Global Warming/Climate Change Discussion Thread

[Touring Mars]

Exactly my point about global mean temperatures - the further back you go, the more inaccurate you get, until we (very quickly) reach a point where we can say roughly what the climate was like, but not what the temperature was to any degree of meaningful (well, when placed into the context of a 1.0 degree variance over 125 years) confidence...

You do have a very good point and I accept that there are limitations to what the paleoclimate data tells us. The observed data, however, is far more accurate - atleast an order of magnitude more accurate (+/- 0.05 degree variance), and this actual (real) warming is not predicted by any climate models that fail to factor in anthropogenic forcings. (Note also that other paleoclimate data sets have significantly smaller errors associated with the last 400 years than the graph I posted the other day...)

But it's a highly dangerous position to take. We don't have a workable mechanism for how carbon dioxide causes global warming. We don't have a working knowledge of how much carbon dioxide causes how much heat absorption. We don't know how increasing carbon dioxide results in increasing global mean temperatures. The assumption is that it just does and that, since we can't explain what we're seeing from what we understand of natural sources (not much) then what mankind is doing must explain the rest. Even though we don't understand how what mankind is doing can explain the rest.

Although we could always use a better understanding of the mechanisms of how long-lasting greenhouse gases generate their warming effect, it's important that we don't ignore the fact that LLGHG's do increase radiative forcing. We also now know that certain aerosols have an opposite effect. Just because we may not know precisely how something is happening does not justify saying that it isn't happening. There is evidence to show that warming is happening and also evidence to show that the major natural forcings - solar variance and explosive volcanism - are not wholly responsible.

That notwithstanding, carbon dioxide emissions rose between 1940 and 1980 at almost the exact rate they have since and global mean temperature dipped. Meanwhile Amazon deforestation was at a higher rate, year-on-year, than at any point in history, save for 1995, 2002, 2003 and 2004. So that's unprecedented levels of anthropogenic forcings (GHG emissions, deforestation) but reduction in global mean temperatures...

We have to remember that what we are dealing with is a cumulative effect here. You are right to say that 1940-1980 had unprecedented levels of anthropogenic forcings - at the time. The fact that unprecedented levels of anthropogenic forcings also accompanied a period where little or no warming happened does not mean that anthropogenic forcings don't (or even can't) affect the climate. You'd also be right to say that 1970-2008 has seen unprecedented levels too - even compared to 1940-1980.

Comically, 1980 saw a massive drop in global mean temperature, following an unprecedented (outside of historical and geological record) natural release of carbon dioxide into the atmosphere as Mount St. Helens went mental.

A massive global drop... really? Although any explosive volcanic eruption invariably releases a massive amount of CO2, it also releases plenty of other things, including vast quantities of aerosols/sulfates/dust which are known to exert a negative radiative forcing effect. In the 20th Century, explosive volcanism has actually exerted an overall negative forcing effect on global climate.

The mountain still releases over 1,000 tons of carbon dioxide into the atmosphere every day - the equivalent of 6.4 MILLION vehicle kilometres for average cars. Every day.

However true this might be, all volcanoes are only responsible for the equivalent of ~1% of the CO2 from human activities. That said, natural CO2 emissions in total are far larger than CO2 from human activity - but that doesn't mean that CO2 from human activity is insignificant. Indeed, even if we produced zero GHG emissions (not possible or true, but bear with me), we'd still be able to change the CO2 content of the atmosphere by removing the ability of the Earth's surface to remove CO2 i.e. by deforestation. Contrary to what you said earlier, deforestation is a major anthropogenic forcing - so much so that deforestation accounts for something like 20-25% of all anthropogenic forcings... removing the Earth's ability to regulate CO2 levels via deforestation is a major problem IMO - far more serious than persuading a few residents of Chelsea to ride a bike.

 

[Famine]

Although we could always use a better understanding of the mechanisms of how long-lasting greenhouse gases generate their warming effect, it's important that we don't ignore the fact that LLGHG's do increase radiative forcing. We also now know that certain aerosols have an opposite effect. Just because we may not know precisely how something is happening does not justify saying that it isn't happening.

Which is something I've been careful not to do...

There is evidence to show that warming is happening and also evidence to show that the major natural forcings - solar variance and explosive volcanism - are not wholly responsible.

The assumption from this is that if what we know of non-anthropogenic factors cannot expain it, then anthropogenic factors must, and we backwards engineer just what those factors are from that point of view without any understanding of how that might happen or whether there are other non-anthropogenic factors we don't yet know of.

It's like Pluto. What we knew of the 8 planets at the time didn't explain the motion of the Solar System. So we came up with the idea that there was a 9th planet. And we discovered it. And then we learned more and found out that the 9th planet wasn't actually a planet at all, and there were other factors at play which explained the motion of the Solar System without the need for a 9th planet.


It is fundamentally dangerous to assume that the unknown factors total 1 - anthropocentric activities - and reverse-engineer the numbers to fit that assumption. That's not to say mankind bears no responsibility for effects we are seeing, but that if mankind does, we don't know to what level our responsibility is.

We have to remember that what we are dealing with is a cumulative effect here. You are right to say that 1940-1980 had unprecedented levels of anthropogenic forcings - at the time. The fact that unprecedented levels of anthropogenic forcings also accompanied a period where little or no warming happened does not mean that anthropogenic forcings don't (or even can't) affect the climate. You'd also be right to say that 1970-2008 has seen unprecedented levels too - even compared to 1940-1980.

Quite - but the increase is pretty much linear. The year-on-year increase through 1940-1980 is almost identical to the year-on-year increase from 1980 to the present day. The rate of deforestation is decreasing year-on-year - from 1970-1987 the deforestation stood nearly 20% higher than the 1988-2006 deforestation.

I guess the point is that global mean temperature increased in line with carbon dioxide and deforestation rates up until 1940 and then, despite still-increasing carbon dioxide and deforestation rates, it dropped. Then, in 1980, despite still-increasing carbon dioxide and decreasing deforestation rates, global mean temperature increased again. This would lead me to think that the three factors are not directly connected - at least in a causative correlatory relationship - unless you ascribe to a "threshold" or "tipping point" viewpoint, where a set carbon dioxide level and deforestation level must be reached. In which case, we're boned.


Still, all the forests on Earth don't have even a tenth of the carbon dioxide absorption potential of the seas. And since sea levels are slated to rise, win. More sea = more carbon dioxide absorption.


Incidentally, I don't really have time to hunt nor access it here, but wasn't there data which seemed to show that global mean temperature rises preceded atmospheric carbon dioxide concentration rises, rather then the other way round?

A massive global drop... really?

0.5 degrees in one year. Compare to a 0.6 degree rise since 1980...

sulfates

IUPAC be damned!

