The Climate Conspiracy Exposing the deceit of climate scientists

More on Bad Temperature Data

In a previous post I discussed the poor quality temperature data which is found in the United States and showed that the data is manipulated to give it a "warming" trend. Today, I am going to show how this bad data compares to good data.

There are basically two primary ways that we collect current temperature information from the earth:

The first is based on land temperatures, in which we use thermometers to measure the temperature. The temperatures are gathered from stations and ships all around the world and then the results are compiled into different data sets. As we previously saw, these data sets are then adjusted and manipulated in order to account for missing data, and to impose a warming bias into the data. One such data set is the HADCRUT3 data set. It is maintained by the Hadley Centre Climatic Research Unit. Yes, these are the same people implicated in ClimateGate.

The second is based on satellite temperature measurements of the lower troposphere. Satellites do not measure temperature as such. They measure radiances in various wavelength bands, which must then be mathematically inverted to obtain indirect inferences of temperature. A good thing about satellite data is that it can't be affected by a guy BBQing underneath the thermometer, nor can it be affected by being placed in the middle of an asphalt parking lot. One such data set is from the University of Alabama in Huntsville (UAH), primarily overseen by Dr. John Christy.

Today I will demonstrate some of the effects of the poor temperature data. We will look at the period of warming that alarmists claim was caused by human produced greenhouse gases. (We already demonstrated that the warming in that period occurred at the exact same rate as as the warming from 1910 - 1940.)

The satellite data starts in 1979 and runs through 2010. We will look at the "warming period" from 1979 - 1998, comparing the slopes of the land-based HadCRUT3 data to the satellite-based UAH data.

Land HadCRUT3 vs Satellite UAH (1979-1998)

Data Set	Slope ('C/year)	'C/decade	'C/century
HadCRUT3 (land)	0.011	0.11	1.1
UAH (satellite)	0.004	0.04	0.4
Possible Surface Error (HadCRUT3 minus UAH)	0.007	0.07	0.7

Comparing the slopes of the land-based HadCRUT3 data to the satellite-based UAH data.

(If there are any concerns about cherry picked data, rest assured. Both data sets have been cherry picked for the same years. Unfortunately we can only go back to 1979 because that is when the satellite data starts.)

If we look at the slopes, we can see that the HadCRUT3 data has a slope that is 2.75 times greater than that of the UAH data. This means that the HadCRUT3 data indicates a rate of warming 2.75 times greater than that of the UAH data for this period. This allows us to clearly see the influence of their adjusting earlier station records downwards to give the appearance of a greater warming trend.

In the bottom row of the table I calculated the possible surface temperature error. To do this I subtracted the UAH data from the HadCRUT3 data. This difference is the amount of warming that has possibly been introduced by all of the problems with station data. For this period of time, we can clearly see that the possible surface error is on the order of 0.7'C per century, which is greater than the amount of warming we have supposedly seen this century.

Note, there are other analyses like these out there. All of which highlight the differences between the land based and satellite based data. There is a very comprehensive one at WUWT, an excellent one at Warwick Hughes, and a very well explained one at Carlin Economics.