Start of Tawny Frogmouth nesting

I have been monitoring the breeding activities of a pair of Tawny Frogmouths since 2008 - a period of 10 years.  I have good measures of when nest building started for 7 of those years.  Of the other three years:

• 2008 was the first year and I didn't realise the nest existed until I spotted the male in incubation mode;
• 2011 the nest site moved to another tree, and again I didn't realise what was going on until the male went missing and I found him incubating.;
• 2014 had - I now think - a false start, in that some twigs were blown into the nest site and gave a misleading picture.

For this year the birds seemed very late in starting their activities but on 6 September 2017, when we returned from Mallacoota, there were a few twigs in the nest fork.  On the morning of 7 September more twigs had arrived, but they hadn't yet been trampled ...

... so I am taking 6 September as the start of nest building.  (If they had started while we were away I think it would be no more than one day earlier.)  This is the latest date I have recorded: counting back from the date of fledging chicks 2008 may have been later but making assumptions like that is a very slippery slope.

This year has been pretty dry and - in terms of average minimum temperature - quite cold.  I wondered how these factors went in explaining the relatively late start on nest building.

I have daily data on rainfall, maximum and minimum temperatures for the period 2009 to the present.  I used this to build some data for analysis:

• total rainfalls and average extreme temperatures for official Winter (June to August) and my version of Winter (July and August); and
• number of days since 1 January to start of building

for each in-scope year.  On looking at the correlation coefficients it seemed that the 2 month weather data correlated more closely with the date of starting than the three month values so the shorter period data form this table (actually an image of a table)

The correlation coefficient for rainfall with date of starting building is quite strongly negative, suggesting that the lower the rainfall, the higher the number of days (ie the later the start of nesting).  The correlation for minimum temperature with date of starting is not so strong but again suggests a cold year gives later nesting.  The coefficient for maximum temperature is too weak to be considered.

The data can be plotted on a graph.

From my inexpert view this reinforces the idea that rainfall in July and August is the most important factor (of those I have assessed) in determining the start of nesting.

As the breeding event proceeds I shall look at other benchmarks (start of incubation and nest being empty).