Skip to content

Latest commit

 

History

History
42 lines (35 loc) · 3.71 KB

nadir_alongtrack.md

File metadata and controls

42 lines (35 loc) · 3.71 KB
Raw

Extracting alongtrack nadir SSH from eNATL60-no-tide in the EUROSEA target regions

Stephanie Leroux, UPDATED 2022-03-14.

What?

  • Along-track sampling of the model SSH and SLA(*) (eNATL60-no-tide) in the MEDWEST subregion.
  • Sampling along the satellite tracks of Jason3, Altika, Cryosat-2 (c2n), H2B, Sentinel-3A, Sentinel-3B
    • From 2020-07-01 to 2020-12-31.
    • From 2021-01-01 to 2021-06-30.
  • (*) The model SLA(*) is computed by removing the time mean over the 1-year period (2020-07-01 to 2021-06-31) and also then removing the spatial mean over the basin at each hourly model output as a proxy for filtering out large scales (as requested by SOCIB/IMEDEA/CLS).
  • The along track interpolation is made with the software gonzag (https://github.com/brodeau/gonzag) from Laurent Brodeau, based on the Akima interpolation method.
  • Each file is 6-month long.
  • In each *SLA* file you’ll find:
    • sla_unfiltered (the real obs value (SLA)
    • slam (Sea Level Anomaly from model after removing time mean and spatial mean, interpolated at all alongtrack locations)
    • slam_np (Sea Level Anomaly from model after removing time mean and spatial mean, at nearest model grid point )
  • In each *SSH* file you’ll find:
    • sla_unfiltered (the real obs value (SLA)
    • sossheig (SSH from model, interpolated at all alongtrack locations)
    • sossheig_np (SSH from model , at nearest model grid point )
  • Important: Note that the the real obs value ('sla_filtered') is not expected to be consistent with the interpolated model values (slam) since the model was run in 2009-2010 and the real obs are taken in the years where the satellites exist (2020-2021).
  • Important too: If you need to retrieve the time-mean and spatial mean that was substracted from the model SSH, you can substract both interpolated files: SSH_interpolated - SLA_interpolated. You can also look directly at the time-mean model gridded field (on the opendap).

Where to download the data?:

  • On the opendap: https://ige-meom-opendap.univ-grenoble-alpes.fr/thredds/catalog/meomopendap/extract/lerouste/Eurosea2022/pseudo-ssh-alongtrack/catalog.html
  • In INTERP_RESULTS you'll find the interpolation results (for SLA and SSH).
  • In DATA_src you'll find all the input data used to perform the interpolation:
    • along track real obs downloaded from CMEMS portal and reformatted (see github notebooks for more details): jason3, sentinel3a, sentinel3b, altika, h2b, cryosat2.
    • model gridded SSH and SLA (SLA: time mean and spatial mean removed): caution: in both files the variable is named sossheig.
    • model time-mean over the 2020-07-2021-06 period : *_TM.nc (look for the sossheig_mean variable)
    • model spatial mean in the west med region at each hourly output: *_spmean.nc

How is this done? (step by step documentation)

  • Step1. Notebook explaining how to download and prepare the satellite files before applying the interpolation tool.
  • Step2. Notebook explaining how to prepare the model files before applying the interpolation tool.
  • Step3. Notebook demonstrating the interpolation step.
  • Step4. Notebook demonstrating how to read and plot the interpolated data and compare to the gridded model fields.

subregions
Fig.1 Example of interpolated SSH along jason3 tracks (3 day period)