Background The Qinghai-Tibet Railway (QTR) runs 1956 km from Xining to Lhasa. It is the longest altiplano railway in the world and at the highest altitude. It is also a central project in the strategy of "Western Development" in China. However, the QTR has been under constant threat from aeolian sand since its opening, and the Nanshankou section is one of the worst areas along the railway line. In order to build a scientific and efficient protection system against wind-sand disaster in Nanshankou, we conducted field investigation and wind tunnel experiment to reveal the mechanism of wind sand disasters.

Methods The allocation model of protection system and volume of sand sediment around railway line was measured through the field investigation. Taking the roadbed and control measures as prototypes, scaled models were made for wind tunnel simulation. The speed of inflow in wind tunnel was kept at 15 m/s, which was highest frequency during the local windy season. And the KIMO hotwire anemometer was used to measure the wind speed in the working section of wind tunnel. The simulation research was made up of three parts:We firstly observed the flow field around the simulated roadbed to explore the vulnerability of the roadbed to windblown sand. Then, the simulated control measures was disposed around the roadbed to reveal the influence of current protection system on flow field. Finally, the efficiency and defects of current protection system were analyzed based on the modification simulation and distribution of sand sediment.

Results The results indicate that:1) Roadbed was an obviously obstacle in Nanshankou. The low-speed area occurred at the foot of windward slope, and the high-speed area occurred at the shoulder of windward side and the top of roadbed. These two areas were respectively responsible for sand deposition and wind erosion, harming the safety of the QTR. 2) In the current protection system, the distance between PE fences were 19 m and 20 m, respectively. It is too far for sediment precipitation. Hence, the sand sediment proportion blocked by the second and third PE fence was only 2.42% and 0.24%, respectively. 3) The wind tunnel experiment recommended that, the distance between each PE fence should be reduced to increase the efficiency. More sleeper typed fences should be added to the west side of current protection system to increase the width of protection distance and block the sand at outer space as much as possible.

Conclusions This study reveal the characteristics of windblown sand near the railway, we put forward and discussed the possible improvement program. These improved information are essential to provide a firm basis for desertification control along the QTR, and the results of this paper may also be used in other areas where experience threats from windblown sand.