Vegetation development of sown grass margins in arable fields under a countrywide agri-environment scheme

Abstract

Farmers were paid to establish 6 m wide sown grass strips in arable field margins under the Countryside Stewardship Scheme in England, UK. One hundred and sixteen sites in eight regions were surveyed to determine if grass margins had been established successfully and the extent to which they were colonised by forbs. Sown margins had more grass and fewer weed species than naturally regenerated sites. Grass margins contrasted with normally cropped sites, having greater species richness of grasses, forbs and perennials and more bird, butterfly larva and bumblebee foodplants. Mesotrophic grassland forbs were scarce in margins established from basic grass seed mixtures but significantly more abundant if included in the seed mixture. Annuals were more prevalent in sites up to 2 years old but species composition was not related to age in sites over 2 years old. Variation partitioning showed that overall species composition was related to seed mixture type, region and soil properties but there was little overlap in the variation explained by these environmental variable sets. Habitat context and management practices did not explain any variation in species composition. Perennial grassy vegetation was established successfully using basic grass seed mixes but only competitive species colonised subsequently. Diverse seed mixtures containing mesotrophic grassland forbs merit support in agri-environment schemes because they do enhance the botanical diversity of sown grass margins.

Introduction

Field boundaries are important linear habitats in lowland arable farmland landscapes. The boundary usually comprises a hedge, trees, ditch, bank or other semi-natural feature but can also include a perennial grass or herbaceous verge next to the crop. The verge can serve many functions, including provision of valuable habitat for flora and fauna and a buffer zone that protects the remainder of the boundary from fertiliser and pesticides applied to the crop (Marshall and Moonen, 2002, Meek et al., 2002). The perennial vegetation also prevents annual weeds from establishing in the boundary and spreading to the crop (Smith et al., 1999b, Moonen and Marshall, 2001). However, in the UK and Western Europe, many field boundaries have been removed as field sizes increased to accommodate modern farm machinery. Contamination of verges by fertiliser and pesticides has also reduced their plant species diversity (Marshall and Moonen, 2002).

To counteract these losses, research has focussed on methods of establishing perennial margins between the remaining field boundary and the crop. Experiments have shown that sown grass species can establish successfully. Species that form a dense, tussocky sward provide habitat for invertebrates and other fauna (Collins et al., 2003). Annual species are usually suppressed after the first year (Smith and MacDonald, 1989, Huusela-Veistola and Vasarainen, 2000, Asteraki et al., 2004) and the risk of weeds spreading from new margins into the crop is low (Smith et al., 1999b). Grass margins have been established successfully in this way in an experimental agri-environment scheme (Critchley et al., 2004). The development of species-rich vegetation is, however, often inhibited by a paucity of colonisation sources and the low dispersal capability of many perennial grassland species (van Dorp et al., 1997, Kleijn et al., 1998). A dense sward also inhibits establishment of additional plant species. Experiments have also therefore focussed on sowing more diverse seed mixtures that include perennial forb species (Smith et al., 1999a, Asteraki et al., 2004, de Cauwer et al., 2005). These have been generally successful in establishing more species-rich perennial vegetation although the mowing regime can affect establishment rates (Smith et al., 1999a, de Cauwer et al., 2005). Margins with perennial forbs can increase arthropod diversity (Thomas and Marshall, 1999) and provide nectar and pollen food sources (Feber et al., 1996, Carvell et al., 2004). Most experiments have been short term (5 years or less) although Bokenstrand et al. (2004) found that the majority of meadow species from a diverse seed mixture were still present after 10 years. However, it remains to be confirmed whether results from these experiments would be replicated if the methods were adopted by farmers at a national scale and over longer timescales.

Since 1993, farmers in England, UK were encouraged to establish grass margins under the Countryside Stewardship Scheme (CSS). In return for payments, farmers were required to establish and maintain a grass strip averaging at least 6 m wide (range 4–12 m) alongside a cropped field. Two establishment options were available to farmers, using either a limited selection of grass species from a defined list or a more diverse mixture that includes a range of grasses and forbs (Defra, 2003). By November 2004, 24,000 ha of grass margins had been established in this way under the CSS. This represents a substantial investment of resources and it is important therefore to determine whether the scheme had been successful in establishing the intended vegetation and maintaining it in the longer-term.

The aims of this project were to assess the plant species composition of 6 m CSS grass margins of varying age and margins sown with additional forbs. Specifically, the following questions were addressed:

• Does the intended perennial vegetation establish in grass margins?

• Do sown perennial forbs establish and persist in grass margins?

• Do perennial forbs colonise margins sown with basic grass seed mixtures?

• How does species composition of sown grass margins vary with region, age, soil properties, management and habitat context?