Run Eulerian RMS

Velocity metrics package illustration: Run Eulerian RMS

Agulhas, BFN-QG

---

Authors: Datlas Copyright: 2024 Datlas License: MIT

Agulhas: BFNQG currents maps

The notebook aims to illustrate how to run the velocity metrics. Here, the example uses surface current maps produced by the BFN-QG in the Agulhas region.

import velocity_metrics.utils.constant as const
import velocity_metrics.eulerian.eulerian_drifters as eulerian

import datetime
import sys
from IPython.display import display, Markdown
import matplotlib.pyplot as plt
import cartopy
import warnings
warnings.filterwarnings("ignore")
sys.path.append('../')

from src import utils

Parameters

Input directories

input_dict = '../dc_data/DC_example_BFNQG_Agulhas/dictionnaries/'
input_drifter = '../dc_data/drifter/'

Output directory

outputdir = '../results/metrics_illustration/'

Evaluated maps parameters

path_dict_product = input_dict+'data_type_metric_illustration_bfnqg.json'

depth = 0

Region parameters

path_dict_region = input_dict+'region_metric_illustration_Agulhas.json'

Drifter parameters

drifter_list = [input_drifter+'Drifters_AOML_region_T1_00m_20190101T000000Z_20200101T000000Z.pyo.gz']

first_date = datetime.datetime.strptime('20190101T000000Z', const.FMT)
last_date  = datetime.datetime.strptime('20190201T000000Z', const.FMT)

1. Eulerian RMS computation

eulerian.run(drifter_list, path_dict_product,
             first_date=first_date, last_date=last_date,
             region=path_dict_region, sdepth=1, output_dir=outputdir)

Percent: [#############################-] 96.77%, ,

Compute and save statistics in ../results/metrics_illustration/ directory

2. Eulerian RMS array outputs (Mean and std)

list_dict = [{'type_stat':'Mean','path':outputdir+'Eulerian_RMS_BFNQG1.pyo','name':'BFN-QG'}]

utils.DictTable(list_dict)

Mean	Norm Quadratic Error (%)	Norm Correlation	Norm Explained Variance (%)	Norm RMSD	Number of Points
BFN-QG	38.848	0.698	61.360	0.317	2890.000

list_dict = [{'type_stat':'STD','path':outputdir+'Eulerian_STD_BFNQG1.pyo','name':'BFN-QG'}]

utils.DictTable(list_dict,['Direction difference', 'Number of Points'])

STD	Direction difference	Number of Points
BFN-QG	45.014	2890.000

list_eulerian_pickles = [outputdir+'Eulerian_BINNED_BFNQG1.pyo']
list_names = ['BFN-QG illustration']

listvar = ('RMSD',  'Correlation', 'Explained Variance (%)', 'Quadratic Error (%)')

dic_p = {'RMSD': {'vmin': 0.1, 'vmax': 0.4, 'cmap': 'YlGnBu_r'},
         'STD': {'vmin': 0.1, 'vmax': 0.4, 'cmap': 'YlGnBu'},
         'Correlation': {'vmin': 0.5, 'vmax': 0.9, 'cmap': 'YlGnBu'},
         'Explained Variance (%)': {'vmin': 40, 'vmax': 80, 'cmap': 'YlGnBu'},
         'Quadratic Error (%)': {'vmin': 40, 'vmax': 80, 'cmap': 'YlGnBu_r'}

        }

proj = cartopy.crs.PlateCarree()

ind = 0

var = listvar[ind]

for var in listvar:
    print('\n')
    display(Markdown(f'\n<br><center><b>Results for diagnostic {var}</b></center>'))

    for eulerian_pickle, name in zip(list_eulerian_pickles, list_names):
        display(Markdown(f'<b>{name}<b>'))
        fig = eulerian.plot_bin(eulerian_pickle, outputdir, var, #box=[-5, 11, 35, 43],
                                vmin=dic_p[var]['vmin'],
                                vmax=dic_p[var]['vmax'],
                                cmap=dic_p[var]['cmap'],
                                proj=proj)
        plt.show()

Results for diagnostic RMSD

BFN-QG illustration

Results for diagnostic Correlation

BFN-QG illustration

Results for diagnostic Explained Variance (%)

BFN-QG illustration

Results for diagnostic Quadratic Error (%)

BFN-QG illustration