Optical calibration, bathymetry, water column correction and bottom typing of shallow marine areas, using
World View 2 at Lee Stocking Island, Bahamas
the "PAN solution" uses the PANCHRO band
bands 1, 2, 3, Panchro, 4, 5, 6, 7 and 8
bands 1, 2, 3, 4, 5, 6, 7, 8 and 9

For other 4SM investigations using the panchromatic band:
see my PANCHROMATIC web page


1 - NO NEED for field data, nor for atmospheric correction
2 - this is demonstrated in this website, using a variety of hyper/multi spectral data
Requirements are
1 - homogeneous water body and atmosphere
2 - some coverage of optically deep water
3 - some coverage of dry land
Problems are
1 - the precision on estimated depth is found wanting, because the noise-equivalent change in radiance  of accessible data is too high for shallow water column correction work 
2 - radiance data should be preprocessed by the provider at level 1 in order to improve S/N ratio
3 - exponential decay: the deeper/darker the bottom, the poorer the performances
I keep digging
until suitable data
become available


All modeling was performed using the "PAN solution".
This means: Purple, Blue and Green bands against PANchro band.
This means: Yellow, Red, RedEdge, NIR1 and NIR2 play no role in modeling.

Retrieved depth

Bottom brightness in the blue band

TCC zoom, TOA

TCC zoom, water column corrected, TOA

NDR, zoom
  • negative values are mapped in shades of green
  • positive values are mapped in shades of purple

TCC zoom, water column corrected, normalized
  • enhances dramatically the subtle nuances of the above water column corrected TCC
  • the spectral contrast is enhanced, irrespective of the brightness

Results by Mobley et al, 2005
using Ocean PHILLS image
  • The profile shows that the oolithic ridge is ~6 m high, from 3 m at the top  to 9 m at the base


Profile Orange

Profile Red
  • The oolithic ridge is located between waypoints 1 and 2 in all four sections.
    • its spectral water column corrected signature is high in the Purple channel
  • The profile shows that the oolithic ridge is ~6 m high, from 5 m at the top  to 11 m at the base
  • Please note the 15-20 m deep through in Profile_Red between waypoints A2 and A3, with an extremely green bottom signature.

Profile Black
  • Please note that the water column corrected reflectances have been divided by 2 in order to accomodate the very bright reflectances at the end of Profile_black (section A2)
System noise
  • these profiles are very noisy, in spite of strong smart-smoothing.
    • this noise is seen to increase steadily from shallow to deeper (much more than with a "Green" solution).
    • this comes with the "PAN solution", and reduces the "precision" on retrieved depths.
The advantage of the "PAN" solution
  • In 4SM, and in the absence of a PAN channel,
    • first the "RED solution" is operated for pixels that carry a Red signal that is stronger than the Red threshold value.
    • then the "YELLOW solution" is operated for pixels that carry a Yellow signal that is stronger than the Yellow threshold value.
    • then the "GREEN solution" is operated for pixels that carry a Green signal that is stronger than the Green threshold value.
  • This means that gaps show off atboth transitons
    •  Red_to_Yellow
    • Yellow_to_Green
  • Using the "PAN solution" for all pixels yields a depth raster without any gap, much to the relief of the end-user!

  • Georeferencing: this image in a geographic projection (Lat/Long):
    • see details in 09DEC07154605-M2AS-052963315020_01_P001.IMD in zipfile
    • UL origin is -76.232682 and 23.871708, as reported by OpenEV for tile R1C1
    • pixel size are 1.7999999948e-05 and 1.80000000012e-05, as reported by OpenEV for tile R1C1
This zipfile contains the following:
  • lsiwvmp_20091209_Zcm.tif  one S16 channel, retrieved depth in centimeters, as is: uncorrected for tide
  • lsiwvmp_20091209_WCC.tif seven U8 channels, TOA water column corrected reflectances, scaled 0-250
    • channels 1 to 6 are water column corrected Purple, Blue, Green, PAN, Yellow and Red
    • channel 7 maps the "solution":
      • 4 for the "PAN solution",
      • a very few 3 for the "Green solution" where the PAN signal is less than the threshold.
  • 09DEC07154605-M2AS-052963315020_01_P001.IMD
  • 09DEC07154605-M2AS-052963315020_01_P001_README.TXT

can someone knowledgeable run this result through seatruth validation?

as I have no access to seatruth data.


Command line
Oct 30th 2013
#nice -20  4SM -Process    -Origin/DigitalGlobe                               
-MakePIX            @-Prepare/Scale                               

nice -20  4SM   -Process  -Origin/DigitalGlobe                       
-B/TclNe/cNorm5.00/cLM2 0.66666 1 .5                   
-Model/v/runWV2/mask_3    -TPs/2/4/Text     -LL10           
-ProfileAB/ZrvWB/profile_black/dTM/chAB_1_2_3 _5_6