Forage and amenity grasses

Cropping practice and use in Denmark

Denmark is the world’s largest exporter of grass seed, and more than 40% of the total grass seed production in the 25 EU countries is located in Denmark. More than half the grass seed production comprises amenity types within the three largest species produced: perennial ryegrass (Lolium perenne), red fescue (Festuca rubra) and smooth stalked meadow-grass (Poa pratensis) and almost the total production of common and creeping bent seed ₍Agrostis capilaris&A. stolonifera var.

palustris₎are for amenity purposes.

Grass seed production is concentrated in areas with favourable climatic conditions and an

infrastructure that is favourable to seed production. Since 2002, the area has increased by more than 20,000 ha, and this development is expected to continue, with a forecasted grass seed area of 100,000 ha in the near future. The production of grass seed for recreational areas (golf courses, sports grounds, parks and lawns) is increasing.

All grass seed production in Denmark is contracted and is carried out according to the Executive order on field seeds⁹ and is thus subject to a range of restrictions in the form of separation distances and cropping intervals.

Crop area, Denmark, 2002 and 2006 (Rounded figures)

Year of cultivation 2002 2006

Conventionally cultivated grazing fields in crop rotation 189,000 ha 224,100 ha Conventionally cultivated permanent grassland 137,000 ha 208,000 ha Conventionally cultivated set-aside areas with grass 192,000 ha 157,200 ha Conventional grass seed production 63,000 ha 85,300 ha Conventional production in total 581,000 ha 674,600 ha Organically cultivated grazing fields 34,000 ha 55,500 ha Organically cultivated permanent grassland 20,000 ha 17,100 ha Organically cultivated set-aside areas with grass 4,000 ha 3,500 ha Organic grass seed production 1,600 ha 2,500 ha Organic production in total 59,600 ha 78,600 ha Grass area in total 640,000 ha 753,000 ha Sources: Kristensen, 2007; The Danish Plant Directorate, 2006a.

In 2006, grass areas constitute a total of approx. 27% of the total crop area, and this area comprises both grazing fields and pastures. A steadily increasing share of “the grazing fields” is used for silage or a combination of silage and grazing. For the major part of the areas, a first or second cutting is made and subsequently the areas are used for grazing for the rest of the growing season.

Approx. 10% of the grass area is organic. It is common that organic farms have both seed production and grazing fields. From 2002 to 2006 the cultivated grass area increased by approx.

18%, which is partly due to the fact that also small grass areas are now registered on application for

9 Danish Plant Directorate Executive Order on field seeds, no. 52 of 24 January 2000 with later amendments

EU subsidies (the single payment scheme). Grass is very common in uncultivated areas (field boundaries, road and wood edges, etc.).

Experience with GM grasses in Denmark

There is no experience with production of GM grasses in Denmark, but the breeding company DLF-TRIFOLIUM A/S, Dansk Planteforædling has in 2006 been granted permission for trial releases with genetically modified ryegrass with increased content of carbohydrates.

Experience with GM grasses in foreign countries

As yet, no countries allow the cultivation of GM grasses. In the EU, 3 trial releases in grass are registered, 1 in ryegrass (changed lignin biosynthesis) and 2 in tall fescue (changed lignin biosynthesis and glyphosinate tolerance). In the USA, an application has been submitted for a glyphosate tolerant creeping bent. There, experience is available from a 450 ha monitored area, in which seeds from the GM grass were propagated in 2003. Restrictions had been set out for the monitored area, such as isolation distance, requirements concerning cleaning of machinery, etc.

Creeping bent is in that area harvested after previously being laid in swaths, but in 2003 part of the plant material left in swaths (including seeds) was dispersed by the wind. Several studies have subsequently shown that there is now GM bent outside the fields in which the seed was originally propagated and outside the actually monitored area (Reichman et al., 2006; Watrud et al., 2004;

Zapiola, 2007). At present, a PhD project is being carried out, which includes a determination of whether plants from seeds sampled from non-GM bent inside and outside the monitored area are GM. That is, the study can confirm or deny whether GM pollen is still being dispersed. Analyses of small plants sampled from non-GM plants inside and outside the monitored area during the years 2003, 2004 and 2005 show low GM presence.

