More on wind

I have posted in the past about a comparison of data from my weather station (WS) and the Bureau of Meteorology site  The conclusion is that my data is reasonable as a general indicator of what the weather is doing (or has done).  To some extent this is a little surprising as the anemometer is mounted about 2m above the ground rather than 10m and is not well exposed other than from about NE - N -WSW.

With that being said, as I have compiled my monthly reports it has seemed to me that the wind direction has changed in the Summer months with a much higher proportion coming from the South East than in other seasons.  So I decided to check out what my ~14 months of WS data had to say about wind direction.

Cutting to the chase, my examination that follows with (unfortunately in some views) several graphs concludes that for 2019 at least there was a seasonal pattern in the direction of the wind.  ESE in January February, then WNW or N until July with WSW-W for the rest of the year.  I took a brief foray into barometer readings to see if they explained anything but have so far failed to find the silver bullet in that regard.

There are several measures of 'amount' of wind available from the WS: average speed over preceding period; high speed over preceding period; and run over preceding period.  As I take a period as 1 hour the wind speed and run are effectively the same.  I will start off using run.

The first chart shows the average hourly run recorded x month.  Note that I have truncated the vertical scale to emphasise month to month differences.  The surprising aspect is that November appears to have the fastest winds.

 Clearly the strongest winds come from the North.  I suspect this may be related to exposure of the site to that direction, but if that is the case I am surprised that the winds from the NNW and NW are not equally strong.

What I was most interested in was whether the apparent shift between segments of the compass was real.  I tried many ways of demonstrating this and concluded that the best way of doing so was to focus on the three segments (ESE; NNW+N; and WSW+W) which together make up 85% of recordings over the year.

My first approach was to look at the % of run recorded from those the segments.  There is a clear Summer peak in East winds, but it is still outweighed by the NNW+N segment in all months with WSW+W becoming dominant in late Winter/early Spring.

 Adopting a slightly different approach and counting the number records from each segment emphasises that pattern with East winds dominant in January and February; the NNW+N ruling until July and then WSW+W taking over for the rest of the year.

 I wondered if I could show the strength of winds to be due in some way to barometric pressure.  My first attempt was to plot daily run against average daily pressure.  While the trend line does slope downward (low pressure is linked to stronger winds) this is not significant.

My second thought was to wonder if the change in pressure (such as might occur when a front passed through) showed a pattern.  My measure of change was the difference between the maximum and minimum pressures in a day.  The trend line is again not significant.