The University of Melbourne
Browse

Data from MIKE 21 models for training and validation of Sparse GP models in "Upskilling low-fidelity hydrodynamic models of flood inundation through spatial analysis and Gaussian Process learning"

Download (17.8 GB)
Version 3 2022-05-17, 00:09
Version 2 2022-02-24, 03:09
Version 1 2022-02-22, 01:01
dataset
posted on 2022-05-17, 00:09 authored by Niels FraehrNiels Fraehr

Data from MIKE 21 models for training and validation of Sparse GP models.

The data is used for publication in "Upskilling low-fidelity hydrodynamic models of flood inundation through spatial analysis and Gaussian Process learning" with the Chowilla floodplain as case study.

The data is structured in three folders:

- The raw data folder contains results for running the hydrodynamic models. One folder for the high-fidelity model (HF) and one for the low-fidelity model (LF). Both folders contain MIKE 21 .dfsu data files.


- The managed data folder is structured in three folders. "Classification_Figures" contain figures generated for the publication. "Events_data" contains the MIKE 21 data in binary format as .npz files to be read via the Numpy package in Python. "SPGP_class_models" contains the trained Sparse GP (SPGP) models, EOF analysis data and categories depending on the binary state of the data on cell level.


- Boundary data folder contain data for the boundaries of the hydrodynamic models. This data is retrieved from the Bureau of Meteorology's online water data platform: http://www.bom.gov.au/waterdata/

Python code is located in the main folder and on https://github.com/nfraehr/Hybrid_LSG_model

History

Usage metrics

    4000 - Engineering Students

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC