South Australian spring rain 3 weeks later than 100 years ago, solar blamed

Posted: May 6, 2014 by tchannon in climate, weather




A detailed analysis of the rainfall of Adelaide has established that periodic changes occur in the incidence and duration of the winter rains. These changes have a period and amplitude of approximately 23 years and 30 days respectively, and superimposed on them is a long-term trend which is manifested by protraction of the latter half of the season, spring rains now occurring about 3 weeks later than they did just over 100 years ago. The total quantity of rain precipitated has shown no statistical significant changes.
paper at ADS



As Talkshop readers will know all too well Tim does tongue in cheek, “Manuscript received January 25, 1954”

This paper has most things including academic sensitivity to criticism, a riposte to past words.


Very surprisingly it tries to link the change to solar magnetic activity and a change in storm paths. What were dubbed solar P cycles and E cycles.

Before anyone dismisses that as fringe crazy I will point out that the finding would not appear in a simplistic data analysis. This is a change in timing and using high resolution data, daily.

A further curious report by Deacon, “and he concluded that a major climatic change has been operating progressively since 1880 (the earliest record he used), contemporaneously with changes occurring in the northern hemisphere.” Written in 1954?

It crossed my mind that here is a test, does the assertion hold good after 1954?

The first hurdle is data and here CSIRO do unlike the UK Met Office supply data. I believe this is the station 23000, Adelaide West Terrace, 1839 to 1980, and yes daily data is available.

I feel this is beyond me so here is an opportunity for someone to follow up. Replicate some of the reported findings and then take it forwards.

Post by Tim

  1. craigm350 says:

    Reblogged this on CraigM350.

  2. Captain Dave says:

    Mods – dual typos in sub headline, Adelaide missing “i”

  3. Ray Tomes says:

    English temperatures also show a cycle of around 23 years. Please look at this analysis where I show that the main period in England is 23.9 years with a possible additional component matching Hale cycle.
    This article refers back to a previous one on this blog:
    In my experience there is a cycle present in a number of different things that is close to twice Jupiter’s period and definitely different from Hale cycle. To understand more about this please see my page on Harmonics theory at

  4. tchannon says:

    There is a double Hale widely, why I keep mentioning ignore the obvious optical sunspot cycle.

    Correlating should be with the known magnetic cycle, which varies somewhat in timing.

    The Australian data which is the subject of this article is daily therefore free from the worst of Nyquist artefacts typical of meteorological monthly. The report is suggesting a subtle time variation with a change in storm tracks as the suggested immediate cause.
    Given for example the recent UK storms, also down to storm track change. there are things worth investigating.

  5. Overlayed the Adelaide rainfall time series by Cornish with the sunspot cycle
    Played around a bit and marked in solar max peaks but was quickly disillusioned.
    However when stepping back and looking at the more general amplitude of solar cycles there appears to be some correlation with solar amplitude and Adelaide rainfall
    When solar amplitude is lower , so is Adelaide rainfall. generally.

    May not be just storm tracks as there are other drivers for SA rainfall like North west moisture infeeds
    We have a local independent researcher in SA. His name is Ian Holton. His forecasts and ideology is strongly solar based and he has conducted extensive research in this area