Partner und Internationale Organisationen
(Englisch)
|
AT, BE, BG, CH, CS, DE, DK, ES, FI, FR, GR, IE, IS, IT, LT, NL, NO, PL , PT, SE, SI, UK
|
Abstract
(Englisch)
|
There is considerable interest to increase resource-efficiency in agricultural systems with the aim to conserve non-renewable natural resources and reduce production costs and emissions to the environment. Forage legumes are a key component of such sustainable production systems. This is due to their ability to fix atmospheric nitrogen symbiotically, their high feeding value and their ability to improve soil fertility. However, annual and seasonal fluctuation of legume proportion in mixed swards restrict the adoption of legume-based systems by European farmers. In low input grassland systems plant species richness and particularly functional group richness often positively influence biomass production. If this would hold also in intensively managed systems, working with grass/clover mixtures combining different functional groups would allow to profit from both symbiotic nitrogen (N) fixation and the positive diversity-productivity relationship. In this study, we manipulated absolute and relative abundance of 4 forage species in mixtures grown under low, moderate and high N supply to determine (1) the effects of combining different functional groups (grasses, legumes, fast and slowly establishing species) on sward productivity, (2) the effects of the initial botanical composition on sward dynamics and evolution, (3) the relative abundance of the species that should be targeted to insure highest yield, (4) the effects of N fertilisation on these diversity-productivity relationship, (5) the influence of the interactions between N fertilisation and botanical composition on symbiotic N2 fixation and fertiliser N recovery, and (6) the effect of botanical composition at different level of N fertilisation on the risk for N losses to the environment in form of nitrate and gaseous emissions. In a three year field experiment, absolute and relative sowing abundance of the four species Lolium perenne (fast establishing grass), Dactylis glomerata (slowly establishing but more persistent grass), Trifolium pratense (fast establishing legume) and Trifolium repens (slowly establishing but more persistent legume) were widely varied. The swards were grown under a wide range of N supply (50, 150 and 450 kg N ha-1 yr-1). The 99 plots were harvested five times per year and the total dry matter yield and the proportion of each species were determined. The influence of the sward composition and N fertilisation on the total annual yield was analysed using multiple linear regression on the yield data. The stability of the swards was assessed by analysing the relative growth rate difference (RGRD) of the sown species and the evolution of the yield proportion of the unsown species. Symbiotic N2 fixation was quantified using the 15N dilution technique. The recovery of fertilizer N in the plants was investigated after a 15N labeled ammonium nitrate application. The nitrate leaching potential of 54 selected plots was measured using ceramic cups placed at 60 cm below the soil surface. To determine N2O fluxes and related parameters, two plots each (high and low fertilisation levels) with pure stands of both T. repens and L. perenne as well as two grass-legume mixtures (at moderate and high fertilisation levels) were selected from the field experiment in the third year after establishment (2005). Combining species from different functional groups in mixtures resulted compared to monocultures in a strong overyielding effect over the three years and at all levels of N fertilisation. But increasing N fertilisation reduced the positive effect of functional diversity, the proportion of legume in the swards, the quantities of N2 fixed symbiotically, as well as the efficiency of fertiliser N. The yield response of the four-species mixtures to N fertilisation was therefore small, especially at optimal legume proportions. It is remarkable that well balanced mixtures fertilised with only 50 kg N ha-1 yr-1 produced a similar forage yield to the highest yielding grass monoculture at 450 kg N ha-1 yr-1. Mixtures also strongly reduced the incidence of weeds in the sward, and when fertilised at low levels of nitrogen showed lower risks for nitrates and nitrous oxide emissions per yield unit than the heavily fertilised pure grass swards. Results from our experiment shows that - in intensive systems - high productivity can be maintained while saving large amounts of fertiliser N and minimising detrimental effects to the environment by the use of well balanced grass-legume mixtures. Under the leadership of Switzerland, Spain and Ireland the international cooperation resulted in a large network carrying out a pan European experiment at 40 sites from the Mediterranean to the North (Iceland Norway, Finland). These experiments showed that the positive effect of functional diversity holds true over a very wide range of environmental conditions (Kriwan et al., 2007). Looking to the future, it seems clear that species mixtures can deliver benefits to forage yield that are of significant magnitude, temporally persistent and spatially persistent at a geographical scale.
|
