UPDATE: A good photo of one of the Russian stations has been found, see below after the “read more” link.
As most readers know by now, the problematic GISTEMP global temperature anomaly plot for October is heavily weighted by temperatures from weather stations in Russia.
GISTEMP 11-12-08 – Click for larger image
Like in the USA, weather stations tend to be distributed according to population density, with the more populated western portion of Russia having more weather stations than the less populated eastern areas such as Siberia. To illustrate this, here is a plot of Russian Weather Station locations from the University of Melbourne:
Click picture for larger image, source image is here
Interestingly, the greatest magnitude of the GISTEMP anomaly plot for October is in these mostly unpopulated areas where the weather station density is the lowest. While I was pondering this curiosity, one of the WUWT readers, Corky Boyd, did a little research and passed this along in email:
…Posters at Watts Up have commented on the ongoing consistently high anomalous temperatures from Russia. I have noticed this too. In light of the erroneously posted data for October, I took a look at the monthly NCDC climate reports back to January 2007. By my eyeball estimate the results from Russia are almost all on the high side. . Some I classified as very highs are massively high. Of the 21 months reported, only 2 appeared to be below average.
Category 2007 2008 (9 months)
Very high 6 4
High 3 1
Average 2 3
Low 0 1
Very Low 1 0
Is there a way to validate or invalidate GISS data by comparing it to RISS? Does it strike you as odd that the verifiably erroneous data has shown up in the same area that was suspect in the first place? Could there be a pattern?
Corky also sent along a series of images depicting global near surface and ocean temperature anomalies from NOAA. Here is the most recent one from September 2008:
I was curious if indeed there was any pattern to the Russian anomaly, so I decided to animate the last year and a half worth of images. You can see this animation below. It is about 1 megabyte in size, so please be patient while it downloads.
What I found interesting was that the January 2007 anomaly (the last time we had a “global heat wave”) was primarily in the northern Russian and Asian. According to January 2007 UAH satellite anomaly data, the Northern Hemisphere had a whopping anomaly of +1.08°C and the “northern extent” was even greater at +1.27°C, the largest anomaly ever in the Northern Extent dataset
Curiously though, the very next month, the Russian anomaly virtually disappears and is replacing with cooling, though a sharp boundary to warming now exists in Asia. It was as if somebody threw a switch in Russia.
Click for larger images
In March 2008, a very large positive anomaly returned in Russia, and again in April evaporated with the same abruptness as the Jan-Feb 2007 transition. Again almost as if a switch was thrown.
Click for larger images
Such abrupt repeated changes don’t seem fully natural to me, particularly when they occur over the same geographic location twice. I realize that two events don’t make a trend, but it is curious, given that we now have had a problem with Russian weather data again that caused GISS to announce the “hottest October on record”.
I also noticed that in the animation from the anomaly maps, there does not seem to be much of an anomaly in the summer months.
This made me wonder what some of those weather stations in Russia might be like. So I went to the Russian Meteorological Institute website at http://www.meteo.ru/english/
I know from John Goetz work as well as this artcle in Nature that Russian weather stations had been closing with regularity due to the trickle down effects of collapse in the former Soviet Union. Though some new ones are being built by outside agencies, such as this one sponsored by NOAA in Tiksi, Russia.
What I found interesting in the NOAA press release on Tiksi, was this image, showing weather stations clustered around the Arctic:
The interesting thing is that all these stations are manned and heated. The instruments appear to be “on” the buildings themselves, though it is hard to tell. One wonders how much of the building heat in this tiny island of humanity makes it to the sensors. The need for a manned weather station in the Arctic always comes with a need for heat.
I was hoping my visit to the Russian Meteorological institute website might have some particulars on the remaining weather stations that have not been closed. I didn’t find that, but what I did find was a study they posted that seems to point to a significant warm temperature anomaly in Russia during winters between 1961 to 1998:
Fig. 1. Linear trend coefficient (days/10 years) in the series of days with abnormally high air temperatures in winter (December-February), 1961-1998.
From the Russian study they write:
For the winter period 1961-1998, most of the stations under considerations exhibit a tendency for fewer minimum temperature extremes. Maximum (in absolute value) coefficients of the linear trend were obtained in the south of the country and in eastern Yakutia.
Whenever I read about elevated minimum temperatures, I tend to suspect some sort of human influences such as UHI, station siting, or irrigation (humidity) which tend to affect Tmin more than Tmax.
In Northern Russia Siberia, I wouldn’t expect much in the way of irrigation. So that leaves station siting and UHI as possible biases. UHI seemed doubtful, given that many of these Russian Stations in Siberia are in remote areas and small towns.
So I decided to put Google Earth to work to see what I could see. One of the stations mentioned in a recent post at Climate Audit cited the station of Verhojansk, Russia, which has temperatures conveniently online here at Weather Underground.
