By Steven Goddard and Anthony Watts
Fort Collins, Colorado is most famous for Balloon Boy, and Boulder, Colorado is most famous for Jon Benet and Ward Churchill.
Both are hotbeds of Climate Science, with familiar names like Roger Pielke (Jr. and Sr.) Walt Meier, William Gray, Kevin Trenberth and Mark Sereeze. Both are of similar size (Boulder 91,000 and Fort Collins 130,000) and located in very similar geographical environments along the Front Range – about 50 miles apart. The big difference is that Fort Collins has tripled in size over the last 40 years, and Boulder has grown much more slowly. Fort Collins population is shown in blue and Boulder in red below.
Sources:
http://en.wikipedia.org/wiki/Fort_Collins,_Colorado
http://en.wikipedia.org/wiki/Boulder,_Colorado
Until the mid-1960s, NCDC temperatures in the two cities tracked each other quite closely, as you can see below. Again, Fort Collins in blue, and Boulder in red – with Fort Collins temperatures shifted upwards by two degrees to normalize the left side of the graph. Since 1965, temperatures in Fort Collins have risen much more quickly than Boulder, paralleling the relative increase in population.
Source: NCDC Boulder Temperatures NCDC Fort Collins Temperatures
The graph below shows the absolute difference between Fort Collins temperatures and Boulder temperatures since 1930. There is some sort of discontinuity around 1940, but the UHI imprint is clearly visible in the Fort Collins record. The Colorado State Climatologist, Nolan Doesken manages the Fort Collins Weather station. He has told me that it has never moved or changed instrumentation. and that he believes the increase in temperature is due to UHI effects.
Roger Pielke Sr. further commented:
“the Fort Collins site did have the introduction of the CSU Transit Center a few years ago, although this is well after the upturn in temperature differences between Boulder and Fort Collins started to increase.”
From the promotional photo on the CSU website, the Fort Collins USHCN weather station (below) seems reasonably sited.
However when you look at the Google Earth street view, you realize that it is surrounded by concrete, asphalt, nearby parking, and a building just 7.5 meters away (By the GE ruler tool). It would rate a CRN4 by the surfacestations rating. It also appears to have been modified since the promo photo was taken as there is a new fence with shrubbery and wood chips surrounding it.
Besides the pressure of CSU expansion, Fort Collins has seen an increase of about two degrees since 1970, corresponding to a population increase of 90,000. This is probably a little higher than Dr. Spencer’s estimates for UHI.
The Boulder weather station is similarly sited since the concrete path is just under 10 meters away.
It is at the campus of NOAA’s and NIST’s headquarters in Boulder. Anthony Watts visited the station in 2007 and took photos for the surfacestations project. Like Fort Collins, it gets similar expansion pressure due to nearby construction as seen in this aerial photo.
Here are the temperature records fro these two USHCN stations:
NCDC Fort Collins Temperatures
There is some UHI effect visible in the Boulder record below, but much less than Fort Collins.
Conclusion:
We have two weather stations in similarly sited urban environments. Until 1965 they tracked each other very closely. Since then, Fort Collins has seen a relative increase in temperature which tracks the relative increase in population. UHI is clearly not dead.
Willis Eschenbach (20:57:19) :
Other Excel questions related to analysis and display of temperature datasets? I’m happy to answer them.
Thank you Willis. I won’t both you at the moment for more, your offer is generous, you have give me enough to absorb for a while (but I might take you up on that later if really get stuck).
Another thing I just learned from you is that returned matrices don’t have to actually exist on the spreadsheet. I didn’t know that either! That is why I have had a bit of trouble when using the higher level functions that return a matrix. I have also written Basic macros and functions but this is new to me.
I am too much a system level programmer. I have programmed things like Excel itself, DirectX itself, I have written a photo manipulation package back in the 90’s similar to PhotoShop. When I hit a problem I deem to deep for Excel I have tended to just jump to a language and write the whole thing from the bottom up, write back to CSV, and import that to Excel. That might take an hour or so but from that point on I never hit limits. Weird, right. My mind is usually sits on the lower levels of software. It has its pluses and minuses. But, that is reinventing the wheel, right, so I greatly appreciate you tips, I will use them! I will take a look at R also.
