Executive summary
1. In response to rapidly declining populations of Starling and House Sparrow, the Department for Environment, Food and Rural Affairs (DEFRA) commissioned this project with the key aim of investigating the trends in abundance, breeding performance and survival of Starlings and House Sparrows in Great Britain in relation to environmental factors and legal control activity. Another aim was to provide clear guidance to DEFRA on the likely causes of the declines and identify further research that would aid the understanding of these causes.
2. To meet these aims an analysis of long-term national datasets held by the BTO and analysis of two long-term intensive studies, undertaken by University of Oxford, RSPB, CSL and WildWings Bird Management were undertaken. Questionnaire surveys of farm owners and occupiers and Local Authorities on the extent of legal control activity were organised by CSL and DEFRA, respectively.
3. Information on Starling and House Sparrow at the national scale came from the following long-term surveys: Common Birds Census and BTO/JNCC/RSPB Breeding Bird Survey - breeding abundance; Garden Bird Feeding Survey and Garden BirdWatch - wintering abundance and seasonality of garden usage; Nest Record Scheme - breeding performance; and National Ringing Scheme - survival rates.
4. The British breeding population of Starling is approximately 8.5 million birds. Farmland supports 30% of the British breeding population; urban/suburban habitats and rural gardens support 57% of the breeding population. The greatest declines (92%) since 1965 have been observed in woodland, but this probably represents sub-optimal habitat. Declines on farmland average 66%, being particularly associated with livestock-based systems.
5. A decline in numbers recorded in gardens during winter may reflect a decline in breeding numbers but is confounded by the presence of a large number of immigrant Starlings. The number of birds using rural gardens has not shown a marked long-term decline, but rather appears to be related to climatic conditions.
6. Starling breeding performance has improved since 1962. Breeding performance was highest in the South West and West of Britain. It has improved most rapidly in the West, North and East of Britain and shown least improvement in the South East, where densities are currently greatest but where declines have been steepest, both historically and currently.
7. An intensive study of breeding performance in nestbox-breeding Starlings in a suburban area of south-east England indicated no changes in breeding productivity per pair despite an abrupt decline in numbers in the early 1990s. Declines in the number of adults and young that returned to breed in the colony and in the number of years in which they bred there suggested that survival rates had declined or that there was a net emigration from the colony. There were no clear environmental correlates of the decline.
8. The survival rate of Starlings in their first year of life is 15% and thereafter is 69% per annum. Changes in first-year survival are likely to have driven population changes of Starlings in the wider countryside. Regional analyses suggested that while this held for Northern and Western Britain, population changes in Eastern Britain were more likely to reflect changes in adult survival; possibly because of differences in habitat availability.
9. There have been a large number of changes on farmland, particularly in pastoral areas, which may have affected Starling populations. These are likely to have reduced the number of foraging opportunities and the prey density. Further work is particularly needed to determine the food availability and resource use of urban and suburban Starlings.
10. A series of recommendations for further research follow from this work. These include the organisation of targeted ringing to contrast juvenile dispersal and habitat use in Eastern and Western Britain, studies to quantify the importance and quality of urban and suburban foraging habitats, and the use of behaviourally-based models to predict the effects of farm management on food supplies and hence population size.
11. We suggest that it may be possible for agri-environment schemes to encourage the inclusion of pockets of sympathetically managed grassland habitat in arable areas and to promote the management of grassland for the production of short, sparse swards, to help reverse the Starling decline. The provision of extra nest sites on farmland may be beneficial but its efficacy would need careful monitoring because of density-dependent processes.
12. The British breeding population of House Sparrows is between six and seven million pairs, having fallen from about 12 million pairs in the early 1970s. Just over half of British House Sparrows occur in southern and central eastern England. About 60% of House Sparrows occur in towns, villages and rural gardens, with about half of these in suburban areas. Because of evidence for fine-scale patterning of population densities in urban and suburban habitats, there is a need to undertake a national survey to investigate this in relation to the fine-scale environmental characteristics of these habitats.
13. Since the 1970s, House Sparrow populations have declined most in South East England but are currently increasing in Scotland and Wales. Population declines began earlier in farmland than in gardens and in towns, but declines have been greatest in suburban and urban gardens. Rural gardens may be the most favoured habitat for the species.
14. The pattern of occurrence of House Sparrows in gardens has changed over the last thirty years. This change is consistent with a decrease in food resources in the wider countryside.
15. Dispersal distances are very limited for House Sparrow, so the main demographic processes behind population declines are some combination of changes in productivity and survival.
16. Breeding performance of House Sparrows per nesting attempt has increased over the past 40 years in all regions. Breeding performance does not differ between farmland habitats, but is higher on farmland than in urban and suburban habitats. Increases in breeding performance have been least in South East England where populations have declined most rapidly and most rapid in the North and West, where some populations have increased.
