The world has never seen such freezing heat

[KrazyHorse]

There is no partial credit. You either write true statements or you write false ones.

[Nikolai]

Originally posted by KrazyHorse
There is no partial credit. You either write true statements or you write false ones.

¨Nobody said that what you've read is wrong.

I've claimed that some of your presentation of it has been wrong.

[KrazyHorse]

Originally posted by Naked Gents Rut
Apparently a team is monitoring the light reflected from Earth onto the dark portion of the moon to track changes in the Earth's albedo. Pretty cool.

They've found that albedo has been increasing since 2000. Clouds might have played an important role in this.

That's interesting, and I will read some of their stuff. My first question is how they accounted for the fact that the moon is in an equatorial (pretty much) orbit whereas the important changes in albedo due to ice loss are in circumpolar areas.

[KrazyHorse]

Originally posted by Nikolai

¨Nobody said that what you've read is wrong.

I've claimed that some of your presentation of it has been wrong.

Yes, and?

[Nikolai]

"Some"

Now I have to sleep. Looking forward to see what you've come up with tomorrow.

[KrazyHorse]

Now I'm beginning to think that you're simply having language difficulties.

[snoopy369]

I am positing that, in an inherently chaotic (over long periods of time chaotic) system, you cannot predict accurately 30 (or more likely 300) years from now what the average temperature over a decade will be, as well as you can predict what the avearge temperature for the next decade (or, frankly, the next week). You are comparing apples and oranges, as you well know

I would say that it's also very probable that a meteorologist could more accurately predict the temperature in Chicago on next Tuesday more closely than he could predict the weather in 2030-2040. Two weeks out, I don't know (simply because I don't know enough about meteorology to know about 2 weeks out predictions, and I don't ever check them that far out myself); but I still believe the accuracy is probably greater at some point, and particularly when you head out a good distance (say, a few hundred years) rather than a nearby year. We simply don't really understand all of the things that affect climate and climate change to accurately predict things at this point. We think we do, of course, but we thought the world was flat at one point, also ...

[KrazyHorse]

Originally posted by KrazyHorse


That's interesting, and I will read some of their stuff. My first question is how they accounted for the fact that the moon is in an equatorial (pretty much) orbit whereas the important changes in albedo due to ice loss are in circumpolar areas.

Okay. Scanned through the most recent paper they have listed

http://bbso.njit.edu/Research/EarthShine/literature/Palle_etal_2008_JGR.pdf

and they simply appear to not claim to measure it...

This may be a useful method to measure the albedo of temperate and tropical regions of the Earth's surface. It appears to be a poor method to measure the albedo of arctic regions.

[snoopy369]

Originally posted by KrazyHorse

Plus I don't buy the premise that he can predict 20 to 30 years from now as accurate as 2 weeks.

Law of large numbers.

Somehow I think the quality of the input, and the degree of chaotic effects, is rather important; as well as the understanding of the potential effects. But you're a physicist so those things don't matter, right?

[KrazyHorse]

Somehow I think you should reconsider engaging somebody who teaches statistical mechanics, has studied things like chaos theory and stochastic calculus and who does research using Monte Carlo simulation in this sort of debate.

[snoopy369]

None of that means you can't be told you're wrong if you're misrepresenting my arguments ...

I also work in the statistical field, as you should know, so I'm not entirely talking out of my ass here. But frankly, what I'm suggesting (which is clearly not what you're suggesting I'm suggesting, of course) is not something it should take a Ph.D. to understand...

[Naked Gents Rut]

My first question is how they accounted for the fact that the moon is in an equatorial (pretty much) orbit whereas the important changes in albedo due to ice loss are in circumpolar areas.

Found this...

Meanwhile, Koonin and Gordon MacDonald, an earth scientist at the University of California, San Diego, began exploring the question of whether earthshine is a reliable gauge of albedo. Koonin says he was skeptical when he first heard about Danjon's work from MacDonald in 1990, while the two of them were overseeing a review of a proposed DOE small satellite program to study climate change. "Basically, the moon samples the light from the equator. Light hitting the polar ice caps can scatter out in ways that never hit the moon at all. You worry about missing that light, and [whether] the earthshine might not be a good measure of albedo." But by building a computer model of how sunlight is reflected off Earth and the moon, says Koonin, "we proved it was. And we showed that if you could measure earthshine to 1%, you would know the albedo to 0.2%, which is a number quite competitive with satellites."

Koonin was also able to reconcile Danjon's calculation of an average albedo of 36% with satellite measurements showing a value closer to 30%. What Koonin knew, and what Danjon had failed to account for adequately, was that the moon demonstrates an opposition effect: Its reflectivity is highest when it is bouncing light straight back to the source, as it does with earthshine. Modern instruments, Koonin realized, provide a much more accurate estimate of the opposition effect. And, indeed, when he fed the modern data into Danjon's calculations, Danjon's albedo numbers dropped into the right ballpark.

[KrazyHorse]

Originally posted by snoopy369
None of that means you can't be told you're wrong if you're misrepresenting my arguments ...

I'm not misrepresenting them. You claimed that evidence of a shortish timescale result which is contrary to the predicted longish timescale trend implies that predicting longish timescale trends is difficult. That is retarded.

I also work in the statistical field, as you should know

I actually have no idea what you do. You haven't made that much of an impression. Sorry.

, so I'm not entirely talking out of my ass here. But frankly, what I'm suggesting (which is clearly not what you're suggesting I'm suggesting, of course) is not something it should take a Ph.D. to understand...

What you appear to be suggesting is that global climate on the timescale of decades is a chaotic system. I'd like to know how an observation on the timescale of a month allows you to claim that.

[KrazyHorse]

by building a computer model of how sunlight is reflected off Earth and the moon, says Koonin, "we proved it was"

Somebody's obviously thought about the issue...but this answer on its own is unsatisfying.

And we showed that if you could measure earthshine to 1%, you would know the albedo to 0.2%

Now, this is really the mind****. To a first approximation "earthshine" is linearly dependent on albedo. How the **** can you get less relative error on albedo than on earthshine? Since earthshine is a direct observation you would expect other errors, which would naively add in quadrature with earthshine error.

[Naked Gents Rut]

I might understand what you were saying if only I had stayed in my science major.

It was biology, though, so I probably wouldn't.