In the context of “ecological civilization construction” in China, it is important to scientifically assess the landscape ecological risks (LER) in eco-sensitive regions. This is crucial to identify spatiotemporal variations and drivers, develop ecological security patterns, and achieve spatial protection and governance of the region. We selected Ankang City in the Qinling-Daba Mountains region and divided the region into 921 ecological risk assessment units using remotely sensed land use data and landscape pattern indexes from the 2000-2020 period. A landscape ecological risk index (LERI) model was constructed using geostatistics, spatial statistics, and boosted regression tree analysis for quantitative assessment of spatiotemporal variations in LERIs and their drivers. During 2000-2020, the land use structure underwent considerable changes in woodland and construction land areas. Woodlands were mainly converted from cropland and grassland, as well as construction land from cropland, grassland, and water bodies; the LERI decreased from 0.1627 to 0.1609. Spatially, the LERI level was higher in the northwest and lower in the southeast, higher in the middle, and lower in the surrounding areas. The LERI gravity centers were detected in Hanbin District, with trends of southeast to northwest shift and concentrated contraction. The contraction trend of the major axis was stronger than that of the minor axis. Mean annual temperature, GDP, mean annual precipitation, drought index, and soil total phosphorus were the primary drivers of LERI. Furthermore, the effects of human economic activities on LERI show a trend of decreasing and then increasing over time. These findings can aid policy and decision-making for sustainable development in the Qinglin–Daba Mountains region.