This study presents daily high-resolution (5 km × 5 km) grids of mean, minimum, and maximum temperature and relative humidity for Germany and its catchment areas, from 1951 to 2015. These observational datasets (HYRAS) are based upon measurements gathered for Germany and its neighbouring countries, in total more than 1300 stations, gridded in two steps: first, the generation of a background field, using non-linear vertical temperature profiles, and then an inverse distance weighting scheme to interpolate the residuals, subsequently added onto the background field. The modified Euclidian distances used integrate elevation, distance to the coast, and urban heat island (UHI) effect. A direct station-grid comparison and cross-validation yield low errors for the temperature grids over most of the domain and greater deviations in more complex terrain. The interpolation of relative humidity is more uncertain due to its inherent spatial inhomogeneity and indirect derivation using dew point temperature. Compared with other gridded observational datasets, HYRAS benefits from its high resolution and captures complex topographic effects. HYRAS improves upon its predecessor by providing datasets for additional variables (minimum and maximum temperature), integrating temperature inversions, maritime influence and UHI effect, and representing a larger area. With a long-term observational dataset of multiple meteorological variables also including precipitation, various climatological analyses are possible. We present long-term historical climate trends and relevant indices of climate extremes, pointing towards a significantly warming climate over Germany, with no significant change in total precipitation. We also evaluate extreme events, specifically the summer heat waves of 2003 and 2015.

这项研究显示了1951年至2015年德国及其集水区的每日高分辨率（5 km×5 km）平均，最低，最高温度和相对湿度的网格。这些观测数据集（HYRAS）基于为德国及其邻国总共有1300多个站点，分两步进行网格划分：首先，使用非线性垂直温度分布图生成背景场，然后使用反距离权重方案对残差进行插值，随后将其相加背景字段。使用的修改后的欧几里得距离综合了海拔，到海岸的距离以及城市热岛效应（UHI）。直接的站网比较和交叉验证在大部分区域内的温度网格误差较小，而在较复杂的地形中偏差较大。相对湿度的内插因其固有的空间不均匀性和使用露点温度的间接推导而更加不确定。与其他网格化观测数据集相比，HYRAS得益于其高分辨率，并捕获了复杂的地形效应。HYRAS通过提供其他变量（最低和最高温度）的数据集，整合温度反演，海事影响和UHI效应并代表更大的面积来改进其前身。使用包括降水在内的多个气象变量的长期观测数据集，可以进行各种气候分析。我们介绍了长期的历史气候趋势和极端气候的相关指标，指出德国的气候明显变暖，总降水量没有明显变化。我们还评估了极端事件，特别是2003年和2015年夏季的热浪。