According to the “exposure-sensitivity-adaptive ability” criterion of vulnerability assessment, the entropy weight method, weighted summation model and spatial regression model are used to explore temporal and spatial differentiation and influencing factors of vulnerability for urban “social-economic” system in the Loess Plateau from 2005 to 2015.The results show that the urban “social-economic” system in the Loess Plateau is moderately fragile and in a benign orbit. The vulnerabilities of the “social-economic” systems of different types of cities show different evolutionary characteristics. The vulnerability of resource-based cities and most urban agglomerations declines. The provincial capitals face strong risk coercion due to the polarization effect. The vulnerability index in the Loess Plateau has gradually increased from east to west. The large topographic unit as an important part of the natural environment has great significant impact on vulnerability differentiation. The hotspots region has always been concentrated in the Gully area of Loess Plateau in the eastern part of Qinghai Province. From the perspective of vulnerability dimension assessment, it is clearly indicated that the core issue of urban development in the Loess Plateau has been transformed from promoting economic growth to improving sustainability, that is, the construction of both software and hardware infrastructure, improving economic efficiency, regulating urban-rural imbalances, and reducing resource and environmental pressures for economic development.

Taking the 341 county-level administrative units in the Loess Plateau as the study subject, the characteristics of geographical differentiation and formation mechanism of the county urbanization from 1990 to 2015 were analyzed systematically using the methods of statistical analysis, zoning analysis and geography detectors. The results show that the urbanization level of the county in the Loess Plateau showed a steady upward trend, and the urbanization rate increased from 23.83% to 54.33% from 1990 to 2015. Geographically, the macro-pattern was low in the west and high in the east and middle regions. The centrality of provincial capitals and prefecture-level cities and other high-urbanized counties was obvious. Area of high urbanization rate centered on the Ordos Plateau being gradually strengthened. Since 1990, the regional differences of population agglomeration and socio-economic development of county have shown a decreasing trend, and the balanced development trend of county urbanization county in the Loess Plateau has been obvious. In geographical division, the regional differences of county economic development in the wind-sandy area of Loess Plateau have increased significantly, while the regional differences in population agglomeration and economic development in other districts tend to shrink. There are many factors affecting the formation and evolution of county urbanization differentiation pattern, mainly including economic development, proportion of non-agricultural industry, per capita total retail sales of consumer goods, per capita output of grain, population density, the shortest travel time to central cities and urban per capita disposable income. Therefore, regional urbanization should be scientifically positioned and promoted according to the geographical differences in natural environment and social economy. The relationship between urbanization construction, agricultural production and ecological environmental protection should be coordinated through regional function optimizing, industrial transforming and upgrading, population migration and agglomeration, infrastructure construction and sharing, and ecological compensation, to achieve the goals of steady improvement of county urbanization and high-quality development of urban and rural construction in Loess Plateau in the new era.

The methods of ArcGIS spatial analysis, nearest neighbor analysis, kernel density estimation and regression analysis are used to study the spatial and temporal distribution characteristics of rural settlements and its influencing factors of Huangling County in Weibei Upland, and reconstruction ideas were put forward for different residential areas, based on the remote sensing images in 2000, 2010, 2015 and DEM data.The main results showed that the number and area of rural settlement patches increase firstly and then decrease, with obvious characteristics of stage change of Huangling County. The rural settlements have obvious spatial agglomeration characteristics, and the agglomeration scope is constantly expanding. The layout of rural settlements has a strong dependence on topography, rivers and roads. At the same time, factors such as population and economic development, industrialization and urbanization, and policy control have increasing influence on the layout of rural settlements. The spatial reconstruction of rural settlements should be based on the objective conditions and functions of all kinds of residential areas. Scattered residential areas in ecologically fragile areas should pay attention to ecological migration. Reserved residential areas should improve the level of intensive land use. Central residential areas should strengthen their hub effect of connection and serve the urban and rural areas.

