Low sustainability and short persistence of renovated pastures are main issues for pastoral production systems. Pasture establishment and the first year of production were evaluated following different pasture improvement methods. On a degraded pasture (control), treatments of fertilized naturalized pasture (FNP), cultivated pasture (CP: Lolium perenne [Lp]) and Trifolium repens [Tr]), direct-drilled pasture (DP: Lp and Tr) and diverse directdrilled pasture (DDP: Lp, Bromus valdivianus [Bv], Holcus lanatus [Hl], Dactylis glomerata [Dg] and Tr) were evaluated within a completely randomized block design. The trial was grazed by sheep. Soil fertility levels, seedling establishment and development, pasture growth, herbage mass and botanical composition were analyzed using ANOVA and LSD. After one year, soil aluminum saturation was low, Olsen-P medium and exchangeable potassium high. At 110 days after sowing, Lp had greater tillering than other grass species and its development at plant level was similar in all sowing methods. Sown species total leaf area was similar for all sowing methods and the improved pastures had a greater maximum pasture growth rate than NFP, but differed in their growth distribution across seasons. The accumulated herbage mass was similar for all improved pasture treatments. In conclusion, direct-drilling and pasture cultivation appeared to give a good pasture establishment. At seedling stage, Lp was faster growing than neighboring species. Increasing species diversity diminished seasonality of pasture growth. All the pasture improvement methods gave a higher total yield than the degraded pasture.