Abstract:Two widely-distributed and efficient sand-fixing shrubs in the Hobq Sandy Land, Artemisia ordosica and Caragana intermedia, have received considerable attention related to their role in the process of vegetation restoration. This paper addresses the natural restoration of A. ordosica and the reintroduction of C. intermedia in the Hobq Sandy Land, begun in 1995. Variations in the soil microorganisms and chemical properties of the rhizosphere and non-rhizosphere soils were compared in areas restored with either A. ordosica or C. intermedia. This was carried out using field investigations, indoor biochemical experiments, and statistical methods. The measured variables included the total abundance of microorganisms, the abundance of soil bacteria, fungi, and actinomycetes, the amounts of C and N in the microbial biomass and measures of soil organic matter, soil pH, total nitrogen, total phosphorus, available nitrogen, available phosphorus, and available potassium. The relationship between soil microorganisms in the rhizosphere and soil chemical properties was also studied. Meanwhile, a synthetic index method was used to evaluate the effects of these two plant species on the recovery of the soil ecosystem, with the goal of providing scientific evidence to support vegetation restoration and redevelopment in the Hobq Sandy Land. Untreated sandy land subject to natural moving sand conditions was used as a control. The main findings were as follows:(1) compared to untreated sandy land, vegetation restoration resulted in varying increases in both the soil microbe abundance and microbial biomass of both the rhizosphere and non-rhizosphere soil of the two types of vegetation restoration analyzed here. Among these variables, the total numbers of microorganisms and bacteria, as well as the quantities of C and N in the microbial biomass of the rhizosphere, were higher in sites restored with A. ordosica than in sites with C. intermedia. In addition, the numbers of fungi and actinomycetes in the rhizosphere of sites restored withC. intermedia were higher than that of sites with A. ordosica; this shows that the rhizosphere environment in areas with naturally restored A. ordosica is more conducive to increases in microbial activity than areas with reintroduced C. intermedia. (2) The soil organic matter, soil total nitrogen, total phosphorus, available nitrogen, and available potassium content of both the rhizosphere and non-rhizosphere soil under these two types of vegetation restoration were higher than in untreated sandy land. Among the soil conditions, the soil organic matter, soil total nitrogen, and available nitrogen content of both the rhizosphere and non-rhizosphere soil were higher in sites with A.ordosica than in those with C. intermedia, showing that the soil nutrient enrichment was significantly higher at sites restored with A. ordosica. (3) The soil factors that affect the quantity of soil microbes present, as well as amounts of C and N in the microbial biomass of areas restored with A. ordosica and C. intermedia, include soil organic matter, pH, total nitrogen, available nitrogen, available potassium, and available phosphorus. Among these, soil organic matter, total nitrogen, and available potassium had a positive effect on the quantities of soil microbes as well as the amounts of C and N in the microbial biomass under the two types of vegetation restoration analyzed here. (4) The synthetic index method was used to calculate a soil quality index that was based on 14 soil quality indices; the order of the values of the soil quality indices was as follows:rhizosphere soil of A. ordosica > rhizosphere soil of C. intermedia > non-rhizosphere soil of A. ordosica > non-rhizosphere soil of C. intermedia > soil of untreated sandy land. The results confirm that the effects of the A. ordosica community on the soil during natural restoration were better for soil improvement than those of the C. intermedia community, making the A. ordosica community more appropriate for use in the ecological restoration of the Hobq Sandy Land.