Feb 15 2008
A Surface Station You’ll Never See Profiled at Climate Audit
Anthony Watts posts at Climate Audit about the surface station at Lampasas, Texas. It seems he’s found another station that is biased by the microclimate effects due to station sitting. The station is located near a parking lot and a building - something which always causes a great kerfuffle among the surface station crowd. Watts notes that while this station may measure the weather of downtown Lampasas well, it may not accurately represent the regional climate. This post looks at the raw data from this station, and the raw data from another station in an area not influenced by humans. [Added: By "this post" I mean on this site, not the post on ClimateAudit.]
There seems to be at least two contentions with the surface station record: the location of the actual station including movement of the station, and any adjustments made to the data. While these two issues are related, especially if the the adjustments are done well, it is possible to include “corrections” that degrade the data. I think they are concerned with both issues with this station, as articulated by Watts:
But the big surprise of just how bad this location is came from the GISS plot of temperature. It clearly showed the results of the move to this location, causing a jump in temperature almost off the current graph scale. Note that before the move, the temperature trend of Lampasas was nearly flat from 1980-2000.
USHCN Raw Data vs. GISS
This could very well be. However, we also know that GISS takes the available temperature data from the USHCN and modifies it. Therefore, if one wishes to discuss changes in temperature due to station moves, it would be preferable to look at the raw data. Fortunately, I have the raw data from the USHCN for the Lampasas station.
|
As can clearly be seen, the raw temperature at this station has clearly been decreasing since the 1930s (top panel). This raw data has a first order correction that automatically looks for outliers and reports those as bad values. The bottom panel shows the USHCN filnet data which they describe as filling in missing data based on surrounding station data.
The GISS temperature for this station has clearly “hockey-sticked” since the year 2000. I do not know the cause of this. But it is clearly a problem / feature of their algorithm and not an artifact in the data due to a station move. If the problem occurs at the same temporal location as the station move, it is possible that GISS does not account for station moves in one or more of their algorithms.
A Look at a Good Station
This discrepancy between the raw temperatures in the USHCN and those at GISS got me looking at other surface stations. I was looking for stations that were not influenced by microscale features and that were cataloged by surfacestations.org. In addition to those constraints, I was looking for stations that were classified by the USHCN as being not influenced by land use change. They offer a product that displays the land cover within certain radius’ from the station. I wanted stations that were far away from human-made structures at all three distances they report. This meant I wanted a number between 1 and 4 (natural surfaces) out to 10 km from the station.
|
The first one I found was Miles City, Montana. USHCN reports the land cover for all three distances from the station to be open farmland, grassland, or tundra. Thanks to surfacestations.org we can see what the station looks like.
|
Surface station photo courtesy of Anthony Watts, www.surfacestations.org and Craig Limesand.
I think everyone would agree that this is a nicely-sitted station. So I went to see what surfacestations.org rated this station. I was expecting it to get the best rating. What I found was something completely different.
|
This station, although it had been surveyed, did not have a ranking. Let me help you out: it’s a 1. I know that looking at the trend of one station doesn’t tell us much, but other folks like to do it. So here’s the temperature from this nice station in Montana. The trend after adjusting for missing values is almost exactly the same as reported by RSS. Also notice how the adjustment reduces the temperature trend.
|
Conclusion
Individual surface stations may be influenced by their local environment. However, this effect seems to be small given that the Lampasas station is heavily influenced by humans and is experiencing a decrease in maximum temperatures over the last few decades. Adjustments made by GISS to this station seem to add an artifact that shows spurious warming since the year 2000. Other stations such as Miles City, MT show a temperature increase without the influence of humans.
(No Ratings Yet)
Loading ...Related Posts:
23 Responses to “A Surface Station You’ll Never See Profiled at Climate Audit”
To reduce spam, comments are automatically closed 30 days after the last comment. If you would like to comment on any closed thread, please use the contact form at the top of this page.
Atmoz, Good find!
A couple points and then a bet
You highlighted the Miles City site. A fine RURAL site. You wonder why Anthony Watts has not “rated it.” Well, let me speculate and later on I’ll call him and confirm if you like. The Survey was completed on 1/9/08. Typically Anthony will wait until he has a fair number of sites before rating them. Simply, he doesnt rate them as they come in. He does periodic updates, not real time updates. However, you were able to see from the pictures that it was a WELL SITED SITE. So you recognize the value of verifying the noaa
description of the site.
