In the framework of the innovations proposed in the Catalogue of Innovations for Multifunctional Olive Systems and in the Catalogue of Innovations for Grazed Woodlands, the LIVINGAGRO project identified two different forage seed mixtures specifically for two Mediterranean agroforestry systems: one for the multifunctional olive groves and the other for the grazed woodlands. To this end, two experimental field trials were launched in autumn 2021 in northern Sardinia. In the silvo-pastoral system (grazed woodland), three different treatments were compared: i) natural pasture; ii) improved pasture using a commercial seed mixture; iii) improved pasture using an innovative legume-based seed mixture. In the chosen multifunctional olive system, the same treatments of soil management practices were compared (natural cover crop, cover crop with innovative seed mixture and cover crop with commercial seed mixture), with the addition of a fourth treatment: shallow soil tillage as traditional practice.

The data collection on grassland sward conducted in both systems, below and outside the tree canopy, took into account the following parameters: establishment (1st year) and autumn re-establishment (2nd year); visual assessment of soil coverage and spontaneous species (%); dry matter production, botanical composition and bromatological analyses of the sampled biomass. In the olive grove, the effects on olive trees were determined by monitoring: i) water status and chlorophyll index; ii) Normalized Deficiency Vegetation Index (NDVI) and other indices to assess the physiological response; iii) olive production and drupe quality.

In the olive grove system (Figure 1), the results referring mainly to the establishment year showed a higher winter and spring soil coverage of the natural cover treatment compared to the others, with a lower presence of legumes. Soil coverage and dry matter yield were higher for the innovative seed mixture than for the commercial one, which showed a higher presence of spontaneous species. Negligible differences were observed in terms of autumn sward re-establishment between the two mixtures. With regard to the water potential of olive trees, during the period of greatest water deficit, the traditional soil tillage showed lower levels of water potential, while the natural cover showed a significantly lower photochemical efficiency. Moreover, natural cover favoured smaller fruit size, while no significant differences emerged between traditional soil tillage and treatments with sown mixtures.

In the silvo-pastoral system (Figure 2), preliminary results showed a higher soil coverage in natural pastures than in both improved pastures in the first autumnal phase after sowing. However, higher winter and spring dry biomass production was observed in both improved pastures compared to natural pasture. The presence of legumes was negligible in natural pasture, whereas in the improved pasture with innovative and commercial seed mixture, legumes species and varieties accounted for more than 50% of the biomass composition. Furthermore, the improved pasture with innovative seed mixture produced more dry biomass with a higher presence of legumes than the improved pasture with commercial seed mixture. In natural pasture, negligible differences in terms of botanical composition were observed between the areas below and outside the tree canopy, whereas, in both improved pastures, lower legume presence and lower biomass production were observed below the tree canopy than outside. Finally, the natural pasture showed a higher presence of legumes than the other two treatments in the autumn sward re-establishment.