Contrary to what you said earlier, deforestation is a major anthropogenic forcing - so much so that deforestation accounts for something like 20-25% of all anthropogenic forcings... removing the Earth's ability to regulate CO2 levels via deforestation is a major problem IMO - far more serious than persuading a few residents of Chelsea to ride a bike.

Considering that forests have about 10% of the planet's carbon dioxide sink potential, that really puts just how insignificant anthopocentric sources really are. We remove about 0.5% of the planet's forests each year, and that accounts for at least one fifth of our climate changing ability?


Leading question: Wouldn't removal of green cover increase the planet's albedo and reduce the amount of absorbed radiation?

 

[ExigeEvan]

Leading question: Wouldn't removal of green cover increase the planet's albedo and reduce the amount of absorbed radiation?

In snowy areas, yes.
http://news.bbc.co.uk/1/hi/sci/tech/6540119.stm

 

[Sureboss]

Afaik isn't plankton a big 'absorber' of Co2? Ergo, couldn't we create more of it?

 

[Stevisiov]

Afaik isn't plankton a big 'absorber' of Co2? Ergo, couldn't we create more of it?

Yes, there is a big plankton factory in Swansea.

Alternatively we could declare open season on blue whale, although I doubt that would get us very far...

 

[ExigeEvan]

Yes, there is a big plankton factory in Swansea.

Alternatively we could declare open season on blue whale, although I doubt that would get us very far...

Is there!? I'll be damned if I know about it. Though it's probably part of the University.

 

[Stevisiov]

Is there!? I'll be damned if I know about it. Though it's probably part of the University.

Also goes by the name of DVLA Headquaters, though I am not sure if that is true...

 

[ExigeEvan]

Also goes by the name of DVLA Headquaters, though I am not sure if that is true...

Hey! My girlfriend works there!

 

[Touring Mars]

The assumption from this is that if what we know of non-anthropogenic factors cannot expain it, then anthropogenic factors must, and we backwards engineer just what those factors are from that point of view without any understanding of how that might happen or whether there are other non-anthropogenic factors we don't yet know of.

It is fundamentally dangerous to assume that the unknown factors total 1 - anthropocentric activities - and reverse-engineer the numbers to fit that assumption. That's not to say mankind bears no responsibility for effects we are seeing, but that if mankind does, we don't know to what level our responsibility is.

Although I agree that mishandling/misrepresenting the data is wrong in principle, I think that is pretty far from what is happening in the scientific literature. Significant progress has been made by the scientific community in recent years in all aspects of climate change - including uncertainties. The latest 1000 page dossier published by the IPCC - Climate Change 2007: The Physical Science Basis challenges the idea that we don't have "any understanding" of these processes, their effects, significance, magnitude or source. On the contrary, our understanding of all factors known to affect climate is increasing all the time, as is the confidence level in the assertion that anthropogenic forcings are mostly (not wholly) to blame for the current trend in warming.

I presume you mean it is wrong to assume that total forcings = known forcings, therefore it would be wrong to assume that the unknown component is insignificant? However true that may be, the reverse is also true.... assuming that there are significant unknown factors at play is a major assumption in itself, and one not founded on any evidence - indeed it is an assumption based entirely on a lack of evidence! There may well be natural (and anthropogenic) forcings that we are not aware of, and we also need to remember that what we do know carries with it a level of uncertainty. But conjuring up uncertainties out of thin air is as valid or as sensible as rejecting evidence just because it is not "100% certain".

Although I accept that some unknown forcings might exist, it's decidely Rumsfeldian to worry about them whilst ignoring the elephant in the room...

Quite - but the increase is pretty much linear. The year-on-year increase through 1940-1980 is almost identical to the year-on-year increase from 1980 to the present day. The rate of deforestation is decreasing year-on-year - from 1970-1987 the deforestation stood nearly 20% higher than the 1988-2006 deforestation.

I guess the point is that global mean temperature increased in line with carbon dioxide and deforestation rates up until 1940 and then, despite still-increasing carbon dioxide and deforestation rates, it dropped. Then, in 1980, despite still-increasing carbon dioxide and decreasing deforestation rates, global mean temperature increased again. This would lead me to think that the three factors are not directly connected - at least in a causative correlatory relationship - unless you ascribe to a "threshold" or "tipping point" viewpoint, where a set carbon dioxide level and deforestation level must be reached. In which case, we're boned.

I'd say that deforestation, CO2 levels and climate are intrinsically linked, but that deforestation is only a major problem when atmospheric CO2 levels are abnormally high (like now) and that deforestation continues to outstrip reforestation (reducing the CO2 scrubbing potential of the Earth's surface). Although deforestation rates may have slowed down, the amount of forest area continues to fall. Similarly, although CO2 emission rates may not have increased (AFAIK), CO2 levels are still increasing unabated.

Still, all the forests on Earth don't have even a tenth of the carbon dioxide absorption potential of the seas. And since sea levels are slated to rise, win. More sea = more carbon dioxide absorption.

And warmer water absorbs less CO2, so at the very least, there is a trade-off there...

Incidentally, I don't really have time to hunt nor access it here, but wasn't there data which seemed to show that global mean temperature rises preceded atmospheric carbon dioxide concentration rises, rather then the other way round?

The reality is that both things can happen - rising temperatures can increase C02 emissions (from natural sources) and CO2 emissions also increase temperature. The difference is that natural processes often have a long lag phase but anthropogenic CO2 emissions since ~1800 are much faster and larger, hence their effect has been larger and faster than natural processes.

0.5 degrees in one year. Compare to a 0.6 degree rise since 1980...

I don't think that was Mount St. Helens, but I do accept that it is possible for a volcanic eruption (such as the El Chichon eruption) to have a big affect on global climate... the fact that a single volcanic eruption can have such a massive and abrupt impact is undisputed - it depends entirely on the amount of stuff it spews out, and generally results in a cooling affect. We also have to be careful with phraseology here... The El Chichon event was reported to have had a cooling effect on global climate of between 0.2-0.5 degrees... but it didn't cause the globe to cool by 0.2-0.5 degrees in 1981! Yes, volcanoes can exert a big influence - but the effects are transitory. Despite several major eruptions in the 20th Century, the net (cooling) effect on global climate by explosive volcanism has been small.

IUPAC be damned!

Please accept my sincere apologies!

Considering that forests have about 10% of the planet's carbon dioxide sink potential, that really puts just how insignificant anthopocentric sources really are. We remove about 0.5% of the planet's forests each year, and that accounts for at least one fifth of our climate changing ability?

