Is there any proof that PMOD is a valid correction to apply to the ACRIM dataset?

Question

Trevor, it's very simple what I am asking, if you choose to look at the Acrim Composite there is a (+.04%/decade) increase in TSI. Pmod shows a (-.01/Decade) gradient. ACRIM says that the warming observed over the last thirty years is primarily because of solar, and PMOD say's that it is probably Anthropogenic.

Answer 1

Scientists don't like words like "proof". How about "demonstration" or "indication"?

Satellite observations of TSI, while hugely better than anything done on earth, do need proper calibration to become useful data. Individual instruments on individual satellites are slightly different, which must be accounted for; and there is also a "break-in" period in the first few months in orbit when data from a new instrument is a bit flaky.

As it happens, two different scientists have done these calibration corrections in two different ways. Richard Willson has calibrated the data in a way that shows a (slight) gain in TSI from the 1987 solar minimum to the 1997 solar minimum. Claus Froelich and Judith Lean have looked at the data differently, and have calibrated the data in a way that shows no difference between these two minima.

Two key points here: although there is a difference in solar minima using the two different datasets, both datasets show a (slight) reduction in the solar maxima from 1980 to 1991 to 2000.

Second, even if Willson is right, and even if we ignore the decline in solar maxima and look only at the slight increase in solar minima, the difference is still not enough to account for the recent warmth. Willson's data shows the 1987 minimum at 1365.4 and the 1997 minimum at 1365.9 Wm^-2, a difference of 0.5 Wm^-2 per decade. (By way of comparison, a typical solar maximum has been about 1367 Wm^-2.)

Even if we were to extrapolate this single-cycle difference over 30 years -- which itself is a hugely invalid procedure, given that no two solar cycles are alike -- that would only amount to 1.5 Wm^-2 difference during that period. But that's measured in space, normal to the Sun; extrapolated over the earth's entire surface, we divide by 4 to get a difference of less than 0.4 Wm^-2 down here -- and that could account for no more than 0.2°C in temperature rise over 30 years, while the actual rise during the past 30 years has been more than twice that.

Answer 2

I'm not really sure what you're asking here, I don't know how you would prove the validity of a correction

PMOD and ACRIM are separate TSI composites. In bridging the Acrim Gap it could be said that ACRIM was applied to PMOD in that the relationship to the ERBE was applied thus creating a correlation.

There is a downward step between ACRIMs 1 and 2 of approx 0.5W/m2/yr at the point of the bridge which isn't apparent in IRMB or PMOD but the overall trends in the datesets are unchanged.

Whichever dataset(s) was used ro repair a two year break in any of the other datasets would have little difference unless you spliced someting to or from IRMB in the last few years.

- - - - - - -

RE YOUR ADDIT DETAILS

Following the Challenger disaster there was a delay in launching ACRIM 2, this led to a break in the ACRIM dataset of something like two years. The Acrim Gap has been bridged using data from other datasets, it wasn't that a section of PMOD was spliced into ACRIM. From your original question I thought you were implying that PMOD had been used to 'correct' ACRIM.

It seems what you're asking is which is the most reliable - ACRIM or PMOD.

It's not an easy question as the difference in readings is a margin of less than one thousandth and until ACRIM I it wasn't known that total solar radiant energy was a variable. We don't have much data to work with when trying to evaluate which is the most reliable, but in any event, the study of both datasets places the primary cause of global warming on factors other than the sun. Even using the ACRIM a maximum value of 25% is placed on solar activity and this is just speculation.

Bottom line, taking even the most significant increase in TSI shown by any dataset would only account for a small proportion of global warming.

Answer 3

Great question. Sensible skepticism, data based.

This particular detail is currently beyond me (but not likely beyond Trevor).

But here is an interesting site that claims it doesn't matter as much as you say, that either shows a mostly anthropological cause, although they do differ in the exact percentage.

http://tamino.wordpress.com/2007/07/24/pmod-vs-acrim/

And part II (the link is upper right) goes into more details about which this guy thinks is better.

