Search

Questions: Due to their high ecological and agronomical variability, borderland regions offer an excellent opportunity to study assembly patterns. In this study we compared the influence of various factors on summer annual weed communities consisting of both native and introduced species. Location: The borderland region of Austria and Hungary. Methods: We assessed the abundance of weed species in 300 fields of six summer annual crops, and collected information on 26 background variables for each plot. We applied redundancy analysis (RDA) to estimate multivariate species responses and variation partitioning to compare the relative importance of three groups of variables (environmental variables, management variables, and country as a singleton group), and we also checked for statistical association between country and the predictors of the other two groups. Results: The full RDA model explained 22.02% of the variance in weed species composition. Variation partitioning showed that environment and management had similarly high (~8%) influence on weeds, while country had a modest yet substantial (~1%) effect, and there was relatively little overlap between the variance attributable to the three groups. Comparing the individual variables, country ranked third (after preceding crop, and actual crop). The effects of 15 further variables were also significant, including seven management, and seven environmental variables, as well as the location of the sampling plots within the fields. Comparisons between the countries showed that farming type, preceding crops, tillage system, tillage depth and field size were significantly different between the countries. Conclusions: Country exhibited a small but significant influence on weed community composition, which could not be explained with easily accessible management and environmental variables. This suggests that the distinct historical agronomical background of the two countries, possibly involving some legacies of the former Iron Curtain period, still has an impact on the weed species composition of arable fields. [ABSTRACT FROM AUTHOR]

Floristic composition and conservation value of the weed vegetation of winter cereal fields on base-rich soils in western Hungary was compared among three management systems: (i) small, extensively managed fields; (ii) small, intensively managed fields and (iii) large, intensively managed fields. Data were analysed by principal components analysis (PCA) and redundancy analysis (RDA). Most variation in species composition was explained by the differences between the three field types, particularly by the difference between the extensively managed fields and the two types of intensively managed fields and by crop cover. There were several species positively associated with small extensively managed fields, but no species associated with either of the two types of intensively managed fields. Logistic regression was made of the occurrence of species groups important for nature conservation on the first RDA axis, representing the distinction between the extensively and intensively managed fields. This showed that Red List and insect-pollinated plant species occurred more frequently in small extensively managed fields, but there was no difference between fields in the frequency of bird seed-food species. [ABSTRACT FROM AUTHOR]