AND, its rural. according to NOAA and according to HANSEN.
One other thing we know. Hansen says that URBAN SITES
are adjusted to match RURAL SITES. The code is posted you
can go see this.
Since this site is rural, The we have a GOOD TEST of GISS Urban adjustments. Correct?
Its simple. We start with this file:
GISS Stage 1: GHCN + USCHCN adjustments:
http://data.giss.nasa.gov/work/gistemp/STATIONS//tmp.425742300020.0.1/station.txt
That is Miles city data as NASA inputs it from GHCN and USHCN. They INGEST the USHCN adjustments for TOBS, Station moves, and
Filnet. Again, you can go read the source code and see this.
THEN, GISS adjust the data for HOMOGENIETY. Essentially, Urban adjustments. Now, the code for this is available. And papers have described it. URBAN SITES are adjusted to match rural sites within 1000km.
Miles city is Rural as you note. So, do you think it would OK to
adjust its temperature UPWARD based on comparions with nearby
sites? It’s class 1. Its Rural. What would you bet? It’s rural. It meets
standards, you would think that other sites adjust to it. You wouldnt adjust this site one way or the other. You’d adjust the
bad sites.. right?
The next data file to look at is the GISS ADJUSTED. What does nasa do to a class 1 site? Guess?
I dont know. I havent looked at it. It’s a good blind test.
http://data.giss.nasa.gov/work/gistemp/STATIONS//tmp.425742300020.2.1/station.txt
If GISS adjustments for UHI correct errors in URBAN sites, Then
It would be a good bet that this site, since it is RURAL, would undergo no adjustments. After all, What would the sense be in adjusting a Class 1 site.
So, Take the annual temperature averages for Miles City AFTER adjustment and then subtract the Raw+USCHN data. Tell me what you see. I’m curious.
lastly, funny that the site ENDS in 2006.
It’s a good site. Looks like it’s been dropped from the network
As I understand it, data from a rural site could be adjusted if there are nearby rural sites that differ.
[...] upon a previous post about the surface station at Miles City, Montana, this entry will again look at that station, but [...]
Normally I do a group of sites about once every month or so, since it is more efficient to do it that way as I have to have a number of resources simultaneously available to complete the ratings. I didn’t know Miles City was missing a rating. So its just a simple oversight. I’ll get it with the next batch which I plan to do shortly. I think in Craig’s case he also had a transfer failure with this survey, and it wasn’t complete which is why it got passed over on the first pass.
Thanks for catching it.
As for ASOS stations in general, some like this one do in fact get better site ratings, some, like Reno, far worse. I concur with you this Miles City site would rate a CRN1 based on a cursory review.
But there is another complication now for ASOS sites. The HO83 hygrothermometer which has been proven to be unreliable and biased. See this post:
So even though Miles City is an excellent exposure, it is hard to say how accurate it’s record is not knowing how much warming bias the HO83 added to it. The HO83 only adds a warm bias, and has never shown to be cooler by any studies. Measurements from HO83 sensors still remain in the climate records, and no attempt has been made to correct them.
Unfortunately I don’t have access to maintenance and replacement records. Note that Miles City has had the HO83 upgraded to a dual sensor system, with newer Vaisala equipment piggy backing on it (it is the instrument package to the immediate right of the blue/green rain gauge wind shield).
So it is possible that Miles City warming you note may have a root in the HO83 problems.
BTW since you’ve made a recent habit of using my posts for your commentary, how about adding me to the blog roll?
Thanks for your consideration.
Oh forgot to mention:Your sentence: “This post looks at the raw data from this station, and the raw data from another station in an area not influenced by humans.”, isn’t exactly true. The survey from the Blanco TX station (the “another station”) is now online. While it is more rural than Lampasas, it certainly wasn’t and isn’t free of human influence.
http://gallery.surfacestations.org/main.php?g2_itemId=7154
[Reply: By "this post" I meant on this site not on ClimateAudit. The two stations being Lampasas ("bad") and Miles City ("good").]
The first two plots in the post indicate a cooling trend at Lampasas, if I’m reading them correctly.
But if I correctly understand the original Lampasas post at CA, http://www.climateaudit.org/?p=2731, GISS/NASA has a warming trend.
