Google Earth KML files – spot the global warming

[Updated – pictures added for those who don’t want to download Google Earth – VJones]

I’m putting up this new thread so that I can initially let visitors here share in benefit of some work I’m been doing recently in reproducing a piece of work done by Ken Mankoff back in February 2008.

Steve Mosher put me on to this work in via a comment he posted on another thread here on DITC, so cheers Mosh!

Now back in February 2008, Mosh and other regulars on CA were contributing to the following thread on CA.

http://climateaudit.org/2008/02/23/code-1-stations-the-top-guns/

A commentor called Mike posted this comment  

http://climateaudit.org/2008/02/23/code-1-stations-the-top-guns/#comment-138744

which referred to work that had been done by Ken Mankoff at Columbia University. Ken managed to produce some early versions of Google Earth KML files that interfaced with output from a software data visualization package that he was developing at the time. In Ken’s case he was using the GISS dataset. In my case I’ve used the GHCN V2 dataset (and very shortly the GHCN V3 beta dataset) and I’ve linked to trend charts that I’ve created using a piece of software I’ve developed called TEKTemp.  So in effect I’ve replicated Ken’s earlier work but have updated it to the latest available data.

The net result is some very useful Google Earth KML files that I’d like to share just now with visitors to DITC.

I strongly recommend that you download and install Google Earth on your PC/laptop as it will be well worth a few minutes of your time. Google Earth can be downloaded by clicking on the following link.

http://www.google.com/earth/download/ge/agree.html

If you don’t want Google Chrome as part of the download don’t forget to untick the tick boxes at the top of the agreement page.

Next click on the following link to download the KML files I’ve produced so far

http://www.climateapplications.com/kmlfiles.asp

Thus far I’ve provided two KML files.

The first link listed on that page shows a coloured dot for stations in the GHCN V2 that have at least 10 years of data recorded (not necessarily in consecutive years) any at time during the 1880 to 2010 time period.

The second link listed on that page shows a coloured dot for stations in the GHCN V2 that have at least 50 years of data recorded (not necessarily in consecutive years) any at time during the 1880 to 2010 time period; all other stations i.e. with less than 50 years of data are coloured with a white dot.

Key to colours used in the KML files


Canada (trends for minimum 10 years)


Canada (Trends only for 50+ years data)

It’s very interesting to load up and contrast the two KML files. The first shows lots of ‘dark red dots’ (which means a warming trend in excess of 5 deg.C/century) especially just north of the 49th parallel in Canada. But when you look at the same stations in the second KML file, pretty much all of them are coloured with a white dot i.e. for nearly all of these (future CAGW so ‘we must act now’) stations there is less than 50 years of data available in the GHCN v2 dataset for these stations. In particular what data is available is mostly for the post 1950 period so most importantly not many of these Canadian stations have much if any data available during the 1910 to 1940 warming period, which means that this important warming period can’t be directly  contrasted with the 1970 to 2010 warming period for the same stations.

Similarly just look at how many white dots there are in the other continents, especially Africa and Australia. It seems that there are very few long lived stations in either of these two key continents.

About these ads

About KevinUK (DITC)

Ex-nuclear physicist now self employed software developer searching for plausible evidence as to whether or not mankind is responsible for the late 20th century's (non-)global warming trend.
This entry was posted in Climate Cycles, Mapping, Station Data, Trends and tagged , , . Bookmark the permalink.

5 Responses to Google Earth KML files – spot the global warming

  1. ken mankoff says:

    Good stuff. Thanks for replicating the work. My KML file can be found here: http://kenmankoff.com/2010/01/24/gistemp-stationdata

    Other KML files here: http://kenmankoff.com/tag/KML

    And you shouldn’t distribute your KML as zipped files. Zip them but make the extension KMZ and then Google Earth knows how to handle that format.

  2. KevinUK says:

    Ken,

    Thanks you for your comment and advice. I didn’t know that Google Earth could read zipped KML files, so I’ll make that change just now.

    I think the work you did back in 2008 is excellent and I replicated largely so that it could be updated so that people can use the updated KML files as a resource for viewing the warming/cooling trends that are very evident throughout the world during the different warming/cooling periods from 1910 to 1940 (warming), 1940 to 1970 (cooling) and 1970 to 2010 (warming).

    I’m in the process of persuading Anthony Watts to let me do a bit of development which will enable anyone and not just mean to interface with the photos and station metadata stored on the surfacestations.org web site (as you had tried to do back in 2008).

    Have you been in a position to look at the GHCN V3 beta dataset yet a I’ll be putting up a series of threads on the difference between the V3 and V2 datasets shortly which will include some V3 KML files. Rather annoyingly NCDC have used the USHCN V2 station ids in the GHCN V3 beta dataset rather than the normal WMO Station code/imod combinations. As best as I can make out at this stage there is a one to one relationship between the USHCN V2 stations IDs and the WMO station code/imod combinations in the GHCN V2/V3 station inventory files and I’ve been given a translation table that relate sthem but haven’t had time to use it.

    I trying to set up a nice generic third normalised form relational database that can be used to store all the different global temperature indices datasets e.g. GHCN V2, GISS V2, USHCN V2, HADCrut etc which will then enable differences between the different datasets of raw/adjusted data for individual stations and their fitted warming/cooling trends. I hope at some point to make this available as a zipped MySQL dump file so that anyone who wants to further analyse the data can do. I got quite a bit to do yet before that can be the case as normalising all this climate data is a challenging task.

    I’ve also got another project on the go that will involve getting a lot of this warming/cooling trend data into GapMinder so that once and for all the very obvious (to me) natural cyclic warming/cooling trends can be visualised in such a way that they can’t be ignored.

  3. ken mankoff says:

    Regarding data formats and databases, I suggest you look into the data formats from Clear Climate Code http://clearclimatecode.org/ project. They have re-implemented the GISS GISTEMP algorithms (binary compatibility) in Python with the aim of code clarity.

  4. Doug Proctor says:

    We’re locked into interpretations based on linear trends when climate research continuously shows the temperature (and other) trends to be cyclic. So many of the stations that show on the Google Earth map have limited data that look to my untrained eye to be unsuitable for linear trends, but would be suitable for portions of cyclic patterns.

    We throw them out or add them in. Looks like a terrible exercise based on forcing ourselves to use linear models with datasets shorter than the 60 or greater cyclic patterns. Would cyclic interpretations of the less continous or complete datasets be able to better pick up the patterns and general temperature changes?

  5. KevinUK says:

    Doug

    “Would cyclic interpretations of the less continous or complete datasets be able to better pick up the patterns and general temperature changes?

    The answer to your question is very simple. Yes! Its very clear that where we have ‘long lived’ stations i.e. where we have data that spans the majority (say 90 years) of the whole 130 years period there is clear evidence of cyclic warming/cooling trends that are well correlated to known ocean cycles such as the PDO and AMO. These multi-decadal warming/cooling trends and in turn superimposed on a modest warming trends that is due to the earth’s recovery from the nadir (lowest point) of the ‘Little Ice Age’ (LIA) i.e. it a wholely natural multi-centennial warming trend that was preceded by a multi-centennial cooling trend when the Earth cooled from the Medieval Warm Period (MWP) to the LIA.

    This is why Jonathan Overpeck knows that it is so important to ‘get rid of the Medieval Warm Period’.

Comments are closed.