Sun's Activity Levels Reconstructed

neutron_p writes "An international team of scientists has reconstructed the Sun's activity over the last 11 millennia and forecasts decreased activity within a few decades. The activity of the Sun over the last 11,400 years, i.e., back to the end of the last ice age on Earth, has now for the first time been reconstructed quantitatively. The scientists have analyzed the radioactive isotopes in trees that lived thousands of years ago. As scientists report in the current issue of the science journal Nature, one needs to go back over 8,000 years in order to find a time when the Sun was, on average, as active as in the last 60 years."

Comments

Innumeracy warning!

[Engineer-Poet]

The Sun is approximately 4.5 billion years old. An increase in the average brightness of 30% over that time is equal to 6.7% per billion years, or .00000067% per century. If you think that this will become a measurable change in the next century, you obviously never learned anything worthwhile from a lab course (and nobody should trust you with numerical methods, either).

Re:Error Bars

[Idarubicin]

m not sure I trust their error bars (they appear on the second plot). Since they're using 10-year averages, they should be removing the effects of the solar cycle. But their sunspot number curves drop below 0 sunspots in several places. A negative number of sunspots is, obviously, unphysical. Also, their data is pretty wildly varying over short timescales (again, solar cycle should be removed) and doesn't match the actual sunspot records from 1610 on very well, either.

Unfortunately, most people don't have online access to Nature for the full text of the original article. (For those who do, it's here. If you actually check the data--they have a plot of just the last thousand years comparing various methods--the match with the records from 1610 on is actually quite good.

They also comment that, "The slightly negative values of the reconstructed SN [sunspot number] during the grand minima are an artefact; they are compatible with SN = 0 within the uncertainty of these reconstructions as indicated by the error bars." I'm not surprised that their calibration might be a bit off when they've had to extrapolate sunspot numbers lower or higher than we've seen with firm data in the last four hundred years.