Average, Average 0, Average 6, Average 12, Average 18,
0.14, 0.06, -0.04, -0.01, 0.59
Max
4.85,2.05, 0.50, 0.50, 4.85
Min
-0.50, -0.50, -0.50, -0.50, -0.50
STDEV
0.73, 0.41, 0.32, 0.30, 1.28

The table above shows the Average, Max, Min and StDev of the differences for each recording hour. The worst correspondence is for the 18:00 observations. It could be if these are different locations there is a shadowing that affects the 18:00 observations but not the others.

I think I have looked at this data enough. Strangeness abounds. Why the double readings? Why the changing GISS correction. Why does this data show no change over time but the GISS data shows warming?

Do enjoy reading your articles, especially on the weather station sites.

During the time period after the double readings stopped (6/14/04 on) there were not fewer 0:00 readings as I supposed in my last post but more 6:00 readings (6 more to be exact) than the 0;00, 12;00 and 18:00 readings.

For the average of the Max and min I just averaged the max and min data for each day then plotted versus the date. This would have been easier if I had sorted the data to collapse the Max min data, but I just cut and pasted an average formula. For most of the data there were four hourly observations and the max min so it was regular, near the end of the four years there were more than the 4 daily observations. Like two 6 am observations with slightly different temperatures, Some days were missing data as well. I think the time frame, even 8 years, is too short to make too much of the trends. The difference in the GISS data I do not understand, but then I do not know what they are correcting for. I would assume correcting for altitude and such would give more of a constant change in the numbers not the decreasing correction seen.

Looking at the every 6 hour data it looks like data is not collected automatically especially since it appears that more midnight data is missing than others (would have to check to verify that), When they were doing duplicate data for each observation it is like they were changing thermometers or station sites perhaps and wanted a little bit of time to overlap the data to check for consistency.

Can you share your method and or files, such as an Excel file or other tool?

- Anthony

Sure looks strange. If anyone from GISS, NOAA or RC is reading this can you explain why there is such a big difference.

Year GISS Data Daily Data
2001 -16.46 -13.35
2002 -14.41 -12.28
2003 -14.51 -12.32
2004 -15.87 -14.38
2005 -13.19 -12.16
2006 -13.82 -14.42
2007 -13.12 -13.42

Obviously the GISS corrections change the data trends.

Trend GISS +0.45 C/yr, Trend Daily data -0.12 C/yr

The min and max data is only for the first 4 years. A trend on all the ave temp data using the max min data was -2.1 C/yr. Using only the March to March data it was -3.125 /yr. Using the eight years of 4 times a day observations gives a trend of +0.3 C/yr, the March to March 4 times a day observation for the eight years gives a trend of -.84 C/yr.

Doesn’t match the end of the yearly plot very well. Will look at the other data set.

I would assume the person reading the temp comes up and turns on the light just before doing the reading.

“Expanding forests and changing growth forms of Siberian larch at the Polar Urals treeline during the 20th century”

The aim of this study was to reconstruct stand structure and growth forms of Larix sibirica (Ledeb.) in undisturbed forest–tundra ecotones of the remote Polar Urals on a centennial time scale. Comparisons of the current ecotone with historic photographs from the 1960s clearly document that forests have significantly expanded since then.

However:
Because thousands of more than 500-year-old subfossil trees occur in the same area but tree remnants of the 15–19th century are lacking almost entirely, we conclude that the forest has been expanding upwards into the formerly tree-free tundra during the last century by about 20–60 m in altitude.

So once upon the time, there was no tundra, but forest…

Having been around a lot of “natural fertilizers” I can assure you that it breaks down into large quantities of “greenhouse gasses” including many hydrocarbon containing aromatic ones! BTW, where do the sequestering trees go in 25 to 50 years? Or do we need to keep doing this on new land each 50 years (I assume they would be using a very high growth tree species that reaches maturity in 25-50 years for max sequestration.)

