A STUDY OF THE IMPORTANCE OF GARDEN HABITATS FOR TREE SPARROWS

Melvin Butler

(The following is from a personal project undertaken for the University of Birmingham Diploma in Ornithology)

Aim of project
To investigate the use made of gardens by Tree Sparrows Passer montanus by looking at their occurrence in gardens at different times of the year. This would indicate importance for either winter survival and/or breeding success. Local records from north Worcestershire are compared against national data bases.

Introduction
Distribution and behaviour.
There are between 20 and 25 species in the genus Passer with the higher number being based on the phylogenetic species concept. The most widespread species are House Sparrow Passer domesticus and Tree Sparrow. Both are found throughout the Palearctic and Oriental regions and they have also reached other parts of the world, often with the help of man, either by deliberate introduction or accidentally through ship assisted passage.

Both Tree and House Sparrows show a habitat preference for both the built-up environment and the surrounding agricultural land in situations where they do not occur together. "In those parts of the world where the Tree Sparrow is not in competition with its larger and dominant congener, the House Sparrow, it fills the urban role successfully. Yet where the two species overlap, as in Britain and most of continental Europe, the Tree Sparrow is virtually excluded from towns and villages, and occupies instead the rural niches of farmland and lightly wooded countryside. At the northern limit of their range, in Scandinavia, northern Germany and northern Russia, the separation is much less clear and both species tend to occur in towns and villages." (Marchant et al 1990)

In this country, the Tree Sparrow is mainly to be found on farmland. It breeds in loose colonies and individuals remain faithful to that colony for life, although apparently thriving colonies can dwindle and disappear. A number of studies referred to in Birds of the Western Palearctic (Cramp, 1998) describe the feeding and foraging behaviour as "Foraging flocks stay close to cover, with birds flying into cover every few minutes without obvious reason and then drifting back to the feeding area" and "if there is sufficient snow to cover the food plants, flocks break up into smaller groups that scatter to search for food, particularly near human settlement"

Tree Sparrows had high population levels in this country around the late nineteenth and early twentieth centuries, followed by a longer period of decrease and a contraction of the range. This became very marked by 1930, but the late 1950s and early 1960s saw an increase and expansion. Common Birds Census data collected from 1962 saw population levels reach a plateau in 1976/77. A steep decline followed, and the population is now below the level prior to 1950. (Marchant et al 1990). These oscillations in the population are not easily explained. It is suggested that the declines (before 1950) followed a significant period of agricultural decline in the 1930s. There was a lag of approximately 15 or 20 years in the resurgence of agriculture in the 1940s and then an increase in the population of Tree Sparrows in the late 1950s. Numbers grew and remained high during the main period when organochlorides were used (1956 -1963), but numbers began to fall some time after that. According to Marchant (Marchant et aI 1990) the decrease in population is not likely to be related to Dutch Elm disease as Tree Sparrows do not depend on Elms for nest sites. There is an alternative view about this in O'Connor and Shrubb (1986) which says that there may have been an effect on the population of Tree Sparrows caused by the loss of Dutch Elms which reduced the availability of nest sites for a number of species. Such trees provided large cavities suitable particularly for species like Kestrel Falco tinnunculus , Stock Dove Columba oenas and owls, but they also were used by other species like Tree creeper Certhia familiaris and Tree Sparrow. Elms also often were the most mature trees in hedgerows and Tree Sparrows are known to use hedgerows and hedge intersections (Lack, 1992). O'Connor and Shrubb refer to Forestry Commission statistics which show that some 16% of the non-woodland elms in southern England were felled in the mid to late 1970s; any loss of such trees near to urban areas may also have affected the occurrence of Tree Sparrows in gardens. Species including Tree Sparrow "showed a short term decrease on CBC (Common Birds Census) plots affected by the disease." (O'Connor and Shrubb, 1986)

However, the decline in the population may be related to the wide spread use of herbicides in weed control on farmland which reduces the availability of Tree Sparrows' preferred food, that is, smaller seeds.

Gardens and wildlife.
There has been an increasing interest in gardening throughout the twentieth century. More recently this interest has developed to include gardens as important habitats for birds and other wildlife. A general increased awareness of conservation issues has also encouraged the development of more wildlife gardening with more trees and shrubs being planted and significantly more food being put out during the winter. In particular, garden habitats are seen to provide extra natural food, nest sites, and roost sites. There are also a significant number of gardens where supplementary food, nest boxes and roost sites are provided. (Glue and Muirhead, 1991)

Methods
A literature search was carried out which included the Internet.

