Opened 8 months ago

Last modified 6 months ago

#725 new flight processing

FIDEXII/23, flight day 226a/2023, Canada

Reported by: mark1 Owned by:
Priority: immediate Milestone:
Component: Processing: general Keywords:
Cc: Other processors:

Description

Data location: /users/rsg/arsf/arsf_data/2023/flight_data/canada/FIDEXII-2023_226a_Canada

Data arrived from NERC-ARF via hard drive September/October 2023
PI: Martin Wooster
Sensors:
Fenix1k
IBIS

Change History (18)

comment:1 Changed 7 months ago by mark1

Fenix Line 19 has no darkframes (autodarkstartline is at end of file) - it's a night flight so assuming this is correct as hard to tell.

comment:2 Changed 7 months ago by mark1

Fenix and Ibis lines have been processed with all bands. Can't check geolocation accuracy as night flight and data is rubbish. The Ibis and Fenix do not appear to line up against each other when comparing the bright "fire" patches - a few pixels offset.

comment:3 Changed 7 months ago by wja

Fenix1k & Ibis Processing

mark1 has completed processing all bands.

Making deliveries.

comment:4 Changed 7 months ago by wja

Ibis Delivery

Ready for check.

comment:5 Changed 7 months ago by wja

Fenix1k Delivery

Ready for checking.

Note that line 19 doesn't have darkframes and has been omitted (there's a note in the ReadMe).

comment:6 Changed 7 months ago by asm

Ibis DC
Started

comment:7 Changed 7 months ago by asm

Fenix1k DC
Started

comment:8 Changed 7 months ago by asm

Ibis DC

As it was flown at night, there is no way to check how good is the geocorrection. It is mostly noise but some fire is visible. This is captured in the readme.
All possible checks have been performed and passed.
Mapped files are ready to be zipped.

comment:9 Changed 7 months ago by wja

Ibis Delivery

Zipping mapped files

comment:10 Changed 7 months ago by asm

DC 226a Fenix

Same that for ibis, here is no way to check how good is the geocorrection. It is mostly noise but some fire is visible. This is captured in the readme.

  • Please remove from readme the flightline 19 from the fenix file identifier and change the naming to fenix (“f” instead of “i” for ibis).

All possible checks have been run, only waiting for check_apl_cmd that is still running. The sensor_fov filename was different as this was mounted backwards so it is re-running. Once done, if the output is correct, it will be ready for zipping the lines and deliver.

Last edited 7 months ago by asm (previous) (diff)

comment:11 Changed 7 months ago by wja

Ibis Delivery

All lines mapped.

comment:12 Changed 7 months ago by wja

Ibis Delivery

Delivered via FTP slot 5. PI notified.

comment:13 Changed 7 months ago by wja

Fenix1K Delivery

Delivered via FTP slot 5. PI Notified.

Fenix1K and Ibis deliveries finalised.

comment:14 Changed 7 months ago by wja

Issue with delivered data

Both Fenix1K and Ibis deliveries appear not to be flipped as they should be (apart from line 1 for some reason). I think this was caught out during the recent changes to the code handling the APL config file's behavour with the arg_default and args_extra commands, where the aplcal -FLIPSAMPLES was within the args_default, but was being overwritten by the args_extra being used for the swathmask?

Anyway, these are reprocessing on the grid now, hopefully they'll run through by tomorrow.

comment:15 Changed 6 months ago by wja

Fenix1K Delivery

Three lines need reprocessing (timing out on the grid). Currently running.

Ibis Delivery

All lines have been reprocessed. Have moved these into delivery (previous flightlines directory moved to flightlines_old for now). Need to check outputs look appropriate.

comment:16 Changed 6 months ago by wja

Fenix1K & Ibis Delivery

Have re-sent the data via FTP with the correctly 'flipped' data.

comment:17 Changed 6 months ago by wja

Fenix1K Processing

Flightline 19 failed to process because it was missing darkframes. Have been requested to process this using closest lines's darkframes (in this case line 18).

Running now

comment:18 Changed 6 months ago by wja

Fenix1K Processing

Flightline 19 processed using darkframes from line 18. Outputs look fine (mapped fined compared to overlapping lines, all bands looked at, didn't look unusual despite there not being any features other than a fire).

Have added to delivery and FTP, notified PI.

Note: See TracTickets for help on using tickets.