I see what you are doing on Steve’s Boulder – Ft. Collins data. So ~1940 the difference in the x-aligned trend points before and after jump off the chart! That is the discontinuity. I am going to take some time to absorb my new Excel skills, thank again!
Willis Eschenbach (20:57:19) :
Just re-read my reply and I better rephrase
“things like Excel itself, DirectX itself”
to
“things similar to Excel itself, similar to DirectX itself”
🙂 No, I haven’t worked for Microsoft!
Willis,
Your claimed pre-1941 trend of 0.20 has an r^2 of less than 0.02. It has zero statistical significance.
You need to put your entire spreadsheet online, not post images of it. You harshly criticize Jones and Mann for the same behaviour.
There are many kinds of discontinuities which you seem to be unaware of. One example (of many) would be a tree growing and gradually shading the screen over a period of months. This would not show up as a single month discontinuity, yet it has the same effect over a year. You can not discover the history of the Boulder station from te simple statistical analysis you are attempting.
Steve Goddard (00:28:16), welcome back. You say:
Say what???
The r^2 value does not measure the statistical significance of a trend. Statistical significance is calculated by dividing the trend value (0.019°C/yr 1897-1941) by the error in the trend line (0.003°C/year). This gives 5.4.
To determine if the 5.4 is significant, you must perform a “T-test” for the particular significance level desired. This uses the number of degrees of freedom. If we want 95% significance, and we have in this case 534 degrees of freedom (months of data – 2), the T test gives a value of 2.25. Since 5.4 is greater than 2.25 the trend is indeed significant. The Excel help file explains all of this under the LINEST function, viz:
You also need to adjust the degrees of freedom for autocorrelation, which is beyond the scope of this discussion. See here for details. The trend is still significant after adjustment for autocorrelation. The number of adjusted degrees of freedom is 253.
I’m beginning to see why you are having problems with your claims …
I was trying to encourage you to do the math yourself, but I’m starting to see that may not be possible. In that case, my spreadsheet is here. It depends on an external gaussian averaging function and some other external functions, so parts of it won’t work. What does work should be enough, if not let me know.
I’m not aware of other types of discontinuities? You’ve taken up mind reading? I have discussed a host of different types of discontinuities on WUWT.
What I said was that you can’t just say, as you have done, ‘let’s add 0.7°C to the period 1942-1953″. You need to have evidence of a discontinuity to remove it. I have given such evidence. I have made no claim that there are not other types of discontinuities in record, there may well be. But until we have evidence of them, we have to use the record as it sits.
Willis,
You are shouting louder and louder.
There is very clear evidence of a discontinuity in 1920. Linest from 1895-1919 shows a negative slope. Linest from 1920-1941 shows a zero slope. In 1920, Ft. Collins minus Boulder temperature differences jumped up two degrees in 1920, and then dropped by nearly an equal amount in 1942. It is abundantly clear that there are two discontinuities.
https://spreadsheets.google.com/oimg?key=0AnKz9p_7fMvBdElxNDA4Vlh2OGhvOUdEX1N0bm1CeWc&oid=2&v=1268855768008
Your dependence on spreadsheets over observation is not good. The neural network in the human brain is vastly more powerful at identifying patterns than are computers. A baby can easily identify the difference between a man and a woman’s face, but it is an incredibly difficult problem for a computer.
Steve Goddard (12:59:03)
A change in slope doesn’t indicate a discontinuity. Squinting at a graph and mumbling about neural networks doesn’t indicate a discontinuity. That’s why special mathematical methods have been developed to show discontinuities. If we could find them by just looking, there’d be no need for the mathematical methods.
If you don’t want to use math, fine. But good luck convincing scientists by saying “Look here, the trend changes, my human brain neural network sez it must be a discontinuity, look real close, honest, it’s there …”
Math says no discontinuity in 1920. I know you don’t like that, but there it is.
Steve Goddard (13:28:36)
Trend 1895-1919 (not including 1920) 0.04°C/decade
Trend 1920-1941 (not including 1920) 0.09°C/decade
In other words, there is a trend both before and after 1920.
I’m afraid your eyeball has misled you again. There is a change in the trend in 1920, but a change in the trend != a discontinuity.
Also, please recall that your claim was that
When I showed there was a trend post 1941, your new claim was that there was no trend pre 1941. But please note:
Your original thesis was that the trend only existed post-1970, and was driven by the differential post-1970 population growth.
