Annual Weather Report 2023

 A rather damp year overall, with a very dry patch in the middle.  Overall a little warmer than usual due mainly to higher daily minima.  I have tried to talk about interesting matters below rather than a simple of recitation of orders of rank etc.  Enjoy!

A summary of kay data variables for the year, from my Weather Station is in this Google Sheet.

Rain

We ended the year with 1006mm recorded at my Davis weather station.  I also have a Nylex plastic gauge which agrees closely with the weather station.  I place that total in context it the table below, showing 2023 totals for a few other sites of interest to Mallacoota.  (I have included Bombala AWS as it is close to the source of the Genoa River.)

    Angophora Drive    1006mm

    BoM Airport              898mm

    BoM Gabo                 937mm

    Clarke St                  1198mm

    Bombala AWS          592.2mm

The first chart is a time series of annual falls at Mallacoota.  For 1981/2 and 1995 no monthly data is available from BoM as no information was collected in a few months.  For 1983/4 several readings are accumulations over more than one day so daily readings are misleading, but the annual totals are accurate.

Looking at the cumulative rainfall through 2023 shows quite reasonable falls until May when a very dry period commences, lasting until October.  A major rain event in late November and early December transformed the year from 'dry' to wet.  (The level of Bottom Lake in that time was higher than the author had ever seen it, including periods when the Mouth was closed.  Anecdotes suggest it was higher at other times - possibly before the BoM Airport site began operations - commenting, for example, that the water reached to the bakery steps.)

A second way of looking at the rainfall through the year is the progression of pro-rata estimates of the annual rainfall.  This process uses the average % of annual rainfall recorded to a date divided into the cumulative rainfall year to date.   (For example by 16 April, on average, we have received 26.496% of our annual rain.  In 2023 we had recorded 275mm of rain in total, so the estimate of annual rainfall on that date is 275/0.26296 = 1037.9mm.)  The next chart plots the daily estimates through the year from 18 January: as there was a heavy fall earlier in January estimates before that date simply expand the vertical axis.  The dry period from May to September is reflected by the steadY downwards slope of the line through that period.

My final look at the year-total material is the number of days with rain.  Again there are gaps in the record, in this case also excluding 1983/4 as the data for individual days in not complete.  I have no great insights from this!

Finally a view of this year compared with median monthly fall.  This merely shows the material referred to above.

Temperature

My overall assessment of temperature is the anomaly: the difference between the average temperature for Year to date  and the long term average temperature for that period.  The following chart shows the way that calculation evolved through the year, together with the daily difference between the actual average temperature and the long term average for each day.  The anomaly ends at +0.47C, so the year overall was warmer than average. 

Heat Waves and Cold Freezes

Both minimum and maximum temperatures are used in determining whether a heat wave or a cold freeze has occurred, since (eg) if a minimum temperature is not also significantly high there is relief from the heat.  It has proved a little interesting to investigate this: I have put some definitions etc in an addendum at the end of this report and the underlying data is in the page 'Heat Wave Data' in this Google Sheet.  

I have also included a summary of the number of significantly hot or cold days in the page 'Heat Wave Summary'.  There were only 7 Cold Days in the year not of them in an extended period.  There were 32 Hot Days, of which 7, in September formed a Hot Spell (3 of the 14 values were less than 1 SD above the long term average maximum for the month).

Minimum Temperatures

The next chart shows the long term mean minimum temperature and the actual minimum value for each day of 2023.  

I have also compiled a time series of Average Minimum temperatures for each year.  The simple linear trend appears to be close to showing an upwards trend.

Maximum temperatures

The next chart shows the long term mean maximum temperature and the actual maximum value for each day of 2023.  The hot spell in September, referred to above shows well in the 2023 data.

The time series for maxima has been adjusted to allow for the values in years before 2019 coming from the airport and thus being more impacted by cooling sea breezes than my Weather Station.  I don't think that explains the very low readings in 1995-96: in those two years nearly every month was below average and a very similar pattern is evident for Gabo Island (with even lower monthly values).  In plain terms, that was a very cold period.

Humidity

Very limited background data as the BoM releases next to no data without charge.    The key points shown here are:

• the afternoon reading is usually (see below) lower than the morning reading  (on average the afternoon reading is 85% of the morning one;
• there is no significant trend over the year (at least for 2023).  In other words the relative humidiity is constant across seasons.

I was intrigued about the relationship between the two readings.  As expected most of them are grouped closely together.  On the few occasions where 1500 is higher than 0900 (ie % >100) it is mainly because the morning reading is very low: for example on 20 Sept the morning reading was 40% under a Northerly wind and by the afternoon the wind was Southerly boosting humidity to 77% which is very close to double. 

Wind

Again no (free) help from BoM.  The first chart shows the daily run: note the horizontal axis labels approximate the first day of each month: with months of variable length its beyond my skills and/or pay rate to get them precise.

As the series peaks around the end of the year I have compiled the last chart on a year ending June.  The key lines are the red dots (polynomial of the AVG series) and the green dashes (first half of the current period).  

The current 6 months' readings are surprisingly low given the amount of storms we have had - possibly reflecting my WS poor exposure to Easterly winds.  However I have looked at the maximum gust recorded by BoM on Gabo Island for that period and the overall shapes of the curves are surprisingly (understatement) similar.

Addendum: definitions of Heat Waves etc

The BoM defines a Heat Wave as a period of 3 or more days of significantly high maximum and minimum temperatures,  I have defined "significantly high" as 1 Standard Deviation above the long term average temperature for the month in question.

I have also defined a Hot Day (HD) as a single day in which both min and max are significantly high, so my definition of Heat Wave is 3 or more consecutive HD.  The days within a Heat Wave are rated as Heat Wave Days (HWD).

In looking at data from month to month I have also found periods in which there are several HDs but with a day in which either min or max fails to be significantly high, thus breaking the series.  It is still an unpleasantly hot period! I have decided to define such periods as a Hot Spell with the constituent days being Hot Spell Days (HSD).  I have not yet developed rigorous rules for limiting the length of a Hot Spell but it must be at least 4 days, with at least 3 of them being HDs.  (If days either side of a Heat Wave would meet the conditions for being HSD they will be counted as such.)

The same thinking can be applied to low temperatures, although I am not aware of the BoM having done so.  I have come up with the following definitions:

• Cold Day (CD): both min and max >1 SD below the long term average for the month;
• Cold Snap (CS): a run of several days in which most of the days are CD;
• Cold Freeze (CF): three or more consecutive CDs.