Depends on how you look at it - I'd say it shows how drastic an effect small, manmade changes can be. Yes, deforestation is ~20-25% of anthropogenic forcing - but tree-removal is but one aspect of deforestation. Tree burning is a massive source of CO2 emission, on top of the removal of CO2-scrubbing effect tree removal has. Changing land-usage is another major factor bundled with deforestation.

Leading question: Wouldn't removal of green cover increase the planet's albedo and reduce the amount of absorbed radiation?

It could, but there is little evidence to suggest that this is a significant factor. Deforestation has multiple effects, and on balance, the increase in radiative forcing attributable to decreased CO2 uptake by plant matter, increased CO2 emissions from tree-burning and change of land-use as a direct result of deforestation (increased argiculture/habitation etc.) is much higher than the relatively insignificant changes in albedo caused by deforestation.

 

[Danoff]

So what's this I'm hearing about cooling until 2015? Some climatologists are claiming that we're entering a cooling trend (I gather this is a retroactive attempt to explain why temperatures have gone the opposite direction from what was expected).

Anyone else heard about this?

 

[Sureboss]

I have, can't remember where though. Something about an 11 year solar cycle?

 

[Touring Mars]

No, I don't think this has anything to do with solar cycles... a paper published in this week's Nature says that:

Our results suggest that global surface temperature may not increase over the next decade, as natural climate variations in the North Atlantic and tropical Pacific temporarily offset the projected anthropogenic warming.

Although it appears to support a viewpoint that projected increases in global temperature are wrong, it's not a cause for anthropogenic climate change deniers to rejoice - ironically, this new model which predicts a temporary offset is also based on/incorporates projected anthropogenic forcings... and as such, they predict that this temporary offset will end around 2014-2015. Effectively they are saying that there are some predictable short term variations which will reduce the overall effect of current and projected anthropogenic forcings, but that these variations are only short term and will not change the overall, long term effects of anthropogenic forcings..

What the paper doesn't say is that we can expect "cooling", but a temporary cooling effect to offset predicted warming effects... (i.e. a relatively short period of less or no overall warming where a warming trend was previously predicted) but not a stoppage or even a reversal of the overall warming trend.

 

[ExigeEvan]

Global warming may 'stop', scientists predict
By Charles Clover, Environment Editor
Last Updated: 6:01pm BST 30/04/2008

Global warming will stop until at least 2015 because of natural variations in the climate, scientists have said.

Researchers studying long-term changes in sea temperatures said they now expect a "lull" for up to a decade while natural variations in climate cancel out the increases caused by man-made greenhouse gas emissions.


Melting icebergs: The study predicts the IPCC's 0.3ºC temperature rise for the next decade may not happen
The average temperature of the sea around Europe and North America is expected to cool slightly over the decade while the tropical Pacific remains unchanged.

This would mean that the 0.3°C global average temperature rise which has been predicted for the next decade by the UN's Intergovernmental Panel on Climate Change may not happen, according to the paper published in the scientific journal Nature.

However, the effect of rising fossil fuel emissions will mean that warming will accelerate again after 2015 when natural trends in the oceans veer back towards warming, according to the computer model.

Noel Keenlyside of the Leibniz Institute of Marine Sciences, Kiel, Germany, said: "The IPCC would predict a 0.3°C warming over the next decade. Our prediction is that there will be no warming until 2015 but it will pick up after that."

advertisement
He stressed that the results were just the initial findings from a new computer model of how the oceans behave over decades and it would be wholly misleading to infer that global warming, in the sense of the enhanced greenhouse effect from increased carbon emissions, had gone away.

The IPCC currently does not include in its models actual records of such events as the strength of the Gulf Stream and the El Nino cyclical warming event in the Pacific, which are known to have been behind the warmest year ever recorded in 1998.

Today's paper in Nature tries to simulate the variability of these events and longer cycles, such as the giant ocean "conveyor belt" known as the meridional overturning circulation (MOC), which brings warm water north into the North East Atlantic.

This has a 70 to 80-year cycle and when the circulation is strong, it creates warmer temperatures in Europe. When it is weak, as it will be over the next decade, temperatures fall. Scientists think that variations of this kind could partly explain the cooling of global average temperatures between the 1940s and 1970s after which temperatures rose again.

Global warming forecast predicts rise in 2014
Writing in Nature, the scientists said: "Our results suggest that global surface temperature may not increase over the next decade, as natural climate variations in the North Atlantic and tropical Pacific temporarily offset the projected anthropogenic [manmade] warming."

The study shows a more pronounced weakening effect than the Met Office's Hadley Centre, which last year predicted that global warming would slow until 2009 and pick up after that, with half the years after 2009 being warmer than the warmest year on record, 1998.

Commenting on the new study, Richard Wood of the Hadley Centre said the model suggested the weakening of the MOC would have a cooling effect around the North Atlantic.

"Such a cooling could temporarily offset the longer-term warming trend from increasing levels of greenhouse gases in the atmosphere.

"That emphasises once again the need to consider climate variability and climate change together when making predictions over timescales of decades."

But he said the use of just sea surface temperatures might not accurately reflect the state of the MOC, which was several miles deep and dependent on factors besides temperatures, such as salt content, which were included in the Met Office Hadley Centre model.

If the model could accurately forecast other variables besides temperature, such as rainfall, it would be increasingly useful, but climate predictions for a decade ahead would always be to some extent uncertain, he added.

Make what you will of it.

Personally, the earth is merely doing what it has done since its creation, changing.

 

[Danoff]

Although it appears to support a viewpoint that projected increases in global temperature are wrong,

It actually does support that viewpoint, given that it wasn't long ago that projections for 2015 were gloom-and-doom. It supports the viewpoint that pretty much all previous projections were wrong given that pretty much all previous projections were for an increase.

Not that I trust this report either. Just that it does support that viewpoint.

this new model which predicts a temporary offset is also based on/incorporates projected anthropogenic forcings... and as such, they predict that this temporary offset will end around 2014-2015. Effectively they are saying that there are some predictable short term variations which will reduce the overall effect of current and projected anthropogenic forcings, but that these variations are only short term and will not change the overall, long term effects of anthropogenic forcings..

And where was this last year? We have a cooler year and suddenly we get downturn in the forecast. Coincidence?

What the paper doesn't say is that we can expect "cooling", but a temporary cooling effect to offset predicted warming effects... (i.e. a relatively short period of less or no overall warming where a warming trend was previously predicted) but not a stoppage or even a reversal of the overall warming trend.

As of last year they weren't predicting this. This year, the temperatures don't turn out exactly they way they were expected so someone tweaks their model slightly and we get a cooler forecast. It's all reactionary. Where is the empirical validation of the models? Or are we just going to generate enough models that one of them is bound to be right?

 

[Touring Mars]

It's all reactionary.

In many regards, all climate models are - or atleast should be - "reactionary" in that they should be informed by the very latest observations... however, I don't think that this model is somehow an attempt by the scientific community to backpedal on the longer term predictions of climate change. What is fairly clear from the article is that this new model merely predicts that a well-known (short-term) periodic variation may offset the predicted warming (atleast in the Northern Hemisphere), but at the same time does not disagree with the longer term predictions of an increasing warming trend governed largely by anthropogenic forcings.

Where is the empirical validation of the models?

In this case, the model uses different initial states from which climate modelling begins. By using earlier states, they can make "predictions" about what should happen to global climate, and then test the model against actual observations. Subsequently, using current conditions as their initial state, they can make predictions about what hasn't happened yet. Although it has to be said, predicting short-term climate variability is inherently prone to unknowns. Also, their model is based on some fairly large assumptions, so it will be interesting to see just how accurate they turn out to be. Suffice it to say that it is just one more model, and even if they are bang on the money, it is still not great news for us.

Personally, the earth is merely doing what it has done since its creation, changing.

The question is, how much of an effect are manmade influences also having, over and above what we know nature can do all by itself?

There are any number of natural processes that cause variations in climate - some periodic and predictable, like the solar cycle, and some which are not predictable but whose net previous effects are quantifiable, like volcanic activity. For example, you cannot predict what influence a volcano will have (or when it will have it), but we can quantify how much of an effect previous volcanic activity has had on current climate... In order to "predict" the unpredictable, climate models may assume a "Business As Usual" approach, such that in the case of volcanic activity, there is no reason to expect that 2008-2108 will have any more or less climate forcing from volcanoes than say 1908-2008. But there could be a volcanic eruption every day from now until next year, and that would have a dramatic effect on climate that no current model could possibly have predicted. However, observations tell us that although this is possible, it is highly improbable...

Although certain things are inherently unpredictable, nature does react to short-term forcings and the net long term effects of volcanic activity or solar cycles is, in reality, only very small. That's not to say that individual events cannot have huge/dramatic effects of global climate (and there is plenty of evidence of such things in the past), but it is to say that in a period of relative normality in terms of the behavioural patterns of natural phenomena such as these, the current trend of warming can only be accurately modelled when manmade influences are also taken into consideration. Unfortunately, manmade influences on climate are a double whammy, since we not only produce a continuous positive forcing (by emitting greenhouse gases, and at an increasing rate) but are also reducing nature's capability of dealing with it at an increasing rate i.e. by removing the ability of the planet to remove excess levels of CO2 through deforestation etc.

 

[Danoff]

In many regards, all climate models are - or atleast should be - "reactionary" in that they should be informed by the very latest observations... however, I don't think that this model is somehow an attempt by the scientific community to backpedal on the longer term predictions of climate change.

Agreed. This is not a backpedal on longer term predictions. It's a backpedal on mid-range predictions. At work I perform a very similar function to what you're discussing. I do routine position/velocity determination of spacecraft. Periodically I get a new batch of data, and I have to use that, combined with all past data to update my prediction of where the spacecraft is headed. It's very much like getting new temperature readings and updating your prediction of where the climate is headed - even if you're not changing your model.

But at some point you have to validate your model. At some point you have to say "how well did we predict this". It's hard to do any worse than getting the sign wrong. I very strongly believe that we need to validate our models before we follow them.

What is fairly clear from the article is that this new model merely predicts that a well-known (short-term) periodic variation may offset the predicted warming (atleast in the Northern Hemisphere), but at the same time does not disagree with the longer term predictions of an increasing warming trend governed largely by anthropogenic forcings.

If it was so well-known, why is it suddenly changing the predictions so radically?

In this case, the model uses different initial states from which climate modelling begins. By using earlier states, they can make "predictions" about what should happen to global climate, and then test the model against actual observations. Subsequently, using current conditions as their initial state, they can make predictions about what hasn't happened yet.

I understand this. I literally do it every day. But they never hold their current predictions to the fire. They're constantly changing and refitting. The fact that they haven't been able to determine the sign of the trend for a ~10 year period casts a great deal of doubt on their models.

...their model is based on some fairly large assumptions, so it will be interesting to see just how accurate they turn out to be.

Not only interesting - crucial for establishing credibility.

But there could be a volcanic eruption every day from now until next year, and that would have a dramatic effect on climate that no current model could possibly have predicted. However, observations tell us that although this is possible, it is highly improbable...

Granted that we're dealing with a chaotic system. All the more reason that it's difficult to understand that system well enough to extract meaningful estimates of our influence on it.

the current trend of warming can only be accurately modelled when manmade influences are also taken into consideration.

...based on our understanding of the natural influences - which have been shown to be incorrect time and again.

Unfortunately, manmade influences on climate are a double whammy, since we not only produce a continuous positive forcing (by emitting greenhouse gases, and at an increasing rate) but are also reducing nature's capability of dealing with it at an increasing rate i.e. by removing the ability of the planet to remove excess levels of CO2 through deforestation etc.

Most of the world's oxygen production doesn't come from forests (as I'm sure you know). So deforestation may play a minor role. My contention is that we don't really understand the natural phenomena well enough to determine our perturbation to the end result. We could be having a massive effect on the climate, or we could be having a nearly insignificant effect on the climate. But until we can start actually validating these models with predictions that actually come true (rather than "validating them" by fitting past data), I'm don't see any reason to trust their predictions. Especially when they've proven that they can be wrong not just significantly, but as wrong as is actually theoretically possible - by getting the direction of the trend incorrect over a period of time that lasts for about half of the length of time that we've got accurate data.

No, I'm not going to put faith in these people. They need to demonstrate some sort of credibility.

 

[Touring Mars]

If it was so well-known, why is it suddenly changing the predictions so radically?

What I meant by that is that the phenomena are well known, and are understood to play a role in short term climate variability. Clearly, making predictions about how these phenomena affect climate is not well understood, however this model is an attempt to assess whether or not they are able to predict future climate forcings from these period cycles...

The predictions made by this single group suggest that short-term variability within these systems may produce an offsetting of the predicted warming for the next few years, but then warming will resume - that's not "radically" different to other, existing predictions.... It reinforces the fact that short-term variations can influence climate quite markedly, whilst remaining distinct from (or largely unaffecting of) long term trends...

they've proven that they can be wrong not just significantly, but as wrong as is actually theoretically possible - by getting the direction of the trend incorrect over a period of time that lasts for about half of the length of time that we've got accurate data.

No, I'm not going to put faith in these people. They need to demonstrate some sort of credibility.

