Sunspots In The Atmosphere | Watts Up With That?

Guest Post by Willis Eschenbach

Buoyed by the equal parts of derision and acclaim for my previous post, “CEEMD vs Joe Fourier“, I thought I’d take a look at a place where we know there is a solar effect.

How do I know there’s a solar effect somewhere? Because I’m a ham radio operator. Here’s my license, my call sign is H44WE, “Hotel 44 Whisky Echo”.

I got my license in the Solomon Islands at age 41, part of my commitment to life-long learning … plus I was living on a very remote tropical island and that was one of my few means of communication …

And any ham will tell you that long-distance ham radio transmission is strongly affected by the sun’s activity. Solar energy affects the ionosphere, the layer of the atmosphere that reflects radio waves back to the earth. So we know that the sun affects the very top of the atmosphere … but what about the lower levels?

To investigate this question, I thought I’d take a look at the solar effects at different layers of the atmosphere. The UAH MSU satellite temperatures cover four different levels of the atmosphere.

Figure 1. Regions of the atmosphere covered by the UAH MSU satellite temperature measurements. T—Temperature. LS—Lower Stratosphere. TS—Tropopause. MT—Middle Troposphere. LT—Lower Troposphere.

I started at the highest altitude, the lower stratosphere data, because that’s where I expected to find the solar signal. And indeed, both CEEMD and Fourier analyses verify that the sun affects the temperature of the lower stratosphere.

Figure 2a. CEEMD empirical mode containing any cycles around 11 years, for both the SILSO sunspot data and the UAH MSU lower stratospheric temperature data.

Figure 2b. Fourier periodograms for both the SILSO sunspot data and the UAH MSU lower stratospheric temperature data.

As you can see, both the CEEMD and the Fourier data clearly establish a relationship between variations in solar activity and variations in the lower stratospheric temperature. Both show peaks at around six and 11 year cycles, and the CEEMD data shows the effect on temperature varies with the strength of the solar signal and that the temperature variations lag the solar variations.

From there, I moved to the next lower layer, the tropopause. This is the layer dividing the stratosphere from the troposphere.

Figure 3a. CEEMD empirical mode containing any cycles around 11 years, for both the SILSO sunspot data and the UAH MSU tropopause temperature data.

Figure 3b. Fourier periodograms for both the SILSO sunspot data and the UAH MSU tropopause temperature data.

Here at the tropopause, while we still see the temperature peaks in the Fourier periodogram at six and eleven year cycles, they are much smaller than up in the lower stratosphere. Plus there are more cycles, at three and four years. And the CEEMD analysis shows that temperature and solar activity are much more loosely related.

Now, there’s not much atmospheric exchange between the stratosphere and the troposphere. They experience different winds and weather. Here’s the same analysis as above, but this time for the middle troposphere.

Figure 4a. CEEMD empirical mode containing any cycles around 11 years, for both the SILSO sunspot data and the UAH MSU middle troposphere temperature data.

Figure 4b. Fourier periodograms for both the SILSO sunspot data and the UAH MSU middle troposphere temperature data.

Once we’ve crossed the tropopause and are in the troposphere, the solar signal is basically gone. The CEEMD analysis shows no relationship between the two, and the Fourier periodogram shows the strongest temperature peak at four years. There’s no signal at either the six or the eleven year periods.

Finally, here’s the lowest level of the UAH MSU data, the lower troposphere.

Figure 4a. CEEMD empirical mode containing any cycles around 11 years, for both the SILSO sunspot data and the UAH MSU lower troposphere temperature data.

Figure 4b. Fourier periodograms for both the SILSO sunspot data and the UAH MSU middle troposphere temperature data.

Total disconnect. Here, there’s absolutely no relationship between the lower tropospheric temperature data and solar activity. There’s no sign that the variations in the sun are having any effect on the atmosphere near the surface.

CONCLUSIONS

First, this verifies that the two methods I’m using, CEEMD and Fourier analysis, are both able to reliably find a solar signal in climate data if one exists.

Second, this shows that the solar signal decreases as we move lower in the atmosphere, and by the time we reach the lowest level of the troposphere, the solar signal has disappeared entirely.

Further affiant sayeth not.

Best to everyone,

w.