search

GEUS-SICE / SICE

Operational Sentinel-3 snow and ice products (SICE) - Geological Survey of Denmark and Greenland (GEUS)
http://snow.geus.dk
GNU General Public License v3.0
5 stars 2 forks source link

Merge SNAP8 updates #31

Closed BaptisteVandecrux closed 3 years ago

BaptisteVandecrux commented 3 years ago

The current version (master at time of pull request) was designed to work with SNAP7. It is the only release where per-pixel geocoding is possible, but it cannot read S3 data after 2019. It has been stored in this release: https://github.com/GEUS-SICE/SICE/releases/tag/v2.2 This version (and therefore the release has been used for Adrien's work.

The new version (branch SNAP8 at the time of pull request) was designed to work with SNAP8. It can read more recent data but only tie-point geocoding is possible. It has been used in Adrien and Jason work and is stored here: https://github.com/GEUS-SICE/SICE/releases/tag/v2.3 I did some further cleaning and called for a pull request so that we keep the latest version in master.

The next version under development is in dev-bav and will include higher spatial resolution and several enhancement.

BaptisteVandecrux commented 3 years ago

And as a warning, the coefficients of the empirical equation for BBA in both master and SNAP8 branches are:

https://github.com/GEUS-SICE/SICE/blob/e72e59a2d370d309d631f1ab791dea936107af43/dm.grass.sh#L179

I thought they had been updated but apparently not.

Adrien, you mentioned that for your GEUS bulletin paper you had calculated the BBA_emp separately with the good coefficients. What about the reprocessed data produced with the master branch and the NRT production with the SNAP8 branch? Do they have a separate calculation of BBA_emp?

AdrienWehrle commented 3 years ago

And as a warning, the coefficients of the empirical equation for BBA in both master and SNAP8 branches are:

https://github.com/GEUS-SICE/SICE/blob/e72e59a2d370d309d631f1ab791dea936107af43/dm.grass.sh#L179

I thought they had been updated but apparently not.

Adrien, you mentioned that for your GEUS bulletin paper you had calculated the BBA_emp separately with the good coefficients. What about the reprocessed data produced with the master branch and the NRT production with the SNAP8 branch? Do they have a separate calculation of BBA_emp?

The current master branch has the updated coefficients used in our GEUS bulletin paper:

https://github.com/GEUS-SICE/SICE/blob/e1d06f8c917f4d85a26867048d50a25c931c56d4/dm.grass.sh#L148

But indeed, the SNAP8 branch does/did not have them. The reason is I have been running the SNAP8 version through another repository where I develop and store all the codes linked to NRT computation and release (and compute the BBA combination in post-processing with the updated values for BBA_emp). One goal for next year's melt season or before is to make those NRT-related codes publicly available here.