Abstract: During aeolian processes, the two most critical factors related to dust emissions are soil particle and aggregate saltation, which greatly affect the vertical profiles of near-surface dust concentrations. In this study, we measured PM10 concentrations at four different heights (0.10, 0.50, 1.00 and 2.00 m) with and without continuous and simultaneous aeolian saltation processes on a Gobi surface in northwestern China from 31 March to 10 April, 2017. We found that the vertical concentration profiles of suspended PM10 matched the log-law model well when there was no aeolian saltation. For the erosion process with saltation, we divided the vertical concentration profiles of PM10 into the saltation-affected layer and the airflow-transport layer according to two different dust sources (i.e., locally emitted PM10 and upwind transported PM10). The transition height between the saltation-affected layer and the airflow-transport layer was not fixed and varied with saltation intensity. From this new perspective, we calculated the airflow-transport layer and the dust emission rate at different times during a wind erosion event occurred on 5 April 2017. We found that dust emissions during wind erosion are primarily controlled by saltation intensity, contributing little to PM10 concentrations above the ground surface compared to PM10 concentrations transported from upwind directions. As erosion progresses, the surface supply of erodible grains is the most crucial factor for saltation intensity. When there was a sufficient amount of erodible grains, there was a significant correlation among the friction velocity, saltation intensity and dust emission rate. However, when supply is limited by factors such as surface renewal or an increase in soil moisture, the friction velocity will not necessarily correlate with the other two factors. Therefore, for the Gobi surface, compared to limiting dust emissions from upwind directions, restricting the transport of suspended dust in its path is by far a more efficient and realistic option for small areas that are often exposed to dust storms. This study provides some theoretical basis for correctly estimating PM10 concentrations in the Gobi areas.
摘要：The shear stress generated by the wind on the land surface is the driving force that results in the wind erosion of the soil. It is an independent factor influencing soil wind erosion. The factors related to wind erosivity, known as submodels, mainly include the weather factor (WF) in revised wind erosion equation (RWEQ), the erosion submodel (ES) in wind erosion prediction system (WEPS), as well as the drift potential (DP) in wind energy environmental assessment. However, the essential factors of WF and ES contain wind, soil characteristics and surface coverings, which therefore results in the interdependence between WF or ES and other factors (e.g., soil erodible factor) in soil erosion models. Considering that DP is a relative indicator of the wind energy environment and does not have the value of expressing wind to induce shear stress on the surface. Therefore, a new factor is needed to express accurately wind erosivity. Based on the theoretical basis that the soil loss by wind erosion (Q) is proportional to the shear stress of the wind on the soil surface, a new model of wind driving force (WDF) was established, which expresses the potential capacity of wind to drive soil mass in per unit area and a period of time. Through the calculations in the typical area, the WDF, WF and DP are compared and analyzed from the theoretical basis, construction goal, problem-solving ability and typical area application; the spatial distribution of soil wind erosion intensity was concurrently compared with the spatial distributions of the WDF, WF and DP values in the typical area. The results indicate that the WDF is better to reflect the potential capacity of wind erosivity than WF and DP, and that the WDF model is a good model with universal applicability and can be logically incorporated into the soil wind erosion models.