Providing evidence to improve practice

Individual study: Recreating species rich hay meadows using regional seed mixtures in the White Carpathians Protected Landscape Area, Czech Republic

Published source details

Jongepierova I., Mitchley J. & Tzanopoulos J. (2007) A field experiment to recreate species rich hay meadows using regional seed mixtures. Biological Conservation, 139, 297-305


Since 1950, almost a third of the area of ancient herb-rich semi-natural grassland have been lost in the Czech Republic thorough agricultural intensification, or abandonment of traditional grazing and haymaking. In the White Carpathians Protected Landscape Area (48°54′ N, 17°25′ E; eastern Czech Republic) in an attempt to reinstate some areas of former meadows on arable land, the use of an appropriate regional seed mix was trialled. The cost of seeding large areas with seed entirely of native species is high, therefore the efficacy of a cheaper technique, sowing strips within a commercial grass mixture, was also investigated.


Study area: The experiment was undertaken on a 3 ha arable field (a former flower-rich meadow ploughed and sown with arable crops from 1970 to 1998) in the Protected Landscape Area.

Experimental design: In spring 1999 (subsequent to autumn ploughing and spring harrowing to reduce weeds), four methods of establishing hay meadow vegetation using an appropriate regional seed mixture (7 grass and 20 herb species; seed rate of 2 g/m²) were compared in four replicated 55 m × 20 m plots for each treatment:

i) sowing over the whole plot;

ii) sowing 2.5 m-wide strips of the seed (excluding grasses) into a matrix of a commercial grass mixture;

iii) sowing 2.5 m-wide strips (excluding grasses) into natural regeneration;

iv) natural regeneration, no seed sown.

Plots were mown each July or August and the hay removed. Monitoring was carried out in June 2000-2004 to assess establishment and cover of all vascular plant species.


Establishment of sown regional grasses: All seven grasses established, by 2004 their cover in the whole sown plots averaged 55%; the most abundant species were upright brome Bromus erectus, Festuca rupicola and red fescue F.rubra. The grasses gradually colonised the other treatments; by 2004, sown strips in natural regeneration averaged 9.4% cover; sown strips in commercial grass 6.4%; natural regeneration (unsown) treatment 2.2%.

Establishment of sown herbs: In the whole sown plots, 19 of the 20 sown species established (the one that failed, common rockrose Helianthemum nummularium, was probably due to poor seed quality), the most abundant were brown knapweed Centaurea jacea, ribwort plantain Plantago lanceolata, hedge bedstraw Galium album and selfheal Prunella vulgaris. By 2002 the herbs had reached a fairly constant cover of about 30%. By 2004, the other three treatments had average cover of less than 5%.

Colonisation of some plots by unsown desirable weed species occurred.

Conclusions: The most effective treatment for establishing hay meadow vegetation was when the seed mix was sown over the entire plot; sowing in strips provided a lower cost option but establishment and spread of the desired species was slower.

Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at: