Correction of US Weather Station Temperature Data

[LowDown]

Here is the raw data from US stations compared to the corrected data:



You'll note that the effect of corrections is to emphasize the increase in temperatures. The raw data shows temperatures in the 1930s as being higher than they are now. The corrected data shows the opposite.

The NCDC tells us what the corrections are based on. What they won't show us is exactly how they were done. This process of revising older readings downward is something that has occured over a number of years, and it has been done station by station.

Raw station data can be downloaded from NASA. What is remarkable is that the uncorrected data simply averaged without regard to station location and so on is almost identical to the results obtained after the data is gridded and smoothed. After that, how the data is treated to get these corrections isn't clear. Efforts at reverse engineering provide no explanation for these corrections. There is no shift of station locations, instrumentation, etc. that would justify it. The heat island effect should have gotten stronger with time, but there appears to be no correction for that.

Gridding and smoothing is when temperatures from stations at various locations are combined and reconciled and then used to determine the representative temperature for a grid square on the map. Some grid squares may have only one station and others may have several. In some cases a grid square temperature is extrapolated from surrounding grids. All the grids are weighted equally when determining the continental or global mean.

But this is nothing compared to the job they did on sea level data and deep ocean temperature data.

 

[longview]

Here is the raw data from US stations compared to the corrected data:

View attachment 67163989

You'll note that the effect of corrections is to emphasize the increase in temperatures. The raw data shows temperatures in the 1930s as being higher than they are now. The corrected data shows the opposite.

The NCDC tells us what the corrections are based on. What they won't show us is exactly how they were done. This process of revising older readings downward is something that has occured over a number of years, and it has been done station by station.

Raw station data can be downloaded from NASA. What is remarkable is that the uncorrected data simply averaged without regard to station location and so on is almost identical to the results obtained after the data is gridded and smoothed. After that, how the data is treated to get these corrections isn't clear. Efforts at reverse engineering provide no explanation for these corrections. There is no shift of station locations, instrumentation, etc. that would justify it. The heat island effect should have gotten stronger with time, but there appears to be no correction for that.

Gridding and smoothing is when temperatures from stations at various locations are combined and reconciled and then used to determine the representative temperature for a grid square on the map. Some grid squares may have only one station and others may have several. In some cases a grid square temperature is extrapolated from surrounding grids. All the grids are weighted equally when determining the continental or global mean.

But this is nothing compared to the job they did on sea level data and deep ocean temperature data.

The text in the table data from GISS is

using elimination of outliers and homogeneity adjustment

They do not go into detail on how outliers are determined, or how through the homogenization.
I suspected something when the 1880 numbers changed between Sept 2013 and Dec 2013.

 

[flogger]

Well you can fool some of the people some of the time I suppose

Meanwhile here is a fairly compelling map of US NOAA max temperature records by state, which backs up your OP



Note that 23 of the US State records are STILL those set in the 1930s despite all the hype

 

[Lord of Planar]

I winder what their reasoning is to correct the data that way, other than to show a warming?

 

[LowDown]

The text in the table data from GISS is

They do not go into detail on how outliers are determined, or how through the homogenization.
I suspected something when the 1880 numbers changed between Sept 2013 and Dec 2013.

Yeah, but the methods they describe don't get the results they claim. Elimination of outliers and homogenization doesn't explain the downward revision of earlier data.

 

[Deuce]

The correction is in opposition to my beliefs, so it must be wrong!

 

[Lord of Planar]

The correction is in opposition to my beliefs, so it must be wrong!

What do you mean?

Seem to me the corrections support your beliefs. Without the corrections, you have nothing to stand on.

 

[flogger]

The correction is in opposition to my beliefs, so it must be wrong!

Without a rational explanation of why the earlier ones were all wrong why were they made ? To most objective observers it would seem rather suspicious that all the errors were in the same direction

 

[Deuce]

Without a rational explanation of why the earlier ones were all wrong why were they made ? To most objective observers it would seem rather suspicious that all the errors were in the same direction

Not if the reason for the error would provide a bias in a particular direction. In that case, the correction would be in the opposite direction.

 

[flogger]

Not if the reason for the error would provide a bias in a particular direction. In that case, the correction would be in the opposite direction.

What reason would that be then because correcting for modern Urban Heat Island effect would show the opposite adjustment to the one that has been made ?

 

[Deuce]

What reason would that be then because correcting for modern Urban Heat Island effect would show the opposite adjustment to the one that has been made ?

Presumably that is not the only error that can exist.

 

[flogger]

Presumably that is not the only error that can exist.

Well until we are enlightened as to what these 'outliers' alluded to in the OP represent we'll never know how much subjective interpretation was involved in these alterations

 

[Deuce]

Well until we are enlightened as to what these 'outliers' alluded to in the OP represent we'll never know how much subjective interpretation was involved in these alterations

And until evidence of fraud is presented to me I will remain skeptical of the accusations.

 

[flogger]

And until evidence of fraud is presented to me I will remain skeptical of the accusations.

Until I see verifiable evidence of the need for these alterations and why they were done I'll remain skeptical of the motives

 

[Lord of Planar]

And until evidence of fraud is presented to me I will remain skeptical of the accusations.

Admit it.

You're a denier of facts.

 

[code1211]

The correction is in opposition to my beliefs, so it must be wrong!

The correction is in opposition to common sense, but does support the Hansenization of data, so there's that.

When Hansen started his little project to correct for heat island distortions, nobody with any sense would have thought that the already too warm readings due to heat island artificial warming would have been showing a cooling bias due to the the heat island artificial warming.

Hansen managed to warm the readings taken after the heat islands were in place and cool the readings before the heat islands were in place.

This makes no sense whatever.

 

[code1211]

Not if the reason for the error would provide a bias in a particular direction. In that case, the correction would be in the opposite direction.

So you assert that all of the land station data is garbage and you agree with Hansen that it should be discarded and replaced with better data?

 

[Lord of Planar]

The correction is in opposition to common sense, but does support the Hansenization of data, so there's that.

When Hansen started his little project to correct for heat island distortions, nobody with any sense would have thought that the already too warm readings due to heat island artificial warming would have been showing a cooling bias due to the the heat island artificial warming.

Hansen managed to warm the readings taken after the heat islands were in place and cool the readings before the heat islands were in place.

This makes no sense whatever.

It does to con artists and charlatans.

 

[code1211]

Presumably that is not the only error that can exist.

That was the concern that initiated Hansen's exercise and he spent a little effort trying to isolate where the Heat Islands were and how hot they were.

 

[Manc Skipper]

Aaaaand another denial thread spirals back into irrelevance.................

 

[code1211]

And until evidence of fraud is presented to me I will remain skeptical of the accusations.

How much Estimation is too much Estimation? « Climate Audit