ATMFAIL on northernmost 3 lines in OLCI granules
ATMFAIL on northernmost 3 lines in OLCI granules
I recently switched from tag V2022.0 to R2022.14.
In processing OLCI (S3A and S3B) data, I now see ATMFAIL on the northernmost 3 rows of every granule. The only places on these rows where ATMFAIL is not indicated are the columns where NAVFAIL is indicated (columns 0-~150, ~4815-1864). In previous versions, these NAVFAIL columns were listed as CLDICE. However, as far as I can tell, the ATMFAIL rows were not systematically flagged in previous versions (even as recent as V2022.0).
Additionally, if I subset a granule within l2gen (e.g., cut off the northern ~half of the data), I still get ATMFAIL on the now-northernmost 3 rows.
As an example, I downloaded S3B_OL_1_EFR____20221012T142413_20221012T142713_20221012T234532_0179_071_267_2520_PS2_O_NT_003.zip and processed using l2gen with default params (both on V2022.0 and R2022.14).
In the figure below, red=ATMFAIL, green=NAVFAIL, blue=CLDICE, yellow=none_of_these (pseudo band_math: if ATMFAIL then red else if NAVFAIL then green else if CDLICE then blue else yellow). V2022.0 is on top, R2022.14 is on bottom. Note especially the red line at the top of the bottom panel.
In processing OLCI (S3A and S3B) data, I now see ATMFAIL on the northernmost 3 rows of every granule. The only places on these rows where ATMFAIL is not indicated are the columns where NAVFAIL is indicated (columns 0-~150, ~4815-1864). In previous versions, these NAVFAIL columns were listed as CLDICE. However, as far as I can tell, the ATMFAIL rows were not systematically flagged in previous versions (even as recent as V2022.0).
Additionally, if I subset a granule within l2gen (e.g., cut off the northern ~half of the data), I still get ATMFAIL on the now-northernmost 3 rows.
As an example, I downloaded S3B_OL_1_EFR____20221012T142413_20221012T142713_20221012T234532_0179_071_267_2520_PS2_O_NT_003.zip and processed using l2gen with default params (both on V2022.0 and R2022.14).
In the figure below, red=ATMFAIL, green=NAVFAIL, blue=CLDICE, yellow=none_of_these (pseudo band_math: if ATMFAIL then red else if NAVFAIL then green else if CDLICE then blue else yellow). V2022.0 is on top, R2022.14 is on bottom. Note especially the red line at the top of the bottom panel.
Filters:
-
- Subject Matter Expert
- Posts: 271
- Joined: Thu Mar 05, 2009 10:25 am America/New_York
- Been thanked: 2 times
Re: ATMFAIL on northernmost 3 lines in OLCI granules
Yes, we see the same problem.
Thanks, we will take a look.
don
Thanks, we will take a look.
don
-
- Subject Matter Expert
- Posts: 271
- Joined: Thu Mar 05, 2009 10:25 am America/New_York
- Been thanked: 2 times
Re: ATMFAIL on northernmost 3 lines in OLCI granules
I found the problem with the SAFE file format reader. This will be fixed in the next T and R tags. A value needed by the smile correction was not calculated soon enough. Thanks for finding this.
don
don