Before we head down this rat hole would it not be easier to just stop cutting down the existing rainforest? Oh right, since it’s mature it is not a net sequestration so no carbon credits. Guess we need to plant new forests then and let the old ones go… gotta get those carbon credits.

How can these looneytoons in one breath say farming causes GHGs and in the next breath say we can cure GHGs with tree farming. Sheesh.

(Disclaimer: Elsewhere on the net I’ve advocated tree farms for biofuel production and I’ve shown that if we wanted a one time short duration suck down of CO2 we could do in a rapid forest system. But I’ve not been from the “we must sequester CO2 forever” camp, so have no hypocrisy in tossing darts at the idea of trees as long term sequestration.)

Should read, “I’ll have to ponder all this a bit more but thanks for setting me off down that path.” (I don’t think you were headed that way. ) Thanks again.

You wrote in part, “… I had thought that global mean temperatures were based on the mean of daily min and max temps. How does GISTEMP calculate annual mean temp in the absence of recorded minimum and maximum temps?”

That set me thinking. What does the mean of max and min tell anyone anyhow? Doesn’t the length of time that a temperature persists throughout a day count for anything?

Consider a typical Indian Summer Fall day in the midwest. It might start with overnight lows in the 50’s and warm up to 70 degrees F where it remains for most of the day. Then, along about, say, 10:00 pm, a clipper could come through and drop the temperature to let’s say 20 degrees F. The mean for the day would be 45 degrees F but really, the bulk of the day was mostly in the 60-70 degree F range. I’d argue that the better characterization of the temperature for a given day be the time weighted average average for the max and min temps.

I’ll have to ponder all this a bit more but thanks for setting off down that path.

REPLY: I’m not suggesting single pipes 100 yards away are the big difference to what happens inside the screen. What I AM suggesting is that we have a significant city-wide active UHI driven by a power/steam/hot water plant.

Let’s say you have a steam/hot water plant that puts out 5 megawatts, which given the size of the power plant I see in the photos, (the building with blue roof here) it could be a reasonable guess…thanks to the piping system, that heat dissipates over the entire town. In a small town like this, that gives a fairly well concentrated heat source. It doesn’t matter if the pipes are buried or above ground, the heat will dissipate into the area.

Wind will distort the heat plume downwind, as we’ve seen from other UHI studies. So even a weather station on the edge of town (as this one is) could be affected.

As a benchmark, look at this site, Cordova Alaska, a town about the same size, which has a 7 megawatt capable plant:

http://wattsupwiththat.com/2007/12/10/how-not-to-measure-temperature-part-45/

Now look at the size of the Russian town – a couple of square miles, dissipate 5 megawatts over a couple of square miles…that is bound to raise the local air temperature some.

Back of the envelope calcs: 2 (square miles) = 5,179,976.22 square meters or about a forcing of 1 watt per square meter. That is significant.

So if this city wide dissipated heat means the regular difference between -30C and -25C for example, a 5C anomaly will show up as bright red in the GISTEMP analysis, but it will still be bitterly cold…there will be no visible effects, snow and ice will still stay in place (except maybe over/under heat pipes).

Then there is humidity. Depending on how much water is lost from the system in the area can also be a factor. As we know from Christy’s study of the San Jaquin Valley, extra night time humidity can effect Tmin Again with a lot of water vapor being put into the air that may affect the temperatures. Note the plume in one photo on the left side within the station viewshed.

UHI from waste heat and waste water vapor doesn’t have to be “warm” to create an anomaly, and we are talking about temperature anomalies here that are subzero. Without some good data on the powerplant, the town, and the state of the system it will be hard to pin down exactly. But, as they say on Mythbusters: “I think it’s plausible” that we have a UHI effect in Verhojansk.

Here is one place where modeling could actually do something useful…model the town, the power, etc and see where the waste heat goes and how much it raises the air temp. – Anthony

REPLY: I feel a vodka and borscht run coming on. – Anthony