Man Made Global Warming

The ‘metric’ is only computed to support a narrative that the Earth has warmed since the widespread use of fossil fuels, and will continue to do so dangerously if the use of such fuels is not substantially curtailed.

This is sheer conjecture. The metric is computed so that scientists can observe how the earth’s climate state is evolving. And the notion that scientists are simply being bought and paid to commit fraud is quite absurd – the methods are freely published in academic literature. Show us the fraud. And the raw data are freely available, anyone who wants can obtain the data quite easily and compute their own estimate of global temperature change. Show me the global temperature estimate from people you believe are completely untainted by public funding.

I do always find it interesting how easily the contrarian set think people can be bought. A bit of projection of their own ethical fortitude, perhaps? “I could easily be bought, so everybody can easily be bought.”

Alan, we’re talking about many billions of tax payer dollars (and pounds, Euros, etc.) here. Would we have expended this much treasure and diverted so many resources just to compute GATs for the last century if it wasn’t supportive of the CAGW narrative?

NASA and the Met Office do an enormous amount of earth science research, so, yes, they absolutely would still be engaging in initiatives to understand and monitor the earth’s climate.

And, again, the raw data are freely available to anyone, show me a rogue scientist or group of scientists who has produced an independent surface temperature analysis that contradicts those from NASA and other major research orgs. The Berkeley Earth non-profit (founded by climate change skeptics, no less) did an independent analysis and they arrived at the exact same conclusion as NASA, Hadley Centre.

If there’s this gigantic worldwide conspiracy to defraud the public, where is the fraud? Nobody here can say. Nobody can produce the “right” answer that shows the others to be lies. They can’t point to any methodological flaws in the multitude of publications on this topic from Hadley (I doubt anyone in this thread crying fraud has ever even glanced at the papers).

You’re attacking a straw man argument of your own making. Let’s agree to a basic principle that all resources are scarce. So yeah, they would still be researching earth science, but would they really be devoting so many resources to support the CAGW narrative if solid, i.e. unadjusted, evidence to support it were readily available? Why not spend more on, say, better weather forecasts?

You might be surprised to learn that these agencies can work on multiple things simultaneously. Weather forecasts are being improved all the time, often using insights gained from advancements in computational climate modeling. There is no research being done to “support a narrative” – that is a contrivance of your own making – so I disagree with the fundamental premise.

You cover an enormous amount of ground in this paragraph. My response is simple – you can’t create data that doesn’t exist. If series are contaminated by UHI or otherwise impaired, they can’t be used regardless of processing.

That simply isn’t true, since this processing can, and has been rigorously shown to, remove such biases. So the question to you is again, where are the independent temperature estimates from the contrarians? Where are the publications demonstrating how everyone else has got it wrong?

Personally, I don’t think the fact that a scientific subject has become a political hot button issue should deter scientists from conducting research into it. I also don’t believe it has diluted their efforts in the slightest – these scientists are incredibly productive within their fields and highly respected for their contributions. I understand where your bias is coming from – you dislike the politics – but you should recognize that it is you who is injecting political bias into the conversation.

The study you cite isn’t talking about urbanization bias, they are stating that cities are warming faster than rural areas. This urban warming is a real component of the evolving global surface temperature. When scientists talk about urbanization bias, what they refer to is the possibility of sampling this urban warming in rural parts of the network, which is, again, a statistics problem, and one is that is largely solved.

And how, exactly, would one sample urban warming in a rural part of the network? It’s not a statistics problem – if an instrument ‘senses’ the effect of urban warming, it cannot be ‘uneffected’, statistically or otherwise. Including such affected measurements into the gridding process just adds a warming bias to the GAT ‘metric’, which I guess is desirable in some quarters.

Again, urban warming by itself is not a bias – it is a real cause of warming. The surface of planet earth is warmer because of human cities than it otherwise would be. Including such stations in the surface temperature estimate is not only warranted but necessary to get a picture of how the earth’s temperature is changing over time.

The bias arises because we are using stations to represent not just the area in their immediate vicinity, but all of the area in between the station and the coverage of its neighbors. Thus an urban station, with a correctly recorded urbanized temperature trend (along with the background regional climate signal), might be spread over non-urban area. This is what homogenization procedures are designed to adjust for, and research has shown that they are very effective.

I’ll buy that

No one would disagree with this assessment. The hodge-podge collections of weather stations cobbled together to form the historic observing networks were not originally intended for long-term climate observation, they were intended to provide local weather information. That is precisely why scientists have to do so much work designing all of the adjustments that have to be done in order to compile regional or global temperature indexes. That changes in the US with the creation of the US Climate Reference Network, so going forward we do have a pristine climate observing network free of systematic bias, and we can use it to compare to the bias-adjusted network to see how well scientists’ efforts fare. And the answer is: they do a really good job.

Dealing with uneven spatial and temporal coverage is a standard part of working with geospatial data. The solutions to these problems are well understood. UHI isn’t a serious problem, not at the global scale. Most of the planet is covered in water, not cities.

You don’t want to “cull” thermometers in urban areas from the network, because in fact urban warming is a real phenomenon. Urbanization bias is essentially a sampling problem – stations in urban areas need to represent change common to the urban area as well as the surrounding non-urban region. This is the problem that homogenization solves.

But, again, even if UHI is left unaddressed, it is a small issue for the global mean, because land-area is just 30% of the globe’s surface, and only a tiny fraction of that land area is urban.

Both of these issues (UHI bias and lack of spatial / temporal coverage) are problematic in terms of determining GAT anomalies.

Yes, they are potentially problematic, but they are essentially solved problems. And they aren’t that large as problems go. Urbanization is small and restricted to land areas, as noted, and for most of the world we have far more station records than are actually needed to compute a robust global mean estimate (I’ve seen Nick and others show that you can get a very good estimate of the global mean with something like 90 stations).

The real (read political) drive to measure GAT is to drive home the narrative that CO2 emissions, aka fossil fuel usage, is roasting the planet. It is therefore dishonest to include UHI-affected stations.

Not at all, this is nothing more than your own prejudice showing through. Scientists measure the GAT to understand how it is evolving. That means just observing the changes, using good sampling procedures and trying to reduce systematic bias. The part where we identify what is driving the observed trends is left to other areas of science.

The problem with homogenization, and why it’s beloved by CAGW alarmists, is that it doesn’t eliminate UHI, it ‘spreads’ it around, thereby imparting a warming bias to rural stations that have no trend or even exhibit cooling on an unadjusted basis. It’s like Nosferatu, everyone says it’s fixed or is not used, but it’s always there lurking in the background.

This is a myth, oft repeated with no substantiation. Significant amounts of peer reviewed research show that homogenization is effective at eliminating non-climatic biases from the surface station network (see, e.g., Menne et al. 2009). Just look at the US network – the full, bias adjust nClimDiv network, comprised of some 10,000 stations, is perfectly aligned with the unadjusted Climate Reference network, a smaller station network specifically designed and maintained to be free of any source of systematic bias:

If the bias adjustments were not working as intended, these series would be seen to diverge. Either that, or the adjustments don’t work, but the bias is so insignificantly small that it is utterly imperceptible.

Sounds impressive until one remembers it’s just math. Nick could probably compute an estimate with 10 stations, but what would that honestly tell you other than Nick is good at math?

It’s all just math. We have a finite set of observations meant to represent a continuous surface. If you do your sampling carefully, you can minimize the number of samples needed to be truly representative. For the earth, for temperature anomalies, that turns out to be a pretty small number of samples indeed, and we have tens of thousands.