4SM Clipperton atoll

home
Landsat January 19th 2020
Landsat December 2nd 2020
Ikonos

Landsat January 19th 2020, 360*374 pixels, 15 m GSD

TOA TCC	BOA deglinted TCC glint regressions
calibration Nir, Red, Green, Blue	calibration Red, Pan, green, blue
*map of the ratio 100Kblue/Kgreen**	satellite derived bathymetry, in cm
BOA TCC of water column corrected radiances Ublue, Green, Blue RGB

Landsat December 2nd 2020
work done june 2021

Clipperton atoll

Blue waters over the ocean

Ultrablue and Blue bands shine Green, Pan and Red bands are low

Green waters inside the lagoon

UltraBlue and Blue bands are low Green, Pan, and Red bands shine

The purpose of the LUT approach is, among other things,
to account for such variations of the optical properties of the shallow waters
upon operating the water column correction, pixel-wise

Deglinted/smoothed data

With Coastal C5 water type, the central lagoon is most probably optically deep

UltraUBlue
is extinct

Blue
is slightly depressed

Green
glows strongly

Panchromatic
glows

Red
glows slightly

TCC a neat and very small scene image size: 360*374, 15m GSD	Glint regressions based on SWIR shows PAN vs SWIR a good start: regressions are excellent
calibration NIR, Red, Green, Blue full image, clear oceanic waters K2/K3=0.65 related to Oceanic II water type	calibration Red, Pan, Green, Blue full image, under satisfactory control
Map of the ratio Kblue/Kgreen oR1	TCC BOA deglinted
SDB in cm Relates reasonably well with existing bathymetric maps	TCC BOA WCC Ublue, Green, Blue RGB TCC of water column corrected radiances: the ultimate goal

A 1371*1160 IKONOS 4 bands image of Clipperton atoll

Wavelengths at midresponse curve are 480, 551, 665 and 805 nm
Strong sky and sun glints. Clouds not masked. Confined fresh waters in the lagoon
Work done in 2000, updated in 2005, image courtesy of S. Andrefouet

Raw image, BGR TCC
histeq enhancement

Raw image, full resolution
histeq enhancement

Deglinted image, BGR TCC
histeq enhancement
No smoothing over the deep ocean

Deglinted image, full resolution
histeq enhancement
No smoothing over the deep ocean

image_Z

image_B

image_LBS normalized

image_LBS
histeq enhancement

Blue vs Green
natural

Blue vs Green
linearized

Optical calibration
of outer slopes

Fairly straightforward, although
2K[red]=0.38 at 665 nm is a physical nonsense,
therefore the range of computed depths is problematic

Wavelengths must be set so that all ratios Ki/Kj fit nicely the slopes observed in this calibration diagram
Kblue/Kgreen=0.762
Kblue/Kred and Kgreen/Kred are very easy to check
Quite likely WLgreen must be increased and WLred must be decreased

Optical calibration
inside the lagoon

Quite difficult, as the lagoon bottom substrates
would appear to be extremely dark

The following is adopted:
Jerlov water type Coastal 3+0.14
2Kblue~=0.54
2Kgreen~=0.38
Kblue/Kgreen~=1.42