A few points you've just made rely on your rather contentious assumption that the current trend in global mean temperatures is downward... it's not... the 11-year averaged trend since the 1970's is decidedly upward, and even with this single new prediction that natural variability may offset projected increases for the next 7 or 8 years, it won't stop or reverse the longer term upward trend. One reason for this is that the processes being described/predicted in this new report are known to be periodic, and with little or no net forcing on climate over the long term, but do have noticable short term effects. Anthropogenic forcings, on the other hand, are inherently long term in nature because a) they are not being remedied by nature or man and b) they are not being halted or even reduced...

This makes sense to me from a strictly logical point of view. I also think that the observed data bears this out, and in the process validates the majority of climate predictions made during the last several decades which predicted anthropogenic climate change.

 

[Danoff]

What I meant by that is that the phenomena are well known, and are understood to play a role in short term climate variability. Clearly, making predictions about how these phenomena affect climate is not well understood,

You call it short term, but the cycle of this particular phenomenon is on the order of the length of time for which we have super-accurate climate data. How many other phenomena have we mis-modeled when calculating the human contribution.

What I don't like about this analysis is that there doesn't seem to be a term used to absorb and identify error, or, rather, that term seems to be "anthropogenic forcing". In otherwords, if it doesn't fit into our current models, it's caused by humans. That sort of thing ignores the notion that we might have misrepresented some natural phenomenon.

however this model is an attempt to assess whether or not they are able to predict future climate forcings from these period cycles...

...and... this model claims that the previous model was incorrect in it's usage of this particular climate forcing. At some point one has to wonder whether we understand the natural forcings sufficiently to categorize the human forcing.

The predictions made by this single group suggest that short-term variability within these systems may produce an offsetting of the predicted warming for the next few years, but then warming will resume - that's not "radically" different to other, existing predictions....

Yes it is. The trend over a ~10 year period is suddenly being revised. That means this phenomenon was either neglected, incorrectly modeled, or incorrectly initialized in the previous estimates with a massive change in the end result. Any of those errors can result in an incorrect estimate of human forcing. And one has to wonder how many other phenomena are so poorly understood.

It reinforces the fact that short-term variations can influence climate quite markedly, whilst remaining distinct from (or largely unaffecting of) long term trends...

It reinforces the fact that we don't understand our climate to the decade level well enough to even identify the sign of the temperature change over that time period. But our understanding the climate over a comparable timespan is something we've been relying on to determine anthropogenic forcing...

A few points you've just made rely on your rather contentious assumption that the current trend in global mean temperatures is downward... it's not... the 11-year averaged trend since the 1970's is decidedly upward,

Look at the last ~5 years on page 5 and tell me that you see an upward trend. The current short-term trend is downward, and the current forecast from this particular study is that it will continue downward mildly.

and even with this single new prediction that natural variability may offset projected increases for the next 7 or 8 years, it won't stop or reverse the longer term upward trend.

...assuming that our understanding of the climate data is sufficient to determine how much of an effect that humans are having. I'm claiming that this revision underscores the fact that we don't understand our climate well enough to determine much of anything.

One reason for this is that the processes being described/predicted in this new report are known to be periodic, and with little or no net forcing on climate over the long term, but do have noticable short term effects. Anthropogenic forcings, on the other hand, are inherently long term in nature because a) they are not being remedied by nature or man and b) they are not being halted or even reduced...

This makes sense to me from a strictly logical point of view. I also think that the observed data bears this out, and in the process validates the majority of climate predictions made during the last several decades which predicted anthropogenic climate change.

Show me a study from 1980 that accurately predicted the climate change over the following 25 years. Show me one from 1990. The current thinking is that last year's model of 2015 is wrong. There has not been any prediction-based validation of any of these models that I'm aware of.

Once again, I see no reason to subscribe to their conclusions until they've shown some a larger degree of credibility. Predicting the weather is not easy. I don't envy their task. But if they want to make big world-changing conclusions they need big credentials.

 

[Touring Mars]

You call it short term, but the cycle of this particular phenomenon is on the order of the length of time for which we have super-accurate climate data. How many other phenomena have we mis-modeled when calculating the human contribution.

I think it's important to remember the difference between short-term variability and the long-term trend. Although the behaviour of the Pacific Decadal Oscillation does show some correlation to short-term climate variability, it doesn't show a correlation in the long term trend. This new prediction is saying that we can expect to observe such a short-term effect over the next few years, but won't have an impact on the long term warming trend.

In otherwords, if it doesn't fit into our current models, it's caused by humans.

That's a pretty big misrepresentation of how anthropogenic influences on climate are assigned...

At some point one has to wonder whether we understand the natural forcings sufficiently to categorize the human forcing.

It is not just our understanding of natural forcings that give us an insight into human forcings... Indeed, improved understanding of either helps to better understand the other. The case of atmospheric CO2 levels is a good analogy. CO2 levels have been rising exponentially since ~1800. We know that CO2 levels in the atmosphere vary naturally, sometimes drastically, but generally natural fluctuations are gradual - of course there are short-term variations which can buck the trend (like a large volcanic eruption), but the long term trend remains nevertheless. We also know that human activity adds to CO2 in the atmosphere, and we suspect that human activity is responsible for the majority of the increase in CO2 in the atmosphere in recent times. Since burning fossil fuels results in a detectable change in the isotopic ratio of C13/C12 in the atmosphere, we have a means of quantifying how much atmospheric CO2 we are directly responsible for, arguably a far more tangible method than the simple assertion that "Total - Natural = Us"...

Yes it is. The trend over a ~10 year period is suddenly being revised. That means this phenomenon was either neglected, incorrectly modeled, or incorrectly initialized in the previous estimates with a massive change in the end result. Any of those errors can result in an incorrect estimate of human forcing. And one has to wonder how many other phenomena are so poorly understood.

It reinforces the fact that we don't understand our climate to the decade level well enough to even identify the sign of the temperature change over that time period. But our understanding the climate over a comparable timespan is something we've been relying on to determine anthropogenic forcing...

Look at the last ~5 years on page 5 and tell me that you see an upward trend. The current short-term trend is downward, and the current forecast from this particular study is that it will continue downward mildly.

A couple of significant things to note about that graph is that a) 10 years is the minimum period by which a "trend" can be established and b) trends are established by using 11 year averaged means... this is why the trend lines in Figure 1 (Page 3) stop with 5 years left unplotted... Despite the appearance of the graph you highlight (which supports your claim better than the other two graphs in Figure 1), the trend as far as it can be accurately ascribed is still upward! This is a good example of why we need to be careful about what we call a trend... Your analysis of that graph gives the misleading impression that there is a downward "trend", when infact there isn't one. And besides, the years 2001-2007 were the hottest in 150 years (except for 1998)..., not exactly a ringing endorsement of the claim that 1998-2007 represents evidence that global warming is in decline!

