Abstract: [Objective] Approximately 22.5% of the total distance of the Qinghai-Tibet railway has been affected by sand damage, a major threat to the safe operation of the railway. In this study, we calculated the volume of sand deposition under each sand control measure through field investigation, and analyzed the influence of their spatial distribution on sand blocking capability, aiming to provide a scientific basis for deploying effective sand barriers to reduce the impact of sand damage to the railway and to guarantee the safety of railway operation.[Method] In this study, We focused on Nanshankou section of the Qinghai-Tibet railway, which was partially deployed with sand control measures. Four sand control measures and a control with no measures were established to compare their effectiveness. Area Ⅰ with no sand control measures was used as control, area Ⅱ was deployed with PE (polyethylene) checkerboard barriers (PCBs); area Ⅲ with sand fences (SFs) and PCBs; area Ⅳ with SFs; and area Ⅴ with sand blocking PE nets (SBPNs) and SFs. By measuring cross-section elements and depth of the sand deposition, the volume of sand deposition in different areas were calculated, respectively. The effectiveness of the sand control measures were evaluated using annual deposited-sand volume per unit width in different areas with different models of sand control measures as the main indicator, and the influence of their spatial distribution on sand blocking capability were analyzed. [Result] 1) Our investigation showed that no sand accumulation on the railway track and no evidently visible abrasion of the rail in the area I, indicating that the degree of sand damage in Nanshankou section was mild and no need to have sand barriers. Inappropriate sand barriers may lead to extra sand accumulation and causing damage to the railway embankment. 2) Annual deposited-sand volume per unit width in the area Ⅲ was 1.125 m³·m^-1, the highest among the 4 areas, indicating minimum average annual sand flux at Nanshankou area was 1.125 m³·m^-1. 3) Annual deposited-sand volume per unit width in the area Ⅳ was 0.978 m³·m^-1, accounting for approximately 87% of those in the area Ⅲ, indicating that SFs was an effective single sand control measure; 4) Annual deposited-sand volume per unit width in the area Ⅴ was 0.782 m³·m^-1, accounting for 70% of those in the area Ⅲ, indicating SBPNs did not enhance the amount of sand deposition, even weaken the effects of sand blocking of SFs,the main reason was that they were settled too close to each other and to the railway. 5) Considering the sand blocking effects and economic costs, our investigation showed that the optimum width of the PCBs in Nanshankou section was 13 m; and single measure of SFs could play an important role in the sand control of Nanshankou railway section, which cost less. [Conclusion] The sand barriers in Nanshankou area:the large width of the PCBs resulted in a waste of material and the optimum width of the PCBs was 13 m; SBPNs did not enhance the amount of sand deposition, even weaken the effects of sand blocking of SFs, because they were settled too close to each other and to the railway; SFs was an effective single sand control measure in the study areas. Moreover, our investigation showed that the degree of sand damage of Nanshankou section was mild and no sand barrier was needed. If necessary, single measure of SFs can be used for sand control for Nanshankou section with a low cost. The result can provide a basis for scientific arrangement of mechanical sand barriers in mild sand damage railway sections.

Key words: Qinghai-Tibet railway, sand control measures, sand accumulation, effectiveness of sand control, deposited-sand volume per unit width