Let’s start at the beginning.
The greenhouse effect has nothing to do with the way greenhouses work, which is that they trap warm air and prevent heat from dissipating through convection. Instead the greenhouse effect is about molecules in the atmosphere absorbing energy from the sun and retransmitting it as heat. All gases in the atmosphere do this, but the heavier ones do it more because they have bigger particles and so absorb more photons. The trouble with carbon dioxide is that while it is heavier than oxygen and nitrogen, there isn’t very much of it. There are only 380 parts per million of carbon dioxide. This makes it a trace gas. It simply doesn’t absorb enough heat to raise the temperature of the earth by the required amount.
Climate scientists get round this by positing feedback effects. They say that if carbon dioxide causes a small amount of warming, that warming causes other effects that create yet more warming. There is much debate about what these feedback effects might be, and it means no-one is sure how sensitive the climate is to carbon dioxide. What evidence supports the assumption that this sensitivity is significant? Well, geological records show a correlation between carbon dioxide in the atmosphere and temperature. But we all know that correlation does not mean causation. Sure enough, it turns out that changes in carbon dioxide concentration lag changes in temperature by about 800 years. In other words, the atmosphere warms up and then 800 years later carbon dioxide concentration increases. That doesn’t sound like carbon dioxide is causing the increase in temperature to me. Climate scientists say that something else caused a small temperature increase, that increased carbon dioxide, and then the carbon dioxide caused even more temperature increase. That is not the simplest explanation. Many other hypotheses fit the data, including “CO2 is driven by temperature”.
It is certainly true that in industrial times the concentration of carbon dioxide has increased very rapidly. So if temperatures increased rapidly at the same time that would point towards man made global warming. That is what Mann’s hockey stick graph claimed to show. But there was a problem: the graph contained data from temperature proxies like tree rings and ice cores spliced with modern surface temperature measurements made with thermometers. To do this, Mann used a complicated computer program. Steve McIntyre found that even if he fed this computer program with random data, it still produced a hockey stick shaped graph. There is also much debate about whether bristlecone tree rings used in producing the graph are a valid temperature proxy; this proxy diverges from known temperatures in the last 25 years.
So evidence that carbon dioxide causes increased temperature is poor. That leaves us with evidence that the planet is warming at an unprecedented rate during the 20th century. What evidence do we have?
There are surface temperature measurements. But these have not been kept very accurately. The thermometers are accurate to within +/- 0.5 degrees but we are trying to measure changes of tenths of a degree per decade. Furthermore, surface temperature measurements are corrupted by local human activity, like building cities. A big problem is urban heat islands. Concrete absorbs heat from the sun and emits it, warming the air around. So where there is concrete the air is warmer. To get around this problem, surface stations are supposed to be located away from concrete. Anthony Watts has been finding that a lot of them are near buildings and car parks. Corrections are applied that are larger than the effect we are trying to measure. Complicated computer programs are used to apply correctsions and errors are found in these. I am not convinced that the surface temperature record shows much more than incresing urbanisation. Certainly, urban stations nearly always show warming and rural ones often show cooling. There is room for suspicion.
What about satellite records? These are much more accurate and it is much easier to get a global temperature because we can measure the whole planet. One problem is that this record does not go back very far. But it doesn’t really show much evidence of global warming anyway. It shows that 1998 was unusually warm, and since then temperature might have risen a bit or might have remained about the same. Time will tell if the warming will continue.
What about sea levels? Sea level has been rising very slowly for the last 10,000 years because of our emergence from the last cold glacial period. To show that sea levels are affected by man made global warming, climate scientists need to show that the the rate of sea level rise has increased in industrial times. They have not shown this; the rate seems constant and not very alarming.
What about ice? Some ice is getting thicker and some is melting. Some glaciers are receding and some are advancing. This has always been the case. The pictures of collapsing ice you see on TV were taken in the summer. The satellite records tell a somewhat less dramatic story. Polar bears are not drowning.
What about all the predictions that the temperature will rise by so many degrees in so many years? These predictions come from computer models. The models are made by taking a best guess about how the climateworks, incorporating assumptions about CO2 sensitivity and other unknown factors, comparing the results to historical temperatures, and tweaking until they match. But no model has ever accurately predicted the future. Mostly they don’t try to; they are scenarios based on the assumptions. There are questions about how well clouds and water vapour can be modelled. The models’ predictions vary widely and the media focus on the most dire predictions.
When I look around I don’t see the incontrovertible evidence we are supposed to see. I don’t see the obvious link that proves CO2 causes warming. I don’t see much, if any, actual warming, or any changes that can’t be explained by natural decadal climate cycles. Even the climate scientists are preparing their excuses for the cold period we’re likely to go through for the next 10 years or so. All I see is article after article in the media taking climate change for granted. These endless news reports are just proof by repetition.
Out of interest, what would change your mind? On any of these things: that global warming is happening or that it is caused by human activity.
The idea that carbon dioxide drives climate just can not be reconciled with the historic record. It is possible, even likely that a temperature increase will produce a rise in CO2 levels but the converse on which the entire panic is base is supported by the underlying data.
To put it in simple terms, in the past much higher CO2 levels did not cause large scale warming so why would anyone expect todays modest levels to have some big effect.
Patrick: Sorry your comment was smitten. Good question.
That warming is happening: satellite data that showed a convincing trend. Maybe we’ll have this in ten years or so.
That it is caused by human activity: more tricky as we only have one planet so it’s unprovable. I’d be convinced if there was unprecedented warming, I don’t think we’ve seen that yet. Better understanding of the climate would help a lot, e.g. if climate models starting making accurate predictions.
This pretty well sums it up.. global warming is a manufactured crisis.
Rob: What satellite data? Please provide a link. I thought global temperature had stayed even or slightly decreased in the last ten years. It’s also interesting that the past few months have been cooler than normal on a global scale at about the same time the sun decided not to get the next cycle cranked up when scientists predicted it would.
Rick: I’m referring just to the UAH and RSS data, a graph of which I already linked to.
Looks inconclusively noisy to me. You could probably argue all manner of trends.
Hi. How would you address my residual correlation analysis? Specifically, I’ve found that temperature fluctuations lag cummulative CO2 emission fluctuations by roughly a decade throughout the last 150 years. I can say this with statistical confidence and then some.
Joseph: ok, I know little of statistics. But have various questions: What are you using to get temperature? What exactly are you saying? I was under the impression that CO2 emissions have gone in only one direction: up. Whereas temperature goes up and down. Why would temperature go down when the rate of CO2 increase goes down, and what does this mean for CO2 sensitivity?
It seems that CO2 goes up and down during the year. William M. Briggs says there is reason to think that CO2 lags temperature by about a year.