De. Chamberlain et Rj. Fuller, Contrasting patterns of change in the distribution and abundance of farmland birds in relation to farming system in lowland Britain, GLOBAL EC B, 10(4), 2001, pp. 399-409
Patterns of change in distribution (presence/absence) and abundance since t
he late 1960s were examined in 20 species of farmland bird in southern Brit
ain in predominantly arable (eastern), predominantly mixed (central) and pr
edominantly grassland (western) regions. Comparisons were made between chan
ges in distribution and in abundance to determine whether these measures sh
ow similar relationships to environmental change. Local extinctions of sele
cted species and reductions in species richness were significantly greater
in the predominantly grassland region. Decreases in abundance were greatest
in seven species in the predominantly arable region, two in the mixed regi
on and nine in the grassland region. Changes in distribution and abundance
showed consistent patterns in three species, turtle dove Streptopelia turtu
r L., yellow wagtail Motacilla flava L. and reed bunting Emberiza schoenicl
us L. In another four species, grey partridge Perdix perdix L., lapwing Van
ellus vanellus L., tree sparrow Passer montanus L. and corn bunting Miliari
a calandra L., decreases in abundance were greatest in the arable region, y
et declines in distribution were lowest. For other individual species, chan
ges in distribution were too small to draw any conclusions in relation to f
arm type. We suggest that modern grassland systems are suboptimal habitats
compared to arable or mixed agricultural land for many farmland species tha
t occur at relatively low density in the more western, grass-dominated regi
on. Declines in abundance are therefore more likely to lead to local extinc
tion in these areas than in eastern areas where abundance is higher. Howeve
r, the role of changes in grassland management on bird populations requires
further research. It is suggested that conclusions drawn from changes in d
istribution alone, in the absence of supporting data on changes in abundanc
e, may be misleading where the aim is to assess how large-scale spatial dyn
amics of populations relate to environmental change.