There is scientific proof that human emissions of CO2 are the main cause of climate change.

There is a lot of buzz these days (2007) about climate change, and in particular about how humans are responsible for it.

I am not going to try to bring any argument to make a case either for or against climate change or its causes, or either about its good or bad effects for human life, or about how "somebody has to do something".

I will argue that it's false to claim that "there is scientific proof that human emissions of CO2 are the main cause of climate change".

The scientific proof for the cause of a phenomena can be claimed to exist only when there is a scientific theory which produces a verifiable effect (in the future). Science is done with reproducible events and predictable outcomes, not with "consensus". Consensus is an indication that there is no proof, so a number of people claim that they know the truth.

There may be several causes of a phenomena, but in order to say that one of them is the main one, the one which overwhelms the other possible causes, it's necessary to quantify all of them, put them in equations and then see if the theory leads to a verifiable effect. If that is so, it can be said that the theory is correct.

Supporters of the human-made climate change say to have the data and the theories right, say that nothing else can be responsible, but they can't say what will be the global annual average temperature for the next year (with a margin of error smaller than that produced by statistics).

They complain that it's difficult to predict the temperature with such a precision because of the internal variability of the climate. But you know what, difficulty is not a measure of proof.

If Kepler were to say "I can't predict its orbit, but I know it's circling the Sun, it can't be anything else", he wouldn't be in the history books as a scientist.

So how does this apply to climate change? In order to be able to have scientific proof that humans, specifically their CO2 emissions, are mainly responsible for climate change, the theory must quantify all the known factors responsible for climate change and produce a verifiable effect. What is the effect of climate change? An increase of the average temperature of the Earth. So, the theory must produce a verifiable increase of the average global temperature.

Practically, the theory should predict the average global temperature for the next year, with a precision better than that produced by statistics.

Some people have pointed out that this is impossible due to the many tiny factors which influence the climate all the time. However, on a global scale, all these tiny factors compensate one another and produce a value which varies little from year to year.

The inability to predict the average global temperature represents an inability to consider all the influencing factors and their relevance, when creating a climate model.

For instance, in the usual climate model, one can scientifically say that volcanic eruptions alter the climate, but can't scientifically say to what extent that happens because that would require to quantify all relevant factors. The inability to do so only shows that there is no scientific proof that volcanic eruptions drive global temperature changes. Other relevant factors could overwhelm the effect of volcanic eruptions.

If changes in the amount of atmospheric CO2 would indeed be the main drive in global temperature changes, all the other factors would have to have only a (much) smaller effect. Consequently, a prediction could be made about the future. If this doesn't happen, even though it can be said that it's scientifically proven that CO2 alters the climate (because of small scale experiments and observations), it can't be said that it's scientifically proven that CO2 is the main drive for climate change.

But is it practically possible to predict the average global temperature? For this we have to check some data. You can calculate that the average global temperature change from year to year, starting from 1990, is about (±) 0.1 C per year. Here is a chart which shows that.

So, any prediction must give a smaller margin of error, else it's just statistics based on historical data.

You can see that it's currently possible to estimate the average global temperature with a 0.1 C margin of error: "Averaged over all forecast lead times, the RMSE of global annual mean Ts is ... 0.105°C for DePreSys".

For predicting the temperature we only need to correctly state its future value from two possible ranges: T...T - 0.05 and T...T + 0.05. If the temperature would be predicted with this precision for 5 years in a row, the chance of simply guessing it would be 1 in 32 cases.

If any climate model is to be accepted as good, it has to predict the temperature for several years in the future. We would then see how much is prediction and how much guessing.

But currently, there is no such verifiable prediction. Consequently, the claim that "there is scientific proof that human emissions of CO2 are the main cause of climate change" is false.

One word of caution. Even though a theory could predict the average global temperatures, the equations and data must still be peer reviewed to see that the claim that CO2 is the main cause of climate change, does indeed result from the theory.

I have certain concerns about the validity of such a theory because of this statement "A statistical forecast method (18) is also able to capture the trend and interannual variability of Ts for the coming year (green triangles in Fig. 2A). The statistical method accounts for interannual variability using predictors based on the state of El Niño and recent volcanic activity."

Basically, I am led to believe that this climate model is as good as statistical forecast rather than the quantified effect of the CO2 level in the atmosphere.

License | Contact