Comments on: How not to measure temperature, part 87: Grilling in the cornhusker state

By: Michael D Smith

Michael D Smith — Thu, 11 Jun 2009 18:30:02 +0000

Dave the Denier (19:18:14) :

I’m saving up to buy the official GISS BBQ grill and nearby rusty trash-burning barrel to bookend my Stevenson screen.

Now that’s funny! Wouldn’t it be fun to make up a GISS BBQ logo, and make up an infomercial for the new grill, complete with a British pitch-man, and catchy new recipes for dishes like “Urban Heat Polar Bear with Hinge Point Marmalade, on a bed of Stevenson Greens… Anyone have any others, I think a short You-Tube video of this would be fun to make.

For good product placement, we could feature Dr Hansen’s Hellacious Habanero sauce, with a Gore “Storm Surge” chaser. For later, be sure to try Cap’n Trader’s sequestration sorbet! Lose the handy “Scientific Method Skewer”, get a new one free, just pay separate shipping and handling!

By: pwl

pwl — Tue, 02 Jun 2009 10:36:27 +0000

For those new to the whole concept of USHCN stations, the NOAA thermometer is the white slatted object on the post in the center of the photo. It is known as an MMTS thermometer and a cable goes from it into the home where the volunteer observer will write down the high and low into the B91 logbook and send in the report once a month to the National Climatic Data Center (NCDC).

WTF? It’s not automated? How absurd. How many of the stations are manual, or involve some manual processing of the data by humans?

The fact that some of the stations are not automated raises the red flag of not just HUMAN ERROR but willful deceit for various reasons.

That’s not even counting the scientists who massage the data later in the process.

It’s still the 1950′s in climate science technology it seems.

It also doesn’t seem like the the climate scientists have learned anything from the quality control experts in high tech manufacturing.

By: Mark

Mark — Mon, 01 Jun 2009 12:50:24 +0000

Hey Anthony,

You’re mentioned in my local newspaper today!

http://www.ocregister.com/articles/temperature-stations-global-2433763-heat-watts

By: Pat

Pat — Mon, 01 Jun 2009 10:27:18 +0000

According to the weatherman tonight in Sydney, Australia, May 2009 was 0.8c above the “long term average”. Not sure how this “long term average” was derived however.

By: bill

bill — Sun, 31 May 2009 15:27:26 +0000

OK to finally put to bed these idiotic and disingenuous statements that GISS are “adjusting” the temperatures to prove global warming here is a random selection of locations with their adjustments.
To me they look rather like someone trying to correct inconsistencies NOT someone trying to depress early temperatures.
I have plotted all I looked at (both up and down adjustments) – no cherry picking other than sites with long and recent measurements.

http://img11.imageshack.us/img11/7440/gissrawtemps.jpg

By: Dave Wendt

Dave Wendt — Sat, 30 May 2009 20:19:36 +0000

BTW, can anybody explain to me what’s going on with the adjustments to that ” Huron/ Huron R” station that bill linked?

By: Dave Wendt

Dave Wendt — Sat, 30 May 2009 20:06:16 +0000

bill (04:00:34) :

I’m sorry to be so slow responding to your comments, but I’ve been otherwise occupied. I’d have to grant you 4 out of 7 for meeting my criteria of down trend enhancing adjustments. The last 2 only work if the adjustment profile is a record of adjustments made in the past and not a present day revision of the historical record and Cooperstown is marginal. I suspect a full 1 degree reduction in the start point more than compensates for a 0.1 degree bump in the middle. But, though I’m not sure I’ve ever used the phrase “the adjustments are always up” in the past, if I have occasion to refer to this situation in the future I will do my best to insert an almost. Personally, I suspect that the entire surface station network is so throughly compromised that no amount of statistical manipulation, no matter how diligently it is performed, is likely to produce data that has real relevance to what is occurring in the actual world. My own objection to the AGW agenda has little to do with whether and to what degree warming has occurred in the past or will occur in the future but to my own conclusion that if humanity proceeds to produce and distribute as much energy as we can to as many people as possible the future that develops will be infinitely preferable to the one provided by following the carbon demonization scenario, no matter what the climate may be in a hundred years.

