Optical calibration, bathymetry, water column correction and bottom typing of shallow marine areas, using passive remote sensing imageries
busy? see 4SM slides

Breaking news:
using a panchromatic band
home

 
 


 
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
 
So
I keep digging
until suitable data
become available
 





Calibrating the PAN solution

the ratio Kblue/Kpan

 


the empiric way is impractical

 
  • The empiric way is far from easy, as the effective diffuse attenuation coefficient for the panchromatic band decreases steadily
    • from very high values at very shallow depths
    • to Kpan ~1.7*Kblue in very clear waters at extinction for a WV02 image.
  • Once the Red, Yellow and Green solutions are calibrated to satisfaction, the image is processed in order to derive the experimental law of variation of  the ratio Kblue/Kpan as a function of bottom depth for the current image.
  • Now that this experimental law is available, the PAN solution is applied to all shallow pixels which exhibit suitable signal.
    • as all this refers to a "ratio method" where radiances remains in units of uncalibrated DNs,
  • Isobath curves at 1 m interval are shown
  • Note that the experimental BPL is "bumpy"
    • this yields "bumpy" depth results
 



As opposed to the empiric way,
the optical way is ideal

 
 
  • As of august 26th 2014
  • This this only uses the radiance response curve for the PAN band along with the optical calibration model for the Multi/Hyperspectral bands of the current image.
  • Therefore, it is a generic process which makes it quite easy to add the PAN band to the Multi/Hyperspectral dataset for shallow water processing of any image.
  • Still not quite satisfactory here using a WV2 image at Princess Cays!
  • Isobath curves at 1 m interval are shown
    • note that they are packed very closed to each other deeper than 5-6 m: results shall become quite noisy at depth
  • Note that the BPL model is very smooth: 
  • this yields smoother results 

see Landsat 8 PAN response curve
This applies to Landsat 8 OLIP as well
as illustrated below

Note that the Landsat 8 Panchromatic band
  • only covers the Green_to_Red range
  • this excludes the Blue and the NIR ranges





Calibrating the PAN solution
for a hyperspectral image
  • All visible bands in a hyperspectral image may be binned together to produce a synthetic  Panchromatic band.
  • This was done with complete success for the Heron Island study case, then on Hyperion and HICO data
2K Jerlov


 
Results
  • The whole depth range is processed in a very homogeneous way:
    • processing time remains the same: a same day service can be achieved.
    • no more of these disturbing and unacceptable gaps.
  • The overall accuracy on retrieved depth shall be shown to have distinctly improved.
  • The precision on retrieved depth shall be shown to have
    • increased dramatically wherever the GREEN solution was the only recourse (think of Stumpf et al's method).
  • The homogeneity shall be shown to be outstandingly improved.
  • Processing pan-sharpened images might become routine:
    • please improve the radiometric quality of the image, rather than indefinitely reducing the pixel footprint.

 

Using HYPERION at Fakarava atoll

Blue vs PAN for a Hyperion image
of Fakarava atoll, French Polynesia

work done october 2014



see also HICO at Lee Stocking Island