Must be that ‘relative to’ thing?
Steve Bloom,
I could be wrong, but rurals are not adjusted to other rurals.
Could you site the hansen paper which states this. Also, thats
not what I recall from the code. futher the adjustments in this site
dont appear to follow hansens described algorithm
Atmoz: “If the problem occurs at the same temporal location as the station move, it is possible that GISS does not account for station moves in one or more of their algorithms.”
Does “temporal location” mean time? If it does, we can reduce your sentence to “If the problem arises simultaneously with the move, it is possible that GISS does not account for the move.”
Why on earth would GISS behave in such an irresponsible way?
Bloom. hansen 2001:
In the prior GISS analysis the time series for temperature change at an urban station
was adjusted such that the temperature trends prior to 1950 and after 1950 were the same as the mean trends for all
“rural” stations (population less than 10,000) located within 1000 km (with the rural stations weighted inversely
with distance). In other words it was a two-legged adjustment with the two legs hinged at 1950 and with the slopes
of the two lines chosen to minimize the mean square difference between the adjusted urban record and the mean of
its rural neighbors.
The urban adjustment in the current GISS analysis is a similar two-legged adjustment, but the date of the
hinge point is no longer fixed at 1950, the maximum distance used for rural neighbors is 500 km provided that
sufficient stations are available, and “small-town” (population 10,000 to 50,000) stations are also adjusted. The
hinge date is now also chosen to minimize the difference between the adjusted urban record and the mean of its
neighbors. In the United States (and nearby Canada and Mexico regions) the rural stations are now those that are
“unlit” in satellite data, but in the rest of the world, rural stations are still defined to be places with a population less
than 10,000. The added flexibility in the hinge point allows more realistic local adjustments, as the initiation of
significant urban growth occurred at different times in different parts of the world.
H2001 makes no mention of adjusting rural stations. The description above matches the code. I’ll keep looking, I could be wrong
Has anybody counted how many adjustments or adjustment algorithms were mentioned in the article obove?
Each station has its own very special characteristics, behaving in winter different from summer, at night different from from daytime, depends on air condition habits of the owners of the site, their vacations, on business sites from business cycles, from change of business and customer frequency …… One fits all? Or two or three fit all? I don’t believe that adjustments can be done on objective criteria.
This reminds me to all the adjustments the medival astronomers made to the calculations of planet courses or orbits - just to fit their calculations to their earth centric thinking.
Why are the graphs you posted for the Lampasas site truncated? They don’t show the sudden jump in temperature that was the actual topic of discussion at Climate Audit.
Miles City is “bright,” as is Lampasas. Unless I am completely missing something, they are both urban stations as far as GISTEMP is concerned.
http://data.giss.nasa.gov/gistemp/station_data/v2.temperature.inv.txt
“However, we also know that GISS takes the available temperature data from the USHCN and modifies it. ”
So I’m looking at the USHCN data that you have presented - raw and processed, and I look at the GISS data, and I’m trying to figure out where USHCN data became GISS raw data. Does GISS use USHCN raw or adjusted as it’s raw data? Frankly, when I look at the raw GISS data, there seems to be little correlation with either USHCN data set. It almost looks like a different data set. Can you help me with this?
mosh, I’m unfamiliar with the details, but my understanding is that there is a method to catch extreme outliers even if rural. Without such a method, there would be no way to remove rural station with, e.g., readings that go out of whack due to a bad instrument. OTOH, is it possible that GISS simply drops such stations when they spot them?
Re cce’s comment, wouldn’t that “bright” rating mean they don’t get used?
Re cce
“Unless I am completely missing something,”
Looks like you are. Both places are listed as Rural in the link that you gave us.
Bloom and cce. to clarify:
in the US Hansen 2001 uses nightlights to put cities into three bins: unlit, dim,bright. Miles city is in the dim category. Its population <10000 puts it in the rural category. In the rest of the world hansen2001 uses population to determine rural.
Both dim and bright are adjusted to match unlit sites with 500km
or 1000km if necessary. If there are less than 3 rural sites within
this radius the site is dropped.
so in the US he uses nightlights. in the ROW population.
Nightlights is a proxy for urbanization. the satillite photo was taken in 1995, the pixel extent is 2.7km. Since then other reasearchers have employed some techiques to get the resolution down to 1km
there are other stdies that might be better to determine urbanization, I’m research some now. But i’ve found urban sites with lights =0 and rural sites with lights in the dim category.
