A minority of FBI species had (mostly weakly) significantly higher densities on organic farms, although the majority of species' densities were greater on organic farms and there was significantly greater total density on organic farms. Only lapwing Vanellus vanellus (in 2001) showed a (non-significant) trend towards higher density on conventional farms. The information-theoretic approach suggested that both habitat extent and FTYP were important predictors for starling and greenfinch, but otherwise there was no evidence that FTYP was an important predictor when accounting for other habitat variables.
Habitat structure is a principal correlate of spatial variation in bird abundance on farmland (Fuller et al. 1997
). Norton et al. (2009)
found a greater extent of non-crop habitats and more heterogeneous land use in a sample of organic farms in the UK (of which this study is a subset). Variation in structural habitat is likely to be a key factor in explaining the organic–conventional contrast in birds. For example, variables that are known to systematically vary between farming systems (Norton et al. 2009
) and which were strongly linked to winter bird density included hedgerow density (AFBI
), proportion of arable area at the farm scale (stock dove, jackdaw) and grass : arable ratio at the landscape scale (woodpigeon, jackdaw), although there were some models where no consistent effects of any variable were found (table S2 in electronic supplementary material). Similar effects of cropping patterns and landscape complexity, coupled with relatively weak effects of organic management, have been found for breeding birds (Piha et al. 2007
; Kragten & de Snoo 2008
). Furthermore, landscape structural heterogeneity is a key component of overall system differences for invertebrates (Schmidt et al. 2005
; Rundlof & Schmidt 2006
). Similarly, the results presented here suggest that the ‘physical’ habitat of farmland is likely to explain much variation in winter bird abundance between systems.
Availability of food resources in winter is likely to be a key limiting factor for many FBI species, especially granivorous passerines, and their decline is strongly linked to loss of key foraging habitats such as stubbles (e.g. Gillings et al. 2005
). Our observations suggest that this group of species may not benefit (in winter) from wider adoption of organic farming practices. The general lack of farm-type differences could be strongly influenced by stubble availability. Although higher arable weed abundance on organic farms (Fuller et al. 2005
) may be expected to increase winter food resources, at the time of the study stubbles were more prevalent on conventional farms (Norton et al. 2009
), as organic farmers cannot afford the resulting weed burden. The majority of species identified in this paper likely to most benefit from organic farming practices in winter were increasing species and as such are not of conservation priority. Starling was the only decreasing species (Gregory et al. 2004
) to show strong effects of FTYP when accounting for broad-scale habitat variation, possibly owing to better foraging provided by organic grass management (temporary grass leys and application of farmyard manure).
We conclude that variation in broad-habitat extent is a better predictor of bird abundance and richness than FTYP per se
. As well as lack of pesticide and synthetic fertilizer use, organic farms differ from conventional farms in terms of a range of habitat variables and management practices (Norton et al. 2009
), which vary in the extent to which they could be considered intrinsic to the system. Organic farming has clear benefits for a range of taxa (e.g. Hole et al. 2005
), but some aspects of organic farming may not currently provide significant benefits to bird species that are limited by winter seed resources. However, a reduction in stubbles on non-organic farmland, as has recently occurred with the phasing-out of set-aside, could result in organic farms becoming more heavily used by some granivorous species.