arheem / wave-modelling

Wave modelling with SWAN. Includes data cleaning, munging and visualisation.

wave-modelling

Notebooks and functions used for wave modelling with SWAN. Includes data preparation, cleaning, munging, and visualisation.

The script spec2d.py returns the 2d-spectrum from Hs, Tp, pdir and wave spread. The script spec2d_era.py returns the 2d-spectrum from Hs, Tp, pdir and cossene power.

The notebooks have the following objectives:

• buoy_csiro_correl and buoy_era_correl:

  • Read .csv files containing buoy observations
  • Read netcdf file of CSIRO reanalysis containing mean wave period
  • Clean and prepare buoy data
  • Select ERA5 timeseries with the same coordinates and date/times of buoy observations
  • Correlate ERA5 wave period to buoy observations
  • Plot the correlations as scatter plots

• create_specs and create_specs_era:

  • Calculate wave spectra from wave parameters of ERA5 partitions
  • Write it to text file to use as input for SWAN

• create_wind_and_timepoints:

  • Writes wind and timepoint inputs for SWAN from netcdf files

• get_ww3:

  • Download wave watch 3 dataset from noaa repository via web scraping

• netcdf_xyz:

  • Writes gebco netcdf bathymetry to a .xyz file

• plot_specs and plot_specs_datasets:

  • plot_specs plots 2d-spectrum reconstructed from CSIRO wave partitions to a 3d plot
  • plot_specs_datasets plots 2d-spectra from three different sources: (1) downloaded directly from ERA5, reconstructed from ERA5 partitions, and reconstructed from CSIRO partitions. The plots are in polar coordinates. Also plots spectra-extracted Hs as timeseries for comparison purposes

• swan_out_processing:

  • Processes the output of SWAN. Extracts data from selected coordinates (wave buoy coordinates for correlation purposes)
  • Saves timeseries to numpy array