I'm claiming that this revision underscores the fact that we don't understand our climate well enough to determine much of anything.

Evidently, I don't share your skepticism/pessimism! What this new prediction says to me is that our understanding of climate is improving all the time - that we are on the brink of being able to predict short-term variability as well as the effects of short-term variations on long term trends. This model doesn't disagree with the central assertion of AGW theory that the long term trend is upward, despite a prediction that we can expect a short term hiatus.

I get your point that it is damn near impossible to say with much certainty whether 2015 will be warmer than 2008 or not, but it is well known that short-term variations are much harder to predict/model than long term trends. But the inability to accurately model short term effects doesn't mean that long term predictions are wrong...

Show me a study from 1980 that accurately predicted the climate change over the following 25 years. Show me one from 1990. The current thinking is that last year's model of 2015 is wrong. There has not been any prediction-based validation of any of these models that I'm aware of.

How about Broecker, Science, Vol. 189, 1975... In his paper, there is an interesting table which predicts atmospheric CO2 levels between 2000-2010 to be between 373 and 403 ppm... current levels are ~383 ppm... His paper also specifically stated the following:

a strong case can be made that the present cooling trend will, within a decade or so, give rise to a pronounced warming induced by carbon dioxide.

Is this not an example of a validated prediction?

 

[Danoff]

That's a pretty big misrepresentation of how anthropogenic influences on climate are assigned...

Is it? I don't see an error term in these analyses, but there has to be one. I don't see "unmodeled forcing" anywhere. Where do you think it gets put?

but it is well known that short-term variations are much harder to predict/model than long term trends.

Interesting statement given that long term trends have never been accurately predicted.

Is this not an example of a validated prediction?

The prediction (which was vague) was validated, but I was looking for something a little more concrete than a prediction that the sign of the trend would change at some point in the future (which was very safe). I was looking for something that might really lend credibility to our climate models.

Edit:

b) trends are established by using 11 year averaged means...

I wasn't using "trend" in the technical sense.

 

[Touring Mars]

Is it? I don't see an error term in these analyses, but there has to be one. I don't see "unmodeled forcing" anywhere. Where do you think it gets put?

There are errors associated with all types of climate forcings, but it's important to clarify what is meant by "unmodeled forcings" nevertheless. As I mentioned on the previous page, it is not valid to produce uncertainties out of nowhere. The fact that uncertainties exist is only to be expected, but they are taken into account by all current attribution studies. As a result, conclusions drawn from climate change attribution studies come laden with likelihoods, not simply bold statements of facts. Reading the summary of "Understanding and Attributing Climate Change" by Hegerl and Zwiers (2007) shows that recent observed changes in climate are best explained by the inclusion of the predicted influence of anthropogenic forcings... Although there may well be "unknown" and/or "unmodeled" factors influencing global climate, they have to be put in perspective and measured against what we do know.

The prediction (which was vague) was validated, but I was looking for something a little more concrete than a prediction that the sign of the trend would change at some point in the future (which was very safe). I was looking for something that might really lend credibility to our climate models.

I think vague is the wrong word - simple perhaps, but not vague. It was an unequivocal prediction, that projected rises in CO2 levels (which were surprisingly accurate) due to anthropogenic emissions would result in a climate forcing capable of becoming the dominant climate forcing, and resulting in a warming trend that otherwise we wouldn't have expected to see.

 

[Danoff]

I think vague is the wrong word - simple perhaps, but not vague. It was an unequivocal prediction, that projected rises in CO2 levels (which were surprisingly accurate) due to anthropogenic emissions would result in a climate forcing capable of becoming the dominant climate forcing, and resulting in a warming trend that otherwise we wouldn't have expected to see.

Does that do it for you? Is that enough validation for you? Because it seems pretty weak to me.

 

[Touring Mars]

Does that do it for you? Is that enough validation for you? Because it seems pretty weak to me.

It may seem "weak" to you, but it begs the question what wouldn't seem weak to you... that was an example of a prediction, from 1975, that you specifically asked for as an example of a climate prediction that proved to be correct... "weak", "vague" or otherwise, it answers the question you asked, does it not? And significantly, the literature shows that even as far ago as the late 70's, the majority of scientific models were already predicting a warming trend, so the argument that "one of them was bound to be correct" is not valid... If anything, not only have the majority of predictions made in the 1970's turned out to be correct (in that the majority of papers published in the 70's on climate trends predicted a warming trend rather than otherwise), but the accuracy of predictions since then also has increased...

In terms of the prediction made by that model alone, yes, it is a validation.. that's not to say that this single validation means that AGW theory is entirely correct. However it does show that predictions based on models that do not take anthropogenic forcings into consideration are considerably more unreliable than models that do...

 

[Danoff]

It may seem "weak" to you, but it begs the question what wouldn't seem weak to you...

Oh I don't know... a mean temperature prediction for even a long time period to within 10% of the predicted change. Hell, I'd take 50%. What would seem weak to you? Because they had a 50/50 shot at getting the sign right based on a dartboard.

that was an example of a prediction, from 1975, that you specifically asked for as an example of a climate prediction that proved to be correct... "weak", "vague" or otherwise, it answers the question you asked, does it not?

Technically. Not really meaningfully. The spirit of the question was the search for some sort of model validation. A paper in the 70's saying they expected warming of an undisclosed magnitude at some undisclosed time in the future doesn't really do that. Especially given that according to current models we're entering into a mild cooling period - which the folks in the 70's would not have predicted based on our carbon output. I can say that with some confidence given that the folks from the 2007s also didn't predict it.

And significantly, the literature shows that even as far ago as the late 70's, the majority of scientific models were already predicting a warming trend, so the argument that "one of them was bound to be correct" is not valid...

All of them anthropogenic? Everyone thought that humanity was causing global warming in the 70s? Because I seem to remember there being a cooling scare from the 70's.

If anything, not only have the majority of predictions made in the 1970's turned out to be correct (in that the majority of papers published in the 70's on climate trends predicted a warming trend rather than otherwise), but the accuracy of predictions since then also has increased...

To what exactly? Because so far the accuracy I've seen has been quite poor. The 10% I was looking for is nowhere in sight. Not only do they not have a stranglehold to within 10%, they don't have it to within 100% and can't even decide on a direction.

And it is not uncommon for climate scientists to not agree on the direction that a particular phenomenon will send us. I attended a research talk by a notable climate expert who explained that they're still uncertain whether clouds have an overall cooling or warming effect on the climate.... clouds. That's pretty freaking basic.

I wonder if you're defending against my position and so I don't get to hear your concerns about some of this research, of if you truly believe that these guys know exactly what they're doing. But from my point of view, the science is far from settled. I'll know when it's settled because I'll stop seeing their predictions proven wrong.

Way back it was global cooling - wrong.
More recently it's been global warming - temporarily wrong*
Hurricanes will be more common and more severe - totally wrong
Not to mention that some of the hottest days on record turned out to be erroneous.

*according to some research

 

[Touring Mars]

Because they had a 50/50 shot at getting the sign right based on a dartboard.

If by that you mean any climate model has a 50% chance of being right, then how do you explain the fact that the majority of climate models as far back as the 70's agreed that the trend would be upward and not level or downward, and that the latest models (90's and 00's onwards ) exclusively predict a long term upward trend? You mistake the fact that any single random prediction has a 50/50 chance of getting the sign right or wrong, and the fact that the majority of climate models since the 70's have got the general prediction right...

Technically. Not really meaningfully. The spirit of the question was the search for some sort of model validation. A paper in the 70's saying they expected warming of an undisclosed magnitude at some undisclosed time in the future doesn't really do that. Especially given that according to current models we're entering into a mild cooling period - which the folks in the 70's would not have predicted based on our carbon output. I can say that with some confidence given that the folks from the 2007s also didn't predict it.

Not only does the Broecker paper put meaningful (and subsequently validated) numbers on CO2 emissions, they also put constraints on the levels of warming predicted from CO2 forcings aswell, and in a very specific time frame to boot. Not one current climate model predicts a cooling period, even a mild one, that I know of, not even the model described in the Nature article that sparked this debate, which in reality predicts nothing more than a possible short-term offsetting of the overall warming trend... Nowhere does it say anything about a "cooling period", or anything that resembles a cooling trend. The key point of the Keenlyside paper is that although they are attempting to model short-term variations in climate (unlike long-term models), but regardless of their fledgling attempts to model short term climate variability, their long-term predictions do not differ from other long-term predictions made by other climate models in that they too suggest a pronounced overall warming trend for decades to come...

All of them anthropogenic? Everyone thought that humanity was causing global warming in the 70s? Because I seem to remember there being a cooling scare from the 70's.

There wasn't really a cooling scare in 70's... indeed, a recent review by Peterson et al. shows that the majority of scientific studies on the subject of climate change in the 1970's actually predicted a warming trend rather than a neutral or a cooling trend... The Broecker paper being one specific example of such a paper from that era...

I wonder if you're defending against my position and so I don't get to hear your concerns about some of this research, of if you truly believe that these guys know exactly what they're doing. But from my point of view, the science is far from settled. I'll know when it's settled because I'll stop seeing their predictions proven wrong.

In many respects, I agree with you that alot of the science is not settled... it's only natural when dealing with a situation as complex as climate change. But at the same time, if one disagrees with the predictions made by the vast majority of climate models (which predict a warming trend) and that these models largely rely on the incorporation of anthropogenic forcings to explain why the long-term projections are of the direction and magnitude that they are, then the onus is on you to explain where the discrepancy between observation and theory lies...

 

[Danoff]

This article seems to be helpful: http://ams.confex.com/ams/pdfpapers/131047.pdf

In December 1968, a group of scientists
convened in Dallas for a “Symposium on Global
Effects of Environmental Pollution” (Singer 1970).
Reid Bryson showed the panel a remarkable
graph showing the correlation between rising
levels of dust in the Caucasus and the rising
output of the Russian economy over the previous
three decades. It was the foundation for an
argument leading from human activities to dust to
changing climate. Atmospheric pollution caused by
humans was sufficient, Bryson argued, to explain
the decline in global temperatures identified earlier
in the decade by J. Murray Mitchell (Bryson and
Wendland 1970).
Also on the symposium panel was Mitchell
himself, and he disagreed. Mitchell’s calculations
suggested that particulates added to the
atmosphere were insufficient to explain the cooling
seen in his temperature records. But he raised the
possibility that, over time, cooling caused by
particulates could overtake warming caused by
what he called “the CO2 effect” (Mitchell 1970).
S. Ichtiaque Rasool and Stephen
Schneider in 1971 wrote what may be the most
misinterpreted and misused paper in the story of global cooling (Rasool and Schneider 1971). It
was the first foray into climate science for
Schneider, who would become famous for his
work on climate change. Rasool and Schneider
were trying to extend the newly developing tool of
climate modeling to include the effects of aerosols,
in an attempt to sort out two potentially conflicting
trends − the warming brought about by increasing
carbon dioxide and the cooling potential of
aerosols emitted into Earth’s atmosphere by
industrial activity.
The answer proposed by Rasool and
Schneider to the questions posed by Bryson and
Mitchell’s disagreement was stark. An increase by
a factor of four in global aerosol concentrations −
“which cannot be ruled out as a possibility” − could
be enough to trigger an ice age (Rasool and
Schneider 1971). Critics quickly pointed out flaws
in Rasool and Schneider’s work, including some
they acknowledged themselves (Charlson et al.
1972, Rasool and Schneider 1972). Refinements,
using data on aerosols from volcanic eruptions,
showed that while cooling could result, the original
Rasool and Schneider paper had overestimated
cooling while underestimating the greenhouse
warming contributed by carbon dioxide (Schneider
and Mass 1975, Weart 2007). Adding to the
confusion at the time, other researchers concluded
that aerosols would lead to warming rather than
cooling (Reck 1975, Idso and Brazel 1977).
It was James Hansen and his colleagues
who found what seemed to be the right balance
between the two competing forces by modeling
the aerosols from Mount Agung, a volcano that
erupted in Bali in 1963. Hansen and his
colleagues fed data from the Agung eruption into
their model, which got the size and timing of the
resulting pulse of global cooling correct. By 1978,
the question of the relative role of aerosol cooling
and greenhouse warming had been sorted out.
Greenhouse warming, the researchers concluded,
was the dominant forcing (Hansen et al. 1978,
Weart 2007).

This is the problem with trying to interpret scientific history. The events that followed color our view of the events. How does the layperson who doesn't feel like pouring through every single climate change article verify that the question had been "sorted out" by 1978. We know that a group of researchers had predicted that C02 would outstrip aerosols to have an overall warming effect. But we don't know how widely accepted that conclusion was.

Note that this article indicates that throughout the 1970s researchers argued about whether mankind's contribution to the climate was going to result in either a net positive or negative influence. By ~1980 a group of researchers came to the conclusion that the net effect would be positive - and we can only presume that the scientific community embraced that conclusion. I'm willing to bet that plenty of others disagreed.

Regardless, the number of papers predicting anthropogenic warming does seem to be significantly greater than the number predicting cooling. This is from the same article.

