See also the related story on the new Google Earth historical imagery tool here.
I’m pleased to announce that due to the help of many volunteer surveyors, the surfacestations.org project has now reached the 70% mark for stations that have been surveyed. 854 out of 1221 USHCN stations have been surveyed. In addition, Thanks to the splendid work of volunteers Gary Boden and Barry Wise, a new Google Earth KML file has been released that not only shows what stations have been surveyed and their ratings, but now includes numbered icons, and embedded links to the surfacestations.org online gallery for that USHCN station.
This is what the USHCN looks like as of this writing:
Click for a larger image-
Stations marked with a (?) remain to be surveyed. The site ratings system is borrowed from NOAA’s Climate Reference network which you can read about here in section 2.2.1
Climate Reference Network Site Handbook
Alternately, the simplest measure of quality is NOAA’s older 100 foot rule, which is essentially equal to a CRN rating of 2 above
You can download the updated Google Earth (used to make the map above) KML file here.
[UPDATE: I’m having a number of problems with the gallery server this week, and the KML file downloads are no exception. WordPress.com free hosting service for WUWT won’t allow me to upload the KML or even zip files, I can’t put it here. So to reduce loading on the surfacestations gallery server I’ve taken the file off that server tonight. I can make it available on request via email.]
A freeware download of Google Earth 5.0 is available from http://earth.google.com/
The new Google Earth KML file now makes it easier to:
- Determine USHCN station locations
- Decide if they have been surveyed or not
- See what USHCN stations that have been surveyed look like with the embedded image link
- Read a brief description of the station location
- Get the Lights=X value used by NASA GISS to determine the “urbanization” level of the station
There a a couple of caveats. First the lat/lon coordinates used are mostly from NOAA, so some may not be exact. Some that are not precise could be as much as a kilometer off. Also bear in mind that while we do have the stations in hand, and they are used to create the map above, there are a few (about a dozen) that are not yet uploaded to the surfacestations gallery server yet, so you may encounter those few that have a rating but no pictures. This is due to me falling behind because of problems with the server maintenance in the last two weeks, mostly due to rotating IP bot attacks from Beijing, China for some odd reason that cause the server to slowdown and throw a lot of errors during upload attempts. I’m working on it, and these stations will be added in the next few days.
Here is what the survey coverage looks like so far:
Note that stations that have been surveyed but are closed often don’t get a rating if we cannot determine exactly where the thermometer was placed, which is why the pie chart reflects “841 stations rated as of 02/05/09”.
For those of you that wish to help, there are still several hundred stations that remain to be surveyed. Especially in South Texas, Missouri, and Illinois. All you have to do to help is to visit www.surfacestations.org and go through the simple signup and follow the instructions.
I hope to have this project wrapped up by the end of summer, so if your vacation or travel plans include driving, why not help get a station or two?
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I have only recently found this site and the associated surfacestations.org, but have found both interesting.
Have you tried any analysis on the data yet? I think I read somewhere that none had yet been attempted. I am very curious to see a 30 year temperature anomaly analysis on stations rated 1-3 verses 4-5 stations. As well as an estimated temperature deviation as a result of the stations poor siting.
G
REPLY: Having reached 70% I’m starting some analysis efforts now in parallel. However I will not be publishing those analyses here (beyond the simple census counts you see above) until the paper being written is accepted by a journal – Anthony
By clicking on the top image [then clicking on it once again when it comes up], we get a clear picture of the really large number of stations that are in the 2 – 5 degree+ error category.
[What is the explanation for the stations with question marks, versus the white unrated stations?]
Nevermind, I figured it out. Sorry.
Time is on a crunch!
TX for the effort
So we’re flipping out over 0.6C of warming in the 20th century as told to us by a surface station network that, from the looks of it, isn’t even accurate to within 2C?
Sounds about right. Hansen and Gore are using the same math banks used to determine who to give home loans to.
How can we get Anthony to testify before Congress? Shouldn’t Congress be made aware of data integrity issues before deciding legislation that involves predictions based on said data?
The link:
http://www.surfacestations.org
Does not seem to be working…?
REPLY: Works for me, must be a local routing issue. Update: wordpress problem, found and fixed. – Anthony
Where’s the link for the new KML file? All i see on surfacestations.org is one dated 01/25/2009
REPLY: Try reading the article, link there
I imagine John V. is pretty much chomping at the bit by now.
REPLY: He’ll have to wait. If somebody publishes an analysis on the data I’ve gathered, by many journal rules it will preclude publishing there because they have policies to not publish prior published works. I figure that I have the right of first publication. Of course once accepted for publication, I’ll release the data and code so that it can be replicated. – Anthony
I did read the article. Missed the first two times around. I read “You can download the updated Google Earth” as meaning you can download the latest version of Google Earth (which I had already done), and not the KML file. I just think it could have been a bit more obvious, like in the first paragraph where it was mentioned.
Anthony writes: […]”Of course once accepted for publication, I’ll release the data and code so that it can be replicated. – Anthony”
Hey… there’s a novel concept. Maybe it’ll catch on in climate science circles ;0)
There sure is a whole bunch of orange in western NC where I live….
This is fantastic news. I can’t wait to see the results.
Thanks everyone.
DT: ” Shouldn’t Congress be made aware of data integrity issues before deciding legislation that involves predictions based on said data?”
This survey is just for the US. I’m sure the rest of the world (ROW) stations are in better shape.
