Phenological trends of European beech stands along the Carpathian arc: a 20-year MOD13Q1/MYD13Q1 based analysis

European forests are severely threatened by rapid global climate change. Predicting the effects of climate change on the future performance of tree species can be enhanced by using geographical gradients as space-time proxies. This study focuses on analysing seasonal and inter-annual phenological trends for eight sites of European beech (<i>Fagus sylvatica</i> L.) in the Carpathians along such geographical gradient. Using time series of the Enhanced Vegetation Index (EVI) from 2003 to 2022, derived from 16-day MODIS Terra and Aqua vegetation products, combined with meteorological parameters (temperature and precipitation) interpolated for a grid and Land Surface Temperature (LST) data from Landsat, we compared annual forest development at the study sites and extracted key phenological metrics (start, length, and end of the growing season) using the phenofit software package for R. Our results show varied responses of the studied forest sites to climate change. The end of season extended significantly at Soveja (slope = 0.50, <i>p</i> = 0.023). Spring temperatures had negative correlations with the start of season (<i>r</i> = -0.37) and autumn temperatures positive correlation with the end of season (<i>r</i> = 0.43). Understanding these patterns is crucial for developing adaptive forest management strategies in Carpathians.

欧洲森林正遭受全球快速气候变化的严重威胁。借助地理梯度作为时空替代指标，可有效提升对气候变化影响树木物种未来生长表现的预测精度。本研究聚焦于分析喀尔巴阡山脉内8处欧洲山毛榉（<i>Fagus sylvatica</i> L.）样地沿此类地理梯度的季节间与年际物候变化趋势。本研究采用2003至2022年的增强型植被指数（Enhanced Vegetation Index, EVI）时间序列数据，该数据源自16天合成的MODIS Terra与Aqua植被产品；同时结合网格插值得到的气象参数（气温与降水量）以及Landsat遥感影像反演的地表温度（Land Surface Temperature, LST）数据，对各研究样地的年际森林生长动态进行对比，并通过R语言的phenofit软件包提取了关键物候指标：生长季起始期、持续时长与终止期。研究结果表明，各受试森林样地对气候变化的响应存在显著差异。其中Soveja样地的生长季终止期显著延长（斜率=0.50，*p*=0.023）。春季气温与生长季起始期呈显著负相关（*r*=-0.37），秋季气温与生长季终止期呈显著正相关（*r*=0.43）。明晰此类物候变化规律，对制定喀尔巴阡地区适应性森林管理策略至关重要。