Welcome to September wet or dry

[Julian Mayes]

After an August having an enhanced climatic contrast between north and south-east, I wonder what the prospects for September are. The past is not necessarily a guide to the future but here's a graph that shows how much September rainfall has dropped over the South East since 2000. Below it is the equivalent graph for Scotland which shows a smaller drop. Of more concern to Bruce and Jack might be the increase shown in Scotland for most other months.  The previous period had very wet Marches so a small decline there is maybe not surprising. In the South East, there's been a recovery from the very dry Julys of the late 20th century. Also shown is the September time-series from 1931. In each case these are the Met Office regional rainfall series obtained from their website. https://www.metoffice.gov.uk/research/climate/maps-and-data/uk-and-regional-series

I got a bit ahead of myself with the key - 2019 months only up to April - I cannot forecast to the end of the year! 

And here's the September time-series for SE / Cen S England.....

We don't seem to get wet Septembers any more in the SE. 

This is called tempting fate. 

Julian

Molesey, Surrey.  

[Len]

No trend to dryer Septembers since 1985 in Wembury, SW England.

In fact highly variable.

[Julian Mayes]

Len, I think there is a trend - your linear trend line just defies the trend; graph shows the step-down after 2000 (which happens to coincide with my change in averaging periods). A 5 to 10-year moving average would show this. 

Julian  

[xmetman]

Using the gridded regional values from 1931 you can see a drier linear trend during September in all regions but most pronounced in the south.

But if you take a look at the 1910 series the linear trend is more or less flat although the 10 year moving average is way down.

But if you go back as far as you go with the EWP to 1766 the linear trend is sharply down (-21.3%)

That's the thing with linear trends they're a bit like Forrest Gump's box of chocolates.

I wrote a story about them the other day - you can use them to prove just about anything if you select the right period - but usually longer should be best, but 253 years might be too long.

There's no doubt that in recent years September have become drier and in some regions noticeably so.

[Len]

I just use the raw data, No averaging or smoothing.

Since 1985 there is no significant trend in Sept rainfall in Wembury.

The variability is high and statistical test t test shows no significant trend.

However you can play around with different data and time periods and show some significance elsewhere.

It's all smoke and mirrors. A bit like Brexit.

Len

Wembury

[Smartie]

See-

https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/joc.6208

Murphy, C, Wilby, RL, Matthews, TKR, et al. Multi‐century trends to wetter winters and drier summers in the England and Wales precipitation series explained by observational and sampling bias in early records. Int J Climatol. 2019; 1– 10. https://doi.org/10.1002/joc.6208

D

[xmetman]

So don't trust the EWP series before 1820.

I never realised that Manley developed a sleet and snow index.