search

euroargodev / BlueCloud

Working space for BlueCloud demontrator 3 - Lops Task
0 stars 1 forks source link

Try chorophyll data in Ocean Regimes notebook #48

Closed AndreaGarciaJuan closed 3 years ago

AndreaGarciaJuan commented 3 years ago

Try chorophyll data in Ocean Regimes notebook

AndreaGarciaJuan commented 3 years ago

I am using one year of OCEANCOLOUR_GLO_CHL_L4_REP_OBSERVATIONS_009_082 product (daily fields)

AndreaGarciaJuan commented 3 years ago

A new notebook has been created for testting Chlorophyll data: OceanRegimes_CHL.ipynb (for the moment in the plotter_or branch)

AndreaGarciaJuan commented 3 years ago

The mask is created using another CMEMS product in the mediterranean: MEDSEA_ANALYSIS_FORECAST_BIO_006_014. I create a mask by using this product and I apply it to the dataset in order to mask the black sea, the Atlantic Ocean and the lakes.

Dataset before mask: Capture d'écran de 2021-02-01 16-58-39

Dataset after mask: Capture d'écran de 2021-02-01 16-59-45

AndreaGarciaJuan commented 3 years ago

Figures using k=7 and a GMM: Capture d'écran de 2021-01-27 10-43-22 Capture d'écran de 2021-01-27 10-43-36 Capture d'écran de 2021-01-27 10-43-50

You can also see the comparation with the results in D'Ortenzio et al here: https://drive.google.com/file/d/1pzj_vixyjDkmiBGGXrB-muTDEBlGVlBj/view?usp=sharing

AndreaGarciaJuan commented 3 years ago

I tryed to use a Kmeans (used in D'Ortenzio et al): Capture d'écran de 2021-01-25 14-41-34

Results with GMM are more similar to results in D'Ortenzio et al

AndreaGarciaJuan commented 3 years ago

BIC index have been also calculated for Chlorophyll dataset: Capture d'écran de 2021-02-01 16-29-35