A temporary drop in first-year survival coincided with the period of steepest decline, but changes in breeding performance, especially reduced nest failure rates at the chick stage, appear to have driven a levelling-off in the long-term population trend (Freeman & Crick 2002). Over the period 1968-2009, brood size has decreased (see above) but there has also been a decrease in nest failure rates at the egg and chick stage, so the number of fledglings per breeding attempt has shown a net increase. Further evidence for the role of changing survival in House Sparrow declines has been provided by Hole et al. (2002), who found no evidence of significant differences in most breeding-ecology parameters in declining and stable populations in a farm-scale comparison, while Siriwardena et al. (1999) found that national survival rates were lower during the period of decline in the CBC index. Crick & Siriwardena (2002) used NRS analysis to show that breeding performance per nesting attempt has increased and was positively correlated with population growth rate in the wider countryside (although there was no such correlation in gardens).
There appear to be different processes affecting urban and agricultural populations. On farmland, changes in farming practices due to intensification of agriculture and the tidying of farmyards have reduced the winter seed available to farmland populations of House Sparrows, which has resulted in a reduction in survival rates (Siriwardena et al. 1999, Chamberlain et al. 2007, Hole 2001), specifically of first-year birds (Crick et al. 2002). This is supported by a positive effect of supplementary seed in winter on farmland House Sparrow population trends in a landscape-scale experiment in East Anglia (Siriwardena et al. 2007). House Sparrows have probably been deleteriously affected by the decrease in the amount of grain spilt around farm buildings and during the process of harvesting in recent years (O'Connor & Shrubb 1986). The decrease in spring-sown cereals has meant that cereal stubble has become much rarer, reducing food resources over winter, although Robinson et al. (2001) found no influence of spring cereal on House Sparrow abundance in predominantly pastoral farmland. Conversely, breeding performance is worse where there is more spring cereal (Crick & Siriwardena 2002), although this may reflect geographical associations with areas where spring sowing remains in the UK (the west and north) rather than direct effects of cropping.
Recent declines have been particularly severe in urban areas (Robinson et al. 2005b, Chamberlain et al. 2007). Increased predation by cats and Sparrowhawks, lack of nest sites, loss of food supplies, pollution and disease have all been cited as factors possibly depressing populations in towns (Crick et al. 2002), but supporting evidence for these is mixed. Within urban areas, Shaw et al. (2008) reviewed available evidence and hypothesised that House Sparrows have disappeared from more affluent areas, where changes to habitat structure such as planting of ornamental shrubs and increased demand for off-street parking is likely to reduce the amount of habitat available to House Sparrows and influenced foraging and predation risk. The conversion of private gardens to continuous housing has also had a negative effect on House Sparrow abundance (Chamberlain et al. 2007). Vincent (2005) found that annual productivity among suburban and rural human habitation in Leicestershire was lower than that measured on farmland House Sparrows in Oxfordshire, the main cause of the difference being starvation of chicks. Low body masses at fledging, and consequently low post-fledging survival, were also recorded in Leicestershire. Although only a two-year study, Peach et al. (2008) measured reproductive success in a declining House Sparrow population along an urbanisation gradient in Leicester and also found that a year in which reproductive success was too low to sustain the population was characterised by lower chick survival and body mass at fledging (a predictor of post-fledging survival). However, there is no direct evidence that invertebrate food supplies have declined in these areas and variation in survival has not been investigated.
Negative correlations between indices of Sparrowhawk presence during its recolonisation of the UK and House Sparrow abundance from the Garden Bird Feeding Survey have been interpreted as evidence that increasing predation rates are depressing House Sparrow populations (Bell et al. 2010). However, more sophisticated analyses of large-scale and extensive national monitoring data provide no evidence that House Sparrow population declines were linked to increases in Sparrowhawks (Newson et al. 2010b).
