Today’s tide at surfacestations.org brought in an eastern site,Stamford Ct, courtesy of Kevin Green. A couple of interesting features, including something really weird with the GISS adjustments.
First, it shows modern values greater than the 1930s. There was no observable trend through the mid-1980s. The documentation is inconsistent as to whether there was a station mvoe in 1985. In the CDIAC information, the station history shows that it was located at Norwalk from 1892 to 1985, at the Norwalk Gas Plant from 1956 on. At the new NOAA MI3 website, the Stamford 5N station information shows the station at Stamford from 1955 to 1985, but also shows a Norwalk station from 1955 to 1985. My guess is that CDIAC information is primary for the older information and the MI3 information is incorrect.
USHCN Adjustments
At the time of the station move from (presumably Norwalk to Stamford) in 1985, the time of readings was changed from 8 am to 5 pm. This change in observation time is plausibly related to slightly warmer readings (the time of observation adjustment), and a downward adjustment of 0.5 deg C is made effective 1985 (see top row). Other than this, no adjustment is made for the change in station location, although one would presume that the characteristics of the present Stamford site differ from the previous Norwalk site.
Figure 2. USHCN Adjustments
GISS Adjustments
TGhe next figure shows the GISS adjustments in stages relative to the three USHCN series. The top row shows that the GISS dset=0 series (“GISS raw” in my usage) is equal to the USHCN raw series after 2000, while the third row shows that the GISS dset =0(raw) series is equal to the USHCN adjusted prior to 2000. This switch in source (which is not reported in any GISS publications to my knowledge) results in about a 0.5 degree upward jump (bias) in the GISS version from 2000 on. I’ve noticed this in other series and it is extremely hard to think of a justification. It looks like a programming error of some kind (though doubtless, if it is, Hansen will say that it doesn’t “matter”). And it might not be a pervasive problem, as I’ve not noticed a consistent pattern. But from a programming point of view, it’s hard to understand why it would happen in some series and not others.
The big problem here seems to be the change in location plus observing time in 1985, coincident with a quasi-jump in the temperature series. I don’t see how you can disentangle these changes such that you can estimate differences between the 1930s and the present from this particular dataset with any security, which is too bad.
Climate Audit 
29 Comments
Steve McIntyre:
There’s something that I don’t understand about changing Time of Observation. If the system only records maximum and minimum temperatures what difference does TOBS make? The same high temp and the same low temp are recorded (although they might appear on different days). Over a month the average high and the average low temp should be very nearly the same with only one measurement changed in a thirty day period. Over a year there would be no discernible difference.
Steve,
If you are sitting in a restaurant and you see a roach crawling up the wall, you do not think to yourself, “Ah, there goes the only roach.” And, Chef Hansen’s and Chef Schmidt’s assurances that it doesn’t matter would not be comforting.
Re #1,
Bob,
Temperatures near the time of observation often
affect the readings for two days, so that if the
time of observation is at 8 AM, a usually relatively
cool time of day, there would be a cool bias, whereas
at 5 PM, a usually relatively warm time of day, there
would be a warm bias.
For more on the subject than you may ever want to know,
see
TOB stuff .
I grew up in that area… in fact, our house was two miles north of the site, which is on the grounds of the Stamford Museum and Nature Center. There’s been a temperature station there since at least the early 60s; I won’t swear it’s the same one as in the pictures, but it could be.
According to my parents, Stamford in the 30s and 40s would have been a small town. Population approximately doubled between the 40s and the 70s, as there was a lot of development – the center of the city spread out and many high rise commercial buildings went up.
The northern part of the city, which is where the museum is located, would probably have been fairly rural in character before the development started: when our house was built in the mid-50s, it was surrounded by (untilled) farmland. When I was a kid, we had a rule of thumb that the temperature in our neighborhood was always at least a couple of degrees cooler than reported downtown.
The area is still very suburban, mostly one-acre zoning. One side of the museum is fronted by a very busy road that carries a lot of rush-hour traffic. Along another side there’s an old school building with a blacktop playground and a small shopping plaza. However, the property also abuts the Bartlett Arboretum, a heavily wooded area. There’s also a large reservoir nearby.
All in all, it seems like it should be a pretty good site: it should be easy to construct a record going back at least till the 60s with few if any station moves. The conflicts and contradictions in the record just don’t make sense.
According to the USHCN history, the Stamford data historically has come from (at least) 3 locations with smaller relocations around two of the earlier locations. It is true that data from the Nature Center is available since 1955. For some reason, only the data from 1985 are used in the USHCN dataset. My guess is that the algorithm that compiled the histories started with the oldest records (from Norwalk) and used the closest station to that oldest one when the first station was shuttered, and then when that one ceased, the Stamford station was chosen. That is just an educated guess and nothing more. At any rate, data from Norwalk and Stamford can be compared.
