从植被类型、植被物候、气候因子时滞效应等方面研究 2001 年以来河北山区归一化植被指数（normalized difference?vegetation?index，NDVI）演变规律，并将气候因子时滞效应引入多元线性回归方法中，识别气候因子和人 类活动对河北山区 NDVI 演变的贡献。结果表明：2001—2022 年河北山区 NDVI 呈持续增长趋势，平均增速为 0.0037/a，其中乔木林、灌木林和草地 3 种植被类型增速分别为 0.0035/a、0.0040/a 和 0.0038/a；受气候变化影响， 河北山区植被生长季变长，生长季开始时间平均提前 9d，而生长季结束时间仅提前 1d；降水对植被 NDVI 的影响 主要发生在当月，而气温和潜在蒸散发对植被 NDVI 的影响存在 1 个月的滞后性，考虑气候因子的时滞效应后，气 候变化和人类活动对河北山区 NDVI 演变的贡献分别为 39% 和 61%，与不考虑气候因子时滞效应对比，多元线性 回归的决定系数由 0.80 提高到 0.87。人工水保措施是河北山区植被 NDVI 增长的主导因素，对快速改善山区生 态环境至关重要，同时 2001 年以来降水、气温等气候因素也有利于植被恢复，对植被条件较好的区域，应以自然 恢复为主，实现自然条件下的生态平衡。研究结果对优化山区水土保持工作方案，提高植被修复措施效果具有重要意义。

Since 2000, large-scale comprehensive control of soil and water loss has been carried out in the mountainous areas of Hebei Province. Through measures such as enclosure protection, afforestation and grass planting, and return of farmland to forest, the ecological environment in the mountainous areas has been significantly improved, especially the vegetation coverage in the mountainous areas has been significantly increased. However, the current research has not systematically analyzed the vegetation evolution law and causes in the mountainous areas of Hebei Province. It was difficult to scientifically evaluate the comprehensive control of soil and water loss in mountainous areas of Hebei Province. When analyzing the causes of vegetation evolution, the traditional multiple linear regression method usually adopts the climate variables and vegetation variables corresponding to each month, and fails to consider the "time-lag effect" of climate factors, which may lead to underestimating the contribution of climate factors. The evolution of NDVI in mountainous areas of Hebei Province was systematically studied from the aspects of vegetation types, vegetation phenology and time delay effect of climate factors. On this basis, a multiple linear regression model was established considering the time delay effect of climate, which is expected to answer the cause of NDVI evolution more accurately. With precipitation P, temperature T and potential evapotranspiration PET as independent variables and vegetation NDVI as dependent variables, a multiple linear regression model was constructed to improve the traditional residual analysis method, and the effects of climate change and human activities on vegetation NDVI changes were stripped away. The results showed that NDVI in mountainous areas of Hebei showed a continuous growth trend from 2001 to 2022, with an average growth rate of 0.0037/a. The growth rates of woodland, shrub and grassland were 0.0035/a, 0.0040/a and 0.0038/a, respectively. Using the Mann-Kendall trend test method to analyze the significance of NDVI change, it was found that 78% of the regions showed a very significant increase trend, 11% showed a significant increase trend, and only less than 2% showed a significant decrease or an extremely significant decrease trend, indicating that the vegetation recovery effect of the whole Hebei mountain region after 2001 was remarkable. Due to the influence of climate change, the growth period of vegetation in mountainous areas of Hebei Province became longer, and the start time of the growing season was advanced by 9 days on average, while the end time of the growing season was advanced by only 1 day. SOS in Bashang Plateau, Yanshan District and Taihang Mountain area advanced by 6 d, 11 d and 8 d, and EOS advanced by 2 d, 1 d and 1 d, respectively, meaning that the whole growing season increased by 4 d, 10 d and 7 d, respectively. The influence of precipitation on vegetation NDVI mainly occurs in the same month, while the influence of air temperature and potential evapotranspiration on vegetation NDVI has a one-month lag. After considering the time-lag effect of climate, the contribution of climate change and human activities to the evolution of NDVI in mountainous areas of Hebei is 39% and 61%, respectively. After considering the delay effect of climate factors, the coefficient of determination R2 of multiple linear regression increases from 0.80 to 0.87, the root mean square error decreases from 0.086 to 0.071, and the average absolute error decreases from 0.108 to 0.090. The performance of the improved regression model is better than that of the original model.Since 2000, the main contribution of NDVI growth in mountainous areas of Hebei Province has been artificial vegetation restoration, but at the same time, the lengthening of growing season caused by climate change is also an important reason. Studying and identifying the impact of climate change and human activities has important guiding significance for further rational and efficient soil and water conservation in mountainous areas.