Climate Change skeptics? [ot]

For example, if aerosol levels increase like they did in the 2000s, global warming will likely be muted. However, if they decrease like they did in the 1960s, global warming could accelerate.

The buffer effect of aerosol begs an interesting question: could man manipulate global temperatures by artificial techniques like putting aerosols in the atmosphere?

Scientists are already thinking about this possibility. They even have a name for it: geo-engineering.

The prior distribution p(q) that is used in this calculation is chosen to reflect prior knowledge and uncertainty (either subjective or objective) {This is indicating that there is indeed uncertainty, how much uncertainty is unclear from this first sentence} about plausible {This term indicated that they are operating within a set band of parameters that could be deemed plausible, this is not a definite term} this parameter values, and in fact, is often simply a wide uniform distribution. Such a prior indicates that little is known, a priori, about the parameters of interest except that they are bounded below and above {again if little is known except that they are bounded, it makes me suspicious of their level of certainty}. Even so, the choice of prior bounds can be subjective {and even if they are bounded (everything is bounded), priors by others have been subjective}. In the case of climate sensitivity, a uniform prior with a lower bound of 0°C and an upper bound between 10°C and 20°C has often been used (see Table 9.3) {this is a nice way of saying others before have taken a somewhat blind guess and gone with something simple because a lack of understanding}. However, observable properties of the climate system do not necessarily scale with equilibrium climate sensitivity (Frame et al., 2005) {this statement says that things are not as simple as researchers once assumed}. Imposing a flat prior on an observable property, such as the climate feedback or transient climate response, is equivalent to imposing a highly skewed prior on the equilibrium climate sensitivity, and therefore results in narrower posterior likelihood ranges on the climate sensitivity that exclude very high sensitivities. Alternatively, expert opinion can also be used to construct priors (Forest et al., 2002; 2006). Note, however, that expert opinion may be overconfident (Risbey and Kandlikar, 2002) and if this is the case, the posterior distribution may be too narrow. {the last couple sentences say to me that the models are very sensitive to the prior probability distribution which were oversimplified in the past, and many others have gotten it wrong in the past. they have overestimated based on their limited understanding} Also, the information used to derive the expert prior needs to be independent from the information that is used to estimate the likelihood function. However, prior belief about the climate system tends to be shaped by observations of that system, and thus it is difficult, maybe even impossible, to develop truly independent prior distributions. {This is saying that it is extremely difficult to establish an accurate prior distribution, because it is hugely influenced by the bias of the person determining it. I think the authors are also saying that they too had to guess to some degree in determining how they were going to construct their prior probability distribution...In other words...others were wrong, and we may be too}

I, personally, do not find media analysis to ever be “independent.” People call Fox news right leaning and find the NYT impartial, and people find Fox impartial and the NYT left leaning…I don’t much care. Its not important.