Soil aeration is an important factor influencing soil biological and biochemical processes because it directly affects soil microbial activity and respiration. As an important living skin of the soil, biocrusts strongly impact various soil properties and play a vital role in maintaining surface soil structure and multifunctionality. Although there are several studies concerning the effects of biocrusts on soil respiration processes, it is still not well understood how biocrusts affect air permeability (K_a) and soil aeration in drylands. In a semiarid climate region of the northern Chinese Loess Plateau, the K_a and air-filled porosity (ε_a) of moss-dominated biocrusts established on aeolian sand and loess soil were measured as a function of soil water content under laboratory and in situ conditions and compared to measurements of bare soils. The relative air permeability (K_ra) of biocrusts and bare soils was further analyzed and estimated with commonly applied K_ra models that are based on the soil water retention curve (SWRC), and the performance of these models was specifically assessed for biocrusts. The in situ measurements yielded a 51.8% and 76.7% higher K_a for the biocrust-colonized aeolian sand and loess soil, respectively, when compared to the bare soils. Correspondingly, in comparison to the bare soils, the ε_a of the biocrusts also increased by 38.0% for the aeolian sand and by 52.4% for the loess soil. The laboratory measurements showed an 8.5%–9.3% increase of K_a for the biocrust-colonized soils from wet to dry conditions when compared to their bare counterparts, which was mainly attributed to their higher fine particle and organic matter contents as well as to the moss rhizoids. Moreover, a positive linear relationship (R² > 0.43) was found between the log K_a and log ε_a for both biocrusts and bare soils. Additionally, the biocrusts exhibited higher water content at almost all matric potentials in comparison to the bare soils. The K_ra of biocrusts was 22.0% higher for aeolian sand and 42.1% higher for loess soil when compared to the bare soils. The van Genuchten-Burdine and Brooks and Corey-Burdine models performed best with regard to the estimation of K_ra (RMSE ≤ 0.100, bias ≤ 0.018). In general, biocrusts may greatly increase surface soil aeration due to their higher K_a and ε_a, and the soil aeration status of biocrusts can be indirectly assessed through their effective degree of soil water saturation. Our findings suggest that biocrusts should be carefully considered in further studies of soil aeration related to biological processes and gas and vapor transport in dryland ecosystems.