Anthony, hopefully I can get Beeville and Boerne, TX to you this week.
The UK attempt has gone missing, web site giving 404.
Was here http://www.wacv.co.uk
Anyone know what has happened?
REPLY: Well its a weekend and they’ve had major snowstorms creating infrastructure havoc there. Let’s give it a couple of days. – Anthony
Anthony, I predict that your paper, once published, will be the straw that breaks the camel’s back. Best of luck with this.
REPLY: It is here that it is important to note that at this point, I do not know what the outcome of the data analysis will be. There are some who argue that siting doesn’t matter at all, and that the temperature data, filtered and homogenized reflects the real trend appropriately, even though the magnitude of a number of stations may in fact be thrown off by siting issues. So it’s possible that there is no appreciable difference between trends derived from well sited stations and poorly sited ones. It is also possible that siting matters significantly to trends, and that the different sorts of analysis techniques planned will illustrate that. The Baltimore rooftop station that NOAA closed after their own analysis shows that siting is a contributor to absolute temperature numbers, but we don’t yet know th answer for long period (the 30 year climate period) trends. All I can say for certain at this point is that the majority of stations don’t meet NOAA’s own 100 foot rule which has been in place for many years, and most don’t qualify to be climate monitoring stations under the new CRN siting criteria which I used as a rating system. – Anthony
I’m going to assume an average rating of surface stations to be at an error rate of +1.5C, so the .6C of warming currently is modified to -1C, and that we’ve already lost 1C to the fall of the Modern Maximum.
Raging fires in Melbourne aside (my heart goes out to them), the mass of the planet has cooled. Land, oceans, air is all colder.
Aah….Aaahh…Aaaaaahhhhh Chu !
Anthony, How long until you can click on a specific station on that map and have the referenced picture/data pulled up. That would be a powerful tool for the users of this website to be able to access.
Hi all,
I have never posted before but want to say how wonderful your site is Anthony and how much I look forward to the final results of this fantastic project.
I know this is off topic but you may be interested in this :
http://blogs.news.com.au/heraldsun/andrewbolt/index.php/heraldsun/comments/wrong_answer/
As you may be aware Australia is in the middle of some interesting weather. Massive fires have killed over 100 people and destroyed at least 750 homes in Victoria. Melbourne on Saturday recorded it’s hottest ever day of 46.2C. ( but of course this was at the rather badily sited weather station you have featured in Melbourne.
The second big weather event is the vigorous moonsoon and low pressure systems in Northern Queensland. I live just North-west of Townsville and we have received over 1500mm or 60 inches of rain in the last month (Townsville yearly average about 40 inches). Ingham about 100kilometers noth of Townsville has been isolated for the last eight days with rainfalls in excess of 2000mm recorded in some areas.
Of course the usual suspects have come forward and said it is all to do with CO2, Global warming and Climate Change.
I would like someone to point out to me how these weather paterns as seen in this 7 day animation http://www.bom.gov.au/cgi-bin/nmoc/latest_fwo_loop.pl?IDCODE=IDY00030&number=28 have been caused by CO2.
The second challenge: If CO2 did cause these weather patterns how much different would this weather animation look if CO2 had never risen.
Keep up the good work Anthony.
I’m going to assume an average rating of surface stations to be at an error rate of +1.5C, so the .6C of warming currently is modified to -1C, and that we’ve already lost 1C to the fall of the Modern Maximum.
Raging fires in Melbourne aside (my heart goes out to them), the mass of the planet has cooled. Land, oceans, air is all colder.
Slow down a bit. You are confusing offset and trend.
Actually, the offset is closer to 2.0°C. But we do not know what the offset was 100 years ago. (Not all the stations were well sited back then.) The difference of what the offset was back then and what the offset is now – including whatever “adjustments” the NOAA, et al., see fit to include tell us the real story.
I doubt the US has cooled. But I also heartily suspect that the degree of warming has been exaggerated by, oh, say, a factor of two. Or so suggest LaDochy (12/2007) and Michaels & McKitrick (2008).
Remember that a heat sink not only exaggerates a warming trend, but likewise exaggerates a cooling trend (as the effect “undoes” itself).
Maybe you should not divulge anything at all prior to publication. I could see some well funded individuals using your map above as a guide and publishing a quick evaluation of their own.
Possibly you could preempt such idea “plagiarism” by publishing based on currently available data and updating with a second publication when you get the rest of the data — certainly normal procedure in my field.
After all, the actual number of stations is happenstance — not a fundamental constant — i.e. missing a few would not discredit your work. Moreover, with 70 percent of the data so well spread over the entire USA, I imagine the results with current data would be similar to results with all the data.
Just an opinion.
I think your study will be an extremely important and, perhaps, unique contribution — regardless of the findings. I am really looking forward to this study — its the one someone should have done first.
It all depends on the exact nature of station history. Unfortunately much of the data necessary to calculate it has been lost or was never recorded in the first place.
There is little-to-no doubt in my mind that stations siting is much worse today than it was at the outset. Not only has a spaghetti of asphalt roads (malls, parking lots, etc.) and general Ex-, sub-, and urbanization overtaken many of the sites, but the MMTS cabling fiasco has created a massive number of CRN4 violations.
But calculating it all is not an easy task.
This survey is just for the US. I’m sure the rest of the world (ROW) stations are in better shape.
You forgot the /sarc off tag?
Allen63 makes a good point, Anthony.