By: bill

bill — Sat, 30 May 2009 16:38:14 +0000

Carsten Arnholm, Norway (03:57:49) :

I’m not sure where the code came from but it is not the same code as giss use.
giss changes on years other than 1997 and will change either up or down – which e.m. smith claims is impossible.

There is a website devoted to converting the hodge podge of languages to python (just do a search on “dif.ushcn.ghcn.f” there are not many results) – If you would like to help out you can sign up!

By: bill

bill — Sat, 30 May 2009 11:00:34 +0000

Dave Wendt
Here are a couple more + the data from hartington (just to prove I’m doing the right thing)
To me it looks as if GISS are trying hard to correct measurement errors – not trying to falsify the temperature record.

http://img521.imageshack.us/img521/3074/gissraw2.jpg
the one I did earlier:
http://img23.imageshack.us/img23/5382/gissraw.jpg

Also of interest to E.M.Smith (11:02:57) : is that 1979 hinge does not always occur and certainly is not always negative!

By: Carsten Arnholm, Norway

Carsten Arnholm, Norway — Sat, 30 May 2009 10:57:49 +0000

E.M.Smith (11:02:57) :

That FORTRAN code is seriously poor style. I have done professional programming in FORTRAN for ~15 years (until about 1995). Such chaotic coding is always, always a sign of poor design and understanding. It is probably riddled with algorithmic bugs (in addition to the fact that the algorithms are wrong in the first place, as you indicate). The style of the coding looks like it has been done by someone taking a crash course in Fortran some time in the mid 1970s. The FORTRAN 66 standard (from 1966) would produce similar code (except for the lowercase and enddo’s). The FORTRAN 77 ANSI standard (it is still in my bookshelf) would at least let you program with almost no GOTOs, if-then-else and such.

Today, such code should be properly designed using a modern language and people with some competence in software design and implementation. Especially when decisions costing trillions are based on it.

By: Keith

Keith — Sat, 30 May 2009 04:34:49 +0000

“That might make sense, except that GISS doesn’t use satellite data, only surface data. – Anthony”

Sorry if I wasn’t clear, but I know that. About a year ago, I compared the monthly GISS global anomaly against both RSS and UAH, and once you correct for the differences in the base periods, the anomalies match quite closely for the period post 1979. My point was that the GISS team can’t play too loose with the data from 1979 onwards since it is easy to verify against a satellite record. Prior to 1979, however, they could engage in all kinds of shenanigans to make the past appear slightly cooler than it actually was.

By: rbateman

rbateman — Sat, 30 May 2009 03:48:14 +0000

For my own town (Weaverville RS), GISS has July 1986 beating out August 1932, for which there is no equal in the actual data record. There is no record for July 1986 in the data, so therefore it is assumed to be through the roof by GISS.
I see what you mean by GISS barbecuing where there is nothing to fry.

By: Nasif Nahle

Nasif Nahle — Sat, 30 May 2009 03:11:16 +0000

I’ll try to get some pictures of the International Airport Climate Station. I hope not being arrested…

I’ll send you the pictures as soon as I have them. :)

By: Dave the Denier

Dave the Denier — Sat, 30 May 2009 02:18:14 +0000

I’m saving up to buy the official GISS BBQ grill and nearby rusty trash-burning barrel to bookend my Stevenson screen.

By: Old PI

Old PI — Sat, 30 May 2009 01:56:35 +0000

What’s really needed is a complete revamping of the surface temperature station network. The network administrators need to do some site surveys to ensure that future sites are better sited than current ones. Secondly, the entire system should be automated, with sensors collecting temperature, humidity, and barometric pressure and uploading that data once a day to an orbiting satellite. That would remove several factors from consideration: poorly sited instrumentation, human error, and (at least for a short time) eliminating the need to correct for extended human activity. I’m sure a fully-modernized, solar-powered unit could be constructed for around $2000 each. A satellite is expensive, but there’s no reason NASA can’t put a package together that could be added to a previously-scheduled satellite. Since the areas populated by these sensors would only be in the northern hemisphere, a polar orbit would not be necessary. No one can make the types of decisions that we, the people of the US, are being required to make without first ensuring the truth of the data we’re relying on to make those decisions. That degree of integrity is not present in the currently publically-available data.