June Global Temperature falls 0.04C

The global averaged surface temperature for June 2019 was 0.62C using my spherical triangulation method merging GHCNV3 with HadSST3. This is a further drop of 0.04C from May 2018. The discrepancy with GHCNV4 is however growing. V4C calculated in exactly the same way gives a June temperature of 0.75C, a rise of 0.03C,  and 0.13C warmer than V3. This difference is statistically significant.

Both the V3c and V4C spatial  distributions for June are shown below. Two warm zones are visible across Europe and  Eastern Russia sandwiching a cool zone in central Asia

The annual temperatures, including the first 6 months of 2019, looks as follows.

There is a large and growing discrepancy between V3C and V4C, that begins only after 2002. The 6-month 2019 average temperature for V4 is 0.84C,  some 0.07C warmer than V3C. This looks suspicious to me. Although V4C nominally has far more stations than V3C (17378  versus 7280), I am discovering that many of those new stations are actually sub-composites of V3 stations. This introduces an element of double and triple counting, which I hope to write this up in a post fairly soon. 

About Clive Best

PhD High Energy Physics Worked at CERN, Rutherford Lab, JET, JRC, OSVision
This entry was posted in AGW, Climate Change, climate science, NOAA and tagged . Bookmark the permalink.

3 Responses to June Global Temperature falls 0.04C

  1. Nick Stokes says:

    “(17378 versus 7280)”
    Should that be (27378 versus 7280)?
    Double counting shouldn’t matter if you are doing area weighting properly. With tiangulation, if you include a station twice, the weight should halve.

    • Clive Best says:

      No, because I only include stations which have data within the normalisation period (1961-1990). I count 27410 unique stations IDs in V4 of which 17378 have sufficient data from 1961 to 1990 to calculate normals. So nearly 10,000 stations cannot produce anomalies.

      What I have discovered though is that stations which in V3 are composites of of several stations within a single location eg. Darwin (PO, airport1 and airport2) have the individual components presented as well in V4. Even worse the raw V4 data for Darwin Airport includes the merged PO data in the airport site and reproduces PO a second time as a different station. This is double counting.

      I am suspicious of V4 because it has added a huge amount of warming to V3 after 2002. It is impossible that this is due to higher statistics.

      • Nick Stokes says:

        “I am suspicious of V4 because it has added a huge amount of warming to V3 after 2002.”

        I don’t agree with that at all. I have written a post on it here. With unadjusted data, I get virtually no disagreement in recent times, which says that the raw data isn’t causing any differential. With adjusted data, I get a small rise of V4 relative to V3 since 1975, but very little effect recently, with V3 sometimes having the (slightly) greater trend. Here is my plot of the comparison of unadjusted data since 2000:

Leave a Reply