Taking Huangling County in Weibei Upland as a case study area, based on the data of questionnaires and in-depth interviews, the similarities and differences between land-lost and normal farmers′ sense of place in land dependence and local attachment under the background of land circulation were compared, and the influencing factors of their sense of locality were analyzed from the perspectives of demographic characteristics and external environment. The results show that farmers understand the place relying on land, and the land dependence and local attachment of normal farmers are stronger than those of land-lost farmers, but the factor scores of each dimension are different. The core of the perception network of land-lost and normal farmers lies in the dimension of land dependence, land-lost farmers perceive negatively, while normal farmers perceive positively. There are significant differences between land-lost and normal farmers in influencing factors of sense of place. Increase in income will help promote sense of place in land dependence, and improvement in education and external environment will help promote local attachment.

Based on the data of industrial enterprises above designated size in China, using spatial statistical methods such as standard deviation ellipse, Ripley′s function analysis and nuclear density estimation, the spatial differentiation patterns of foreign investment enterprises in the Loess Plateau in 2006, 2010 and 2015 were measured.Geographic detector was used to analyze the geographical factors of the spatial distribution of foreign investment enterprises in the Loess Plateau. The results show that the spatial distribution of foreign investment enterprises in the Loess Plateau firstly spreads to the ellipse center and then spreads along the long axis.The distribution center has been moving to the southwest, and the central point has been in the territory of Shaanxi Province. The scope of agglomeration is first increased and then decreased, and the strongest range of agglomeration trend of enterprise points is also a process of increasing from small to large.High-value core areas and higher-value sub-core areas are gradually expanding, forming a “several” word. The spatial distribution pattern of the type, and the adjacent areas in the periphery of the core area, there is a distribution of the sub-core group in the strip. The factors affecting the spatial layout of the foreign investment enterprises in the Loess Plateau have experienced from the single factor of the openness to the multiple factors, such as, the level of industrialization and technology, innovation capability and openness.

The “Hu Line” is a significant geographical boundary of China′s population and economy distribution. Tourism has strong liquidity and economic driving ability. Whether domestic tourism can take the lead in breaking through the "Hu Line" is an important step to achieve balanced economic development between China′s eastern and western regions. Based on the tourist traffic data of National Day Golden Week tourism and annual domestic tourism in 2005-2017, through the quartile space grading, comparing concentration and difference index and measuring the track of gravity center, the spatial structure changes of National Day Golden Week tourism and annual domestic tourism are analyzed. The results indicate that National Day Golden Week tourism and annual domestic tourism present obvious “Hu Line” spatial distribution characteristics. Over the years, the tourist size of both increased and it has demonstrated a trend of expanding westward. The regional gap of National Day Golden Week tourism and annual domestic tourism tends to be flat, and the latter flattening trend is more apparent, but the cumulative market share difference on both sides of the “Hu Line” is still relatively large. The gravity centers of National Day Golden Week tourism and annual domestic tourism are shifting to southwest region, and the annual is sharper than the Golden Week. Combining population, economy distribution and their trade-offs, three standards for domestic tourism to span the “Hu Line” are set. According to time series analysis and forecast, domestic tourism has crossed the economy boundary (4%) in 2016. It is expected that it will break through the population-economy balance boundary (5%)in 2020-2021, and it will surmount the population boundary (6%) in 2023-2024.

The concept of international influence of outbound tourism was put forward.Based on tourist flows and outbound tourism expenditure of China in recent 20 years and the tourism data of 130 destination countries for China′s outbound tourism in 2012-2016, taking the market share and growth contribution as indicators to measure the spatial and temporal changes of China′s outbound tourism influence. The results found that in the past 20 years, market share and growth contribution of China′s outbound tourism have been continuously improved. After 2010, it surpassed the traditional outbound tourism powers and ranked first in the world. Among the destination countries, the countries with higher market share and growth contribution of Chinese tourists are mainly distributed in the Asia-Pacific region, the market share is above 10% and the growth contribution is over 20%.The phenomenon of spatial agglomeration is obvious. The strong influence areas of China′s outbound tourism are mainly distributed in the Asia-Pacific region, which has the greatest influence on the surrounding countries, especially Japan and South Korea, ASEAN 10 countries, it also has some influence on the United States, some European and African countries, and has a general influence on Western Europe, Southern Europe, America and African countries.