Rainfall 1766 onwards

Posted: April 27, 2014 by tchannon in Analysis, climate, Natural Variation, weather

mo-1766-general-small

Since oldbrew nicely points at Paul Homewood I am taking the opportunity to add a quick post.

I am very close to posting an article on the 17 rainfall data series from 1910 onwards, been working on this since January, took half hour to drop in the long Met Office England and Wales time series 1766 onwards.

I post an output without much explanation.

If you want to really look the PDF is much better provided you know how to magnify, you can zoom in on the timeseries plot. No actual markers, keeps the file size down.

met-office-1766 (PDF 84kB)

Talking about seasonal is no good unless Slingo puts up the full data.

Lets see the daily data, with station data from 1766. For that matter same from 1910 will do.

Nyquist is real Ms. Slingo. Yes there is randomisation in time… if you admit it then why don’t you and the Met Office walk the talk?

Post by Tim

Comments
  1. oldbrew says:

    The ‘Annual Variation’ bar chart for England and Wales could be said to look like a wave.
    I wonder what the chart for Scotland in the same period would show?

    The reason for asking is, if the jet streams have tended to move further south, has this dragged some of Scotland’s usual rainfall with it?

    The Telegraph has an interactive graphic: ‘UK rainfall in every year since 1910’, where you can select on ‘average rainfall’ (instead of ‘year’). 2000 onwards has 7 in the first 20, 1 in the last 20. (1998 and 1999 are also in the top 10).

    http://www.telegraph.co.uk/topics/weather/9777749/Interactive-graphic-UK-rainfall-in-every-year-since-1910.html

  2. Roger Andrews says:

    Tim: If I read your results correctly then all five of the driest months since 1766 occurred in October. Is there something meteorologically different about October?

  3. tchannon says:

    More regional information will be in the 1910 onwards data. (probably post the whole article later today)

    What I am showing specifically excludes seasonal with the exception of indicating what is removed from analysis:- seasonal is not random, is not noise. In the context of Queen’s court soothsayers (Slingo etc.) declaring the sky is falling from an office driving experience of the world not knowing the fact that it rains a lot late in a year is pathetic. Hence here I am showing how things look when normal change is ignored. The AGW claim is after all an irregular change.

    An additional problem which is also it seems outside of Slingo’s expertise is the effect of 2D data being treated by professionals such as the Met Office staff as 1D data. That is specifically Slingo’s mistake, noise occurs in time too. She is complaining about 1D maths not taking into account 2D variation. Publish the daily and station data then it can be worked out. As it stands we are stuck with bad food from bad cooks who won’t let us get at the raw materials.

    If the context is drought and flood, subtly different from rainfall, it involves the accumulation into a regime, which seasonal will not sort out.

    The 1766 data is dubious. Scotland data pre-1990 is dubious and messing up UK data.

    The UK weather split is western coast vs. eastern side, with some north/south variation. UK, or Scotland or England, etc. is lumping together different regimes.

    Basically the west is wetter. Atlantic maritime.

    All areas show a sharp drop in rainfall at the onset of cold winter weather and the rest of the year the rainfall ramps back up. The 1766 data shows a weak variant of this.

    The wet winter 2013/2014 was about a particular storm pattern where the excess came up from the south with widdershins storms passing west to east spinning rain into the southern regions, exactly what turns up in the data. The whole thing rotating the other way. I don’t know how this fits into the concept of jet-stream.

  4. tchannon says:

    RA, the abnormally dry for October months, annual variation for October is excluded.

  5. Roger Andrews says:

    Tim: According to your z-score plot, which shows the five dry Octobers, annual variations have been removed.