Note that they key paper mentioned in the article does not attempt to forecast the actual global climate - only which direction man is sending the climate. It appears to me that by ~1980 man had settled on the notion that C02 output would outpace aersols in terms of climate forcing. That says absolutely nothing about the amount of climate forcing, let alone predicting where the climate was actually headed.

In terms of figuring out where the climate is actually headed, scientists are STILL in disagreement about the sign over the next 10 years. The latest IPCC report still shows the usual alarmist warming trend through 2015 and beyond, while other reports are now indicating mild cooling.

I think the default position here is skepticism. The default is that the assertions are not validated. Not until I see a chart from significantly far back accurately* report the change in global mean temperature will I start to think these guys have their act together. I don't think it's too much to ask.

*Within ~10% of the predicted change over the time period. Or within 2-sigma of uncertainties that are reasonably small to show that the prediction was warranted from the data.

Edit: BTW - I fully admit that I did not know that the global cooling thing was a big media scare and I stand corrected.

 

[Touring Mars]

As scientific historian Naomi Oreskes has addressed, the public perception of what a "scientific consensus" is and what a real scientific consensus is, are two very different things... But as her 2004 study showed, there is infact a clear consensus, and she has recently updated those findings (here) to take into account that the "debate" within academia over the role of mankind in global warming is effectively over, and has moved on to the arguably more important questions about what we can expect and what we could possibly do about the problem... hence, the most recent IPCC book does concern itself with predictions of climate trends rather than the relatively "solid" conclusion that human activities are largely responsible for the present warming trend.

I think the default position here is skepticism.

In the context of this debate, the term "skeptic" generally applies to someone who doesn't agree with the consensus opinion as agreed by the majority of scientific academies from across the globe. Although I think skepticism is a crucial part of science, I wouldn't endorse the view that the default position in this debate should be to disagree with the consensus view. Technically, the default position, the starting point for someone who hasn't explored any or much of the scientific literature on the topic, should not be one of endorsement or skepticism of the consensus, but of neutrality... Id' say that someone who has read a fair bit of the scientific literature on the subject, however, wouldn't agree that the default position is to doubt AGW theory. Similarly, of those who are actually writing the scientific literature on the subject, endorsement of the consensus is far, far stronger than skepticism of it.

 

[Danoff]

Technically, the default position, the starting point for someone who hasn't explored any or much of the scientific literature on the topic, should not be one of endorsement or skepticism of the consensus, but of neutrality... Id' say that someone who has read a fair bit of the scientific literature on the subject, however, wouldn't agree that the default position is to doubt AGW theory.

Not until I see a chart from significantly far back accurately* report the change in global mean temperature will I start to think these guys have their act together. I don't think it's too much to ask.

We appear to be at an impasse.

 

[Touring Mars]

Not until I see a chart from significantly far back accurately* report the change in global mean temperature will I start to think these guys have their act together. I don't think it's too much to ask.

I thought that Broecker's 1975 projection was pretty good...

This figure shows the current date (1975) marked on the graph, and projected global mean temperature change as the bold broken line. His projection shows a predicted rise in GMT of about 0.6-0.65 deg C between 1975-2005, about 0.7-0.75 deg C between 1975-2010, and about 1.0 deg C between 1975-2025... in other words, a decadal mean rise of ~0.2 deg C. What have observations shown? According to NASA's own data (below), there was a rise of ~0.6 deg C in GMT in the period 1975-2005... a decadal mean rise of ~0.2 deg C...

The claim that early climate models fail to accurately predict climate change in the following decades is not entirely justified. For one thing, it ignores the fact that quantitative climate modeling is still a relatively new area, hence the further back you go, the less likely you are to find what you are asking for. However, Broecker provides a good example of a model from the early days of quantitative climate modeling which did not fail to predict that increasing GHG levels would lead to a warming trend. I'm aware that you can argue that if you make enough predictions, one of them is bound to be right - but the question is why was this particular model right? Was it just a random guess, or was a specific assertion being made about the underlying cause of the projected trend? In the case of the Broecker study, he did explicitly single out CO2 levels as being an increasingly significant factor. I'm also aware that just because Broecker's predictions turned out to be mostly correct, it doesn't necessarily follow that his assertion toward the cause must also be right, but it does mean that it cannot be ruled out, which is highly significant... This begs the question "Where are the predictions/models which also got the direction/scale/speed of recent climate change trends correct but propose an alternative reason?"

Until such a time that someone can show me evidence of what is responsible for the current warming trend, I'm happy to accept that the current scientific consensus explanation is more credible than simply asserting that "something else" has to be responsible, especially when that "something else" isn't defined.

 

[Danoff]

First of all, let me thank you for humoring me. I appreciate your attempt at validating an older model, and I agree that comparison of the 2008 point appears to be quite convincing - so I plotted the Broecker data against what I found on wikipedia. I used this chart:

I tried to read data points from the wikipedia chart off by eye and from recreate the Broecker curve by eye. I then scaled the wikipedia values so that they match the Broecker temperature at 1975. I also plotted the Broecker predictions adjusted to be 10 years earlier. Here's what I get.

The agreement is better than I expected it to be for the original Broecker data, so thanks for drawing it to my attention. The agreement is especially good for the year you took - this year. But it's actually quite poor for most of the time period between 1975 and now (far below the 10% or even really 50% accuracy I was hoping for). And recent research indicates that agreement is going to get worse, not better. So if we see mild cooling between now and 2015 imagine holding the Broecker data up an accurate prediction or model. Notice that by moving the Broecker data 10 years earlier you can actually fit the 90s data really well (the period where most of the warming occurred). But after that it doesn't work. If I kept playing with it though, I'm sure I could have adjusted it to fit the data better. No change in the model, just a shift in the initial conditions (though, admittedly to 0th order).

Edit:

For one thing, it ignores the fact that quantitative climate modeling is still a relatively new area, hence the further back you go, the less likely you are to find what you are asking for.

Surely that calls for further consideration before grasping at conclusions.

I'm aware that you can argue that if you make enough predictions, one of them is bound to be right - but the question is why was this particular model right?

That's mostly fair. It depends on how much of a shotgun approach was taken.

I'm also aware that just because Broecker's predictions turned out to be mostly correct, it doesn't necessarily follow that his assertion toward the cause must also be right, but it does mean that it cannot be ruled out, which is highly significant... This begs the question "Where are the predictions/models which also got the direction/scale/speed of recent climate change trends correct but propose an alternative reason?"

Also fair. I suppose it depends on how you approach the problem. In my case, I approach it as waiting for validation before accepting credentials. Your approach seems to be more like waiting for a debunking theory. I'm not really sure either of us is wrong in general. I suppose it depends on what this is going to be used for. For general purposes, your approach may be better than mine. For making policy, I like my approach better.