Requests were made in Spring of 2001 for local records of Tree Sparrows occurring in gardens to the Worcestershire Wildlife Trust and the West Midlands Bird Club using their newsletters.

Data were also obtained from the British Trust for Ornithology's Garden Bird Watch scheme.

Results
Local records -Worcestershire and the West Midlands.
The majority of the published local records for Tree Sparrows refer to known breeding or to the observation of winter flocks. Historical records for Worcestershire area refer to Tree Sparrows being a resident species (Willis Bund, 1891) and as "...common in old orchards where it nests in old tree holes. A few colonies may also be found in pollard willows along river banks..." (Harthan, 1947). A revised list of the birds of Worcester shire by Harthan in 1961 still recorded it as a "...local but generally distributed... " species (Harthan, 1961)

The annual reports from the West Midlands Bird Club throughout 1995-1999 refer to Tree Sparrows as a " Fairly common, though declining resident". Breeding records in Worcestershire again appear in the Annual reports from 1995 to 1999 as follows:
"Some breeding records reported throughout Worcestershire" 1995
"Small numbers bred throughout the County" 1996
"Breeding reported near Bordesley at Rowney Green, Bittell, and Berrow Hill (Redditch), and Wythall" 1997
"Breeding at Brockhill Farm (Hewell Grange area)" 1998
"Breeding at Bittell, Tardebigge" 1999 (note: the picture on the front cover for this report is of a Tree Sparrow).

The occurrence of Tree Sparrows is also reported in local newsletters, for example, "A flock of Tree Sparrows at Lea End (Hopwood) numbered 20, and there was another of unspecified size at Wiggins Hill" (West Midlands Bird Club Bulletin, 2001)

Definite breeding has been recorded throughout the West Midlands according to Harrison (Harrison, 1982). There are records in 99% (76) of the 10km squares in each year between 1966 and1972. Harrison also looked at the regional plots for the Common Birds Census in 1978 which showed 10.8 pairs per sq. km. in woodland, whilst on farmland there were 6.92 pairs per sq. km. compared to 4.76 nationally. More information was available on winter flocks where the commonest size reported was between 50 and150, with the mean size of flock being 164 and the largest being over 1,500 at Blithfield in 1961/62.

Some results on Tree Sparrows from Garden Bird Watch were reported by Mead (1998) to show an increase in the recording of Tree Sparrows in gardens, especially in the Midlands and southwest Scotland. This may be linked, according to Mead, to the increased availability of supplementary food, especially the smaller black sunflower seeds. Tree Sparrows are known to have a preference for the smaller weed seeds in the farmland habitat.

Local and national garden records:
see Table 1 for the local data. Garden Bird Watch data for 2000-01 is not reproduced here.

Location No. Winter Spring Sum Aut Gardentype Food Br
                 
Bordesley 8 Y/O Y/O N N R/M Y N
Upton 1 N N N Y/R R/L Y N
Shenstone 40 Y/F Y/F N N R/M Y Y
Wythall 1 N Y/O N N S/M Y N
Draycott 18 Y/F Y/F Y/F Y/O R/L Y N
Cleobury 7 Y/O Y/F Y/F Y/O R/L Y Y
Belbroughton 7 Y/O Y/O Y/O Y/O R/L Y Y
Knowle 6 Y/F Y/F Y/F Y/F R/L Y Y
Norchard 5 N Y/F Y/F N R/L Y Y
Fairfield ? Y/F Y/F N N R/M Y N
St John 9 N Y/F Y/F N U/L Y Y
Aldridge 8 Y/F Y/F Y/O Y/R R/S Y Y
Shropshire 23 Y/O Y/F Y/F Y/O R/L Y Y


Table 1. Local Garden Records of Tree Sparrows. Key: Y = yes, N = No. Frequency: F = frequent, O = occasional, R = rare. Garden type: U = urban, S = suburban, R = rural, S = small, M = medium, L = large. Sum = summer, Aut = autumn, Br = breeding

Discussion
Recording Tree Sparrows in gardens

The Garden Bird Watch handbook (Cannon, 1998) acknowledges the potential inaccuracies in recording accurately data on similar looking species such as House and Tree Sparrows, particularly by volunteers with a spread of experience. I believe that it is quite probable that Tree Sparrows are under-recorded as garden birds. They are not distinguished by less experienced, and even by some experienced observers (based on personal comments with bird watching friends and neighbours). Tree Sparrows were identified as a separate species in the first century by Alexander the Myndian. However, it was not until the seventeenth century that Willoughby and Ray drew attention to it as a separate species from the House Sparrow. (Summers Smith, 1995). The Garden Bird Watch scheme, however, feel that the size of the sample would still allow any significant variations in the population to be monitored.


