Give me your opinions about global warming

[iNow]

It is known that phytoplankton release dimethyl sulphide, much of which enters the atmosphere, and forms nucleation centres for cloud formation. Satellite studies show that there is extra cloud formation over parts of the ocean with a high chlorophyll count. ie. where there are a lot of phytoplankton. Warmer oceans lead to more phytoplankton growth, thus more clouds, thus more cooling. This is a simple negative feed-back. Would you care to tell me how a computer model could have discovered this?

 

Models don't generally discover things. By their very definition, they model them. However, models can be used to make predictions and anticipate future outcomes based on current knowledge, information, and observations.

[ghstofmaxwll]

Sorry, ghstofmaxwll, but neither. What iNow posted was the conclusion from scientific evidence. It's not an assertion. It's a conclusion from data.

 

In August 2007 Scientific American had a summary article of the data to back the conclusion. Included in Figure 1 are the contributions of solar forcing, volcanic activity, man-made greenhouse gasses, etc. You need to read the article and then go back to the references in the article and read those. And the references in the references. THEN you will have all the data and will realize that iNow gave a conclusion from data.

 

I have a PDF copy of the article. If absolutely necessary, I will post it as an attachment so everyone has it. I can probably get away with the copyright issue by calling this forum a "class" and distributing material to the class.

 

No! A scientific conclusion would be something a long the lines of: The historical data of (?) year climate trends are consistent with a large increase in atmospheric compounds(green house gases). From this we can indirectly conclude that its possible increased industrial atmospheric compounds( in particular CO2) are the cause of an upward trend, but until we have the ability to measure the direct input of manmade compounds against the output of an increased greenhouse layer and also taking into account the time delay of such a system - We conclude that although this is compelling, we must be extremely cautious in jumping to conclusions of fact in any way.

[Fred56]

we must be extremely cautious in jumping to conclusions of fact in any way.

How cautious is extremely cautious? You mean we should wait for Greenland's ice sheet to start falling apart? Or until global temp. goes up by 4-5 C?

[ghstofmaxwll]

How cautious is extremely cautious? You mean we should wait for Greenland's ice sheet to start falling apart? Or until global temp. goes up by 4-5 C?

....No! just to be cautious so as to find the real mechanism for Greenlands vicious circle. Not just make data fit like a glove into the popular culture of the time.

[SkepticLance]

iNow said

 

"Models don't generally discover things."

 

I suspect your view of models is fairly close to mine. They can be useful tools, but ultimately science depends on objective and empirical data. The conclusions of models must, like any hypothesis, be tested. This is done through prediction, and an attempt to disprove the prediction, using real world experiment or observation of a novel kind.

 

However, Bascule does not seem to need those models to be tested. Instead, the mere assertion that they come from IPCC is enough. The rationalisation for this is that most climate scientists agree.

 

There was another incident in recent science that shows how poor an argument this is. For most of the 20th Century, medical science believed that stomach ulcers were caused by excess acid. This was backed up by the fact that anti-acids and similar medication relieved the symptoms. The consensus of medical science was solid. Then some researchers found a bacterium associated with ulcers - Helicobacter pylori. They carried out a little lateral thinking and wondered if this bacterium might cause ulcers.

 

They tested the hypothesis. Prediction : If the bacterium causes ulcers, then killing the bacteria should permit ulcers to heal. They used antibiotics to do this, and found the ulcers did heal. In fact they applied all of Koch's Postulates to the issue, and ticked off each one. This is supposedly definitive proof, or as near to it as science permits. Yet the consensus remained, and it took over another decade before medicine as a body came around.

 

My conclusion: Consensus as such is meaningless. Human intellectual inertia often reigns supreme.

 

I have pointed out to bascule repeatedly, that the solar forcings calculations he has so much faith in, cannot be reliable, since the mechanism by which sunspot activity influences global temperature is not properly understood. If you do not understand how something works, how can you calculate its effect? Clearly you cannot.

[iNow]

Just to clarify my position, models come in different flavors. However, the enormity of the data which is feeding current climate models is more than sufficient (for me) to trust their predictions as being... what's the scientific term?... pretty f%#king close.

 

Also, I think it might be a stretch to suggest that Bascule does not think the models need to be tested. I think his position (and I apologize if I'm mistaken) is that they've already passed every test thrown at them, that they align with the empirical data quite well, and they serve as an incredible tool for us to use to look forward with a fair degree of accuracy.

