Skip to content
  1. A C Osborn permalink
    February 2, 2015 3:55 pm

    It’s worse than we thought.

  2. Dave Ward permalink
    February 2, 2015 6:49 pm

    Paul, it’s worth remembering that there are several ways of expressing a map location – the ones you’ll see in Google Earths “Options” page are Decimal Degrees, Degrees, Minutes & Seconds and Degrees & Decimal Minutes. It’s not that difficult to mix the second and third up, which will make a difference to the exact fix. And the position will (naturally) be more accurate when more details are provided. Your example goes to 2 decimal places – this will only put you within a radius of several miles. Look at the bottom of the GE screen while you move the cursor about from a particular spot to see what I mean. Then select the various display options to see how similar they appear (particularly in small type face on an “HD” screen).

    There is another problem – which Map Datum is used. There are literally hundreds of these covering the globe. Most have fallen into disuse, but even those that are common will introduce further errors. For instance, the UK Ordnance Survey sheets and CAA navigation charts employ the OSGB 1936 standard, but the GPS system (and I assume) GE employs the newer WGS 84. I learned to fly back in the 90’s, and had to remember all this in order to pass the Nav exam. I subsequently bought an early handheld Garmin GPS, and spent many hours programming it with waypoints. I experimented with different datums, using the exact same co-ordinates, and it was surprising how far off the map would move.

    I’m not trying to “stick up” for BEST & GISS, but getting an accurate fix is not always as simple as it might seem…

    • February 2, 2015 9:26 pm

      It’s worse than we thought!

      It’s easy to see why BEST get different coordinates and assume station moves. It is probably just the observer looking at a different map

      • Shub Niggurath permalink
        February 2, 2015 10:58 pm

        BEST reports all their data in a degree + decimals of a degree uniform format.

  3. Bloke down the pub permalink
    February 2, 2015 6:53 pm

    One error that creeps into co-ordinates is caused by the different ways of treating fractions of a degree, ie either as a decimal or as minutes and seconds(and decimal of seconds). This becomes a big issue if a pilot uses the wrong conversion and ends up dropping ordnance on the wrong place. As far as I’m aware, the US Navy and US Air Force still use different methods.

  4. John F. Hultquist permalink
    February 2, 2015 7:34 pm

    I just ran into this issue while looking for a snow measurement site (SNOTEL) in the US state of Oregon. Coordinates for ANEROID LAKE #2 are given as
    45.21 , -117.19
    There is nothing there to be seen – using Google Earth.
    Look 650 m. to the WNW and there is the correct location.
    45.212818, -117.197275

    I would round the Longitude up to 117.20 and that gets a littler closer, and also closer to the Lake. Still, with just 2 digits, that is the best that I can do using Google Earth. The site was installed in “Water Year 1981” and coordinates taken from a map and rounded to 2 digits. A USGS 7.5 minute quadrangle (paper map) is a standard map used at that time, and thus, likely the source.
    I think Dave Ward makes additional useful comments.

  5. catweazle666 permalink
    February 2, 2015 10:07 pm

    Paul, you might find this post on Prof. Curry’s Jan 31 ‘Week in Review’ blog interesting, as it attempts to gloss over the adjustments of the past temperatures.

    angech2014 commented on Week in review.

    in response to Tonyb:

    Morph The met office. We just discovered that the 1659 start date of CET has altered recently. It has been adjusted upwards by 0 .3 degrees. Why 350 year old records needs to be adjusted perhaps Nick can explain. It is one of many changes over the last five years. Tonyb

    Tonyb | January 31, 2015 at 4:42 pm |
    “Perhaps you could explain why, for example, a record from 1659 should need to be adjusted?”

    This should help.
    It incorporates Mosher’s and Zeke’s explanations
    Basically the further back in time you go the more you have to adjust temperatures downwards.
    Zeke and Nick Stokes know this, it’s common knowledge. Why are you surprised?

    “Zeke (Comment #130058) June 7th, 2014 at 11:45 am
    Mosh, Actually, your explanation of adjusting distant past temperatures as a result of using reference stations is not correct. NCDC uses a common anomaly method, not RFM.
    The reason why station values in the distant past end up getting adjusted is due to a choice by NCDC to assume that current values are the “true” values. Each month, as new station data come in, NCDC runs their pairwise homogenization algorithm which looks for non-climatic breakpoints by comparing each station to its surrounding stations. When these breakpoints are detected, they are removed. If a small step change is detected in a 100-year station record in the year 2006, for example, removing that step change will move all the values for that station prior to 2006 up or down by the amount of the breakpoint removed. As long as new data leads to new breakpoint detection, the past station temperatures will be raised or lowered by the size of the breakpoint.

    An alternative approach would be to assume that the initial temperature reported by a station when it joins the network is “true”, and remove breakpoints relative to the start of the network rather than the end. It would have no effect at all on the trends over the period, of course, but it would lead to less complaining about distant past temperatures changing at the expense of more present temperatures changing.

    • Mikky permalink
      February 3, 2015 10:55 am

      I would have thought that if it takes 100 years of data to detect a breakpoint then the break in question would be highly suspect and possibly just due to a general change in land use in the area of the station. Such general changes in land use should not be corrected as they produce genuine changes in the local climate over large areas.

  6. Dave Ward permalink
    February 3, 2015 2:22 pm

    @ Shub Niggurath – That still leaves the possibility of different Datums. If the researchers were using a local map, particularly an older one, and then transferred co-ordinates to Google Earth (for instance), variations will almost certainly occur. Even when using GPS there will be variations in position. Whilst Selective Availability was turned off many years ago, the number and position of satellites visible to the receiver is important. High quality units will show the level of error (this is required for commercial aviation use). Most consumer level units will not, although the old Garmin 45 I still have displays a simple graphic of the position and elevation of which satellites it is tracking. From this one can make a reasonable estimate. DGPS* is used in situations requiring a high degree of accuracy, but needs a very precise fix to start with.


    The ONLY way to maintain accuracy is to use ONE co-ordinate display method AND Datum, and stick to it…

  7. February 3, 2015 2:30 pm

    Weather data was collected in Paraguay by the Navy until 1970. I would presume that the station location data presented at NCDC is the location of the first known station of that name – hence the locations are marked on water. A little online historical research can work wonders in solving such apparent anomalies.

  8. Eliza permalink
    February 5, 2015 5:34 am

    That picture is of Asuncion Airport Silvio Petirossi., Not Roberto Fuster I live here BTW.I might go and have a look or ask around. Im sure ALL the stevenson Boxes are on Ground, not water but i’ll see if I can check this out with the met offuice here

Comments are closed.

%d bloggers like this: