There's always something. This time it's over USHCN, in 2013 in Luling, Texas. And yes, I've been arguing. But it's actually quite interesting.
It started, as it seems to lately, with Steven Goddard, who has a new name. Paul Homewood joined in the excitement and looked into Luling, which he says is the first thing he came across. His posts are (here and here). And an account that is actually very helpful from tchannon.
The basic story is that in 2013 USHCN discarded the raw data from Luling, a Coop station in Texas, and replaced it with infill from neighbouring stations. That is the standard response to missing data. But the excitement was that the raw data was there and was quite a lot cooler. Here is the table that he calls "shocking":
Update: mesoman notes in a comment below that there was a cable fault which caused low temperature readings which was repaired on Jan 14th 2014. Looks like problem solved. The system did the right thing.
| Actual F | Actual C | Bias Adjusted C | Diff | |
| Jan 2013 | 50.3 | 10.17 | 10.79 | 0.62 |
| Feb | 54.2 | 12.33 | 13.48 | 1.15 |
| Mar | 58.1 | 14.50 | 15.33 | 0.83 |
| Apr | 63.4 | 17.44 | 18.30 | 0.86 |
| May | 70.7 | 21.50 | 22.64 | 1.14 |
| Jun | 80.2 | 26.78 | 27.52 | 0.74 |
| Jul | 79.7 | 26.50 | 28.46 | 1.96 |
| Aug | 81.9 | 27.72 | 29.23 | 1.51 |
| Sep | 76.1 | 24.50 | 25.99 | 1.49 |
| Oct | 63.6 | 17.56 | 20.51 | 2.95 |
| Nov | 51.6 | 10.89 | 13.09 | 2.20 |
| Dec | 46.1 | 7.83 | 8.86 | 1.03 |
| Annual 2013 | 64.7 | 18.17 | 19.52 | 1.35 |
| Annual 1934 | 70.9 | 21.61 | 20.72 | -0.91 |
There are nowadays lots of sources of information. All USHCN stations are now GHCN too, so you can look at the GHCN details. They don't help much. Paul linked the metadata, which I'll refer to. There are some other tabs there which may help.
An alternative account which is well worth checking is BEST, which I noted at WUWT and Paul's. It includes this useful plot of the difference between raw values and the regional average:
Note the recent dive and the red markings, which are what BEST understands to be station moves.
This starts to look like an explanation. A station move followed by a marked cooling relative to the region is exactly what homogenization is about. And if the program believes there was a move which changed things, then the right thing to do is exactly to replace the data with a regional estimate until there is enough history to estimate the effect of the change.
Paul posted an update, noting that the metadata did show a change of coordinates at that time, but with a note to say that no equipment had moved. They were just improving the accuracy. Still, it's quite likely that the computer program took the change as confirmation of the inhomogeneity of the sudden dip.
Blogger tchannon found lots of useful information at the site of the Foundation Farm which hosts the station. He noted some equipment issues which he thought might have triggered the computer's response.
If there wasn't an actual move, the sudden dip at Luling doesn't have a clear explanation. It's real, though. GHCN has the same raw data, and you can see my shaded plot of it here. These are plots of anomalies, which I have calculated as described here. The shaded anomaly plot is actually a very good way to spot issues with data, as I describe in that post and some of its links.
I have extracted some of the key months here. The extreme of Paul's table above was October, and here is what my plot shows:
 Anomaly=-2.8°C |
The black dots are stations with data. The deep blue dip is Luling. It is a clear outlier. On my plot you can shift-click for details and it shows the anomaly of -2.80°C. Not coincidentally, this lines up with the -3.95°C in Paul's table (I took the extreme case).
Here are some plots of other months. In each case the blue dip id Luling:
 July 2013 Anomaly=-0.95°C |  August Anomaly=-1.24°C |  Sept 2013 Anomaly=-1.08°C |
 November Anomaly=-4.05°C |  Dec 2013 Anomaly=-2.52°C |  Dec 2012 Anomaly=2.31°C |
November looks extreme, but it was a cold month everywhere there. I've included December 2012 to show that it does seem to be a recent issue that arose some time in 2013. Making the pics is a bit tedious, so I'll leave it there, but you can make your own here.
So something seems to be going on at Luling; it's not just a computer glitch.
Posts
Nick, as I noted in a comment on Paul's blog, the anomaly was created by MMTS cable damage which caused the temperature display unit to incorrectly calculate temperatures with a large cold bias. It was repaired on January 18, 2014.
ReplyDeleteThanks, Mesoman,
DeleteI read what you aid there, but for some reason missed the significance. That seems to fit the pictures quite well. It's a bit odd that it wasn't noted on the metadata (or was it?). tchannon saw a report of an electrical problem in Feb 2011 - I wonder if that was related.
When repairs are made to a COOP site, it is usually only noted on an internal inspection report. The public wouldn't know about it unless it involved the replacement of equipment that has a serial number or the moving of equipment. So, because cable repairs don't meet the criteria, even if the problem was causing a significant temperature issue, it wouldn't be noted in the metadata because such things don't require an update to the B-44 form, which is where the publicly available metadata comes from.
