How biodiversity affects terrestrial productivity is important to the maintenance of ecosystem services under global change. Although the crucial role of plant–soil feedbacks (PSF) in determining diversity–productivity relationship has been increasingly recognized in recent years, its legacy effects on subsequent diversity–productivity relationship are still unclear. We conducted a classic PSF experiment to assess how plant diversity‐conditioned soils influenced subsequent plant diversity–productivity relationships, where three plant diversity levels (1, 4 and 8 species) were planted in soils conditioned at three diversity levels (conditioned by 1, 4 and 8 species for 3 years). In addition, to test the role of soil microbial diversity in mediating the effects of soil conditioning diversity, the three plant diversity levels were planted with low, moderate and high soil biodiversity created by a soil inoculum dilution. The results showed that plant productivities were promoted by mixed‐conditioned soils (4 and 8 species) comparing to mono‐conditioned soils (1 species). Productivity was positively related to planted diversity in mixed‐conditioned soils, while showed no relation with planted diversity in mono‐conditioned soils. Productivity was promoted by soil biodiversity when 4 and 8 species were mixed planted, while did not change when 1 species was planted. Synthesis and applications . Our results highlight that PSF is crucial to strengthen the positive effects of biodiversity on productivity, implying that diversifying cropping systems should be encouraged in agroecosystem management to benefit from positive PSF effects. The importance of soil legacy for optimizing plant productivity is particularly important for conditioning soils with an intermediate number of plant species.

Overgrazing leads to the degradation of grazing lands, which seriously threatens the stability of grassland ecosystems. Root-invading fungi, as one of the main influencing factors, can cause plant diseases in grasslands, reduce the proportion of dominant plant species, increase the proportion of invasive poisonous weeds, and further aggravate degradation. In order to predict and improve the effects of root-invading fungi on grassland degradation, we conducted an in situ soil indoor control experiment using soils collected from non-degraded, moderately degraded, and severely degraded areas of Nanshan pasture in Hunan Province, China. We used monoculture or mixed grasslands of dominant plant species, including Lolium perenne, Trifolium repens, and the invasive weed Persicaria hydropiper, and inoculated them with local strains of pathogenic Fusarium species (Fusarium boothii and Fusarium circinatum) and beneficial fungi Arbuscular Mycorrhizal Fungi (AMF) and Trichoderma hamatum to explore how different strains of fungi affect plant growth and community dynamics. The results showed that Fusarium species (Fusarium boothii and Fusarium circinatum), as a major pathogenic fungus, inhibited the growth of the dominant grass Lolium perenne in moderately and severely degraded soils, which provided growth space and resources for invasive weeds Persicaria hydropiper and further aggravated the degree of grassland degradation. However, the collaborative effect of beneficial fungi (AMF and Trichoderma) and their inhibitory effect on Fusarium species (Fusarium boothii and Fusarium circinatum) could promote the growth of dominant plants and weeds in soils with varying degrees of degradation, which is beneficial to maintaining the stability and diversity of grassland plant communities. The collaborative effect of beneficial fungi could also increase the availability of nutrients in severely degraded soils. Therefore, using beneficial fungi (AMF and Trichoderma) for soil improvement and reducing the harm of pathogenic Fusarium species (Fusarium boothii and Fusarium circinatum) to plant growth is of great significance for promoting the protection and management of grassland ecosystems, as well as for the restoration and recovery of grasslands.

Phytoremediation can help remediate potential toxic elements (PTE) in soil. Microorganisms and soil amendments are effective means to improve the efficiency of phytoremediation. This study selected three microorganisms that may promote phytoremediation, including bacteria (Ceratobasidium), fungi (Pseudomonas mendocina), and arbuscular-mycorrhizal fungi (AMF, Funneliformis caledonium). The effects of single or mixed inoculation of three microorganisms on the phytoremediation efficiency of Paspalum vaginatum and Pennisetum alopecuroides were tested under three different degrees of cadmium-contaminated soil (low 10 mg/kg, medium 50 mg/kg, and high 100 mg/kg). The results showed that single inoculation of AMF or Pseudomonas mendocina could significantly increase the biomass of two plants under three different degrees of cadmium-contaminated soil, and the growth-promoting effect of AMF was better than Pseudomonas mendocina. However, simultaneous inoculation of these two microorganisms did not show a better effect than the inoculation of one. Inoculation of Ceratobasidium reduced the biomass of the two plants under high concentrations of cadmium-contaminated soil. Among all treatments, the remediation ability of the two plants was the strongest when inoculated with AMF alone. On this basis, this study explored the effect of AMF combined with corn-straw-biochar on the phytoremediation efficiency of Paspalum vaginatum and Pennisetum alopecuroides. The results showed that biochar could affect plant biomass and Cd concentration in plants by reducing Cd concentration in soil. The combined use of biochar and AMF increased the biomass of Paspalum vaginatum by 8.9–48.6% and the biomass of Pennisetum alopecuroides by 8.04–32.92%. Compared with the single use of AMF or biochar, the combination of the two is better, which greatly improves the efficiency of phytoremediation.

Intense human activities break the grassland–livestock balance and accelerate grassland degradation. We evaluated the use of native dominant species combined with arbuscular mycorrhizal fungi (AMF) in order to recover grassland and restrain grassland degradation. We conducted a full factorial greenhouse experiment to evaluate the interaction effects of native species of distinct traits grass Lolium perenne (L) and legume Trifolium repens (T) with arbuscular mycorrhizal fungi (AMF) inoculation on grass productivity and soil properties across non-degraded, lightly degraded, and severely degraded soils. The grass–legume mixture was manipulated with five ratios (T:L = 1:0, T:L = 1:1, T:L = 3:1, T:L = 1:3, T:L = 0:1). The results showed that L. perenne significantly increased grassland productivity at different grass–legume ratios, regardless of AMF presence or absence. AMF inoculation increased plant N and P content uptake and improved the productivity of degraded grasslands, especially in severely degraded grasslands. The NO3−-N and available P concentrations increased in soil when the legume component increased from T:L = 0:1 (grass monoculture) to T:L = 1:0 (legume monoculture). This may be because the presence of Lolium perenne (L) can promote nitrogen fixation in legumes. Structural equation modeling indicated that grass–legume mixtures directly affected plant biomass, whereas AMF affected plant biomass via providing plant nutrients. A soil quality index based on minimum datasets indicated a significant positive effect of artificial grassland establishment on soil quality. We conclude that planting T:L = 0:1 and T:L = 1:3 combined with AMF inoculation can be used to recover degraded grassland production, and planting T:L = 1:1 and T:L = 1:3 plus AMF inoculation can be applied for grassland nutrient accumulation and stability maintenance.