EUFAR10/03, flight day 197/2010,Lisbon

[ella]

Data location: ~arsf/arsf_data/2010/flight_data/western_europe/EUFAR10_03-2010_197_Lisbon

Data arrived from ARSF via network transfer on 2nd/08/2010.

Scientific objectives: Objective 1.- Development of a reliable tool to estimate coastal water quality indicators by using remote sensing
techniques, both airborne and satellite.
1.1. | Development of a set of semi-empirical algorithms based on hyper-spectral airborne and in-situ data for
estimating the concentrations of the water constituents;
1.2. | Migration of the developed algorithms into the satellite workflow.

Priority: a4h

PI: Jose da Silva

Sensors:

• Eagle (09/03/2011)
• Hawk (09/03/2011)
• Leica LIDAR (03/12/2010)
• RCD Camera (03/12/2010)

[jb362]

Starting hyperspectral processing

[jb362]

Leaving this hyperspectral processing to be picked up again once we have a logsheet for the flight. Applanix navigation has been done.

[jb362]

Starting camera processing.

[ella]

IPAS navigation now processed.

[jb362]

Trying hyperspectral processing again without the logsheet.

[jb362]

Doesn't look like this will work so will have to wait for logsheet still.

[jb362]

Ignore last comment!

[jb362]

The eagle lev3 flight lines for this project do not appear to match up with the kml file when opened in gtviewer. We have decided to wait for the logsheets to see if those can shed any more light on the situation. Also, this flight is almost entirely over the ocean (with only a tiny piece of land visible on one flight line) and therefore (1) will not require a DEM and (2) will be impossible to find SCT offsets for.

[jb362]

Photographs 1-22 removed as they did not cover the study area.

[emca]

Logsheet has arrived and been saved in ~arsf/arsf_data/2010/flight_data/archive-overflow/EUFAR10_03-2010_197_Lisbon/admin

[jb362]

Iopa will be taking over delivery of this for me.

[ella]

RCD delivery check complete, ready to deliver.

[iopa]

RCD camera was dispatched today, 16th September 2010.

[knpa]

Moved to repository, can be rsynced.

[benj]

Corrected PI (recorded wrong on data unpacking)

[knpa]

Appears to be an issue with the bandset labels in the header files; they start at 900nm which is lower than it should be. Possibly incorrect bandset selected or none specified. Will leave this until end of the year and new Hawk calibration.

[iopa]

Taking eagle.

[emca]

Starting LiDAR processing

[emca]

LiDAR processing complete. Ready for delivery check at
~airborne/workspace/EUFAR10_03-2010_197_Lisbon/delivery/20101112/EUFAR10_03

LiDAR data was collected as one continuous line which has been split to match approximately with the Eagle/Hawk data. It was not possible to correct for roll and no DEM was created for the delivery as this project is entirely over water.

[iopa]

LiDAR has been checked,

no errors in the data or the delivery, ready to go.

[emca]

LiDAR and Photography dispatched 03/12/2010

(note: photography is a re-delivery as data sent on 16/09/10 was sent to the wrong PI)

[iopa]

Taking Hawk processing for this as well.

[iopa]

Hyperspectral status:

The flightlines for both Eagle and Hawk have contained very large amounts of turns and to process them correctly they have been split into parts to exclude the major >75 degree turns. The precise splitting points are given in the file line_split_times which contains GPS times that apply to both Eagle and Hawk.

Because of this however, some parts had to be processed with has_sync = false (excluding the parts whose start was the same as their original line's start) and so due to the approximation of their header times there will be some timing errors, which also cannot be corrected to the usual precision since the flight is fully over water.

[iopa]

To utilise those GPS times the command splitline.py was used.

[iopa]

Eagle & Hawk GPS Time splits

[iopa]

Eagle & Hawk SCT offsets

[iopa]

Hyperspectral is ready for delivery check.

While both Eagle and Hawk have been split into parts to be processed with azgcorr, no offsets were applied on Eagle as there are no reference points to de-wobble. Hawk was offset appropriately to match the Eagle, using sun glints, clouds or other odd bits.

[knpa]

Taking HS delivery check.

[knpa]

Delivery check complete.

All looks good apart from the following:

1. DEM directory is empty. If we have one then put it in here, otherwise remove dir and reference in Readme.
2. Something wrong with this sentence in Quality issues section: "During the production of the data, we processed Level 3 imagery . Screenshots of the Level 3"
3. azgcorr command - should have an example of dem even though we don't provide one. Unless it's all over water then obviously they don't need a dem - in which case I don't think we need the -es command either as no geoid-spheroid sep needs to be done? Maybe shouldn't include the misc directory either.
4. Couldn't open mosaic screenshots in eog (too big). Is there a way to reduce their size?
5. Regular dimming and brightening of Hawk throughout all flightlines. Should mention in Readme (and maybe analyse - seen this quite a lot but never heard it mentioned before).
6. Underflow and overflow on parts of all eagle lines, caused by sunglint. Huge underflows on Hawk caused by the above effect. Mention in readme.

[iopa]

Delivery check comments have been amended, a reduced size version for each mosaic has been produced and included as well as the original. More quality remarks have been added.

Delivery has been put on hold to investigate point no 5 of Hawk dimming/brightening.

[iopa]

Hyperspectral has been sent to PI on 9th March 2011.