Feb 26 2008
It’s the Sun. Again?
The sun crowd is back at it again. I guess because we’ve had a cold January and we’re currently at a minimum in the solar cycle that must mean the two are related. That’s bullocks of course. A few months ago it was the sun that was causing the warming. The sun has approximately 11 year cycles, so it can’t be both. So has the sun caused the recent warming, or the recent cooling? Neither or both? This post attempts to find out the answer.
I’ve previously shown that the effective temperature of a planet is equal to
|
where Ap is the albedo, Ip is the irradiance from a sun intercepted by the planet, and σ is the Stefan–Boltzmann constant. Since we have instruments that measure Ip, we can figure out Tp.
According the Greg’s TSI Page, the total solar irradiance since 1980 has been somewhere between 1360 and 1375 W/m2. However, if you look at that image, the measurements from the different satellites don’t match up. This is because making an absolute measurement of TSI is difficult to do. When aligned to a common scale, it look’s much better.
The measurements have been scaled to the TIM instrument. However, other scalings will have little difference than what is shown below.
|
At around the year 1980, the maximum TSI was 1362 W/m2. Since 1980, the TSI has fallen to about 1360 W/m2. Using the equation above, we can calculate the change in effective temperature due to this change in irradiance. In 1980, the effective temperature would be 254.62 Kelvin. In 2007, the effective temperature would be 254.53 Kelvin. This is assuming an albedo of 0.3. Thus, if the sun were the main driver of temperature changes since 1980, we would expect a decrease in temperatures.
That obviously hasn’t happened. By all accounts, the temperature since 1980 have risen. For instance, the group at UAH have shown that temperature have risen by an average of 0.14 degrees C per decade. Other global temperature trends are similar to UAH. I quote theirs because it’s usually the lowest, and it’s the easiest to find on the web. I’ll happily add other trends if provided.
Have I cherry-picked the dates to use as endpoints in the above analysis? Absolutely. But I chose them to show a worst case scenario. In actuality, the variations due to the solar cycle are less than the computed values above. This is most clearly seen when looking at the global average temperature graphs. If the sun is the main cause, we would expect to see a clearly defined 11 year signal. We don’t.
|
Image courtesy Anthony Watts at Watts Up With That.
In January 2008, there was a sharp drop in global temperatures or around 0.6C reported by at least 4 different groups. Using the equation above, we can see what change in irradiance must be required for this to be caused by solar variations. I’m going to estimate that the TSI for December 2007 to be 1360.5 W/m2. This equates to an effective temperature of 254.55K.
What irradiance would be needed to cause the temperature to decrease by 0.6K? This can be computed using the same equation above. I come up with a value of 1347.65 W/m2. When looking at the TSI plot above, we can see this is off the bottom of the chart by almost 10 W/m2. This is almost 5 times more than the variability from each solar cycle. Surely the recent cold cannot be caused by the sun.
Others seem concerned that solar cycle 24 should have started already, and that the lack of current sunspots seems to indicate that we’re doomed to another ice age. I’m not an astrophysicist, so the inner working of the sun are a mystery to me. However, even a cursory glace at the butterfly diagram shows that we aren’t “overdue” yet.
An interesting side note: the above inset image is of a transit of Mercury. The TSI page has total solar irradiance values from a transit of Venus, which show that there was a decrease of about 1 W/m2 during the event. This is about 1/2 the value of the solar cycle.
(No Ratings Yet)
Loading ...Related Posts:
6 Responses to “It’s the Sun. Again?”
To reduce spam, comments are automatically closed 30 days after the last comment. If you would like to comment on any closed thread, please use the contact form at the top of this page.
Atmoz - Thanks for the post, but think there might be a typo in your calculated effective temperatures for 1980 (254.53K) and 2007 (254.62K) since in the text you say 2007 should be lower than 1980 (which makes sense given the slightly lower TSI). Also, is there supposed to be a lag time on the response to changing TSI as there is with CO2? I would assume there is some short-term lag just based on the lag we see on a seasonal basis.
Regards,
Bob
[Reply: Oops. Thanks for the catch. I reversed the years.
Regarding lag time, I would suspect there would be since the oceans have such a large heat capacity. If the sun were a major source of climate change within the last few decades, I would still expect to see an ~11 year cycle in the temperature data - possibly lagged from from the TSI data. But we don't see this in the temperatures.]
It’s clear that the sunspot numbers in the two solar cycles covering the period 1975 to 1995 were higher than in the period 1995 to date.
Temperatures do seem to have increased from 1975 to 1998 and have levelled off since 1998 with a very recent drop.
The solar activity from 1975 to 1998 could well have been enough to produce a net global warming effect peaking in 1998 with the subsequent level of activity only enough to maintain a more or less stable global temperature.
If we now have a quieter solar spell that could indeed result in cooling.
The El Nino /La Nina events seem to follow rising and falling solar trends. Prior to 1975 back to about 1950 there was a slow cooling of the globe at the same time as we had lower sunspot numbers and more frequent La Nina events.
It seems to me that solar activity does seem to be the main driver but it is transmitted to the atmosphere via the ENSO process which smoothes out the 11 year solar cycle due to the delay in water heating and cooling which is much slower than the heating or cooling of air or land.
I don’t think that the absence of an 11 year signal in the temperature record is necessarily significant due to the slow heating and cooling of the oceans and the chaotic nature of the ocean/atmosphere interaction.
It’s the average level of solar activity (especially from 1975 to 1995) that counts. It is the average level from now on that will dictate cooling or renewed warming.
I would just add that, oddly, the effect of sunspot numbers (whether up or down) and therefore the overall activity of the sun does not seem to be fully reflected in the observed changes in the TSI.
I think it likely that the TSI numbers do not give a full account of the heating or cooling effect on the Earth of variations in sunspot numbers.
The reason for the apparent discrepancy requires further research but in the meantime the existence of such a discrepancy renders Atmoz’s initial post less useful for predictive purposes.
It should be possible to get a better grip on the issue if the current solar minmum lasts longer than previous ones and is seen to cause a measurable cooling effect.
The recent drop of about 0.5C globally might be relevant to the issue.
[Reply: The number of sunspots is a proxy for TSI, not the other way around.]
I didn’t think either TSI or sunspot numbers were a proxy for the other.
Can you elaborate ?
I would like to make one comment. You are looking at time domain data and state that clearly there is no 11 year cycle. As you know, any real world signal such as this one, is made up of a series of exponential functions. Often, time domain data will not expose the frequencies buried within. To find the actual frequencies that lie within the data a suggest performing a Discrete Fourier Transform on the data. Such a transform will give all of the frequencies found, and it will probably surprise you. The sampling theorem states that you must sample at 2 times the lowest frequency. So if your data is in years then the lowest detectable frequency will be 2 years.
Some frequencies will be very obvious. For instance, when one looks at the Vostok ice core data they will see a fundamental frequency but there are many other cycles buried in that data, hence the sawtooth configuration. Dr. Richard Muller has done a great deal of work in this area. http://muller.lbl.gov/
[...] drop of 0.6°C would require a dramatic reduction in Total Solar Irradiance (TSI). According to theoretical calculations at Atmoz, TSI would need to fall to 1347.65 W/m2 to produce a global cooling of 0.6°C. In other words, [...]