COAST-HG: A global COastal dAtaset of Storm Tide HydroGraphs

Coastal flooding is driven by both high tides and/or storm surge, the latter being caused by strong winds and low pressure in tropical and extratropical. The combination of storm surge and the astronomical tide is defined as the storm tide. To gain understanding into the threat imposed by coastal flooding and to identify areas that are especially at risk, now and in the future, it is crucial to accurately model coastal inundation and assess the coastal flood hazard. Most models capable of simulating coastal inundation at the global scale follow a simple planar approach, often referred to as bathtub models. The main limitations of this type of model are that they implicitly assume an infinite flood duration and do not capture relevant physical processes. In this study we develop a method to generate hydrographs called HGRAPHER, and provide a global dataset of storm tide hydrographs. These hydrographs represent the typical shape of an extreme storm tide at a certain location along the global coastline. We test the sensitivity of the HGRAPHER method with respect to two main assumptions that determine the shape of the hydrograph, namely the surge event sampling threshold and coincidence in time of the surge and tide maxima. These hydrographs can be used to move away from planar to more advanced dynamic inundation modelling techniques at large scales. The dataset consists of storm tide hydrographs corresponding to the 1-in-100 year RP. The storm tide levels correspond to the COAST-RP dataset. There are two hydrographs available per output location, one that represents average tide conditions and one for spring tide conditions.

海岸洪水由高潮位与（或）风暴潮共同驱动，其中风暴潮由热带及温带（extratropical）区域内的强风与低压环境引发。风暴增水与天文潮的叠加被定义为风暴潮位（storm tide）。为理解当前及未来海岸洪水所带来的威胁，并识别高风险区域，精准模拟海岸淹没过程并评估海岸洪水危险性至关重要。当前多数可用于全球尺度海岸淹没模拟的模型均采用简单的平面近似方法，常被称为浴缸模型（bathtub models）。这类模型的主要局限在于：其隐含假设洪水持续时长为无穷大，且无法捕捉相关物理过程。本研究开发了一种用于生成水文过程线（hydrographs）的方法，命名为HGRAPHER，并构建了全球风暴潮位过程线数据集。这些过程线表征了全球海岸线各点位极端风暴潮位的典型形态。我们针对决定过程线形态的两项核心假设，对HGRAPHER方法的敏感性开展了测试，这两项假设分别为：风暴增水事件采样阈值，以及增水与潮位峰值的时间重合性。依托这些过程线，可在大尺度场景下摆脱平面近似方法，转而采用更先进的动态淹没模拟技术。该数据集包含对应百年一遇重现期（RP，return period）的风暴潮位过程线，其风暴潮位数据与COAST-RP数据集保持一致。每个输出点位均包含两条过程线：分别代表平均潮位条件与大潮（spring tide）条件下的过程线。