That claim has been resoundingly disproven. We’re now discussing other issues about the temperature record.
Steve Goddard posted this on another thread:
Steve Goddard (16:45:44) : edit
Well, let’s see. I am getting my numbers from the GISS site here, using the unadjusted figures. They show the trends I gave above, which were:
Trend 1895-1919 (not including 1920) 0.04°C/decade
Trend 1921 [wrongly noted above as 1920]-1941 (not including 1920) 0.09°C/decade
I have posted my spreadsheet, per your request, above. How about you post yours so we can see where the difference lies? We agree on the post 1920 trend, it is just the earlier one that is in question (note that my trends are per decade and yours are per year).
Perhaps you didn’t start in 1895, perhaps you used annual data rather than monthly data, I don’t know. I just re-ran my numbers ab initio and got the same result.
Finally, you say:
Since the numbers just above are the first numbers I have given for these particular trends (pre- and post- 1920), I don’t know what you mean by the trend I “originally claimed” for those intervals.
Is there a discontinuity in 1920? Possibly. All I’m saying is that you have to show that mathematically, you haven’t done that, and a change in trend is not equal to a discontinuity.
Here’s my data for the period in question, in comma-delimited form for easy import into excel
Year, Ft. Collins – Boulder Temperature
1895.04, 0.47
1895.13, -0.13
1895.21, 1.27
1895.29, 0.67
1895.38, 0.07
1895.46, 1.07
1895.54
1895.63
1895.71
1895.79
1895.88
1895.96
1896.04
1896.13
1896.21
1896.29
1896.38
1896.46
1896.54
1896.63
1896.71
1896.79
1896.88
1896.96
1897.04
1897.13
1897.21
1897.29
1897.38, 1.17
1897.46, 0.97
1897.54, -0.03
1897.63, 0.97
1897.71, -0.03
1897.79, 0.07
1897.88, -1.23
1897.96, -0.83
1898.04, -0.83
1898.13, -0.73
1898.21, 0.27
1898.29, 1.07
1898.38, 1.67
1898.46, 0.47
1898.54, 0.37
1898.63, 0.27
1898.71, -0.03
1898.79, 0.47
1898.88, -1.53
1898.96, -1.13
1899.04, -1.63
1899.13, -2.33
1899.21, -0.33
1899.29, -0.23
1899.38, 0.27
1899.46, 0.67
1899.54, 0.87
1899.63, 0.57
1899.71, -0.53
1899.79, -0.23
1899.88, -1.83
1899.96, -2.33
1900.04, -1.33
1900.13, -2.13
1900.21, -0.23
1900.29, 0.37
1900.38, 0.77
1900.46, 0.57
1900.54, 0.47
1900.63, -0.23
1900.71, 0.17
1900.79, -1.83
1900.88, -1.93
1900.96, -2.03
1901.04, -2.43
1901.13, -1.03
1901.21, -0.93
1901.29, 0.17
1901.38, 0.57
1901.46, 0.47
1901.54, 0.57
1901.63, 0.77
1901.71, 0.37
1901.79, -0.33
1901.88, -2.43
1901.96, -2.03
1902.04, -2.53
1902.13, -1.43
1902.21, -0.23
1902.29, 0.17
1902.38, 0.27
1902.46, 0.07
1902.54, 0.17
1902.63, 0.27
1902.71, -0.03
1902.79, -1.43
1902.88, -1.23
1902.96, -3.13
1903.04, -1.43
1903.13, -2.93
1903.21, -1.63
1903.29, 0.37
1903.38, 0.47
1903.46, 1.17
1903.54, 0.57
1903.63, 0.77
1903.71, -0.43
