By Javier

In a previous article here, I showed how anyone with a computer with internet connection and Excel could check the basis of the 980-year periodicity in solar activity, known as the Eddy cycle, and its excellent match with a North Atlantic iceberg proxy record, known as the Bond series.

In this part I will show you how the same can be done with a proxy for the Asian Monsoon strength based on the deposition of δ¹⁸O in a speleothem from stalagmite DA from Dongge Cave in Southern China.

The scientific reference article for this is: “The Holocene Asian Monsoon: Links to Solar Changes and North Atlantic Climate,” by Wang et al., 2005, can be found here. It is a highly cited article, with 1750 citations in Google Scholar. This is by no means fringe science we are dealing with.

And the data for stalagmite DA can be found here.

A plot of this data with both axis inverted for convenience is shown:

Figure 1. Wang et al., 2005 δ¹⁸O data from stalagmite DA in Dongge Cave.

As you might remember from the previous article, the frequency analysis for this data was performed by Steinhilber et al., 2012 and is available here.

Figure 2. Steinhilber et al., 2012 Lomb normalized periodogram of total solar irradiance (a) and Asian climate record (δ¹⁸O) from Dongge cave, China (b). The horizontal line marks the 95% significance level.

For the analysis, it is better if you detrend the data from its long-term trend, which is caused by slow orbital changes in insolation. There are better ways of detrending data, and Wang et al., 2005 used singular spectrum analysis. But, you can use the Excel trendline tool that gives the formula for a third-degree polynomial, which in this case is:

y = 0.0000000000087*x^3 – 0.000000117*x^2 + 0.00022*x – 7.4517

Figure 3. Polynomial fit to Dongge Cave DA stalagmite data.

So, you now just subtract the polynomial from the data, and its new trend is flat.

Figure 4. Detrended Dongge Cave stalagmite DA data.

At this point Wang et al., 2005 compare their data with the atmospheric Δ¹⁴C record and notice the high correlation, particularly for the 9-6 ky BP.

Figure 5. Solar proxy and Dongge Cave DA stalagmite data comparison. Notice that the x-axis is inverted with respect to the other figures.

But you want to go further and check the millennial periodicity found in the Dongge cave data to see how well it agrees with the other two millennial periodicities, in solar activity and in North Atlantic iceberg activity.

For that you need to reduce the noise in the data, and you should run a 75-point moving average through the Dongge Cave stalagmite DA data. This is equivalent to a 150-225-year average that doesn’t affect millennial variability.

Figure 6. Detrended 75-point averaged Dongge Cave stalagmite DA data.

Now it is time to compare all the data. Remember that 980-yr sine function with the formula y = sin 2π/980(x) that matched solar variability? The function did not match solar variability in the 4100-1800 BP period where the millennial periodicity had low power in wavelet analysis. That period was marked with blue boxes in the figures in Part I. Here I have modified the function to better represent solar activity by decreasing its amplitude during that period. Also, the Bond data has been compressed by 1.7% in the temporal axis to correct for the evident age model drift discussed in the previous article.

Now let’s plot it all together.

Figure 7. The millennial cycle in solar activity (red, the 980-year Eddy solar cycle), North Atlantic iceberg activity (blue, inverted, LHS), and Asian Monsoon strength (black, RHS). Thick arrows show periods when low solar activity, increased iceberg activity, and a weaker monsoon coincided within dating uncertainty. The blue box covers the period when solar activity did not display a 980-year frequency peak. At that time the periodicity became lost also in iceberg activity and monsoon strength. RWP, Roman Warm Period. DACP, Dark Ages Cold Period. MWP, Medieval Warm Period. LIA, Little Ice Age. MGW, Modern Global Warming.

We can follow up on the conclusions from part I:

• There is a 980-year periodicity in solar activity cosmogenic isotope records, known as the Eddy cycle (Part I).

• This periodicity shows an excellent match with North Atlantic iceberg proxy records, known as the Bond series, except for a period ~ 4100-1800 BP (Part I).

• This periodicity also shows an excellent match with Asian Monsoon proxy records from Dongge Cave stalagmite DA, except for a period ~ 4100-1800 BP (Part II).

• The period of poor solar-climate match corresponds to a period when solar activity does not show a strong Eddy cycle periodicity, further reinforcing the solar-climate relationship (Parts I and II).

The evidence supports that the North Atlantic and Southern China display a millennial climate periodicity throughout most of the Holocene. This periodicity agrees well with historical and archeological evidence for the past 2000 years where a millennial periodicity can also be inferred. Reconstruction of solar activity from solar proxies displays the same periodicity in phase with the climatic periodicities. The most reasonable explanation is that small long-term changes in solar output of a periodic nature are responsible for the observed climatic effects. That we do not know what causes the periodicity or how the climate responds to long-term small changes in solar output is no excuse for ignoring the evidence.

The most important consequence for the present is that we are living through a period of several centuries when warming was expected from this millennial periodicity, and warming has occurred. Even if CO₂ emissions since the mid-20th century have contributed to the warming, the implications and consequences of the proposed anthropogenic effect on climate are radically different if a strong cyclical solar effect is responsible for part of the warming.