September 2020 Weather Report

 Overall a fairly mild and dry month.

Rainfall

After 2 months of well above average rainfall September was a drop back to well below median rainfall, recording a total of 28.8 mm (43.3% of median).  It was also well below the fall for September 2019 (itself pretty miserly).

This rather unpleasant outcome seemed to contrast rather badly with a 28 day rainfall forecast I shared to the FB Group early in the month.  These forecasts are the probability of any (ie > 0.2 mm) rain on a day, given as a 4 point scale which I have converted to numbers: Blank = 0; Low = 1; Medium =2; high = 3.  I then did a scatterplot of the numeric scale against actual.

It shows clearly why one has to sign an indemnifier before getting to the data.  As a meteorologist friend once said "trying to forecast weather precisely more than a week ahead is lunacy".  The good news is that on 9 days the forecast and actual were both zero.  Less bad was the second highest daily fall was a day rated High.  The worst news is that on 4 days with high probability there was 0 rain while merely bad news is that the other 2 days of significant falls (>5 mm) were only rated medium.

Despite the low fall for September the pro-rata estimate for the annual total remains quite high.  At the end of September it is 969 mm.  The progress of the estimate through the year is shown below.

Temperatures

My overall measure of the temperature is the anomaly: the difference between the current and long term averages.  I have calculated this, using BoM data throughout, as +0.7C for September.  I do the calculation each day which shows how the measure develops through the month.

Basically the average temperature fluctuated between above and below long term mean with runs of a few days each day until the 23rd when there followed a run of 7 days with average below long term.

A second way of looking at the overall pattern is through the daily extremes. As obvious from the legend I have plotted both the BoM and Weather Station (WS) readings in the following chart.

In both cases the pattern is very similar with correlation of 0.90 for minima and 0.94 for maxima.  The maximum is usually a little lower for BoM, reflecting the impact of the afternoon sea breeze (usually also indicated by the wind direction swinging from NW to the quadrant between NE and SE).  The minimum on 20 September recorded by BoM was 17.2C: this is the highest September minimum recorded there by 0.8C.

Minimum temperatures

For my weather station the lowest temperature recorded was 5.7C and the highest minimum was 15.7C (on the 20th).  The average minimum was 9.8C - the highest ever for September  (the previous highest was 9.72 by BoM in 1999).  As an overview this chart stresses how high the average minimum was.

Maximum temperatures

For my weather station the highest temperature recorded was 29.3C (the 5th highest recorded for September - although noting the above general comments BoM maxima are usually a little cooler than home) and the lowest maximum was an unremarkable 14.1C (on the 26th).  The average maximum was 19.4C.  While this is the highest ever for September it is overstated compared to the BoM site which recorded an average maximum of 18.3 - itself the 5th highest maximum recorded.  

The following graphic shows the WS values for this year and last and the BoM average.  To avoid the overstatement in comparison with historic data caused by the lack of sea breeze the blue lines indicate the BoM average maxima for the years concerned.  This year was still warmer than last year and the average.

Humidity

The first chart shows the humidity at the 2 standard times of 0900 hrs and 1500 hrs.   Aa is usually the case the two series are much closer that for our last place near Canberra.  Slightly surprising me they are not that strongly correlated with a coefficient of 'only' 62.7.

The average morning humidity was much below the value for 2019 but a little above the average from BoM (covering 1993 to 2010).  Comparing the daily values for September 2020 it appears that the 0900  humidity is usually lower at the BoM site, possibly due to a land breeze setting up overnight.  Thus a slightly higher humidity could be expected for the WS.

For the afternoon readings the value was almost the same as 2019 and somewhat below the BoM average from 1993 to 2010.  The comparison of daily values for September shows a very close agreement between the WS and BoM values.

Wind

September is often regarded as a windy month and 2020 was not an exception.  While there were some calmer days the 2 day total of 20th and 21st (682 kms) was the 5th longest 2 day wind run in the 20 months that I have been recording here.  (The longest 2 day run - 883 km - was the same two days in 2019, with the 520km on the 20th being the longest 1 day run).

The total run was almost exactly on the adjusted average from the Bom site (which is always windier than home) but notably less than 2019 ( boosted by the blast on the 20th described above.

The wind rose is interesting in its diversity with 5 rather different directions each contributing more than 10% of the number of readings.  Most months there are only 2 or 3, usually tightly grouped directions (eg NW, N, NE) contributing the bulk of readings.