The application for approval of the GM bent is still being considered by the American authorities.

Sources of dispersal

The commercially used grass species are mainly cross-pollinated and produce a large amount of pollen, which can be dispersed over large distances with the wind. Pollen dispersal can occur between cultivated grass areas for various purposes (grazing fields – seed fields) and between uncultivated areas (perennial grass, set-aside areas, road edges, etc.) and seed fields.

Most cultivated grass species have a large tendency towards seed shedding. Normally, the seeds have a short life (1-3 years), but germinated volunteers can be propagated in the crop rotation (false crop rotation) due to insufficient control of volunteers.

New knowledge

New studies of pollen dispersal were carried out in tall fescue (Wang et al., 2004 a & b), in which gene dispersal frequencies of 5% were found at a distance of 50 m from the donor, approx. 4% at 100 m and approx. 1% at 150 m. No GM grasses were found at a distance of 200 m from the donor, and the study concluded that dispersal of GM grasses can be controlled. In contrast to this

conclusion, it has been shown that GM bent has been dispersed and has established at a distance of approx. 4 km from the growing area (Reichman et al., 2006), and major monitoring and

characterisation of the extent of this dispersal is being carried out (Watrud et al., 2004).

Several publications have appeared concerning the ”dispersal biology” in grasses, for example expressed through ”seed production per plant”, ”floret exploitation” and ”relative fertility” (Cunliffe

et al., 2004). The results of this study confirm a decreasing dispersal for all the three parameters mentioned with increasing distance from the donor (the largest distance from the donor was 140 m).

Furthermore, a study was published, which included the time of flowering in relation to gene dispersal in perennial ryegrass (van Treuren, 2006). A large dispersal in the extent of flowering time among individual plants was recorded, but this is of limited – or no – significance to the actual gene dispersal.

To sum up, it should be pointed out that by and large it is agreed that pollen dispersal in grasses usually is moderate at distances of approx. 200 m from the donor but that climatic and growth irregularities can lead to an extensive gene dispersal and that grasses generally have a good ability of establishing outside and on cultivated areas.

Measures

The most important possibilities of reducing gene dispersal in seed grasses are observation of cropping distances and cropping intervals (cf. the Executive order on field seeds).

Other pollen sources such as grazing and silage fields, uncultivated areas (road edges, set-aside areas, highly sensitive agricultural areas, coverts, etc.) with unknown varieties and wild types of the grass species should be cut before flowering.

Furthermore, it is important to reduce as much as possible the number of volunteer seeds that are incorporated into the soil and in general to control grass volunteers in the crop rotation.

Need for further knowledge

There is a need for determining the extent of pollen dispersal from other sources, for example grass areas covered by special environmental agreements according to which cutting/hay-making is not allowed until 15 July, and from set-aside areas.

There is a need for determining the extent of grass volunteers that produce seeds in other crops (false crop rotation).

There seems to be an increasing need for studying competitive parameters for different biological characteristics in grasses in order to be able to assess the dispersal efficiency of a given trait both in and outside the cultivated areas.

Conclusion

Grass is very common in Danish agriculture, is used increasingly in recreational areas and is common on uncultivated areas and natural resorts. Denmark is in an internationally leading position within the production of grass seed, and the seed production is progressing.

Grass areas constituted in 2006 a total of approx. 26% of the total cultivated area, and 10% of the grass area was organic.

So far, we have only limited experience with GM grass (from the USA), and this experience together with the most recent knowledge in the area confirms that grasses are highly capable of gene dispersal.

All the grasses grown in Denmark are highly capable of dispersal – both on and outside the cultivated areas, and on the present basis of knowledge it is still not possible to indicate guidelines to ensure co-existence in Denmark.

We lack knowledge about the extent of pollen dispersal between fields for different uses, from potential sources outside the field and the extent of grass volunteers, which produce seed in intervening crops.

To assess the ability of a given GM trait to survive and possibly to increase its distribution, an assessment should be made of the competitive ability of the trait both inside and outside the cultivated areas.