The importance of gardens as habitats
While the average size of a garden in this country is only 186 square metres, it has been estimated that the total area that they occupy is at least 485,000 ha. or 3% of the total land area in England and Wales. They often have a large number of plant species and a wide structural variation and many are close to natural and semi-natural habitats such as parks, railway lines, derelict land or local nature reserves. They can therefore be seen to at least have the potential to support a range of wild life. (Good, 2000)

"the conservation significance of private gardens greatly outweighs their absolute value as bird habitat. They have immense potential to educate and engage their users in concepts of species protection and habitat management" (Cannon, 1999). Gardens, although small are not isolated but are part of a continuum of modified habitats. Many birds use gardens not only for the emergency food in hard weather, but also are establishing territories and successfully breeding. This includes "Amber listed" species (of medium conservation concern) such as Blackbirds Turdus rnerula. Some Red listed species (of high conservation concern) also appear regularly in some gardens, for example Song Thrush Turdus philomelos, Bullfinch Pyrrhula pyrrhula, Spotted Flycatcher Muscipa striata and Tree Sparrow (Cannon, 1999). Cannon also supports the view that there is significant value in engaging people in garden bird monitoring programmes, both in terms of the data produced and engaging people in wider conservation matters.

The initial results of some current research on a population of Tree Sparrows at Rutland Water has shown a connection with willows and rushes at wetland edges. The adult Tree Sparrows were found to feed their chicks on the insects which were abundant in this habitat. (Birds, vol. 18, no 8, 2001). These links between breeding colonies and wetland edges supports and explains the description of Tree Sparrows by Harthan "A few colonies may also be found in pollard willows along river banks..." (Harthan, 1947) and by Nicholson (Nicholson, 1951), who says that they are "more garden birds on the continent than they are with us, but they do come freely into gardens and even breed in them, especially where marshes with pollard willows or other specially suitable sites adjoin"

Local garden records (Table 1 ):
Only 13 records were obtained from the requests in the Worcestershire Wildlife Trust and West Midlands Bird Club newsletters. All are for recent (over last two years) or current occurrences. Only 7 of these are for Worcestershire. Although it is too small a sample to analyse statistically, the data do show some interesting features:

All provided supplementary food in their gardens.

Spring is by far the most likely time of year when Tree Sparrows occur in gardens with 12 of the 13 records reporting Tree Sparrows at this time of year, and of these, there were 9 records where they occurred 'frequently'. There were similar levels of reporting for the other periods in the year- 9 in winter, 8 in summer and 7 in the autumn. In choosing the four seasons as categories there may have been some confusion in the returns between, for example, winter and spring. The more detailed records I have for my own garden show more detailed records month by month and that February is the peak month each year that I have observed Tree Sparrows in my garden.

Regarding the type of garden -11 out of the 13 returns classify their gardens as rural and only one each as urban and suburban. Only one return is classified as small, but this also had known breeding nearby.

Due to the habit of colonial breeding, it may have been expected that fewer records of garden breeding would be seen, but four of the records have known garden breeding taking place. A further four records refer to known breeding nearby.
Wide variation in maximum numbers were reported (from 1 to 40) and it is difficult to draw any significant conclusions about the numbers occurring. My own more detailed records for the winter of 1999/2000, however, show that numbers built up from two birds in October to eight birds in February 2000.

BTO Garden Bird Watch data
There is a very low reporting rate for Tree Sparrows in the BTO's Garden Bird Watch scheme and not possible to distinguish data by type or size of garden. For the year 2000 and up to week 13 of 2001, the mean proportion of gardens reporting Tree Sparrows was 0.049 from a mean weekly return of 8752 gardens which means that an average of 429 gardens reported Tree Sparrows each week during that period. However, the peak reporting rate during 2000 (0.057 up to 0.061) occurs in the spring around week 13 (end of March) while the lowest figures (0.034) are in the early autumn at week 37 (mid September).

