Participatory design of improved forage/fallow options across soil gradients with farmers of the Central Peruvian Andes

Land use intensity is increasing in Andean smallholder systems, and innovations are needed to sustain soil fertility and productivity of potato-cereal rotations with shortening fallow periods. In collaboration with farmers in central Peru, we assessed forage-based fallows in 58 fields across three production zones over three years. Fallow treatments selected with farmers tested grass-legume mixtures with different combinations of Vicia Saliva (vetch), Avena sativa (oats), Lupinus mutabilis (Andean lupine), Trifolium pratense (red clover), Medicago sativa (alfalfa), and Lolium multiflontm (ryegrass) compared to an unseeded control fallow with natural revegetation. The ability of fallows to quickly cover soil was tested, as was their biomass production in years one and three. Following the incorporation of fallow vegetation in a sub-set of nine fields, we also tested fallows' effects on soil pH, available phosphorus (P), permanganate-oxidizable carbon (PDXC) and potato yield. In year one managed fallows produced from 1.9 to 5.4 Mg ha(-1) of forage biomass compared to 0.5 to 1.1 Mg ha(-1 )in unseeded controls. Managed fallows also exceeded controls in nutrient uptake, soil cover, and forage quality (lower lignin and higher protein content). First-year biomass of vetch and Andean lupine responded differently to soil pH in fields, indicating that appropriate fallow options likely depend on soil context. After three years, total biomass did not differ among treatments. However, legumes had greater biomass in treatments employing perennial species (0.79-1.18 Mg ha(-1) of legumes) than in controls (0.15 Mg ha(-1)). Potato yield and soil fertility was not reduced in managed fallows compared to the control, and an alfalfa + liming treatment yielded higher than the control (p < 0.05). Diseased tubers were also less prevalent in fallows containing ryegrass and clover, versus other treatments (2.7 % vs. 4.7 % diseased; p = 0.05). In a post-hoc analysis considering 41 treatment plots with contrasting cutting regimes, plots that were cut repeatedly throughout the fallow period had more negative changes in PDXC than those cut initially and then left to regrow (p = 0.04). In evaluation workshops, farmers emphasized forage production, potato yield, and potato tuber health as evaluation criteria for the fallows, and ranked the alfalfa + liming treatment as the best. In Andean communities with shortening fallow periods, forage-based fallows represent a promising, multi-functional option to maintain soil health and productivity while generating additional sources of high-quality forage. However, future research should examine long-term nutrient and carbon balances under different forage removal scenarios, as well as designing fallows for varied agroecological contexts.