Sunday, 1 November 2015

Weather report October 2015

The official weather report for the last 3 months of the year mentions the words "El Nino".  That certainly looks like a fair call thus far.  The last 2 months have been drier than average (as measured by rainfall) and October had a considerably higher set of minimum temperatures than last month.


While the fall was better than in September 2015 it was still well below average.  Probably as a result of the above average Winter falls Whiskers Creek has kept running although it is only a trickle by the end of October.  

Looking at a 12 month moving average (thereby eliminating seasonal variation) the final three months shown downward movement, but the trend of the series is still heading up.  But with a value of R2of 0.61 it isn't a particularly strong result.


The bars showing the direction of change from 0:30 to 23:30 are all near the minimum temperatures, showing the desirable state of warm during the day but cooling at night (= easy sleeping).

The next two charts - sorry to those of you who don't like charts - compare the average monthly maxima and minima with the averages of those.
 The maxima were a little over the average.  This was particularly so early in the month, with a couple of days over 30oC, which I thought a trifle rude!  We went over 25oC on 16 days and all maxima were >15oC.
The minima were well above the average.  No frosts were recorded at all (probably trying to lure the tomato planters into an error before Melbourne Cup day).  The minimum was >10oC on 4 occasions and only went below 5oC for the first three nights of the month.

I have also looked at the average temperature for the month.  I won't use up bandwidth on another chart but most of 2015 has been a little cooler than 2014: in October this has changed with the later year 3oC higher.


As usual I have shown the 3pm level of Relative Humidity.  In the following chart I have also shown the amount of rain in that calendar day (unlike BoM who measure stuff on a day starting at 9am - which I think reflects their traditional working day).
It is probably a bit surprising that there was no rain in the humid days around 11th to 14th and that there a couple of drops in the last week of the month, when humidity was rather low.  

Unfortunately I only have one other year of humidity records for October (ie 2014) but it is of interest that the 2 averages are pretty similar (and thus the average of averages is also similar).


As usual the daily gusts resemble a cross section of the Alps.
 Looking at my average gusts per month - again with only one other year of data - 2015 was a tad less windy.
I have experimented a little with putting a trend line on this graph (to reduce the 'noise' rather than to form a basis for extrapolation etc) and it seemed very sensitive to the type of trendline used.  Using my customary 4th order polynomial gave a 'nice' value of r2 but reckoned the least winy months were March and April which seemed counter-intuitive.  Restricting the trend to a quadratic funcrion gave a less windy period in June/July but only had a poor value of r2.  So I conclude the data isn't up to that level of mucking around.

No comments: