The MMTS system introduced by the National Weather Service in the mid 1980’s continues to be the Achilles heel of the surface observation network. Intrepid surfacestations volunteer Don Kostuch finds another poorly sited USHCN station in America’s midwest.
Click for larger image
Here we see measuring the official temperature for use in the US Historical Climatological Network, in Crosby, ND station ID # 321871, just 5 feet from a building. Yes it’s in the shade, which is great for keeping the sun off the sensor and tempering Tmax, but also traps the longwave IR at night due to the tree canopy, not to mention the effect of building proximity, which boosts Tmin. Then there is the wind sheltering effect.
I keep hoping that we’ll find better stations in the midwest, and while we’ve found a few, stations like this still keep popping up regularly. The MMTS cable issue forces the sensors closer to buildings.
Could the sensor be placed further away from the building? It certainly looks like it. Why didn’t they; Laziness? Obstacle? Mom’s Garden? Who knows.
Click for a larger image
According to NCDC metadata, the MMTS was installed on May 27th, 1987. Here is what GISS shows for their temperature plot.
Click for source image.
Our nation’s USHCN climate network is a mess. Stations like this are now the norm, not the exception. I’m continually amazed and disheartened at the systemic lack of quality control on the part of the NWS deployment of the MMTS system.
I happened to be walking in the city of Redwood City, California today and saw what looked like an MMTS sensor on the roof of the Kaiser Permanente hospital. I don’t know if it was part of any network or not, but I looked at it and shook my head thinking I had spotted yet another rooftop MMTS installation.
Don’t be discouraged Anthony. Nothing can be improved if it is not brought into the light. The truth is until recently nobody realized how critical these sites were going to be, now they are at the center of a world wide political and scientific debate. If it were not for the AGW debate, it would just be another example of inadequate government over site, wow what a shock 🙂
Jerry
Lights = 0!
Have you updated JohnV’s analysis to include new class 1,2 stations?
REPLY: I never did any analysis, he did, and very prematurely at that. Only 17 stations were used. I’m saving my analysis for when a majority of stations exist with representative geographic distribution.
I knew MMTS would break continuity of U.S. COOP Climate records.
But, what you are showing as per site qualty completely invalidates the
USHCN for any serious study of climate change.
Does anyone know how many Cotton Region stations of lengthy duration
remain? It should be a priority to locate such in rural areas. Perhaps the
Ag Stations would be a start.
This is truely sad.
Anthony
I was gone over the 4th. Holiday last five days, and didn’t get back before you closed off your request for help on your project on the previous topic. If you need any more help please count me in on helping if I can. This is a “GREAT SITE”, Thanks.
Question: Who installs these devices? Is it the property owner, a rep from the NWS or someone else?
Surely there must be some guidance of do’s and don’ts sent with the instrument?
Just curious…
REPLY: The local NWS office, a person with the position known as “COOP manager”.
I know Anthony doesn’t want to be premature with an analysis based on site surveys, and I appreciate that, but when the analysis comes I suspect its going to severely rock a few boats.
[…] one of Anthony Watts’ volunteers has discovered another temperature monitoring station that has all sorts of problems in Crosby, […]
Question: Who installs these devices? Is it the property owner, a rep from the NWS or someone else?
REPLY: The local NWS office, a person with the position known as “COOP manager”.
So, in otherwords, someone who should know better? Many thanks for settling my curiosity.
You know, if I were a civil servant who was told that there was a new, huge threat to the world that needed monitoring, I’d like to think that I would arrange for an appropriate monitoring network. The money would be there, after all.
Even if these stations were sited correctly, they were never intended for measuring ‘climate change’ but rather weather. Are the instruments sensitive enough? Why isn’t CO2 measured at the same location? Given the world concern, you would think that an international monitoring standard instrument set could be put in place fairly easily, which would give definitive answers without the need for complex corrections. I know that grafting earlier data on would be problematic, but this system seems to involve making it up as you go along.
Did you see the Register’s take on ice data? It came with an: Editor’s note: This is one in an occasional series examining “PBEM”, or “Policy-based Evidence Making”.
Is that GISS temperature data raw or cooked — I mean adjusted? And those mean temps are really killers, but it is North Dakota after all.
Was there a change in station status in the early 70s? It seems the temps went up after that and carried on at the higher levels after the installation of the MMTS.
Obviously the USHCN surface stations are deeply flawed. This project has been going on for about a year now you would think that the USHCN would have sent out memos to either requite the operators to provide site data to them or more likely correct the stations before Anthony can document how bad the system is.
I remember how early in the process they tried to hide the location data but has anyone heard that they are trying to identify and correct the problems themselves? It seems that they simply don’t care.
I never cease to be amazed!!!!
Bill Derryberry
Question: Who installs these devices? Is it the property owner, a rep from the NWS or someone else?
REPLY: The local NWS office, a person with the position known as “COOP manager”.
Seems to me, he flew the coop on this one!
Okay, okay… the devil made me do it!
Jack Koenig, Editor
The Mysterious Climate Project
http://www.climateclinic.com
Continue the needed work, and thanks from all of us.
I think it clear that this network simply cannot be used for long-term temperature trends. There is no way, even knowing individual site history, to “adjust” for siting problems without introducing a ton of subjectivity.