 

To close this post, I want to respond to your last sentence. We don't understand how gravity works, yet we CAN calculate it's effect. All the best.

[SkepticLance]

To iNow

 

First : about models passing every test. Not so.

The models predict current temperature change quite adequately. Hey, I can do that by mental arithmetic. The world is warming at the rate of 0.16 to 0.2 C per decade. How much will it warm in the next 25 years? I think I can work it out.

 

On other matters, such as the frequency of storms, which areas are becoming more or less arid etc. Sorry. The models sometimes get it right, and sometimes get it wrong. For example : the models predicted mainland Antarctica would be warming. It is, in fact, cooling.

 

And the models have no show at all at predicting anything that depends on off Earth phenomena, which of course relates to exceptional changes in sunspot activity. That is, changes that do not fit into the 11 and 22 year cycles. The models could not have predicted in advance the Little Ice Age, for example, since the Maunder Minimum was not predictable.

 

 

On gravity. I can predict its effects also, very easily. Mostly it is quite calculable using Isaac Newton's formula, which is very simple. Climate is quite different, since it is based on a million different variables, some of which may still be unknown, some of which are estimated, some of which are based on assumptions, and some have significant known error factors.

 

The simple fact that the IPCC values for solar forcings do not tally with the known historical impact of sunspot activity should be enough to tell you just how much to trust their calculations.

[Fred56]

I see someone's found another simple fact. Is there also a simple explanation for it, perchance?

I think I can work it out.

So do a lot of other people...

[iNow]

To iNow

 

First : about models passing every test. Not so.

The models predict current temperature change quite adequately. Hey, I can do that by mental arithmetic. The world is warming at the rate of 0.16 to 0.2 C per decade. How much will it warm in the next 25 years? I think I can work it out.

 

On other matters, such as the frequency of storms, which areas are becoming more or less arid etc. Sorry. The models sometimes get it right, and sometimes get it wrong. For example : the models predicted mainland Antarctica would be warming. It is, in fact, cooling.

 

And the models have no show at all at predicting anything that depends on off Earth phenomena, which of course relates to exceptional changes in sunspot activity. That is, changes that do not fit into the 11 and 22 year cycles. The models could not have predicted in advance the Little Ice Age, for example, since the Maunder Minimum was not predictable.

 

 

On gravity. I can predict its effects also, very easily. Mostly it is quite calculable using Isaac Newton's formula, which is very simple. Climate is quite different, since it is based on a million different variables, some of which may still be unknown, some of which are estimated, some of which are based on assumptions, and some have significant known error factors.

 

The simple fact that the IPCC values for solar forcings do not tally with the known historical impact of sunspot activity should be enough to tell you just how much to trust their calculations.

 

Okay. This is a healthy discussion. What concerns me a bit is how frequently you state that we do not understand the impact of sun spot activity, yet you so vehemently assert that climate modesls must include the historical impact of sun spot activity. If we don't understand the role of sun spots on global climate (or, if there even is one), then I don't see how this is the proverbial nail in the coffin of modern day global climate science.

 

So, let's start here.

 

Which models EXACTLY are inaccurate, and how?

What is the impact and scope of the above referenced inaccuracy in the literature, and how does it measurably effect our understanding of global climate (I'm talking quantitative impact/effect, not qualititative).

 

Then, once that has been established, let's speak of the issue of sun spot activity, what we know, and what aspects of climate change (notice I don't say "global warming") can be attributed to sun spot activity. If I were to guess, I think this impact is a relatively low percentage of the overall factors involved in current global climate change, not to be ignored if it is indeed a factor, but not fully deserving of all our focus and attention when it comes to change.

 

To be frank, I'd like to understand all factors involved in climate change, but I'm not sure that a complete focus on sun spots is a valuable use of our time. We can't change sun spot activity. We can change human activity.

[bascule]

I would be interested to know which part of the historical record you thing were 'so accurately' reconstructed. The last 30 years are easy. That can be reconstructed pretty well with a pen, ruler, and graph paper.

 

So wait, you're saying the last 30 years of the global climate can be simulated using a pen, ruler, and graph paper?

 

Can I ask what the resolution of your graph paper GCM is, and what's the granularity of your timesteps? Are you doing your calculations on a stack of graph paper that could go from the earth to the sun?