Gardens clearly provide some important opportunities for Tree Sparrows, especially where supplementary food is supplied during the winter months. This, of course, will not be sufficient on its own, because the breeding population outside gardens may continue to decline. Tree Sparrows do successfully breed in some gardens. It is known, for example, that they readily take to nest boxes (Cramp, 1998) but the importance of nearby wetland edges to breeding colonies as seen at Rutland Water, may be equally important to those Tree Sparrows breeding in gardens. Further investigation is needed to determine the factors involved in the situations where successful breeding is taking place in gardens. The small, but distinct increases in the occurrence of Tree Sparrows in gardens recorded in the Garden Bird Watch scheme and reported by Chris Mead lead him to pose the question at the end of his article "Could the Tree Sparrow become a garden bird conservation success story of the next few years?" (Mead, 1998)


Summary
The importance of gardens as a habitat for Tree Sparrows is examined, both for winter survival and for breeding success. Data on the occurrence of Tree Sparrows in local (north Worcestershire) gardens are compared to national data bases.

In this country, the Tree Sparrow is mainly a farmland bird with a significantly declining population. Historical records show that Tree Sparrows occur throughout Worcestershire. It is known to occur in gardens, especially during winter months and also to occasionally breed in gardens. Only a small number of local garden records were obtained which did not allow a full statistical analysis but did confirm that Tree Sparrows do occur in gardens in north Worcestershire including occasional breeding records. The majority of the local records obtained were from rural gardens and spring was the most likely time for Tree Sparrows to be observed. Supplementary food was provided by all the people who sent in local records

Historical descriptions and recent research at Rutland Water show that successful breeding colonies are close to water or wetland edges which supply insect food for chicks. The information gathered from local records did include known breeding either in the garden or nearby, but did not include a reference to known water. Further study on the proximity of wetland edges to successful garden breeding for Tree Sparrows is required.


Acknowledgements.
I wish to thank the following who sent me records: Tessa Carrick, Carole Wellings, John Humphreys, Jim Martin, D & G Bennett, D Roe, Mike Southall, Robert McCracken, Patrick Taylor, and anonymous of St Johns (Worcester), Draycott-in-the-Clay, and Aldridge.

References
Birds Tree Sparrows -a watery connection. 2001. Birds, vol. 18, no 8, Winter 2001, p63
CANNON, A.1999. The significance of private gardens for bird conservation. Bird Conservation International, 9:287-297.
CRAMP, S.1998 Birds of the Western Palearctic, Oxford University Press.
CANNON, A.1998. Garden Bird Watch Handbook, British Trust for Ornithology.
GLUE, D & MUIRHEAD, L 1991. Garden Bird Studies: Stroud, D. and Glue, D. (eds). Britain's Birds 1989-90: the conservation and monitoring review. National Conservancy Council and British Trust for Ornithology.
GOOD, R. 2000. The value of gardening for wildlife: what contribution does it make to conservation? British Wildlife 12:77-84, December 2000.
HARRISON, G. 1982 The birds of the West Midlands, West Midlands Bird Club.
HARTHAN, A. 1947 The birds of Worcestershire, Littlebury and Company, (not dated)
HARTHAN, A. A revised list of the Worcestershire birds, Trans. Worcester Naturalists Club, 1961
LACK, P 1992. Birds on lowland farms, HMSO.
MARCHANT, J., HUDSON, R., CARTER, S. &WHITTINGTON P. 1990 Population Trends in British Breeding Birds. BTO and NCC.
MEAD, C 1998 Focus on: Tree Sparrow The Bird Table, No 14, Summer 1998, p5
NICHOLSON, E 1951Birds and man, Collins.
O'CONNOR, R AND SHRUBB M 1986. Farming and birds, Cambridge University Press.
SUMMERS-SMITH, J. D. 1995. The Tree Sparrow, Dennis Summers-Smith.
West Midland Bird Club. Annual reports, 1995 -1999.
West Midlands Bird Club Bulletin, No 407, April 2001, p12
WILLIS BUND, J. 1891. A list of the birds of Worcestershire and adjoining counties. W Leicester.

WBRC Home Worcs Record Listing by Issue Worcs Record Listing by Subject


TRAINING IN BIOLOGICAL RECORDING.

As well as the ornithology courses from which these two articles are derived the School for Professional and Continuing Education at University of Birmingham also offers a range of part-time courses associated with biological recording.

These include Certificate, Diploma and Masters Degree courses. They are organised for part time students, and are residential, either at Field Study Centres on the Birmingham University campus.

If you would like further information contact Sarah Whild, The Gateway Education and Arts Centre, Chester Street, Shrewsbury, SY1 1NB. Tel:01743 355137. email S.J.Whild@bham.ac.uk