I may have misread the post, but was this the historic siting of the station or the post-1987 siting? If the latter, it would seem from a rough eyeball of the data that despite obvious sources of bias, the data record doesn’t line up too terribly with the earlier period, perhaps a +0.5 degree error.
REPLY: A .5 to 1 degree error seems fairly common with these siting issues. Considering that the majority of the network is now MMTS, that is a significant issue.
Clark, I disagree. If the biases of individual stations can become well-understood, then we can better homogenize or adjust the data to reflect the real world. It would take a large amount of work to do this in a satisfactory manner, but it is definitely possible.
Totally off-topic but maybe someone can help. I was reading wikipedia the other day and I stumbled upon this:
“The greatest production of CO2 is not man made but produced by the tectonic movement of the earth’s plates.”
Is this true and if so, is this taken into consideration in AGW models?
It appears that we have a real case of Gavin Schmidt shooting himself in the foot with his RC post “Global trends and ENSO.”
http://www.realclimate.org/index.php/archives/2008/07/global-trends-and-enso/#more-577
Apparently a Thompson et al (2008 ) paper describes a method for extracting the ENSO signal from the SST data. Gavin used the method and then explained how correcting for the ENSO effect continued to produce a positive trend for surface data for the last 10 years.
Gavin:
“Despite our advice, people are still insisting that short term trends are meaningful, and so to keep them happy, standard linear regression trends in the ENSO-corrected annual means are all positive since 1998 (though not significantly so). ”
Gavin didn’t present the charts for the period since 1998 that he is speaking of. He did, however, link a text file for all of the HadCrut3 and GISS data, ENSO corrected and uncorrected, for the entire history of those two records. So I decided to see what the difference was between the ENSO corrected data and the uncorrected data for HadCrut3v since 1998. Note, most of the AGW advocates have claimed that the reason for the flat trend for the last 10 years is due to there being an El Nino at the beginnig of that period and a La Nina at the end.
So, here is my chart of Gavin’s data:
http://reallyrealclimate.blogspot.com/2008/07/gavin-schmidt-enso-adjustment-for.html
The period covered is Jan 1998 to May 2008. The total positive trend for the period using ENSO adjusted data is .0125 C. That is so close to flat that the difference is negligible. The decadal trend due to CO2 warming, according to the warmers, should be about .2 C. So with ENSO now accounted for, the question that people like Gavin must answer is – what happened to the other .187 C. It seems to me that we cannot blame it on volcanoes. The AGW advocates claim that the effects of CO2 are much more dominant than solar. So what are the natual elements of variation that account for the missing .187 C. It appears to me that IPCC greenhouse theory has a big hole and that no one is stepping up with an explanation. At a minimum, the warmer explanation that the recent flat trend is due to a cherry picked set of end points is debunked.
As supplementary information, note that the total divergence for the 125 month period between the ENSO corrected and uncorrected HadCrut3v data is only .0163 C. The uncorrected HadCrut3v data trends very slightly negative. I also took a shot at isolating the ENSO effect in May, here:
http://reallyrealclimate.blogspot.com/2008/05/ten-year-hadcrut3-enso-effects.html
Note that the difference between my hacked approach and Thompson et al (2008 ) is very small. About 0.0294 C. for the 125 month period.
It seems to me that playing the ENSO card actually had a negative benefit for Gavin. He now still has a ten year flat temp trend and one less way to explain it.
“The greatest production of CO2 is not man made but produced by the tectonic movement of the earth’s plates.”
Over geological timescales, this is true, but not relevant to the climate debate, which is about the source of CO2 over the last 150 years. IMO there is no issue that the increase (or at least most of it) over that period is due to human activities.
Good work Tilo. It’s remarkable how many claims in climate science don’t stand up to close scrutiny. In the last day or so, I debunked a claim at CA of ‘drought’ in the Western Australian wheatbelt.
Surely there must be some guidance of do’s and don’ts sent with the instrument?
Oh, yes!
http://www1.ncdc.noaa.gov/pub/data/uscrn/documentation/program/X030FullDocumentD0.pdf
And here is the French study it is based on: LeRoy (1999)
http://www.ccrom.org/documentspublics/2007thematique/documentstechniques/Classification_environnement-note_technique35-2.pdf
Anthony uses these very standards to rate each station from CRN1 to CRN5.
You know, if I were a civil servant who was told that there was a new, huge threat to the world that needed monitoring, I’d like to think that I would arrange for an appropriate monitoring network. The money would be there, after all.
The new NOAA/CRN system will be online this fall. It consist of c. 100 US stations that are (supposedly) very well sited. So much so that there will be no SHAP or FILENET adjustments. And the data will be automatically collected hourly and electronically transmitted, so no missing records (and no “temptations” during data collection) and no questionable TOBS adjustments.
All data to be raw.
Unfortunately this will be a US network, only.
Is that GISS temperature data raw or cooked — I mean adjusted?
If it’s GISS, it’s refried: They take adjusted NOAA USHCN data and put it through their own adjustment mill.
I never cease to be amazed!!!!
What amazes me is that I never cease to be amazed in spite of the fact that over six out of seven stations are CRN3 or worse and over seven out of ten are CRN4 or 5.