Global Atmospheric Rivers catalog for ERA5 reanalysis
收藏资源简介：
The dataset contains hourly detected Atmospheric Rivers (ARs) for ERA5 reanalysis (1979 – 2021). For the detection of ARs, the global detection algorithm introduced by Guan and Waliser (2015) (first version v1), and refined by Guan et al. (2018) (second version v2) is applied. The algorithm considers different requirements: the intensity of integrated water vapor transport IVT, direction, and geometry (described in detail by Guan and Waliser (2015), Guan et al. (2018), and Lauer et al. (2023)). (1) IVT intensity: The IVT must exceed the 85th percentile of IVT for each grid cell and the lower limit of 100 kg m⁻¹ s⁻¹. (2) IVT direction: The IVT direction at individual grid cells have to be coherent, and the direction of object-mean IVT has to be within 45° of the AR shape orientation, with an appreciable poleward component. (3) Geometry: The length has to be larger than 2000 km, and the length-to-width ratio should be higher than two. In case an object exceeds the IVT percentile, but (2) and (3) are not fulfilled, the process is repeated for higher IVT thresholds (up to the 95th percentile with 2.5 steps). Thus, an object surrounded by an increased moisture content can be detected as an AR. To apply the detection algorithm, the input variables - zonal and meridional components of the IVT (IVTx and IVTy) and the IVT percentiles - are first calculated using ERA5 reanalysis. When these variables have been inserted and the algorithm has been applied, an nc-file is output. This nc-file includes the AR shape, axis, geometric characteristics such as length and width, the coordinates of the AR's head, tail, and centroid, mean of zonal and meridional IVT, the direction of mean IVT, and the landfall location (if the AR made landfall).
本数据集涵盖1979年至2021年ECMWF第五代再分析资料（ERA5）的逐小时大气河流（Atmospheric Rivers, ARs）检测结果。大气河流的检测采用Guan与Waliser（2015）提出的全球检测算法（第一版v1），并经Guan等人（2018）优化为第二版v2。该算法需满足三项判定要求：积分水汽输送（Integrated Water Vapor Transport, IVT）强度、方向与几何特征（详细说明可参见Guan与Waliser 2015、Guan等人2018以及Lauer等人2023的研究）。(1) IVT强度：各网格点的IVT需同时超过该网格点IVT的第85百分位数，且不低于100 kg·m⁻¹·s⁻¹的下限阈值。(2) IVT方向：单个网格点的IVT方向需保持连贯，且物体平均IVT方向与大气河流形态走向的偏差需控制在45°以内，同时需具备显著的向极输送分量。(3) 几何特征：大气河流的长度需大于2000 km，且长宽比需大于2。若某一物体满足IVT百分位数阈值要求，但未满足(2)(3)两项判定条件，则将采用更高的IVT阈值（步长为2.5，最高至第95百分位数）重复检测流程，以此可将被湿度含量升高区域环绕的物体识别为大气河流。 为运行该检测算法，需首先基于ERA5再分析资料计算输入变量：IVT的纬向分量与经向分量（IVTx与IVTy）以及IVT百分位数。将上述变量输入算法并完成检测后，将输出NetCDF（nc）格式文件。该文件包含大气河流的形态、轴线、几何特征（如长度与宽度）、大气河流的头部、尾部与质心坐标、纬向与经向IVT平均值、平均IVT方向，以及登陆位置（若该大气河流发生登陆）。
