DeepSurge storm surge predictions for simulated historical and future climates ...

• Main Authors: Rice, Julian, Ticona Rollano, Fadia, Balaguru, Karthik, Xu, Wenwei, Judi, David
• Format: Dataset
• Language: English
• Published: Zenodo 2023
• Subjects: storm surge; tropical cyclones; deep learning; climate change; Pacific; North Atlantic
• Online Access: https://dx.doi.org/10.5281/zenodo.8432850; https://zenodo.org/doi/10.5281/zenodo.8432850

DeepSurge is a newly presented deep-learning approach to modeling the storm surge generated by a tropical cyclone (TC). This dataset is a collection of DeepSurge outputs for synthetic TCs in the North Atlantic generated by the HighResMIP project (Haarsma et al. 2016) for a simulated historical (1950-2014) and future (2015-2050) climate under the climate scenario SSP585. The data generation process and data analysis is detailed in an upcoming publication. The storm surge data presented here intentionally does not include the effects of sea level rise, rainfall, or other factors, in order to isolate the effects of changing TC climatology on future storm surge risk. Dataset format The data comes in the form of maximum surge levels at 2846 near-coastal locations for each synthetic TC. Each TC is defined by the corresponding track in the HighResMIP TempestExtremes dataset (Roberts 2019). The data is presented in NetCDF format, with two dimensions: 'nodes', the number of near-coastal locations, always 2846. ... : This work was supported by the Multisector Dynamics and Regional and Global Model Analysis program areas of the U.S. Department of Energy (DOE), Office of Science, Office of Biological and Environmental Research as part of the multi-program, collaborative Integrated Coastal Modeling (ICoM) project. The Pacific Northwest National Laboratory is operated for DOE by Battelle Memorial Institute under contract DE-AC05-76RL01830. ...