When I get some time, I'll try to educate myself more on this issue. My WAG (wild ---- guess) is that he's at least somewhat right that this is not a make or break deal. My gut tells me there's more to this than the relatively minor difference of ACRIM versus PMOD.

Along that line, this comment struck a responsive chord in me:

"The fact that we’re debating whether the sun is showing a very slight upward trend or a very slight downward trend - in fact, if there’s a trend at all - underscores the stark divergence between steeply rising temperatures and solar activity."

In other words, even if there is a slight upward trend in TSI (per ACRIM), you'd still need some explanation of how it could possibly be enough to cause the observed temperature rise.

But there's more work for me to do on this, and I'm hardly sure about this one. Thanks for raising my awareness of the issue. If all discussions could be at this intellectual level about the nature of the data, we'd start getting somewhere.

EDIT - It seems that the leading proponent of using ACRIM without PMOD, Richard Willson, has published a paper in Science very recently. I can't find a copy (maybe it's not even out yet). But media reports say that it gives an upper bound on solar influence of 30%. Substantially more than the IPCC report, but not disproof of (mostly) anthropological global warming.

EDIT2 - Final answer for now. Major question back; Where did the idea that ACRIM = solar came from? What I get is this:

This is a real debate among climatologists. Those supporting the PMOD dataset seem to be in the majority, but I can't swear to it. This is a topic where you pretty much have to be a climatologist and a satellite data expert to have a meaningful opinion.

But, even the ACRIM dataset doesn't make global warming mostly solar. Even getting to 30% requires a few different error bars to be all considered to be on the solar side.

ACRIM doesn't change the Lockwood and Frolich paper's view that, recently, TSI is going down as temperature is going up, depending on exactly how you define "recently" and do the analysis. A few (3-10?) more years of data should settle that issue.

Still wins my vote as best skeptic question in a long time. I think it should have gotten more stars than just mine.

Answer 4

It is difficult to decide which of these datasets is correct. I'm persoanlly undecided because I think there are plausible reasons to support both.

My main point is really one of does it matter if you are only interested in proving a human influence upon climate? You have said in your question that choosing between ACRIMS and PMOD is one of choosing between a solar or anthropogenically forced climate, but I strongly disagree with this. Neither dataset displays a large trend ie. a trend that could explain much of the observed climatic changes.

You have said that ACRIMS displays a 0.04% increasing trend, but this is only over a period of 10 years which is only a small subset of the data. The trend over the wider dataset is one of 0.008%, which corresponds to a change in total solar irradiance of 0.31 Wm-2, which is climate forcing of 0.075 Wm-2 which, based on a climate sensitivity of 0.75oC/(Wm-2), equates to a temperature rise of 0.06 oC per decade. You therefore can not state that ACRIMS explains the observed warming. Putting your faith in ACRIMS at best describes only a quarter of what has been observed ie. 0.2oC per decade. This still leaves a largely unexplained warming trend if you discount human influences, and I don't.

Answer 5

Looks like you will continue to have a tough sell with those who firmly believe it's anthropogenic global warming. Sometimes it seems that folks treat this like its the US electoral college - it's an all or none proposition, one winner and one loser. Personally, I don't know enough about the collection of TSI data to say whether the correction to ACRIM data is valid, but if there is such a great discrepancy of data, then it seems only responsible to withhold judgment until such discrepancies are resolved. (Especially when it comes to TSI which ULTIMATELY accounts for all energy changes on the planet.)

Even a small increase in TSI has a multiplying effect when you factor in thermal accumulation of the atmosphere (both natural AND anthropogenic greenhouse gases and increased feedback) and ocean/land (both of which have a heat capacity orders of magnitude greater than the atmosphere) as well as orbital/Milankovik forcings.

Answer 6

here is some links:

http://www.pmodwrc.ch/pmod.php?topic=tsi/composite/SolarConstant
http://tamino.wordpress.com/2007/07/27/pmod-vs-acrim-part-2/

Answer 7

My opinion would be "no". You're comparing apples and oranges.