我国东部季风区主要流域气候变化情景下日平均温度和日降水量随机模拟数据集
收藏资源简介：
本数据集是基于历史观测日值资料和耦合模式比较计划第五阶段（CMIP5）的15个全球气候模式（GCM）对历史排放情景和三种未来典型浓度路径（RCP2.6、RCP4.5和RCP8.5）的模拟月值结果，对我国东部季风区海河、淮河、辽河、太湖、渭河和珠江六个流域日平均温度和日降水量进行随机模拟的结果。首先应用广义加性模型（GAM）分别建立各个GCM的统计降尺度模型，然后再运用贝叶斯模型平均（BMA）方法集成15个降尺度模型，最后通过随机模拟得到各流域在历史和未来排放情景下日平均温度和日降水量的样本集合。数据集能够反映局地的极端温度、极端降水等位于气候态分布尾端的小概率事件，可用于水文、生态和环境等领域的工程设计、灾害风险评估等。数据集的空间分辨率为0.5°X0.5°，覆盖区域为：海河流域（Haihe）112.5°E‒119.5°E，34.5°N‒42.5°N（栅格中心点经纬度，下同）；淮河流域（Huaihe）111.5°E‒121.5°E，30.5°N‒36.5°N；辽河流域（Liaohe）116.5°E‒125.5°E，38.5°N‒45°N；太湖流域（Taihu）118.5°E‒121.5°E，29.5°N‒32.5°N；渭河流域（Weihe）104°E‒110°E，33.5°N‒37.5°N；珠江流域东部（Zhujiang-east）109.5°E‒116.5°E，20.5°N‒27.5°N；珠江流域西部（Zhujiang-west）102.5°E‒109°E，20.5°N‒27.5°N。历史排放情景（historical）为1970‒2000年，RCP情景（RCPx.x）为2020‒2050年。数据存储为NetCDF（.nc）格式。数据文件以“[变量]_[流域]_[情景]_[起始年]-[终止年].nc”的格式命名，其中[变量]为日平均温度（tas）或日降水量（pr），以四维数组的形式存储。四个维度分别为集合成员（ens）、时间（time）、纬度（lat）和经度（lon）。每个变量集合均包含50个成员。另提供集合的第5、10、50、90和95百分位数（pr只用后三个），变量名为q5、q10、q50、q90和q95，都是维度为time、lat、lon的三维数组。数据文件可由GrADS软件直接读取。原始文件数56个，数据量25.4 GB（压缩为20个文件，11.3 GB）。
This dataset is derived from stochastic simulations of daily mean air temperature (tas) and daily precipitation (pr) for six river basins in the eastern monsoon region of China, including the Haihe, Huaihe, Liaohe, Taihu, Weihe, and Pearl River basins. The simulations are based on historical daily observational data and monthly simulation outputs from 15 General Circulation Models (GCMs) under the Coupled Model Intercomparison Project Phase 5 (CMIP5) for the historical emission scenario and three future Representative Concentration Pathways (RCP2.6, RCP4.5, and RCP8.5). First, statistical downscaling models for each GCM were established separately using the Generalized Additive Model (GAM). Then, the 15 downscaling models were integrated via the Bayesian Model Averaging (BMA) method. Finally, an ensemble of daily mean air temperature and daily precipitation samples for each basin under historical and future emission scenarios was obtained through stochastic simulation. This dataset can capture rare events at the tails of the climatic distribution, such as local extreme temperature and extreme precipitation, and can be applied to engineering design, disaster risk assessment, and other research in hydrology, ecology, environmental science, and related fields. The spatial resolution of the dataset is 0.5° × 0.5°, and the covered regions are as follows: - Haihe River Basin (Haihe): 112.5°E–119.5°E, 34.5°N–42.5°N (geographic coordinates of grid cell centers, the same below) - Huaihe River Basin (Huaihe): 111.5°E–121.5°E, 30.5°N–36.5°N - Liaohe River Basin (Liaohe): 116.5°E–125.5°E, 38.5°N–45°N - Taihu Lake Basin (Taihu): 118.5°E–121.5°E, 29.5°N–32.5°N - Weihe River Basin (Weihe): 104°E–110°E, 33.5°N–37.5°N - Eastern Pearl River Basin (Zhujiang-east): 109.5°E–116.5°E, 20.5°N–27.5°N - Western Pearl River Basin (Zhujiang-west): 102.5°E–109°E, 20.5°N–27.5°N The historical emission scenario (historical) covers the period 1970–2000, while the RCP scenarios (RCPx.x) cover 2020–2050. All data are stored in NetCDF (.nc) format. The data files are named in the format "[variable]_[basin]_[scenario]_[start_year]-[end_year].nc", where [variable] is either daily mean air temperature (tas) or daily precipitation (pr). The data are stored as four-dimensional arrays with dimensions: ensemble member (ens), time (time), latitude (lat), and longitude (lon). Each variable ensemble contains 50 members. Additionally, the 5th, 10th, 50th, 90th, and 95th percentiles of the ensemble are provided (only the last three percentiles are available for pr), with variable names q5, q10, q50, q90, and q95. These are all three-dimensional arrays with dimensions time, lat, and lon. The data files can be directly read using the GrADS software. There are 56 original data files with a total size of 25.4 GB (compressed to 20 files with a total size of 11.3 GB).
