It is challenging to restore Tibetan Plateau vegetation once it has been damaged. Highway construction, necessary for economic development, have significant negative impacts on the composition, structure, and functioning of the Qinghai-Tibet Plateau alpine grassland ecosystem. However, there is uncertainty regarding the extent of ecosystem damage in this region, such as whether quarries dug during highway construction may be restored and how recovery varies with increasing altitude. We conducted a transect survey of 39 quarry-restored grasslands reseeded with Elymus nutans and adjacent paired natural grasslands on the Gongyu and S308 highways on the Qinghai-Tibet Plateau (2800–5100 m above sea level). By comparing the community height, coverage, and species composition between restored and natural grasslands, we evaluated the degree of recovery and altitudinal gradient pattern of the grassland community and explained the limiting factors that led to the differences in recovery. After two-year restoration, quarry coverage and species composition were not restored to the degree of natural grassland. As altitude increased, the coverage recovery degree of quarry recovery community continued to decline, and the dissimilarity of community composition between the restored and natural grasslands was rising. We showed a breaking point at ca. 4100 m where recovery degree dropped dramatically, indicating a threshold for highway construction. The soil nutrients in quarry decreased significantly with increasing altitude. Structural equation modelling indicated that increasing altitude decreased the coverage of quarry by reducing temperature and soil nutrients. Due to the limitations of single species sowing, low temperature and nutrient availability, the effect of recovery decreased with increasing altitude. Therefore, we propose that during highway construction in the permafrost of the Qinghai-Tibet Plateau, quarries should avoid operation at high altitudes. When the altitude exceeds a certain height (e.g., 4100 m in this study), we do not recommend excavating quarries to destroy alpine grasslands.

The effects of human activities on heavy metal pollution in soil have been less investigated on the Tibetan Plateau. The present study was designed to assess the effects of highway traffic on Cu, Zn, Pb, and Cd enrichments in the 0–60-cm soil profile in the eastern Tibetan Plateau. Soils were sampled at four transects (with an altitude range of 2643–2911 m) across the G212 highway and five transects (3163–3563 m) across the G213 highway. Background concentrations of Cu, Zn, Pb, and Cd to the 60-cm soil depth (measured at each transect 400 m away from highways) varied greatly among transects and between highways. However, this spatial variation in the heavy metal concentrations was not related to the altitude of the investigated areas. On each the left and right sides of G212 or G213, Cu, Zn, and Pb concentrations to the 60-cm depth, at 5, 10, 20, and 50 m away from the highway, were all generally greater than the respective metal background concentrations. Cd concentrations to the 20 cm on G212 or 60-cm soil depth on G213 increased prominently within a distance of 20 m away from the highways, compared to background values in different depths. From the curb to 400 m away from highways, concentrations of Cu, Zn, Pb, and Cd were generally higher in the upper than in the lower soil layers. This may suggest that other factors such as atmospheric deposition were also contributable to the accumulation of heavy metals in soil. The contamination factor (C f ) calculation showed that roadside soils to the 60-cm depth, within a distance of 50 m from the curbs of both G212 and G213, were moderately (1 ≤ C f  < 3) contaminated with Cu, Zn, and Pb. The contamination from Cd mainly occurred (1 ≤ C f  < 8) in the top 20 cm soil with a closer distance from the highways. Our results indicated that traffic effects in enriching heavy metals reached 60-cm depth in roadside soils on the eastern Tibetan Plateau. For assessment of heavy metal pollutions in soil in mountainous areas, it is necessary to in situ identify the background values.