The agro-pastoral ecotone of northern China is one of the important barriers for Chinese ecological security. However, the region is facing the serious contradictions between the economic development and the conservation of the fragile environment. Quantitative assessment on ecosystem services and the influence factors is the significant foundation for ecological management and sustainable development, and is also one of the hot spots for the current research. Taking the central and western Inner Mongolia as study area, which is the core area of the agro-pastoral ecotone of northern China, this study simulated the regional key ecosystem services during 2000-2018 using the remote sensing data with medium or high resolution and the biophysical models. The models included the Carnegie-Ames-Stanford Approach, the Distributed Hydrology Soil Vegetation Model, the Revised Universal Soil Loss Equation, and the Revised Wind Erosion Equation. The Getis-Ord G^* method was adopted to explore the spatial hot and cold spots of the ecosystem services. With the consideration of the spatial autocorrelation in the geographic data, the spatial regression models were selected to explore the influence factors for the spatial patterns of the ecosystem services. The results indicated that net primary productivity (NPP), soil conservation (SC), water yield (WY), soil moisture of top layer (SM1), soil moisture of middle layer (SM2), and soil moisture of bottom layer (SM3) in the eastern area were higher than those in the other areas. While sand fixation (SF) was higher in the southern and western areas, which were usually covered by sandy topsoil. Secondly, the key ecosystem services were not only influenced by changes in climatic factors and landscape patterns, but also by topographical conditions and soil texture. The influences of the factors were different for different ecosystem services. For NPP, the landscape metrics of forests and the Normalized Difference Vegetation Index had more significant effects than the other influence factors. Precipitation was one of the most significant influence factors for WY, SM1, SM2, SM3, and SC. While SF was mainly influenced by wind speed and sand percent of the topsoil. The study revealed the spatial characteristics of the key ecosystem services in the typical agro-pastoral ecotone of northern China, as well as the influence factors behind the spatial changes. The conclusions in this study were able to provide the quantitative scientific support and case reference for how to alleviate the contradictions between the economic development and the environmental conservation in the agro-pastoral ecotone of northern China and other similar areas through landscape planning and land optimization.