Water scarcity is a global environmental problem that jeopardizes human safety and socioeconomic development. In China, it has also become a potentially major obstacle for socioeconomic development, where 164 major groundwater areas are being exploited unsustainably. As a result, the depth of the groundwater table has increased by an average of 1.5 m per year in the arid and semi-arid regions of the north. Since 1952, China has implemented an unprecedented large-scale tree-planting program that focused on arid and semi-arid regions. The goal was to use trees to conserve water and combat desertification. Unfortunately, there is a serious risk that this program could exacerbate water shortages and lower the groundwater table because the trees were not chosen based on local environmental constraints, and their evapotranspiration is greater than the regional precipitation. Since precipitation is the major source of groundwater recharge in these semi-arid and arid areas of northern China, this imbalance will intensify the decline of the groundwater resource. Despite this risk, there has been limited research on the effect of afforestation on the groundwater table in China. In the present study, we selected nine provinces and provincial-level regions (Beijing, Hebei, Henan, Shanxi, Shaanxi, Ningxia, Inner Mongolia, Gansu, and Xinjiang) in China that focused on the tree-planting program, and are environmentally fragile arid or semi-arid regions facing a serious water scarcity. We calculated the influence of the afforestation program on groundwater based on two assumptions using seven evapotranspiration models, without considering evapotranspiration of the underlying vegetation or accounting for differences among the tree species used in afforestation. We confirmed that there is a serious risk that afforestation will cause the groundwater table to decline, independent of any other human withdrawals of this water. Based on our assumptions that afforestation plots had insignificant groundwater recharge or losses as a result of lateral inputs from adjacent land via subsurface or surface flows and that all planted trees survived, the groundwater table would decline most dramatically in Gansu, Ningxia, and Xinjiang regions. When we extended our analysis to assume that groundwater exchanges would possibly occur in land adjacent to the afforestation plots, and that this recharge is affected by the distance between the two plots, we found that the groundwater table in Beijing decreased most. Together, our results suggest that implementing such an afforestation program in arid and semi-arid areas without considering its influence on the groundwater supply will cause potentially severe damage to local ecosystems. In China, this will possibly compromise the goals of the national environmental policy, and damaging local socioeconomic conditions. In contrast, acknowledging the effects of this program will support efforts to utilize groundwater resources more efficiently, decrease the impact on the ecological services provided by these regions, and will mitigate the social and economic impacts on residents in these regions. In planning future ecological restoration practices, the impacts on the groundwater resource must be considered and calculated. In particular, environmental managers should select native tree species with high water-use efficiency and consider alternatives to trees, such as shrubs, sub-shrubs, and grassland or steppe vegetation. In addition, a broader variety of trees and other plant types should be considered; to determine an appropriate scale for afforestation that accounts for local moisture conditions, and determine methods to utilize the groundwater resource more efficiently and sustainably. This afforestation approach will improve the successful ecological restoration and sustainability of arid and semi-arid regions in China.