Hadcrut4 data coverage

Posted: June 19, 2015 by tchannon in Analysis, Dataset

I’ll give the Met Office / Hadley / CRU some due, at least there are signs of actual data.

What-if a list of the dataset grid cells with data from the start Jan 1850 to current is computed? Where are these and what does the data look like?


Figure 1, A fragment in time of the entire consistent Hadcrut 4 gridded data, weighted data as per computing a global mean.

This suggests the reputed bad winters during WWII. Computing the global equivalent would need the weighted mean completing, sum and then divide by the sum of weights. Unweighted (not shown) is similar except the winter 1944 has a hot peak, ie. largely a polar matter. There again without global data is any of this valid?

Here is the global coverage


Figure 2. Unweighted data (as is) for Jan 1942, same coverage always, 1850 or 2015.

This is best viewed as a PDF, which is annotated with the unweighted temperatures and the latitude and longitude centre of each grid cell with data.

Additionally I’ve extracted the time series of each grid cell. Figure 1 is a plotted fragment from that.

This begs the question where are these met sites? Some are easy to guess, others, I have no idea.

Would it be good idea to make this software and data available?

Post by Tim

  1. 8364khz says:

    Could the “Admiralty List of Radio Stations Volume 4” be of any help.
    This book listed all of the reporting stations around the world.
    There was even an entry for Ribble Head Station, about 5 miles north of where I live.
    The main problem here is that out of date volumes were thrown overboard when the replacement arrived.
    Each reporting station had a coded prefix and, if I remember correctly, in those days, contemporary weather reports from these stations could be accessed through the Met Office. When at sea, I often thought of trying to obtain a current weather report for Ribble Head but as far as I can recall, I never did.
    Records were meticulously kept but…
    Now, do I remember correctly, where not all of the maritime observation records “OBS” reports “lost”?
    I still have two significant maritime OBS messages framed on my wall.

  2. tchannon says:

    Annuals of that kind traditionally carry a lot of ancillary information, all of it ought to be available anyway.

    Ho ho.

    I’m only dealing with the published datasets where if you look which hardly anyone does some sad stories emerge.

  3. tchannon says: