>>7628505
Examining the published "proof" that Homogenization is a good algorithm for temperature processing.
"Quantifying the effect of urbanization on U.S. Historical Climatology Network temperature records"
http://onlinelibrary.wiley.com/doi/10.1029/2012JD018509/full
>>7644812
>>>hoisted on their own petard
>>funny you say that, because I just happened to pull up the paper you mentioned. key takeaways:
>>According to these classifications, urbanization accounts for 14–21% of the rise in unadjusted minimum temperatures since 1895 and 6–9% since 1960.
> We look at tampered data and surprise! shrunk the UHI
FTFY
Smoking gun, #1. Pic related. See Fig 3 from the paper, "TOB adjusted" minimum (night) differences between urban and rural data. Black lines for station pairing values. From 1895 to 1960 these values are mostly negative. Yes, that means that the urban minimum (night) temperature data is COLDER than the rural data. Despite the fact that the UHI temp difference is generally strongest between urban and rural minimums (night), the urban being larger. This demonstrates that the TOB adjustments are largely bogus. This is terrible science at best. It is essentially physically impossible.
>>The USHCN version 2 homogenization process effectively removes this urban signal such that it becomes insignificant during the last 50–80years.
> The TOB "adjustments" violate physics; but effectively cools the distant past, creating an increased but fake warming trend.
FTFY
>>7653428
>>7644812
>>And here's the real kicker. What was it you said?
>>7644812
>>the algorithm does not distinguish between rural and urban temp stations...homogenization will normalize to urban stations, Read: add UHI.
>Well it turns out the authors had thought of that possibility, just like good scientists who care about robust methodology should.
Really. Why did you ignore the supplementary material?
Pic related. Fig S-2 in the Supplementary material of the paper, and Smoking gun #2. This graph supposedly "proves" that the homogenization adjustments don't bring in UHI, because using just rural temps works just as well as also using urban temps. But is this true? Remember homogenization "adjusts" the station temp value to the gengeral "trend." But the trend of the majority of rural stations is negative (see the figure for randomly chosen rural temps). Yet the chosen paired rural stations do not have a negative trend. Why? The algorithm looks at all stations (urban and rural) to decide the "correct" trend. In recent years, large numbers of rural stations have been eliminated, there are more and more UHI tainted stations. Thus the trend is UHI tainted, so homogenization chooses the "correct" rural station, the one that most closely matches UHI urban stations. Tampering data with UHI by proxy is still tampering data with UHI.
Overall, using UHI tainted data to define the trend is nothing but Goal-seeked algorithms; it's bad science at best. This fundamental flaw is the hallmark of data tampering. Something a "bad scientist" would do.
http://api.onlinelibrary.wiley.com/asset/v1/doi/10.1029%2F2012JD018509/asset/supinfo%2FHausfatheretalUSHCNUHIsuppliment.docx?l=Cgg2pVVsCMwBVhXLGZM9QYtwQymcmLot19MtWA19H9zPqhrO5lDdMtjRWWQgcm4kWEUZgDG01aDA%0Aqhdk5uEDWsVDIYqqt4qi3VXDQQsjJI8%3D
>>7653437
>>7644812
The NOAA/Zeke are using ridiculous TOB adjustments which yield physically impossible results for the more distant past. See pic for an example of actually deleting the data with bad TOB (reset in afternoon); the correct approach to dealing with data errors, you remove them. The rate of warming is miniscule. (Do you seriously think that in 1930, Sally Brown and her other urbanites always reset their thermometer right before the hottest time of the day?).
>nb4 evil denier graph
Try facts and logic for once.
>>7653439
>>7644812
Zeke and company rely on general definitions of rural and urban areas instead of looking at the actual conditions (micro-environments) of the temperature stations. The latter is what really counts. If the station is close to an air conditioner belching out hot air, it doesn't matter if the station is supposedly a "rural" station. Accounting for micro-environments is what a "good scientist" would do. See pic. Notice that the specific UHI, nearly doubles the temperature growth rate. And NOAA data tampering makes it even worse!
>>7653441
>>7644812
The circular, goal-seeked algorithm. We've seen how Zeke/NOAA used UHI tainted data to define the trend, which in turn was used to homogenize data. This became all-too-clear when despite cooler rural data, the alogirthm chose warming rural data as the homogenization proxy. Homogenization defined the recent UHI tainted urban data as "the trend."
So why do they use a goal-seeked algorithm? Because warmists love to use circular arguments. Homogenization helps find the (global warming) "signal" hiding in all that bad data. Which, as you guessed, leads to global warming results. Thus by creating an algorithm which cools the distant past (via unphysical TOB adjustments) and warms the recent past (via goal-seeked station pairing) the result is achieved. And the circular argument has come full circle.
It is meaningless to talk about finding a "warming signal." The mean value of temperature has a statistical meaning, but no physical meaning. That is because in thermodynamics, temperature is an intensive measure (like density; unlike an extensive measure like length). So to talk about detecting a "warming signal" in a physical sense is so much pseudo-scientific garbage. Instead, the NOAA and Zeke need to start acting like good scientists and let the data do the talking; not the goal-seeked adjustments and algorithms.
http://wattsupwiththat.com/2012/10/15/circularity-of-homogenization-methods/
>nb4 more evil denier
Shut up and lrn2science.
Geez, didn't you get rekt enough in that last failed abortion of a thread? Well, I guess it's easy to think you're right when you don't actually understand the topics at hand. In accordance with your total inability to understand the material that you inexplicably think supports your opinions, I shall be sageing to avoid giving your tired old lines more attention than they deserve. (Apart from the attention I'm giving them because I'm extremely Jewish and hence love to argue.)
>>7653428
>We look at tampered data and surprise! shrunk the UHI
It's easy to dismiss data you don't like when you claim (without evidence) that they're tampered with, isn't it? Unfortunately for you, in the halls of learning we don't get to just say "I don't like it, therefore it's fake".
>Yes, that means that the urban minimum (night) temperature data is COLDER than the rural data.
>This is terrible science at best. It is essentially physically impossible.
You obviously don't know anything about materials science or climatology, and haven't bothered to look. You think it's physically impossible for cities to be cooler than rural areas at night? I guess nobody told you that bare ground and buildings don't hold heat as well as dense foliage does. Seriously? That's your argument? You think that cities can't possibly lose heat at night faster than rural areas? How amateurish.
>>7653437
>This graph supposedly "proves" that the homogenization adjustments don't bring in UHI, because using just rural temps works just as well as also using urban temps.
>But the trend of the majority of rural stations is negative (see the figure for randomly chosen rural temps). Yet the chosen paired rural stations do not have a negative trend.
You didn't actually READ the captions for those figures, did you? Of course not; deniers tend to look at graphs and go with what they think the graphs mean rather than actually getting to the bottom of anything. So, what does the graph represent?
>The cumulative average of the adjustments that are common to both datasets are shown as solid lines and those that are unique are shown as dashed
You're complaining that the dashed green line (adjustments from paired rural stations NOT found in paired urban stations) is different from the solid green line (adjustments from paired rural stations found in paired urban stations). So, the solid lines (which conform to each other fairly well) represent adjustments common to both rural and urban homogenization, while the dashed lines represent adjustments which are found in only one homogenization series. And you're surprised that the adjustments found only in the rural series are different from the adjustments found in both rural and urban series? That is to be fucking EXPECTED!
You REALLY need to learn to read your figures properly.
>>7653428
>>7653437
>>7653439
>>7653441
>>7653445
>>7653489
>>7653504
climatefags on suicide watch
>It's easy to dismiss data you don't like when you claim (without evidence) that they're tampered with, isn't it?
it's the unhomogenized data you moron.
>>7653439
an entirely unsourced graph found only on a RCS post which, surprise surprise, cites no sources.
>>7653441
similarly unsourced
I'm just saying, you could be pulling charts directly out of your ass for all we know. if the data are real, you'll be able to point to a proper write-up of them that describes the methodology. not so hard, is it?
and yet you and the other deniers have been unable to produce such documents. hmm.
>>7653445
>you're adjusting the data and I don't like it!
>no, I don't want to hear about your methodology!
>the adjusted data say things I don't like therefore they must be false!
and whenever someone shows you guys unadjusted data, you complain that the data are all tainted by UHI effects et cetera. one way or the other, m88
>>7653428
Billions of dollars wasted by these hoaxers. Billions. Maybe trillions.
>>7653441
>Zeke and company rely on general definitions of rural and urban areas instead of looking at the actual conditions (micro-environments) of the temperature stations. The latter is what really counts. If the station is close to an air conditioner belching out hot air, it doesn't matter if the station is supposedly a "rural" station. Accounting for micro-environments is what a "good scientist" would do.
So, what makes you think that a weather station that has a bush near it (this is a criterion SurfaceStations and NOAA use for downgrading stations) will record precipitously more warming over several decades than a station without a bush near it? how will that bush make a warming trend appear? be specific now.
the siting of a weather station will affect the reliability temperature (T) recorded, but not so much the trend (dT/dt) recorded. The unspoken and incorrect assumption that this isn't the case is integral to SurfaceStations and their agenda.
>>7653524
you said the data had been tampered with but didn't bring any evidence to that effect.
I said you claimed the data had been tampered with but you didn't bring any evidence to that effect.
what's wrong with that?
>>7653445
And here's where you went full retard:
>It is meaningless to talk about finding a "warming signal." The mean value of temperature has a statistical meaning, but no physical meaning. That is because in thermodynamics, temperature is an intensive measure (like density; unlike an extensive measure like length). So to talk about detecting a "warming signal" in a physical sense is so much pseudo-scientific garbage.
I don't think I need to explain why this is a bunch of meaningless babble. Just for those who are wondering, a "warming signal" is just an observed positive trend in temperature measurements that happens to be caused by actual warming. Nothing more, nothing less.
never heard of signal and noise, m8?
>>7653543
sorry I'm not the OP, but I understand