From the Navy Meteorological exercise I found that Verhojansk has a wide variance in temperature:
Verkhojansk is located in a treeless shallow valley. There is snow on the ground during winter months; it melts in the spring. Verhojansk experiences the coldest winter temperatures of any official weather station outside of Antarctica. Verhojansk has Earth’s most extreme temperature contrast (65oC) between summer and winter. Which of the following indirect factors contribute to this extreme seasonal variation?
From the GHCN station inventory file at NCDC I found that Verhojansk, Russia had a lat/lon of 67.55 133.38 which when I put it in Google Earth, ended up in a mud flat. The Google Maps link from Weather Underground was no better, also off in a field.
Looking in NCDC’s MMS station database yeilded better luck, and I found a more precise lat/lon of 67.55,133.38333 There was very little other helpful information there on the station.
The station appeared to be located in town, though I have no way of verifying the exact location. The lat/lon may be imprecise. But something curious popped out at me as I was scanning the Google Earth image of Verhojansk looking for what might be a weather station – it looks like pipes running across the surface:
These “pipes” appear to go all over town. Here is a closer view, note the arrow to what I think might be the weather station location based on the fencing, objects on the ground that could be rain gauges or shelters, and what looks like an instrument tower:
I was curious about what these pipes could be, it certainly didn’t look like oil pipelines, and why where they so close to houses and building and seem to network all over town. Doing a little research on Russian history, I found my answer in the pervasive “central planning” thinking that characterized Russian government and infrastructure. It’s called “District Heating“
District heating (less commonly called teleheating) is a system for distributing heat generated in a centralized location for residential and commercial heating requirements such as space heating and water heating.
But for Russia there was this caveat:
In most Russian cities, district-level combined heat and power plants (Russian: ТЭЦ, Тепло-электро централь) produce more than 50 % of the nation’s electricity and simultaneously provide hot water for neighbouring city blocks. They mostly use coal and oil-powered steam turbines for cogeneration of heat. Now, gas turbines and combined cycle designs are beginning to be widely used as well. A Soviet-era approach of using very large central stations to heat large districts of a big city or entire small cities is fading away as due to inefficiency, much heat is lost in the piping network because of leakages and lack of proper thermal insulation .
I should also point out that district heating is not limited to Russia, but is in many northern European countries. It seems quite prevalent in cold Euro-climates, and even extends into Great Britain.
So I searched a bit more, and found some pictures of what Russian district heating looks like from the ground. Here is one from Picasaweb from somebody’s trip to Russia:
Click for source image.
Note the pipes in the photo above are not insulated.
I also found a very interesting picture of steam pipes, also uninsulated, from a trip report to the “hot zone” of Chernobyl:
And finally a picture of Krasnoyarsk thermal power station Number 1 that has recently been in the news, according to Reuters due to a burst steam pipe:
Click for larger image – Note the pipes coming out to the left of the power station. You can see steam pipes around the city in this Google Maps view here.
So all this begs the question:
If Russian weather stations are located in cities that have this district heating plan, and a good percentage of the pipes are uninsulated, how much of the waste heat from the pipes ends up creating a local micro-climate of warmth?
Remember when I said that the NOAA map anomalies centered over Russia seemed to be prevalent in winter but not summer? It stands to reason that as winter temperature gets colder, more waste heat is dumped out of these inefficient systems to meet the demand. Basically, we have an active UHI situation in the city that increases in output as temperatures drop.
In the areal photos above of Verhojansk, it appears that some pipes are insulated (white, what appears to be main lines) while others are rust brown, and appear near buildings and dwellings.
I got to thinking about why these pipes might be uninsulated. First there is the classic inefficiency and carelessness of Soviet workmanship, but another thought occurred to me: Russian people might like it that way. Why? Well imagine a place where you walk to the market every day, even in subzero temperatures. Since many of these pipes seem to follow streets and sidewalks, wouldn’t it be a more pleasant walk in winter next to a nice toasty steam pipe?
Steve Mcintyre wrote about this station at Climate Audit, citing a puzzle in the data, here is an excerpt of his post:
Now there are many puzzles in GHCN adjustments, to say the least, and these adjustments are inhaled into GISS. Verhojansk is in the heart of the Siberian “hot spot”, presently a balmy minus 22 deg C. The graphics below compare GISS dset0 in the most recent scribal version to GISS dset 2 (showing identity other than small discrepancies at the start of the segment); the right compares GISS dset0 to the GHCN-Daily Average.
Over the past 20 years, the GISS version (presumably obtained from GHCN monthly) has risen 1.7 deg C (!) relative to the average taken from GHCN Daily results.
What causes this? I have no idea.
Maybe it’s the steam pipes. We need to send somebody to Russia to find out. Of the many station lat/lons I looked at, Verhojansk was the only one I found with enough Google Earth resolution to see the steam pipes. Maybe the heart of our Russian temperature anomaly lies in central heating.
George Costanza could be right.
UPDATE: The photo below shows the Verhojansk Meteorological station and it’s instruments. Hat tip to Jeff C. for the photo below:
Direct URL to the photo above here
Note the cable going to the Stevenson Screen suggesting automated readings. Also note the vertical plume at left.
The station can be seen from Google Earth here