2.4 Data analysis
The wind erosion material in the wind tunnel mainly moved horizontally. Thus, the amount of wind erosion and particulate matter emissions can be expressed in a unified horizontal direction (Panebianco et al., 2016). Affected by the BSC coverage, the amount of wind erosion produced in this research was relatively small. Thus, a high-precision electronic scale was used to measure wind erosion. The total amount of wind erosion was expressed as the difference in weight between the undisturbed soil collection box before and after the test. To calculate the proportion of particulate matter emissions in the total wind erosion, the particulate matter concentration was converted to the particulate matter mass, according to the specifications of the wind tunnel, using the following formula:
\begin{equation} m=\frac{C\times v\times t\times 0.6\times 0.8}{1000000}\nonumber \\ \end{equation}
where m is the mass of particulate matter (in grams); C is the emission concentration of particulate matter (μg/m3); v is the wind speed (m/s); t is the blowing time (s); and 0.6 and 0.8 are the height and width at the middle of the wind tunnel (m), respectively. In order to consider the importance of the influencing factors, this study carried out a linear regression fitting for wind speed, coverage, wind erosion and particulate emissions, and standardized the coefficients as follows:
\begin{equation} B_{i}=b_{i}\times\frac{S_{i}}{S_{j}}\nonumber \\ \end{equation}
where B is the standardization coefficient; b is the unstandardized coefficient; S is the standard deviation; iis the independent variable; and j is the dependent variable.
The linear fitting and coefficient standardization methods used in this study were completed with SPSS 25. Data analysis was completed with Microsoft Excel 2019 and the drawing using Sigmaplot 12.5.