This study addresses the challenge of establishing alfalfa (Medicago sativa L.) in the desertified saline–alkali region of the Qaidam Basin, where seedling survival and productivity are considerably constrained. A two-year field experiment was conducted to identify suitable companion crops by comparing an alfalfa monoculture (control) with alfalfa intercropped with oat (Avena sativa L.) (O+A), forage pea (Pisum sativum L.) (P+A), or triticale (xTriticale Wittmack) (T+A). Agronomic traits, yield, and nutritional quality were measured, and a comprehensive evaluation was performed using the TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) model. The results indicated that intercropping with oat produced the highest fresh and dry matter yields in the establishment year, while triticale and pea treatments exhibited strong yield potential in the subsequent year. Triticale significantly enhanced alfalfa’s nutritional quality by increasing crude protein content and relative feed value (RFV) while reducing fiber content. The TOPSIS model ranked the oat treatment highest for overall performance, effectively balancing yield and quality. In conclusion, oats are recommended as a companion crop to enhance alfalfa establishment in this saline–alkali environment, whereas triticale offers distinct advantages for improving the nutritional value of the forage.

The optimization of mowing management is crucial for establishing a high-yielding “one sowing, two harvests” system for oat (Avena sativa L.) forage production in the alpine regions of the Tibetan Plateau. This study aimed to identify the optimal combination of cutting stage and stubble height to maximize the total seasonal productivity of the system. A two-factor field experiment, arranged in a randomized complete block design, was conducted over two consecutive growing seasons (2024 and 2025) using the local cultivar ‘Qingtian No. 1’. The treatments consisted of two cutting stages (flowering stage, C₁; milk stage, C₂) and three stubble heights (5 cm, H₁; 10 cm, H₂; 15 cm, H₃). Regarding regrowth phase performance, mowing at the flowering stage (C₁) resulted in significantly superior agronomic traits (e.g., stem diameter, tiller number) and photosynthetic characteristics compared to the milk stage (C₂). In terms of forage quality, the C₂H₂ treatment maximized water-soluble carbohydrate content but also led to elevated fiber levels. However, the key metric for system evaluation—the total seasonal dry matter yield—was predominantly governed by the cutting stage. Mowing at the milk stage (C₂) consistently resulted in a higher total yield than C₁, with optimal combinations yielding 1.31 t/ha (C₂H₁) in 2024 and 1.25 t/ha (C₂H₂) in 2025. This advantage was primarily due to a substantially greater first-harvest yield under C₂, which outweighed the regrowth benefits of C₁. A comprehensive evaluation based on total system productivity and forage quality indicates that mowing at the milk stage (C₂) is the optimal strategy for achieving high total output. For ‘Qingtian No. 1’ oat in this region, a stubble height of 10 cm (as in the C₂H₂ treatment) is recommended to ensure robustness across years. This management strategy secures a high total seasonal forage yield through effective biomass allocation between harvests, extending the supply window and offering a sustainable cultivation model for this ecologically vulnerable region.