Subtle differences in birds detected between organic and nonorganic farms in Saskatchewan Prairie Parklands by farm pair and bird functional group

Organic farming may be more beneficial to biodiversity than nonorganic farming but the comparison is often confounded by regional within-farm and landscape differences. We compared breeding bird species composition and abundance on 10 farm pairs of each type matched at the site level for land cover in the prairie parklands of Saskatchewan, Canada in 1990. Land cover was measured around bird point counts at two extents; 'site' (6.3 ha area) and 'field' (16.3 ha area). We pooled species into functional groups; linear mixed models showed no significant differences between farm types for species richness but that all birds, migratory birds using crops and aerial insectivores were more abundant on organic farms. A permutational multivariate analysis of variance demonstrated that farm type did not have a significant overall effect on compositional similarity but that pairwise differences existed between about half of the farm pairs (the direction of differences in beta diversity was not consistent between organic and nonorganic farms according to tests for the homogeneity of multivariate dispersions). Farm-pair differences were more pronounced for all birds and for migratory bird species using crops, migratory birds consuming grains and ground feeders, but not grassland birds. nMDS ordinations suggested that there was more variation in species composition and abundance on organic farms than nonorganic ones but the difference was not significant. Distance-based redundancy analysis (dbRDA) was used to examine the main drivers of bird species composition and abundance and to see which extent was most important; land cover at the field extent was more important than land cover at the site extent or the farming practices measured. The most important field-extent land cover was the amount of native grassland, woodland (including shelterbelts) and wetlands. After controlling for significant field-extent land cover, seed treatment, herbicide use, and number of passes were significant. At the site extent, greater non-crop heterogeneity had a significantly positive effect on abundance and species richness of several groups (e.g., grassland birds, migratory granivores, ground feeders, ground nesters) but a negative effect on richness of woodland birds and abundance of aerial insectivores. Relationships with crop heterogeneity were mostly negative and non-significant. Overall land cover heterogeneity at the site level was positively related to the richness of grassland birds. In contrast at the field extent, non-crop heterogeneity did not have any significant effects on the richness and abundance of any functional group. Crop heterogeneity had a significantly negative effect on aerial insectivore richness and abundance. In the early 1990s, differences in birds between organic and nonorganic farms in Saskatchewan were evident but subtle and variable among farms, and apparently most related to land cover-bird assemblage interactions/relationships.