Global concern about the restoration of vegetation ecosystems has recently increased. Potential natural vegetation (PNV) and climate adaptation concepts should be integrated into revegetation programs to achieve sustainable ecosystems. The Yanhe Basin in the Loess Plateau of China (7687 km(2)) has been subjected to intense human activity for centuries. It was selected as the study area because vegetation degradation and restoration are occurring there. The objectives of this study were to (1) evaluate whether the current vegetation pattern is appropriate, and (2) provide a restoration plan for future revegetation programs based on PNV and habitat suitability patterns simulated by the dynamic vegetation model LPJ-GUESS. Current work focused on the parameter calibration and high-resolution climate data of the model over the study area. The comparisons of model performances in the PNV pattern and productivity indicated that parameter calibration was necessary for the model application. PNV of the Yanhe Basin may shift by 21.37-29.67% from 1981-2010 to 2071-2100, mainly in the southern part. Forests may decrease and steppes may increase as the climate becomes drier in the future. Comparisons between an existing land use map and the current PNV pattern indicated that only 40.8% of the forestland was coincident with the current PNV pattern, whereas grassland was a more suitable vegetation type for the rest of the terrain. In contrast, 83.1% of the grassland aligned with the current PNV pattern. Therefore, 16.9% remains to be forested. Current forestland and grassland patterns should be adjusted to cope with future climate change. Broadleaf summer-green shrubs covered a larger area and had higher habitat suitability than forests; they might be the most suitable woody plants for revegetation of the Yanhe Basin. The applied research approach could be extended to other regions undergoing similar revegetation programs and help promote sustainable vegetation management.