Is There an Ongoing Debate about the Impact of Solar Variability on Climate?

This is part 1 of  review of the claims in Connolly et al 2021. You can see part 2 here. A paper was published in 2021 arguing that scientists have been premature to rule out the possibility that the Sun is the dominant explanation for the current warming trend. One key element in this argument is that there is continued disagreement over the degree of solar variability in reconstructions of Total Solar Irradiance (TSI) since 1610.  Some reconstructions show "low solar variability," and if those are correct, then current warming may be best explained by anthropogenic forcings. However, if "high solar variability" reconstructions are correct, then solar variability may explain most of current warming. So Connolly et al argue that "the answer to our question 'How much has the Sun influenced Northern Hemisphere temperature trends?' depends substantially on which estimate of TSI we choose."[1] But is this correct?

Connolly et al Figure 2

Figure 2 in his paper (above) shows graphs from 8 papers reconstructing TSI from at least from 1850. These generally show the individual solar cycles without 11-year smoothing. At most, these reconstructions show a difference of 2 W/m^2 between the lowest and highest values since 1850. Any 11-year running average would show significantly less than 2 W/m^2 since 1850. Figure 3 shows 8 "high variability" estimates that range anywhere from 2.5 to 4 W/m^2 from the lowest and highest values since 1850. Again, any 11-year running average would show less variability. And all of these reconstructions show a significant decrease in TSI between 1850 and 1900 followed by increases through about 1960 to 2000 and then a decrease following 1960 to 2000. It would be generous to characterize these "high variability" estimates as showing a ~ 3 W/m^2 increase in TSI. In fact, most of them show TSI in 1850 to be not so different from current values - they just vary by a lot more between then and now.

Connolly Figure 3

Because of the disparity between these two sets of reconstructions, Connolly's paper argues that "different TSI estimates suggest everything from no role for the Sun in recent decades (implying that recent global warming is mostly human-caused) to most of the recent global warming being due to changes in solar activity (that is, that recent global warming is mostly natural)." What I want to do is consider whether this makes any sense at all. We can do a simple test to see if this claim holds water.

We've seen almost a 50% increase in CO2 since the beginning of the Industrial Revolution, producing an increase in radiative forcing just over 2 W/m^2 and other greenhouse gases (GHGs) total about half of that of CO2. So the total increase in radiative forcing from GHGs is ~3 W/m^2. This has been partly masked by the increase in anthropogenic aerosols, at ~0.8 W/m^2.[2] So using round numbers, we can say the increase in radiative forcing due to anthropogenic forcings is ~2.2 W/m^2.  If the "low solar variability" reconstructions are correct then these solar forcings are roughly negligible. The sun varies by perhaps 1 W/m^2, sometimes above average and sometimes below. If the "high solar variability" estimates are correct, and if I'm very generous, we can say there has been at most a 3 W/m^2 increase in TSI from solar variability. But in order to convert this increase in TSI to an increase in radiative forcing (RF), we need to do a little math.

RF = dTSI * (1-α)/4, where

    dTSI   = 3 W/m^2 increase in TSI

    α         = albedo = 0.306

RF = 3*(0.694)/4 = 0.52 W/m^2

This would mean the total increase in TSI at most has an effect that is smaller than forcing from anthropogenic aerosols. The total increase in RF would increase to 2.2 + 0.5 = 2.7 W/m^2. It would also mean that at most, solar variability accounts for 0.5/2.7 = 19% of the increase in RF (and therefore temperatures). There's no way that solar variability can be responsible for most of current warming given the empirical evidence we have, even if we grant the "high variability" reconstructions. What Connolly seems to want to argue is that the greater we estimate solar variability, then the climate system must be less sensitive to GHGs, and this is simply bad logic.[3] It requires an assumption that the earth's climate system is more sensitive to an increase in RF from solar forcings than it is to an increase in RF from GHGs. We already know this assumption is wrong. 

And to make matters worse for Connolly's argument, the "low solar variability" reconstructions are far more likely to be accurate.[3] Many of high variability reconstructions are based on outdated science comparing the Sun to other stars. Earlier comparisons concluded that solar variability was higher than what is actually warranted. And as our understanding of the evidence has improved, so has the accuracy of reconstructions. For comparison, see how Lean 2000[6] compares with Lean 2018.[5] There is much less variability in the later reconstruction.

When you plot changes in recent, more accurate TSI reconstructions[4] with temperature (HadCRUT5), both with 11-year smoothing, you can see the departure of TSI from temperature following 1960, as GHG emissions ramped up (see the graph at the top of this post). Solar variability simply can't explain warming since 1960. 

In the graph immediately above I took TSI data from Kopp at Colorado[4] and Lean 2018[5] and plotted them in terms of radiative forcing, where dTSI is the TSI in any given year minus TSI in 1610. I set "0" to be the 1850-1900 mean of these values. I also plotted the change in RF from CO2 alone for increases in CO2 above 280 ppm. As you can see, the overall impact of changes in TSI is negligible, while the impact of CO2 is substantial. I don't think Connolly's paper makes any sense at all.

Expected Temperature Change from Solar Variability Alone since 1610, Assuming ECS = 3 C.

Again above I plotted the radiative forcing from solar variability using Kopp's TSI data in terms of temperature, assuming ECS = 3 C, where 0 is the 1850-1900 mean. What should be clear is that solar variability can be responsible for less than 0.05 C warming. Given we've had about 1.2 C warming since the 1850-1900 mean, this amount from solar variability is basically negligible.

It turns out there isn't really any ongoing debate about the impact of changes in TSI on climate. Solar forcings are small compared to anthropogenic forcings. Of course, that doesn't mean that there isn't much more to learn about the Sun and the impacts on climate that it does have, but the hypothesis that the Sun explains a majority of warming in recent decades is no longer a tenable hypothesis.

References:

[1] Connolly, R., Soon, W., Connolly, M., Baliunas, S., Berglund, J., Butler, C. J., … Zhang, W. (2021). How much has the Sun influenced Northern Hemisphere temperature trends? An ongoing debate. Research in Astronomy and Astrophysics, 21(6), 131. doi:10.1088/1674-4527/21/6/131

[2] I justify these numbers in another blogpost. I'm just using round numbers here for the sake of argument.

[3] "But this is a fallacy. It is equivalent to arguing that if total caloric intake correlates to weight, that exercise can have no effect." Gavin Schmidt. "The Soon Fallacy." Realclimate. https://www.realclimate.org/index.php/archives/2015/02/the-soon-fallacy/

[4]  "Historical Total Solar Irradiance Reconstruction, Time Series." https://lasp.colorado.edu/lisird/data/historical_tsi/

[5] Lean, J. L. (2018). Estimating solar irradiance since 850 CE. Earth and Space Science, 5, 133– 149. https://doi.org/10.1002/2017EA000357

[6] Lean, J. 2000.Evolution of the Sun's Spectral Irradiance Since the Maunder Minimum. Geophysical Research Letters, Vol. 27, No. 16, pp. 2425-2428, Aug. 15, 2000.https://www1.ncdc.noaa.gov/pub/data/paleo/climate_forcing/solar_variability/lean2000_irradiance.txt