To study the tensile strength of rock under different loading rates, direct tensile test is the most accurate method. However, the eccentric tension in the process of rock direct tensile test has a significant influence on the test results. In this paper, firstly, a self-developed centering device for rock direct tensile test is introduced, which can effectively eliminate the eccentric tension in the process of rock direct tensile test. Then, with the aid of the self-developed centering device, the direct tensile tests of red sandstone under the loading rates of 0.001 mm/s, 0.01 mm/s and 0.1 mm/s are successfully carried out. After tests, both the macro failure characteristics and the scanning electron microscope micrograph show that the fracture pattern of the rock is caused by pure tensile loading. The stress-strain curves of the direct tensile test of the red sandstone show that the process of the direct tensile test can be roughly divided into four stages. With the increase of loading rate, both of the tensile strength and the peak tensile strain of the rock increase obviously. The direct tensile test of the red sandstone shows obvious loading rate effect.
Identifying and analyzing the urban–rural differences of social vulnerability to natural hazards is imperative to ensure that urbanization develops in a way that lessens the impacts of disasters and generate building resilient livelihoods in China. Using data from the 2000 and 2010 population censuses, this study conducted an assessment of the social vulnerability index (SVI) by applying the projection pursuit cluster model. The temporal and spatial changes of social vulnerability in urban and rural areas were then examined during China’s rapid urbanization period. An index of urban–rural differences in social vulnerability (SVID) was derived, and the global and local Moran’s I of the SVID were calculated to assess the spatial variation and association between the urban and rural SVI. In order to fully determine the impacts of urbanization in relation to social vulnerability, a spatial autoregressive model and Bivariate Moran’s I between urbanization and SVI were both calculated. The urban and rural SVI both displayed a steadily decreasing trend from 2000 to 2010, although the urban SVI was always larger than the rural SVI in the same year. In 17.5% of the prefectures, the rural SVI was larger than the urban SVI in 2000, but was smaller than the urban SVI in 2010. About 12.6% of the urban areas in the prefectures became less vulnerable than rural areas over the study period, while in more than 51.73% of the prefectures the urban–rural SVI gap decreased over the same period. The SVID values in all prefectures had a significantly positive spatial autocorrelation and spatial clusters were apparent. Over time, social vulnerability to natural hazards at the prefecture-level displayed a gathering–scattering pattern across China. Though a regional variation of social vulnerability developed during China’s rapid urbanization, the overall trend was for a steady reduction in social vulnerability in both urban and rural areas.
