Australian Raven Breeding in the COG AOI
This is another in the series of posts mulling over analysis of interesting data from the Canberra Ornithologists Group (GOG) Garden Bird Survey. One hopes that the analysis is also interesting!
Background
The original catalyst for looking at these data came about from a question raised a reader of the COG publication Birds of Canberra Gardens. For Australian Raven this book commented that "Breeding records also increased over the length of the Survey." Did this comment still apply after a further 10 years of data was available?
An early look at the data suggested the answer to the question was "yes" but showed a perplexing change in the data for the period 2004 - 2007. This note examines that change, bringing in data from the COG Area of Interest (AOI) as a whole to supplement that in the GBS.
Initial Analysis
The initial analyses are shown in the chart below, which is hopefully able to be interpreted at blog-font size.
Whether the three very high years are included or not the number of breeding records shows an upward trend - with a reasonably high value of the correlation coefficient - over the 27 years. As would be expected, the series including the high years shows a faster rate of increase. It must be noted that the definition of 'breeding record' is one or more breeding observations in a site in a year. Thus a higher number means more sites recorded breeding a year and not necessarily that sites made more observations.
A first thought was that the high years followed immediately after the 2003 Canberra bushfires, and this will be discussed further below.
Other data
The Atlas of Birds of the ACT does not include a map of the breeding records of this species. It does note that the species is not suited to the heavily forested areas (such as the gullies in the bushfire affected ranges). It does note that birds of this species maintain a permanent breeding territory - making the change for 3 years seem even more unusual.
Both Atlases of Australian birds do map these records. The first Atlas (based on observations in 1977 - 81) shows breeding records from all 1 degree grid squares in SE Australia. Again reference is made to the maintenance of a long term breeding territory. The second Atlas (observations in 1998 -2002) has a 'gap' in breeding records (but not observation records) covering approximately the high country of Victoria and SE New South Wales.
The Handbook of Australian New Zealand and Antarctic Birds (HANZAB) makes a number of references to the species requiring trees in their preferred habitat. Such habitat includes dry open sclerophyll woodlands - possibly similar to that of much of the burnt out areas of the Brindabellas. Again HANZAB refers to the behaviour of the adult birds "Adult breeding pairs sedentary, occupying permanent territories of up to 120 Ha year-round.... "
COG General Records
The Annual Bird Reports published by COG include some information about the number of breeding records submitted each year for Australian Raven. They are summarised in the graph in this section. Note that in this case a "record" is of a single observation, similar to what I refer to as an observation in the GBS. Again there is a dramatic rise in more recent years, but a little later than for the GBS records. However the picture from the general records broadly supports that from the GBS.
For most years there is only limited detail provided about the general observations, but where details are provided it appears that the majority of the records come from the urban areas. Thus these data do not:
Background
The original catalyst for looking at these data came about from a question raised a reader of the COG publication Birds of Canberra Gardens. For Australian Raven this book commented that "Breeding records also increased over the length of the Survey." Did this comment still apply after a further 10 years of data was available?
An early look at the data suggested the answer to the question was "yes" but showed a perplexing change in the data for the period 2004 - 2007. This note examines that change, bringing in data from the COG Area of Interest (AOI) as a whole to supplement that in the GBS.
Initial Analysis
The initial analyses are shown in the chart below, which is hopefully able to be interpreted at blog-font size.
Whether the three very high years are included or not the number of breeding records shows an upward trend - with a reasonably high value of the correlation coefficient - over the 27 years. As would be expected, the series including the high years shows a faster rate of increase. It must be noted that the definition of 'breeding record' is one or more breeding observations in a site in a year. Thus a higher number means more sites recorded breeding a year and not necessarily that sites made more observations.
A first thought was that the high years followed immediately after the 2003 Canberra bushfires, and this will be discussed further below.Other dataThe Atlas of Birds of the ACT does not include a map of the breeding records of this species. It does note that the species is not suited to the heavily forested areas (such as the gullies in the bushfire affected ranges). It does note that birds of this species maintain a permanent breeding territory - making the change for 3 years seem even more unusual.
Both Atlases of Australian birds do map these records. The first Atlas (based on observations in 1977 - 81) shows breeding records from all 1 degree grid squares in SE Australia. Again reference is made to the maintenance of a long term breeding territory. The second Atlas (observations in 1998 -2002) has a 'gap' in breeding records (but not observation records) covering approximately the high country of Victoria and SE New South Wales.
The Handbook of Australian New Zealand and Antarctic Birds (HANZAB) makes a number of references to the species requiring trees in their preferred habitat. Such habitat includes dry open sclerophyll woodlands - possibly similar to that of much of the burnt out areas of the Brindabellas. Again HANZAB refers to the behaviour of the adult birds "Adult breeding pairs sedentary, occupying permanent territories of up to 120 Ha year-round.... "
COG General Records
The Annual Bird Reports published by COG include some information about the number of breeding records submitted each year for Australian Raven. They are summarised in the graph in this section. Note that in this case a "record" is of a single observation, similar to what I refer to as an observation in the GBS. Again there is a dramatic rise in more recent years, but a little later than for the GBS records. However the picture from the general records broadly supports that from the GBS.
For most years there is only limited detail provided about the general observations, but where details are provided it appears that the majority of the records come from the urban areas. Thus these data do not:
- support the hypothesis that the rise in GBS records is a result of breeding territories in the mountains being burnt out in the 2003 bushfire's; nor
- offer any alternative hypothesis.
Comments
A Ravens are certainly on the increase as a breeder in suburbia as I found out during my 2008 Ainslie survey. 30 years ago there may have been a couple of pairs, now around 8 (final figures as yet not worked out). In the past they were found only in the larger parks/reserves within suburbia, and there in the tallest trees. Now the nest also in small reserves and suburban gardens, provided there is a tall tree.
I also assume that young raised in suburbia may also settle there preferentially. I also have noticed regularly a gang of non-breeders in CIVIC.
I think the initial colonisation of new habitat may take a good number of years, and then it escalates quite quickly. I have read a long time ago, possibly in the BOCa journal, of similar stories with the Little Raven (?) or a crow in other Australian cities.
And they are now regularly recorded in and around my GBS site at any time of year.