DeleteAll this data wrangling over what one can simply discern from what happens to temperature readings when a cable gets damaged.
ReplyDelete"Climate scientists" could get a real dose of reality by getting off their butts and away from their computer screens and do a tour of stations worldwide like I have, though like you, they might "miss the significance" staring right at the damaged cable while wondering why the temperature on the display isn't right.
The metadata has always been hit and miss, and boils down to whether or not the NWS field technician makes full notes or not. Sometimes they are so busy (or have a driving schedule that demands overtime, some have to maintain stations over a two hundred miles or more away from their office) that they often don't note such things. The metadata is better than it was 5 years ago, but still suffers from the random nature of human action or inaction. BEST's detection scheme for station moves, while admirable in its design, is always going to have a fairly large error component just due to the human factor.
'"Climate scientists" could get a real dose of reality'
DeleteThis was a case where climate scientists, or their programs, actually detected the problem and responded appropriately. That sounds like a good relationship with reality. It's not their job to monitor cables. That's a NWS issue, but not a climate science one.
Like Anthony Watts writes, you can never be sure that the station history is complete. That is exactly why you always need to detect and remove non-climatic changes by comparison to neighbouring stations to make the trend estimates more reliable.
DeleteThe main problem of the Watts et al. (2012) manuscript after Evan Jones did a lot of work on it, is that it still makes the implicit assumption that the station history is perfect and that Watts refuses to even detect whether there are non-climatic changes by comparing stations to their neighbours.
He would not even have to correct such stations for his study, but he could at least detect such jumps and remove such stations from his analysis.
Nick, I have a question, how many of the "local" stations that you plotted data for were Estimated values for 2013?
ReplyDeleteI have just looked at the first 10 files in the USCHN Texas zip and 3 (30%) were estimated for 2012 through to the current month.
Also 7 of them had large sections of the beginning of the 20th century estimated, with a couple right up to 1930/40.
All of them had many sections estimated throughout the century.
A C,
DeleteI'm plotting GHCN raw data. None is estimated. You can see from the pictures, or the full world plot if you want, that even with a percentage of stations missing, there is still dense coverage.
You can look up the raw data directly on the CLIMAT forms. Or even rawer, on the B91 forms that Luling puts on its site, which I linked above.
Nick, can you please provide me with the Station names/numbers that you have used in your analysis so that I can check those sheets for myself?
DeleteGood in depth piece. A faulty cable may well explain the adjustments to Luling, Tx data.
ReplyDeletePaul wrote, "From this dataset, I picked the one at the top of the list, (which appears to be totally random), Station number 415429, which is Luling, Texas."
If a random pick led to these discoveries, after more careful analysis of the dataset, how many more discoveries loom?
"how many more discoveries loom"
DeleteWhat have you discovered here? There was a fault for a period, and the much maligned homogenisation software picked it up and quarantined the data, just as it should.
But as I said, the shaded anomaly plot is actually a good way of detecting problems. I spotted that way. You may need to see a few months in succession.
This comment has been removed by the author.
ReplyDeleteLying??
DeleteI think not.
The screenshots of the USHCN database are shown here, as they are listed today.
I cannot say whether they will still be there tomorrow ( and by the way, if you look closely, you will find that USHCN numbers have changed since I printed them off a couple of days ago. Apparently this is "good science").
PS Nick
I suggest you ban any such stupid comments from your website in future)
https://notalotofpeopleknowthat.wordpress.com/2014/06/29/more-news-on-ushcn-temperature-adjustments/
Well I'm using Winrar, so there you go, ordered the way it was packed. So you think NOAA packs their data with 7-Zip? No apologies though as you had an ~50% chance of hitting gold.
DeleteBut wait, see below, I'm now in the process of infilling dummy years to both datasets, then I can subtract monthly values, find the maximum difference, I'm quite sure that I can show a 90%+ return of at least one month being different by 1 degree F from all 1218 stations.
In other words, this is, was and always has been about lying deniers.
Well you too can be an beyond totally incompetent Birther computer programmer for the private sector who has his direct boss check his work 2/3 times per day.
ReplyDeleteHow you ask?
1) Download both the raw and homo data.
2) Pop both datasets into an Excel spreadsheet.
3) Use the SLOPE function on both datasets using the final column annual data (we don't care aboot no stinkin' R^2 dontcha know).
4) Subtract the two slopes (homo - raw) convert to degrees Centigrade/century.
Results?
Median = +0.5 degrees Centigrade/century (N = 606 stations)
N = 300 stations that exceed +1.0 degrees Centigrade/century
Start a blog and post one of the 300 per week in what's known as Goddard's Gish Gallop, that's like 6 years worth of material.
Or heck, just leave it in degrees F (Median = 0.9 degrees F) like Goddard does, then you have N = 606 stations for 0.9 degrees F or N = 571 for 1.0 degrees F.