青海省是中国重要的生态安全屏障。准确评估青海省植被恢复潜力，对生态保护和高质量发展具有重要意义。以植被净初级生产力指数作为评价指标，借鉴相似生境法，通过K-means算法和箱线图分析，确定了各生境单元的植被理想生长状态，对2001—2022年青海省植被恢复潜力的空间分布状况及影响因素进行了深入研究。结果表明：（1）青海省植被恢复潜力指数空间差异性显著，总体呈西高东低、北高南低的分布特征。（2）2001—2022年，青海省植被恢复潜力指数整体呈现下降趋势，植被继续恢复的潜力在减小。（3）柴达木盆地东部、南部及共和盆地的植被恢复潜力指数较高，是青海省未来生态恢复的重点区域。（4）降水量、气温是青海省植被恢复潜力的主要驱动因子，且因子交互解释力大于单因子解释力。

As an important ecological security barrier in China， accurately assessing the vegetation restoration potential in Qinghai Province is of great significance for its future ecological protection and high-quality development. In this study， the vegetation net primary productivity index was used as the evaluation index， and with reference to the Similar Habitat Method， the K-means algorithm and box plot analysis were used to determine the ideal growth state of vegetation of each habitat unit， and investigated the spatial distribution and factors influencing the vegetation restoration potential in Qinghai Province from 2001 to 2022. The result shows that： （1） The spatial variability of the vegetation restoration potential index in Qinghai Province is significant， and the overall distribution is characterized by high in the west and low in the east， and high in the north and low in the south； （2） The vegetation restoration potential index in Qinghai Province shows an overall decreasing trend between 2001 and 2022. As a result， the potential for continued vegetation recovery is diminishing； （3） The eastern and southern parts of Qaidam Basin and Gonghe Basin have a high vegetation restoration potential index and are the key areas for future ecological restoration in Qinghai Province； （4） Driver analysis showed that precipitation and temperature were the main drivers of vegetation restoration potential in Qinghai Province， and the factor interaction explanatory power was greater than the single factor explanatory power.