Similarities in vegetation development of newly established herbaceous strips along contrasting European field boundaries

Abstract

In France, the Netherlands and the UK, arable field boundaries were extended with four metres of crop edge. Plots with perennial, herbaceous vegetation were established by natural regeneration or by sowing grass, both annually cut. Vegetation development was monitored in the first three years after establishment in the original boundary and in the new boundary strip. Species composition, species-richness and biomass production in the new strip were related to those in the original boundary to examine the potential of predicting vegetation development in extended field boundaries from the vegetation composition of the original boundary. Within three years species-richness, biomass production and monocot:dicot ratio in both the grass and regeneration plots converged to the levels of the original boundary in all three countries. Species composition in the new strip was not closely related to the original boundary, however, since only 20–50% of the species encountered in the original boundary at the onset of the experiment had managed to colonise the new boundary strip in the final year. The low similarity was mainly due to low establishment rates of annual and woodland species in the new strip while mobile, perennial grassland species were generally very successful. Sowing grass, which may be preferable with respect to weed control, had adverse effects on species-richness. These relationships, which were found to apply in a broad geographical area and in different boundary types, may aid efforts to restore botanical diversity in arable field boundaries.