1903.79, -0.73
1903.88, -1.73
1903.96, -2.13
1904.04, -1.43
1904.13, -1.53
1904.21, -0.13
1904.29, 0.17
1904.38, 1.07
1904.46, 1.17
1904.54, 0.87
1904.63, 0.37
1904.71, -0.83
1904.79, -1.33
1904.88, -2.83
1904.96, -1.03
1905.04, -0.53
1905.13, -1.13
1905.21, 0.47
1905.29, 0.17
1905.38, 1.47
1905.46, 1.27
1905.54, 0.87
1905.63, -0.23
1905.71, -1.13
1905.79, -0.63
1905.88, -1.33
1905.96, -3.23
1906.04, -1.83
1906.13, -1.33
1906.21, 0.17
1906.29, 0.57
1906.38, 0.77
1906.46, 1.77
1906.54, 1.17
1906.63, 1.37
1906.71, 0.47
1906.79, -0.53
1906.88, 0.07
1906.96, -0.93
1907.04, -1.93
1907.13, -1.53
1907.21, -0.03
1907.29, -0.13
1907.38, 1.27
1907.46, 0.57
1907.54, 0.87
1907.63, 0.27
1907.71, -0.53
1907.79, 0.17
1907.88, -1.63
1907.96, -0.53
1908.04, -2.33
1908.13, -1.93
1908.21, -0.53
1908.29, -0.43
1908.38, 0.57
1908.46, 0.57
1908.54, 1.17
1908.63, 0.97
1908.71, -0.13
1908.79, 1.07
1908.88, -1.53
1908.96, -2.63
1909.04, -1.73
1909.13, -0.83
1909.21, 3.07
1909.29, 2.17
1909.38, 0.97
1909.46, 1.77
1909.54, 1.87
1909.63, 1.17
1909.71, 0.37
1909.79, 0.07
1909.88, -0.53
1909.96, 1.77
1910.04, -0.33
1910.13, -0.03
1910.21, -1.43
1910.29, -0.23
1910.38, 1.07
1910.46, 0.77
1910.54
1910.63
1910.71, 0.47
1910.79, 0.27
1910.88, -0.63
1910.96, -0.73
1911.04, -1.43
1911.13, -0.73
1911.21, -0.13
1911.29, 0.57
1911.38, 0.87
1911.46, 0.07
1911.54, 0.27
1911.63, 0.77
1911.71, -0.73
1911.79, 0.47
1911.88, -0.03
1911.96
1912.04
1912.13
1912.21, -1.53
1912.29, 0.17
1912.38, 0.07
1912.46, 0.07
1912.54, 0.77
1912.63, 0.27
1912.71, -0.13
1912.79, -0.63
1912.88, -1.23
1912.96, -0.53
1913.04, -1.73
1913.13, -0.53
1913.21, 0.07
1913.29, -0.03
1913.38, 1.07
1913.46, 0.47
1913.54, 0.27
1913.63, 0.97
1913.71, 0.07
1913.79, 0.27
1913.88, -1.13
1913.96, -3.23
1914.04, -2.63
1914.13, -2.03
1914.21, -0.53
1914.29, 0.17
1914.38, 0.57
1914.46, -0.73
1914.54, 0.67
1914.63, 0.47
1914.71, -0.63
1914.79, -0.03
1914.88, -1.93
1914.96, 0.37
1915.04, -1.13
1915.13, -0.03
1915.21, 1.27
1915.29, 1.07
1915.38, -0.13
1915.46, 0.17
1915.54, 0.17
1915.63, 0.87
1915.71, -0.33
1915.79, -0.93
1915.88, -0.93
1915.96, -0.23
1916.04, -1.83
1916.13, -0.73
1916.21, -0.03
1916.29, -0.33
1916.38, -1.03
1916.46, -1.13
1916.54, -0.73
1916.63, 1.07
1916.71, -0.23
1916.79, 0.57
1916.88, -2.03
1916.96, 0.17
1917.04, -0.93
1917.13, 0.57
1917.21, 0.67
1917.29, 1.27
1917.38, 1.07
1917.46, -0.23
1917.54, -0.63
1917.63, -0.33
1917.71, 0.17
1917.79, 0.87
1917.88, 0.17
1917.96, -0.63
1918.04, -0.43
1918.13, 0.37
1918.21, 0.27
1918.29, 1.07
1918.38, -0.93
1918.46, 0.87
1918.54, -1.03
1918.63, -1.73
1918.71, -0.73
1918.79, 0.87
1918.88, 0.07
1918.96, 0.17
1919.04, -0.83
1919.13, 0.57
1919.21, 1.67
1919.29, 1.27
1919.38, 0.97
1919.46, 0.47
1919.54, -0.03
1919.63, -1.43
1919.71, -0.73
1919.79, -2.03
1919.88, -1.23
1919.96, -4.03