Seasons in Mallacoota (pt 1)

A key analytic issue is that seasons are best indicated by the regular seasonal changes in the environment.  I don't at present have the data needed to assess this in detail so have labelled the post Part 1.  However my inclination at present is to stick to the 4 official BoM seasons beginning on the 1^st of September, December, March and June.

The best background to my views on this are contained in a post I created for our place in Carwoola.  To summarise the outcome of that for Carwoola my set of seasons there were:

• Spring: September, October
• Summer: November, December, January, February
• Autumn; March, April, May and June
• Winter: July, August

This hints at one of the issues in moving from an astronomic (equinox, solstice) based definition of  the seasons.  That is that seasons vary according to place, with major variables being (IMHO)  elevation, distance from the coast and latitude.  Where an analysis is trying to be all things to all places in (eg) Australia it is probably as well to pick a standard based on astronomy - which obviously does have an impact - as anything else.

Note that I say " a standard based on astronomy" not "the standard based on astronomy" since Australia uses the beginning of months while the actual astronomical events are around the 21st of March, June, September and December. 

A key factor in my variation from the standard was that it didn't seem to fit well to my observations of natural phenomena in Carwoola (nor particularly well to gross changes in the weather over a year).

For Mallacoota, at this stage, I only have a limited number of items to consider.  I do have a series of weather observation data from BoM going back to 1974 (rainfall) or 1993 (temperatures).  I also have a series of observations of birds covering a rather vague period, but heavily biased towards the period from 2014 onwards.

Weather series

Temperature

I have looked at two series about temperatures,  The first shows the monthly average maxima and minima.

 I can see little in the way of abrupt changes in these series, such as would indicate a seasonal break. January and February are clearly the warmest months and July and August are clearly the coldest.  It isn't clear to me that a case exists for preferring December over March as part of Summer, nor preferring June over September as part of Winter.

My second chart looks at the number of days each month with a maximum over 25^oC or a minimum below 5^oC.  (Those numbers were largely chosen by eyeballing the data, but using a limit of 5^oC does give a clearer picture than using 10^oC.)

Concentrating on the 4 "marginal" months December has more warm days than March (and the same  - 0 - cold days).  June is lower than September for warm days and well above it for cold nights.

Rainfall

I looked at the average daily rainfall, and the number of days with >0.2mm of rain, by month.  Neither series showed anything I could pick as a seasonal break.  So I won't waste your downloads with more graphs!

Birds

The main indicative interest of birds as auguries of weather - and I don't want anyone to think that I have been disemboweling them - is the movement of migratory species.  

From a broad scrutiny of the eBird data available to me it seems that most species migrate in the period February to April.  Trying to establish whether which of those months is the month of change is somewhat more difficult in Mallacoota than it was in Carwoola.

• I began by looking at known migrant species and seeing when the number of records dropped.  There were very few species for which April seemed key.  13 species seemed to have headed out in February against 11 in March.
• The most spectacular movements I have seen are Red Wattlebirds and Yellow-faced Honeyeaters flying West up the Inlet - presumably to cross at The Narrows.  This can amount to hundreds of Wattlebirds and thousands of Honeyeaters per day and occurs in late March - April.  However there are many reports of both species right through the year..
• I then examined the number of species with a higher number of reports in February than March (114 spp) and March vs April (143 spp).

Again I don't consider this to be conclusive.