 

To answer your question, model output has successfully been used to replicate climate for the past century, which can be verified against instrumental record:

 

 

Before that, the modellers have been struggling, re-doing models many times to try to get it right. The historical climate record, or course, goes back to Roman times. Are you suggesting that climate modellers can reconstruct it back that far? Not without better references to sunspot effects, they cannot.

 

Paleoclimactic researchers do focus on historical changes in the solar cycle far more than those researching the modern climate do or need to. But that's a red herring...

 

How can anyone calculate margin of error when they do not actually know what the errors are? You can calculate known errors, sure. However, it is pretty arrogant to assume that all factors are known.

 

Present climate science knowledge is sufficient to perform accurate multi-decadal simulations. It's pretty arrogant for you as a layman to say it isn't, especially when you're so clearly clueless about the actual science.

 

Then why have several international conferences on climate modelling discussed this problem, as a problem?

 

There are many potential problems that are discussed at many conferences. Perhaps the fact it's being discussed at conferences should suggest to you that they're addressing the problem? By the way, you misspelled modeling...

 

This statement is utter garbage. In science, it is empirical studies that lead to discoveries such as feed-back loops. Modelling is just a tool, which raises questions that have to be tested empirically.

 

Modeling operates on empirical data, and allows climate scientists to observe the complex interplay of different forces in a complex dynamical system... things that can be directly substantiated by the data, but would be extremely difficult to discover without them. There's not some hard separation between what is empirical and what is modeled like you seem to be suggesting. And what sort of thing would you like climate scientists to "test empirically"?

 

I am not suggesting that models have no value.

 

Actually you just said they could be replaced by a pen and graph paper...

 

*sigh* yeah think it's time for me to resign from this thread again. Arguing with SkepticLance never goes anywhere...

[ghstofmaxwll]

 

 

Actually you just said they could be replaced by a pen and graph paper...

 

 

If you had any scientific training instead of copy and pasting out of context trends, you would know that we can model with paper and pen.

[1veedo]

Not the climate though, you generally need supercomputers for that kind of work.

[SkepticLance]

To bascule.

 

A very nice set of graphs. Thank you for posting them. They pretty much demonstrate what I have been saying all along. That is that climate projections and the correlation with greenhouse gases have been OK since 1976, but start falling apart earlier than that.

 

If you look at the graphs, and screen off with your hand the area to the right of the 1976 portion, the simulations are not terribly impressive. In some cases, such as the global ocean, they are distinctly incorrect.

 

If you wish to challenge my assertions using those graphs, you have to look only at the pre-1976 portions. As I said earlier, getting it right after 1976 does not even need a computer model.

 

And if you go back even earlier, the level of accuracy tumbles. The reason is simple. As I said all along, where sunspot activity becomes important, the models cannot account for it.

 

However, bascule, you should realise that I am not actually challenging your main belief. It is the portion after 1976 which bears out the fact that humans are changing climate, and shows that we should start dealing with that problem. What happened before that is actually irrelevent.

[1veedo]

A very nice set of graphs. Thank you for posting them. They pretty much demonstrate what I have been saying all along.

Lol you wish. You keep trying to "interpret" graphs the way you want to but the fact is actual scientists have already interpreted the data and seem to disagree with your position.

[ghstofmaxwll]

Lol you wish. You keep trying to "interpret" graphs the way you want to but the fact is actual scientists have already interpreted the data and seem to disagree with your position.

You see, the key word here is "interpret". You cant interpret something to an extent of removing doubt. Anything that needs interpretation also need extreme caution when stating a conclusion. real scientists know this.

 

No scientist will ever say the following: "We have interpreted that so and so is fact from our data".

[SkepticLance]

It is actually very straight forward. No great interpretation effort is needed. However, I expect that 1veedo and bascule will see what they want to see in those graphs.

 

The 1976 to present portion of the graphs are pretty simple, and I have not disputed the IPCC version of climate change for that period. My opposition has only been with the time period before 1976. And for that time period, the graphs are seriously off the mark.

 

By way of comparison, check the graph in this link.

 

http://www.mps.mpg.de/images/projekte/sun-climate/climate.gif

 

This shows the effect of sunspot activity on global temperature change. Look at the portion of the graph before 1976, and compare it to the same portion of the equivalent graph as posted by bascule. For bascule's graphs, look at the pink portion compared to the dark line. The blue portion is just a distraction in terms of this discussion.

 

 

If you are honest with yourself, you must admit that the sunspot activity in the graph in my link more closely correlates with actual temperature than the model's predictions in bascules graph.

 

Thus, the IPCC models do not approximate reality for that time period, as well as a prediction based on sunspots.

[swansont]

It is actually very straight forward. No great interpretation effort is needed. However, I expect that 1veedo and bascule will see what they want to see in those graphs.

 

The 1976 to present portion of the graphs are pretty simple, and I have not disputed the IPCC version of climate change for that period. My opposition has only been with the time period before 1976. And for that time period, the graphs are seriously off the mark.

 

By way of comparison, check the graph in this link.

 

http://www.mps.mpg.de/images/projekte/sun-climate/climate.gif

 

This shows the effect of sunspot activity on global temperature change. Look at the portion of the graph before 1976, and compare it to the same portion of the equivalent graph as posted by bascule. For bascule's graphs, look at the pink portion compared to the dark line. The blue portion is just a distraction in terms of this discussion.

 

It actually does not. It shows sunspot activity AND global temperature change. What it doesn't show is how much of a temperature effect each W/m^2 of solar irradiance will have, nor does it show any other effects.

 

If you are honest with yourself, you must admit that the sunspot activity in the graph in my link more closely correlates with actual temperature than the model's predictions in bascules graph.

 

You need to establish that the solar change indicated will actually cause a ~0.4 ºC change, as opposed to stretching a graph to make things overlap well.

 

Thus, the IPCC models do not approximate reality for that time period, as well as a prediction based on sunspots.

 

That will generally be true when cherry-picking data. What is being questioned is the value and validity of such an approach.

[SkepticLance]

To swansont

 

As I have said before, apparently to no avail, the IPCC models do NOT allow for sunspots. They have a variable built in that is called solar forcings. This appears to me to be a result of measures of solar irradiance, which is only ONE of the outcomes of sunspot activity. To my knowledge, there are four major changes during times of high sunspot activity.

1. An increase in solar irradiance.

2. An increase in solar magnetic effects.

3. An increase in the solar 'wind' - the stream of charged particles hitting the Earth, as shown by the aurora's bright displays.

4. An increase in ultra violet, which can be at least 100 times as high when sunspot activity is very high.

 

To assume that solar forcings from irradiance is the whole story is not a good reflection of reality.

 

Swansont said

 

"You need to establish that the solar change indicated will actually cause a ~0.4 ºC change, as opposed to stretching a graph to make things overlap well."

 

If you read my postings, you will see that I have made no claims for quantitative effects of sunspots. Just close correlations between sunspot activity and global temperature change. The 1910 to 1941 warming was 0.4 C, and the record indicates this warming closely correlates with sunspot activity increase. However, there may have been other influences, so I do not claim the whole 0.4 C is sunspot activity. In fact, the whole problem here is the degree of uncertainty.

 

As I have said many times, if we do not understand the mechanism, how can we determine how much?

[swansont]

If you read my postings, you will see that I have made no claims for quantitative effects of sunspots.

 

Since I've pointed this out before, I think you can assume I've noticed. This is precisely why you have to stop after you've said there is a correlation, and can't draw any valid conclusions past that.

[SkepticLance]

To swansont

 

If you cannot draw conclusions from strong correlations, what can you draw them from?

 

There is one thing worse than not quantifying an effect, and that is deriving a false result from a faulty quantifying process. That appears to be what is happening with attempts to quantify solar forcings.

[ghstofmaxwll]

Why was Fred banned?

[insane_alien]

Why was Fred banned?

 

trolling.

[swansont]

To swansont

 

If you cannot draw conclusions from strong correlations, what can you draw them from?

 

There is one thing worse than not quantifying an effect, and that is deriving a false result from a faulty quantifying process. That appears to be what is happening with attempts to quantify solar forcings.

 

Correlation is the beginning of scientific inquiry, not the end.

 

You're convinced the quantifying process is faulty, but I haven't seen you present any evidence that this is so.

[khelben1979]

Hello!

 

I would like to inform about the Climate Prediction.net project. You can find it here: http://www.climateprediction.net/index.php

 

What do you people think about this? (I'm a member of the project)

[bascule]

If you had any scientific training instead of copy and pasting out of context trends, you would know that we can model with paper and pen.

 

So again I ask: what's the resolution of your graph paper GCM, and what's the granularity of its timesteps?

 

If you were replying from any kind of scientific background rather than ignorance, you'd know it's thoroughly impractical to model the climate system without a computer to perform the calculations.