In the Ulan Buh Desert, which is located in a seasonally frozen region, a frozen soil layer can appear in the winter after the wind erosion of dry sand from the surface of a mobile sand dune, thus altering the wind-sand transport process. To clarify the wind-sand transport pattern after the emergence of a frozen soil layer, this study used wind tunnel experiments to study the variations in the wind erosion rate and sediment transport pattern of frozen and nonfrozen desert soil with different soil moisture contents (1-5%). The results revealed that the relationships of the wind speed, soil moisture content and wind erosion rate are in line with an exponential function, and the wind erosion rate decreases by 6-52% after the desert soil is frozen. When the soil moisture content of the nonfrozen desert and frozen desert soil is 4% and 3%, respectively, the wind erosion rate of the soil can be reduced by more than 65% compared with that of natural dry sand (soil moisture content of 0.28%), i.e., the wind erosion rate can be effectively reduced. The sediment transport rate of nonfrozen desert soil decreases with increasing height, with an average ratio of approximately 65% for saltation. The sediment transport rate of frozen desert soil first increases but then decreases with increasing height, with an average ratio of approximately 80% for saltation. When sand particles hit the source of frozen desert soil, the interaction between particles and bed surface is dominated by the process of impact and rebound, so that more particles move higher, and some sand particles move from creep to saltation. In summary, freezing has an inhibitory effect on the wind-sand activity of desert soil, and freezing makes it easier for sand to move upwards.

The damage to microbial solidified engineering residue by freeze-thaw cycles increases the amount of material prone to wind erosion. Microbial solidification of engineering residue was carried out, and freeze-thaw cycle and indoor wind tunnel tests were conducted on the microbial solidified samples to reveal the interaction mechanism between different numbers of freeze-thaw cycles and the wind erosion degree. The test results showed that the larger the number of freeze-thaw cycles, the greater the mass loss of the microbial solidified engineering residue sample due to wind erosion and the lower the surface strength and surface thickness of the sample. However, the surface strength and surface thickness were relatively stable after more than 7 freeze-thaw cycles. The mass loss of the sample was 13 g after 9 freeze-thaw cycles at the maximum wind speed (15 m/s), higher than that of the sample exposed to no freeze-thaw cycles (6 g) but far lower than that of the undisturbed sample (3647 g). The results indicated that the microbial solidified engineering residue had high freeze-thaw resistance. The microbial solidified engineering residue was analyzed by computed tomography (CT) before and after the freeze-thaw cycles, and three-dimensional reconstruction was performed using digital image processing. The microstructure analysis showed that the freeze-thaw cycles did not change the content and spatial distribution of the microbial solidified products but reduced the ability to cement the microbial solidified products and the soil particles. The calcium carbonate inside the hard shell became more fragmented, the equivalent radius of the crystals and the stability of the hard shell decreased, and the porosity increased. However, the microbial solidified engineering residue exhibited high resistance to wind erosion and freeze-thaw cycles.