17. Breeding performance may have had a role in determining the pattern of population changes since 1975 in the wider countryside, but not suburban gardens. High nest failure rates and a lack of any improvement in breeding performance in suburban areas, where populations are declining most rapidly, might be reinforcing the population declines there.
18. Data from a recent intensive study of House Sparrows on Oxfordshire farms was analysed in comparison with similar data from the 1960s. This showed there were no differences in key aspects of breeding performance between the periods and concluded that a reduction in survival rates was probably the demographic mechanism driving population decline.
19. The population decline of House Sparrows in the 1970s appears to have been caused by a decline in first-year survival rates. The decline was halted by a combination of increased breeding performance, most likely through declines in failure rates of nests at the chick stage, and increases in first-year survival rates. Recovery of the population appears to be inhibited by a decline in adult and (probably) in post-fledging survival rates.
20. Arable farmland now appears to be a least favoured habitat for House Sparrows in Britain, despite their previous pest status. Changes in agricultural practice have probably decreased the availability of autumn and winter seed supplies, for all granivores, but House Sparrows are likely to have been strongly affected by stricter hygiene rules governing seed storage.
21. Rural dwellings still support high densities of House Sparrows and studies of the seasonal use of available habitats by House Sparrows around rural gardens would be valuable to understand the factors that promote high population densities. The influence of pastoral land management and livestock husbandry on food availability for nesting House Sparrows is an area which requires more intensive research as declines have been much less marked on pastoral farmland. The effects of reduced immigration from garden populations cannot be definitively ruled out as a factor in the decline of House Sparrows on farmland.
22. Recommendations for further work to elucidate the factors behind the population declines of House Sparrows include: comparative ecological and demographic studies in regions and habitats of contrasting current population change, particularly of breeding performance, but also involving the use of colour-marking studies to determine the patterns of habitat usage in space and time and seasonal patterns of mortality.
23. The strongly contrasting fortunes of House Sparrows in different urban centres urgently needs investigation by comparative studies and surveys of House Sparrow populations. Potential factors that could have led to the declines of House Sparrows in urban situations include predation (by domestic cats and Sparrowhawks); loss of sources of weed seeds (through loss of brown field sites); changes in the suitability of food provided at garden feeding stations; air quality (pollution); disease transmission (via bird feeding stations). Although House Sparrows are not limited to nesting in cavities, evidence from a survey in Bristol suggests that loss of nest sites in urban areas may be important, thus the benefits of nest site provision should be explored experimentally.
24. Although there is a need for more targeted studies to allow the identification of specific factors that can be managed to reverse the declines of House Sparrow in urban and suburban areas, it is highly likely that stricter hygiene regulations have decreased food availability in farmyards. Thus, the provision of tailings as a supplementary food source is likely to enhance House Sparrow populations at minimal cost, although methods will need to be devised to avoid problems from rodent infestation.
25. Questionnaire surveys of farm owners and occupiers and of Local Authorities provided essential information on the extent of current legal control activities. The vast majority of owner/occupier respondents (97%) and all Local Authority respondents, undertook no lethal control. From the results it is reasonable to estimate that around 74,000 Starlings and 16,000 House Sparrows were killed in the survey year (2001). Numbers estimated to have been killed were similar to those reported for the periods one to five years ago and six to ten years ago. Estimates of the number killed more than ten years ago were higher.
26. In comparison with the national population estimates, calculated as part of this project, the numbers culled each year are small. For Starling, assuming a population of 8.5 million birds, approximately 0.8% are culled per year, at current levels. For House Sparrow, assuming a population of 13 million birds, approximately 0.1% are culled per year. The impact on the national House Sparrow and Starling populations is considered to be negligible. Lethal control was considered, in most cases, to be ineffective in eliminating problems in comparison to some of the non-lethal control measures.
27. Neither species currently appears to pose a widespread or serious pest problem; only 3% of owner/occupiers reported Starling damage so severe that farm income was affected (0.5% for House Sparrows), and 5% of responding Local Authorities reported Starling problems severe enough to affect budgets. No Local Authorities had problems with House Sparrows.
28. In conclusion, this report has investigated long-term extensive national and intensive local population datasets to identify the key demographic mechanisms behind the declines of Starling and House Sparrow. It provides recommendations for several key areas of research that are needed to elucidate the environmental factors that could be manipulated most effectively to help reverse the declines and it makes suggestions for conservation actions that are likely to increase the abundance of these species.
Updated 1 August 2002
Return to Wildlife and Countryside Research Programme Index
Return to Wildlife and Countryside Index
Return to DEFRA Home Page