The land ecosystem, the area utilized by man for economic gain, is an essential part of the global terrestrial ecosystem. Anthropogenic changes to land use inevitably result in a variety of changes to the structure and ecological processes of ecosystems, which result in ecological risks. It is therefore important to quantify and evaluate the ecological risks driven by many different sources. Ecological risk assessment can provide a strong scientific basis for future research on the relationship between the ecological environment and anthropogenic effects, especially in the loess hilly-gully region of Mizhi County, Shaanxi Province, China. Land use data for 2009-2015 showed that seven types of land use can be categorized in this area:farmland, forest, grassland, orchard, water body, construction land, and wasteland. A 2 km×2 km grid was established as the auxiliary evaluation unit for this area, and ArcGIS was used as the data integration analysis platform. The agricultural background and socioeconomic development of the research area were taken into consideration and the Agricultural Ecological Risk Possibility (AERP) and Natural-Social Loss Index (NSLI) were used to construct a Comprehensive Ecological Risk (CER) evaluation system. The land use change characteristics in the research area were analyzed quantitatively and maps of comprehensive ecological risk were then generated using spatial interpolation. The spatial-temporal variation for ecological risk was also investigated. Finally, the relationship between ecological risk and terrain relief was examined using the terrain distribution index. Several conclusions were drawn from this analysis:1) although land use in the Mizhi area has experienced considerable change, farmland and grassland still accounted for more than 68% of the land use in this region. However, land use change data showed that there had been a reduction in farmland and an increase of forest and wasteland. The policy of returning farmland to forest or wasteland, a result of various economic factors, has led to a significant transformation of farmland to forests. 2) The study area was divided into five ecological risk classes using the natural breaks method. These were risk region I (CER < 0.095), risk region Ⅱ (0.095 ≤ CER < 0.139), risk region Ⅲ (0.139 ≤ CER < 0.183), risk region IV (0.183 ≤ CER < 0.24), and risk region V (CER > 0.24). Between 2009 and 2015, the comprehensive ecological risk index increased from 0.1466 to 0.1607, which was an increase of 9.65%. The ecological risk levels showed a tendency to increase during the study period, which indicated that ecosystems had become degraded. The ecological risk assessment showed that the comprehensive ecological risk level moved from the Ⅱ/Ⅲ level to the Ⅲ/IV level between 2009 and 2015. Spatial differences in comprehensive ecological risk were also significantly different in the research area. Regions with a higher risk level were mainly located towards the center of the research region, and are the key areas for ecological restoration and pattern regulation. 3) Ecological risk was closely related to terrain relief. Regions with lower risk levels were mainly distributed in areas with a higher relief, and vice versa. The results showed that at risk areas began to move on to the higher relief land during the study period. However, regions with higher risk levels were mainly concentrated in the lower relief areas.