I thing H2001 also has a table on this, at the end
correction; miles is in the bright category, which seems odd. the brightness index is 26. also, some places with brighness index of 0
are called dim. more puzzles
[...] last two posts have been about the surface stations, and this one follows in its footsteps. This is also [...]
Atmoz, you say “The GISS temperature for this station has clearly “hockey-sticked” since the year 2000. I do not know the cause of this. But it is clearly a problem / feature of their algorithm and not an artifact in the data due to a station move. ”
Pardon but to me the jump from the station move seems pretty obvious in both raw and adjusted data.
Secondly, adjustments that in effect downshifts historical values when introducing an modern day outlier would fail exam at my university, at least without very strong motivation. While not completely impossible it is much more likely that a jump today is due to a problem with todays measurement, than an indication that all previous values are wrong. GISS and Mr Hansen has quite a few things to explain to the world.
Hello Atmoz!
Suspecting that the integrety of many land-based weather stations have been comprised over the years and thus produce inaccurate data, the late John Daly analyzed the temperature records of several hundred mostly rual (i.e, remote) weather stations and displayed the results as temperature-time plot of the annual mean temperature. Go: http://www.john-daly.com, scroll down, and click on “Station Temperature Data”.
These results are quite spectacular and in some cases absolutely astounding, e.g., Death Valley whose plot shows a flat lines since 1927. He specifically mentions Alice Springs AU which is located in the middle of continent and in the desert and whose temperature records start in 1879. The plot for Alice Spring show little or no change in the annual mean temperature, i.e., the deviation from the mean is zero for the last few years as compared to the early part of the temperature record.
The results of his analyses confirmed quite conclusively his suspicions, and showed without a doubt that there is no evidence for any global warming. Given the number and world-wide distribution of these sites, I whole-heartedly concur with his conclusions.
Unfortuantely, John Daly passed away quite unexpectedly in 2004 from a heart attack. He was only 61 and many of his plots end in ca. 2000-01. It would be of extreme interest to see addtions of data to end of 2007. I going to do this for Alice Springs, Death Valley, Yuma and Tombstone AZ.
I’m really keen on desert weather stations since due to the low rel. humidty most of the complicating effects water vapor are minimal.
More importantly there is usually little if any human activity to bias
temperature measurments.
Deserts temperature records are the crucial data that can easily shoot down all this climate change claptrap and global wrming gobblygook. Deserts get very hot in the daytime because the heat energy from surface (and plants in some cases) is removed mostly by conduction and convection and to some extent by emission of IR energy. However, after the sun sets the temperature plunges quite rapidly and often to near freezing. This due to the absence of warter vapor. If CO2 has any effect on air temperature, then we would expect the mean minimum temperature, which usually occurs just before sunrise, should show a small but discernable increase
from the ca. 1900 to present and should follow the increase in the concentration of CO2. Since the plots from Death Valley are stratight flat lines, I concluded that CO2 has no role in climate at least at this site.
I posted the ref to John’s website over at RC, but Gavin the Grinch wacked it. He and his crowd at RC are without a doubt aware of John’s work, but for obvious reasons have ignored and suppresed any ref to his site and the results of his research.
-=-Harold Pierce Jr, B.Sc(Hon)., Ph.D.
PS: I’m an organic chemist with ca 45 years at the bench.
Any thoughts on why Mr. Watt’s & Friend’s station “finds” in Lampasas, TX and Miami, AZ do not match the long. and lat. coordinates for those sites as published by NCDC? Mr. Watts claims its because the government’s coordinates are wrong … his are correct.
Atmoz,
What are the units on your Lampasas graph? 80F is around 27C but the GISS data are around 18C so this doesn’t make sense. So what are you plotting?
Thanks for showing the raw data. It is worrying that even the USHCN ‘adjustment’ that is supposed to fill in missing data effectively wipes out the falling trend. And thats before GISS start applying their ‘homogeneity adjustment’ which reduces the early temperatures by a further 0.8C.
It looks from your graph like there is no raw data for 2001 - is that because USHCN identified it as a ‘bad outlier’ and omitted it?
[Reply: I've plotted the monthly mean maximum temperature, as opposed to the monthly mean (mean) temperature. More peculiarities about this station in my next post.]