Species interactions in a grassland mixture under low nitrogen fertilization and two cutting frequencies: 1. dry‐matter yield and dynamics of species composition

Four‐species mixtures and pure stands of perennial ryegrass, tall fescue, white clover and red clover were grown in three‐cut and five‐cut systems at Ås, southern Norway, at a low fertilization rate (100 kg N ha^?1 year^?1). Over a three‐year experiment, we found strong positive effects of species diversity on annual dry‐matter yield and yield stability under both cutting frequencies. The overyielding in mixtures relative to pure stands was highest in the five‐cut system and in the second year. Among the possible pairwise species interaction effects contributing to the diversity effect, the grass–grass interaction was the strongest, being significant in both cutting systems and in all years. The grass–legume interactions were sometimes significant, but no significant legume–legume interaction could be detected. Competitive relationships between species varied from year to year and also between cutting systems. Estimations based on species identity effects and pair‐specific interactions suggested that the optimal proportions of red clover, white clover, perennial ryegrass and tall fescue in seed mixtures would have been around 0·1, 0·2, 0·4 and 0·3 in the three‐cut system, and 0·1, 0·3, 0·3 and 0·3 in the five‐cut system.     

Response of plant species composition,biomass production and biomass chemical properties to high N,P and K application rates in Dactylis glomerata‐ and Festuca arundinacea‐dominated grassland

Little is known about the immediate effect of high nitrogen (N), phosphorus (P) and potassium (K) application rates on sown grasslands cut twice per year. We asked how quickly plant species composition, biomass yield, biomass chemical properties and nutrient balance respond to N, P and K application. An experiment using unfertilized control, P, N, NP and NPK treatments was established on seven‐year‐old cut grassland in the Czech Republic in 2007 and monitored over four years. Annual application rates were 300 kg N ha^?1, 80 kg P ha^?1 and 200 kg K ha^?1. The immediate response of plant species composition to N application was recorded and was found to be different to the response over the four years of the study period. Highly productive grasses (Dactylis glomerata, Festuca arundinacea and Phleum pratense) were promoted by N application in 2008 and then retreated together with legumes (Medicago sativa, Trifolium pratense and Trifolium repens) in all N treatments where the expansion of perennial forbs (Urtica dioica and Rumex obtusifolius) and annual weeds (Galinsoga quadriradiata, Impatiens parviflora, Lamium purpureum and Stellaria media) was recorded. At the end of the experiment, Festuca rubra was the dominant grass in the control and P treatment, and species richness was lowest in all treatments with N application. Mean annual dry‐matter yield over all years was 3.5, 3.9, 5.8, 5.6 and 6.8 t ha^?1 in the control, P, N, NP and NPK treatments, respectively. Concentrations of N in the biomass ranged from 20.0 to 28.7 g kg^?1 in the P and N treatments; concentrations of P ranged from 3.2 to 3.7 g kg^?1 in the N and P treatments; and concentrations of K ranged from 24.1 to 34.0 g kg^?1 in the NP and NPK treatments. The N:P, N:K and K:P ratios did not correctly indicate the nutrient limitation of biomass production, which was primarily N‐limited, and K‐limitation was only recorded for high production levels in treatments with N applications. On the basis of the nutrient‐balance approach, the balanced annual application rates were estimated as 140 kg N ha^?1, 30 kg P ha^?1 and 100 kg K ha^?1. We concluded that high N, P and K application rates can very quickly and dramatically change species composition, biomass production and its chemical properties in sown cut grasslands. High N application rates can be detrimental for tall forage grasses and can support the spread of weedy species.