Runs and Gusts

On our morning walk today it seemed pretty windy.  So I was quite surprised on returning home to see that the wind run for each of the past few hours was only just over 8 km.  The maximum gust was around 30 kph which seemed more like it.  It is noteworthy that as I take records at hourly intervals the run is effectively the average wind speed over the period.

This has led me to explore the relationship between wind run and maximum gust as revealed by my weather station.  This gives me ~10600 observations.  My initial interest was in the ratio of gust: run.  This couldn't be calculated for 1614 observations as the run was 0 (in 368 cases the maximum gust was also 0, so a very calm hour!)

I will start with a chart showing the number of cases in which the ratio of gust:hourly run.

 54% of the ratios are in the range 2 to 3. I have no clear idea whether that is normal.

I attempted to look at some data from my last site in Carwoola but that was very sheltered so the runs were, I think, understated.  The biggest class there was a ratio of 7 -10.  From looking at the highest ratios (>10) at  Mallacoota they all occurred where the run was was very short (1.6 km in the hour) so an understated run might well increase the ratios.

For Mallacoota there is overall a good correlation between run and gust as shown by the quite significant trend line.  The equation for the trendline (Gust = 5.46 + 1.99 Run) indicates that the gust is likely to be twice the run.

I calculated the average ratio by month.  The graph shows a reasonably significant pattern on increase around the middle of the year but none of the values are outside the 95% confidence intervals.

I also calculated the average ratio by wind direction.  With wind direction I am always tempted to display the results as a rose chart.  Doing so in this case produces

1. a pretty chart but 
2. one that doesn't show any great differences!

Possibly the second point is because the differences are quite subtle.

As always, with anything to do with wind I am suspicious about the impact of siting on observations.  In this case I think it quite likely that the location shelters the WS from the strongest gusts frm NE to SE, thus cutting down the ratios in those sectors