Abstract Present investigations aimed at developing a unified cumulative degree day based model approach allowing for (i) a precise simulation of grapevine phenological development of a broad range of cultivars and (ii) a classification of the cultivar specific relative precocity of different cultivars at different stages of grape development. Based on a long-term (7-year) data set of high-resolution phenological observations for 11 cool climate cultivars originating from Remich/Luxembourg, the unified phenological model (UniPhen) was developed using a cumulative degree day approach with three temperature thresholds (lower threshold: 10 °C; upper threshold: 20 °C; heat threshold: 30 °C). The average normalized standard deviation of UniPhen over all stages and cultivars was 5.26 corresponding to 5.26 days at 15 °C. In the cross-validated model, on average 53% of the observations were located in a range of ± 3 days and 82% in a range of ±7 days (assuming daily mean temperatures of 15 °C) around the predicted date. The sequence of the cultivar-specific relative precocity was not stable over the different stages of phenological development. The model approach (i) enables a precise simulation of all 31 BBCH stages between beginning of the bud swell (01) and berries being ripe for harvest (89) for 11 cultivars grown in the climatic conditions of the Luxembourgish grapegrowing region and (ii) is open to being extended to other grape cultivars and re-calibrated under deviant climatic conditions. Due to their instability, a classification of the cultivars’ relative precocity is only reasonable if it relates to a specific phenological stage. UniPhen represents a precise high-resolution phenological model approach and might be applied as a bioclimatic indicator describing the suitability of a location/region for the cultivation of specific cultivars.

中文翻译：

---

UniPhen – 一种统一的高分辨率模型方法，用于模拟广泛葡萄品种的物候发育以及潜在的新生物气候指标

摘要 目前的调查旨在开发一个统一的基于累积度日的模型方法，允许 (i) 精确模拟广泛品种的葡萄树物候发育和 (ii) 不同品种在不同阶段的品种特定相对早熟分类葡萄的发展。基于对源自 Remich/卢森堡的 11 个凉爽气候品种的高分辨率物候观测的长期（7 年）数据集，使用具有三个温度阈值的累积度日方法开发了统一物候模型 (UniPhen)。阈值下限：10 °C；阈值上限：20 °C；热阈值：30 °C）。UniPhen 在所有阶段和品种的平均标准化标准偏差为 5.26，对应于 15°C 下的 5.26 天。在交叉验证模型中，平均 53% 的观测位于预测日期附近的 ± 3 天范围内，82% 位于 ±7 天的范围内（假设每日平均气温为 15 °C）。品种特异性相对早熟的序列在物候发育的不同阶段并不稳定。模型方法 (i) 能够精确模拟在卢森堡葡萄种植区气候条件下种植的 11 个品种的芽膨 (01) 开始和浆果成熟收获 (89) 之间的所有 31 个 BBCH 阶段，以及 (ii)可以扩展到其他葡萄品种并在异常气候条件下重新校准。由于它们的不稳定性，品种相对早熟的分类只有在涉及特定物候阶段时才是合理的。