/kpi_mpc

library to compute the MPC Sentinel-1 KPI

Primary LanguageJupyter NotebookMIT LicenseMIT

Kpi Mpc

libs to compute 2 KPIs for MPC Sentinel-1 project:

  • a) WV VV effective Hs bias (wrt WW3)
  • b) WV VV NRCS (denoised) bias (wrt CMOD-5n)

author:

IFREMER LOPS
antoine.grouazel@ifremer.fr

creation:

2021

usage

python ./src/kpi_WV_nrcs/compute_kpi_1b.py --verbose

installation

conda create -n kpi_conda_env python=3.9
conda activate kpi_conda_env
conda install numpy scipy matplotlib xarray netCDF4 ipykernel
git clone https://gitlab.ifremer.fr/lops-wave/kpi_mpc.git
cd kpi_mpc
python setup.py install 

#data dependencies:

KPI-1b (NRCS SLC)

KPI-1b needs a netCDF file containing the differences of NRCS denoised per SAFE for all the WV since 2015 ( S1%_wv_ocean_calibration_CMOD5n_ecmwf0125_windspeed_weighted_slc_level1_20150101_today_runv16.nc)

KPI-1d (Hs OCN)

KPI-1d is designed to run using Sentinel-1 WV IFREMER L2F daily aggregated nc files.