Europe since 1600. However, its title is “Tree ring isotopes. in Europe indicates a recent anomalous climate” as well as the statement in its Summary that “the recent European summer drought (2015–2018) was highly unusual in the context of many centuries” supported by the data used in the paper.

The lower temporal and spatial resolution of pre-1950 reconstructions compared to SPEIs determined after 1950 casts some doubt if the comparison of a number of years after 1950 with historical values reproduced is appropriate.

MBF23 should be corrected and renamed because several key conclusions, including the main statement in its title, are not supported by appropriate statistical analysis of the SPEI values ​​that our reproducible method they create. The recent European drought to 2018 is still within the natural range.


News7g: Update the world's latest breaking news online of the day, breaking news, politics, society today, international mainstream news .Updated news 24/7: Entertainment, the World everyday world. Hot news, images, video clips that are updated quickly and reliably

Related Articles

Back to top button