These results showed that lipolysis and biohydrogenation of PUFA could be affected by plant phenols, particularly TP.

The effects of restricted time allocation (2, 4 or 6 h/day) to pasture and grazing day (Day 1, initial; Day 4, intermediate; Day 7, final) on feeding behaviour, intake and performance were assessed in Sarda dairy ewes, rotationally grazing Italian ryegrass plots for 7 days, with 21 days of regrowth. A randomised block design with two replicates per access time was used with six groups of six ewes each. The ewes were supplemented daily with 400 g/head of a commercial concentrate at milking, 300 g/head of lupin after grazing and 700 g/head of ryegrass hay overnight. Pasture variables, feeding behaviour, herbage and supplement DM intake, and milk yield and composition were measured on 12 days (4 per target grazing day). Plot average data were analysed by a bifactorial model with interaction, which was not significant. Sward height and herbage mass decreased between Day 1 and Day 4 (P < 0.05). Leaf area index dropped from Day 1 to Day 7 (P < 0.05). Eating time, as proportion of access time, and intake rate were higher in 2 h/day groups than in the others (P < 0.05). Nevertheless, herbage and total intake were higher in 6 h/day than in 2 h/day groups, being 4 h/day groups intermediate (P < 0.05). Herbage intake decreased with grazing period (P < 0.05). Fat normalised milk yield was higher in 6 h/day groups than in the others (P < 0.05) and in Day 1 and Day 4 than in Day 7 (P < 0.05). To conclude, time restriction below 6 h/day and pasture depletion, in terms of herbage quality, constrained intake and performance of rotationally grazing dairy ewes.

The effects of restricted access time to pasture (2, 4 or 6 h d À1 ; 2H, 4H or 6H) on ingestive behaviour and performance were assessed on four occasions per target grazing day (D1, initial day; D4, intermediate day; and D7, final day) in dairy ewes rotationally grazing berseem clover with a 7-day grazing period and a 21-day recovery period. A randomized block design with two replicates per treatment was used. All ewes were supplemented daily with 700 g per head of concentrates and 700 g per head of ryegrass-based hay. Pasture subplot and animal group data were analysed by a factorial model including access time (AT), grazing day (D) and their interaction as fixed factors. Sward height decreased from D1 (P < 0Á001) and green leaf mass from D4 (P < 0Á001) onwards during the grazing period. Grazing time as a proportion of AT was higher in 2H than in 4H and 6H ewes on D1 and D4 but not on D7 (P < 0Á05 for AT 9 D). Herbage intake rate was higher in 2H than in 4H and 6H ewes (P < 0Á001). Herbage and total intakes were higher in 4H and 6H than in 2H ewes (P < 0Á001), with herbage intake varying non-linearly during the grazing period (P < 0Á05). Milk yield was higher in 4H and 6H than in 2H ewes (P < 0Á01). To conclude, despite the evidence of compensatory behaviour, restricting access time to 2 h d À1 constrained intake and performance of dairy ewes rotationally grazing berseem clover.

Grasslands play important roles in agricultural production and provide a range of ecosystem services. Modelling can be a valuable adjunct to experimental research in order to improve the knowledge and assess the impact of management practices in grassland systems. In this study, the PaSim model was assessed for its ability to simulate plant biomass production, soil temperature, water content, and total and heterotrophic soil respiration in Mediterranean grasslands. The study site was the extensively managed sheep grazing system at the Berchidda‐Monti Observatory (Sardinia, Italy), from which two data sets were derived for model calibration and validation respectively. A new model parameterization was derived for Mediterranean conditions from a set of eco‐physiological parameters. With the exception of heterotrophic respiration (Rh), for which modelling efficiency (EF) values were negative, the model outputs were in agreement with observations (e.g., EF ranging from ~0.2 for total soil respiration to ~0.7 for soil temperature). These results support the effectiveness of PaSim to simulate C cycle components in Mediterranean grasslands. The study also highlights the need of further model development to provide better representation of the seasonal dynamics of Mediterranean annual species‐rich grasslands and associated peculiar Rh features, for which the modelling is only implicitly being undertaken by the current PaSim release.

No abstract