In recent years, artificial biological soil crusts (BSCs)—i.e., the inoculation of soil with cyanobacteria—have become one of the most promising biotechnological strategies for preventing soil erosion and restoring soil functionality in degraded drylands. In order to use this biotechnology on a large scale, researchers must explore methods that enable rapid and largescale propagation of soil inoculum, as well as testing methods for developing artificial BSCs in the field. To help do this, we tested the effects of four soil substrates on the development of artificial BSCs in the Tengger Desert. Each soil substrate was collected from a different area in Zhongwei city in northern China and had different particle sizes and nutrient contents. We measured artificial BSCs characteristics according to the coverage, thickness, and wind erosion resistance of the BSCs analyzed. After 12 months, incubated artificial BSC coverage in incubated soils rose from 6% to 20%, and thickness rose from 2.94 to 4.06 mm. Incubated BSC coverage was positively correlated with fine material content (i.e., soil particle size less than 0.10 mm). There was also a positive relationship between coverage and soil organic matter, total nitrogen, and total phosphorus. Artificial BSC inoculation significantly improved soil resistance to wind erosion. These findings indicate that soil substrates improve BSC recovery rate over sand substrates by enhancing sand surface stabilization, increasing initial silt and clay content and nutrient content, especially soil organic matter content. Considering the scarcity of soil resources and the advantages of reusing soil resources, we suggest that left-over soils dredged from irrigation channels and abandoned farmlands can be used as substrate for the propagation of BSC inoculum.