I too noticed the jump up in temperature with the use of the Stamford data and am working to ivestigate. The one thing that came to mind was that the Norwalk Gas Plant data is closer to Long Island Sound, a location that could possibly be cooler due to the moderation of the ocean, but that too is just a guess. More analysis will help.
The MI3 station data doesn’t always mesh with the CDIAC station history. The MI3 version shows Norwalk as starting in 1948, but CDIAC has information back to 1892 (See below). There seems to be some updated information at MI3 that is not available at CDIAC, so, as usual, everything needs to be crosschecked.
name begin_date dist_prev_loc dir_move elev hgt_instr obstime_temp
1 NORWALK (5892) 01 01 1892 NA 999 116 07 21
2 NORWALK (5892) 03 01 1918 0.0 000 116 07 21
3 NORWALK (5892) 09 01 1932 0.0 000 116 07 18
4 NORWALK (5892) 03 15 1955 NA 999 120 05 18
5 NORWALK/GAS PLT (5893) 04 14 1956 1.8 SE 37 05 16
6 NORWALK/GAS PLT (5893) 08 01 1956 0.0 000 37 05 08
7 NORWALK/GAS PLT (5893) 04 11 1957 0.0 NE 37 05 08
8 NORWALK/GAS PLT (5893) 03 24 1960 0.0 000 37 05 08
9 NORWALK/GAS PLT (5893) 07 01 1967 0.0 SW 37 05 08
10 NORWALK/GAS PLT (5893) 04 01 1971 0.0 000 37 05 08
11 NORWALK/GAS PLT (5893) 03 01 1977 0.0 NE 37 05 08
12 STAMFORD (7970) 04 01 1985 NA 999 190 05 17
13 STAMFORD (7970) 05 16 1990 0.0 000 190 05 17
14 STAMFORD (7970) 01 01 1992 0.0 000 190 05 17
GHCN info shows:
ccode id site lat long altitude alt.interp ipop pop topo stveg stloc iloc airport towndis grveg lights
4219 425 72504001 STAMFORD 5N 41.13 -73.55 57 80 U 19342 HI CO 6 x NA COOL CROPS C
4220
I can’t figure out how to extract the separate records during the period of overlap. If anyone has any ideas or information, I’d appreciate it.
I wasn’t familiar with the gas plant, so I did some checking. It is near the water… it’s also bounded on one side by I-95, a very busy highway, and by East Avenue, one of the main roads in Norwalk.
What really surprised me is it appears the weather station there is still in operation. So why was the official site moved to the Stamford station? And if Norwalk is still in operation, shouldn’t it be possible to compare the separate records for Stamford and Norwalk for the last 50 years with each other and with the spliced record to see what adjustments may have been made?
Jim Hanrahan,
Do you have an address for the Norwalk gas station? I went searching for it but couldn’t find it, and I ran out of time. If you have a street address, that would be excellent. The GPS coordinates for the station that I could see are only good to 1 minute of arc and no more. I was under the impression that the station closed, but it could very well be that the station just stopped reporting as a COOP. How do you know the station is still in operation? Do have a contact? A call sign? More information would be greatly appreciated.
Steve,
You can get the historical daily coop data on CD-Rom from NCDC. The full set through 2005 costs $300. Here is the link.
http://ols.nndc.noaa.gov/plolstore/plsql/olstore.prodspecific?prodnum=C00447-CDR-S0001
Free preliminary data are available post 2005, but not all stations are included (sadly, the Stamford 5N station is not included). Some data have to be manually keyed still. Slowly, slowly, the NWS is entering the 21st century.
I’ve looked briefly at the Stamford and Norwalk gas overlap. The Min temp is higher at Norwalk but the Max temp is lower (in the overlap period) and effectively cancel out. That could oceanic modulation. I haven’t looked for secular trends. That part of Norwalk has been more or less urban for over 100 years.
For the meta data, the CDIAC looks to be the most updated. The same history and monthly files are also available on the NCDC site. MI3 has, from my experience, been lacking pre-1948.
#7. That’s interesting that it’s still there.
As to separating the records, it should be possible, but I looked for separate records and couldn’t locate any in the datasets that I’ve used, where the sites seem to have been merged.
Kevin, did you get the overlap data from the CD or did you download it from somewhere?
I got the data from the CD-Rom. You can also download the data from the site…1 month at a time. Good luck.
Steve,
The “GHCN Daily” min/max data include overlap from 1956 to 2004
1956 1987 42500065893 41.1200 -73.4200 11.0 US CT NORWALK GAS PLANT
1955 2004 42500067970 41.1200 -73.5300 58.0 US CT STAMFORD 5 N HCN
Respective URLs could be:
http://www1.ncdc.noaa.gov/pub/data/ghcn/daily/42500065893.dly
http://www1.ncdc.noaa.gov/pub/data/ghcn/daily/42500067970.dly
I guessed that the Norwalk site was still operational when I did a google search on “Norwalk Gas Plant” and found a couple of sites that seemed to indicate that’s where they were getting the local weather info. Here’s an example:
http://www.citytowninfo.com/places/connecticut/norwalk
I can’t seem to find an address. If the map called up by the coordinates I have (41.11°N 73.41°W) is accurate, it should be right off exit 16 of I95 – maybe there’s an entrance off Hendricks Ave. or Sherry St.
Many interested readers may find the CD set at a local library. Nice!
Anyone else getting “Unauthorized Access” at the mi3.ncdc MMS metadata site?
Hope it is just a quirk and they have not tightened the privileges. If so this will effectively end my observations of stations in Georgia.
Maybe what is happening is… If you drill down deeply the MMS site does — I should say did — provide the name of the private station observer. Without getting the phone number from directory assistance or from a local phone book and talking to the individual it is nearly impossible to find these private locations. Heck, even the nonprivate locations will be hard to find w/o the MMS site.
Here is another link to get COOP data.
It includes the Norwalk Gas Plant to 1987 and
the Stamford 5N station through Nov 2004.
#16
Maybe it will work this time.
http://climate.usurf.usu.edu/products/data.php
re 15, my own darn post.
Nevermind. The MMS site is back. Still don’t know what happened. A friend had the same problem as I did.
“Just because you’re paranoid doesn’t mean they aren’t after you.”
Re #12,
Respective URLs should have been:
http://www1.ncdc.noaa.gov/pub/data/ghcn/daily/all/42500065893.dly
http://www1.ncdc.noaa.gov/pub/data/ghcn/daily/all/42500067970.dly
sorry about that.
Kevin Green:
MMS says the location of the Norwalk Gas Plant is at 41.11667, -73.41667. I dropped those coordinates into Google Earth but cannot tell if anything within a few hundred feet of that location is a gas plant or not. Thus far my experience with the coordinates from MMS has been mixed. The Troy location was about 150 yards away, but Falls Village was nearly 0.4 miles away.
Good Luck
RE18 Joel you were not imagining anything the site was partially offline. When it came back up, the “Managing parties” field no longer exists. This filed contained infomation such as the name of the observer or the city entity, such as polic/fire station etc.
I made a query with NCDC yesterday, but so far it has gone unanswered.
Where I come from, if the books and records aren’t available, the company gets a qualified audit report.
#21: Anthony,
Well, I am screwed. Half or more sites in GA are private homes. From recent experience, there will be no way to locate which house among scores of houses within the minimum 1/4 mile coordinate error is the house.
I guess I can understand that they don’t want to show people’s names and all. Yet, this isn’t the NY Times and I can’t help but wonder if this isn’t just a purposeful roadblock.
re #13 (IPCC Review Comments Now Online) & MMS Search.
I just found something odd: I looked at MMS Search for “Vancouver 4 NNE” and found the site listed, but the “MAP” shows it to be miles from where it actually is. The long/lat are correct, though (Maybe the map is showing me where to go to get permission to survey. I’ll find out Friday when I’m back in Vancouver).
Anybody else notice the same discrepancy?
re #24
Try zooming out on the Google Map that comes up after selecting the “Map” tab for the Vancouver 4NNE site. If you look to the north/northeast about 4 miles, you’ll see the blue symbol marking the current location (the red symbols mark previous station lat/longs). The red symbol that comes up may be marking the post office location (hence the 4 NNE station name).
Not knowing the area, I’m not sure about this (all post offices look the same from above).
I think originally (long before Google Earth and GPS), many of these weather stations were located using USGS topographic maps, which historically showed post office buildings. Therefore the “4NNE” designation would show that the station is a radius of around 4 miles to the northeast from the post office. The “4 miles” is probably rounded to the nearest mile, but one conceivably draw these radiuses out from the post office location and get pretty close to the actual location.
As some of the reported lat/longs, from my own experience, can be 0.4 miles off in any direction, this may be helpful.
Hope this helps.
Re the photos at surfacestations.org of the Stamford station. It appears the roof of the station (not well pictured) is gray, unpainted and of a different material than the walls. I don’t know if this is per specs, but might be a matter to be documented at the stations. An unpainted roof would presumably get hotter than a white roof, although this would depend upon how saturated with water it was at the time
Given the interest in this station, I made another site visit and took some more photos of the station. These photos are now uploaded at http://www.surfacestation.org. I specifically took a picture of the shelter roof, and indeed, it needs a paint job to say the least, but it won’t be done as the manual station is being phased out.
Re 27:
Kevin, thanks
The two Norwalk sites are completely different and are not comparable. The earlier site was located further north in a rural area while the “Gas Plant” site was located in a far more urbanized secion of the city. The Stamford 5N site is a completely different site that was operated at the same time as the Norwalk Gas Plant site. There was another Stamford Museum site, located in downtown Stamford from 1950-55. The Connecticut Turnpike came in and forced the Museum to relocate to North Stamford, a more rural and far colder area of the city. Combining the weather records of any of these locations to create a “continuous record” is ridiculous in my opinion, especially in light of changing observation times.