Optical calibration, bathymetry, water column correction and bottom typing of shallow marine areas, using passive remote sensing imageries
busy? see 4SM slides
The stepping stones of 4SM development
4SM has been developped, as a personnal project, then as a hobby, in almost complete isolation, since 1994.
Using first a database of SPOT images, then a very few Landsat TM images.
Then using a number of CASI hyperspectral images, and even an AAHIS image.
Then using Landsat TM/ETM, HYPERION and OLI images, thanks to USGS's Glovis service.
Then using WV2 thanks to Digital Globe.
In 2013-2015, it has matured remarkably with the introduction of several breakthroughs using OLI, WV2, CASI, HYPERION and even HICO data.

Starting in 2009, I was paid good money by a proeminent consultancy firm for projects using Landsat or WV2 in Tanzania, Bahamas, Bahrain, Western Australia, Seychelles, Red Sea, and the UAE.
As for the UAE project, first for a limited AOI in 2013, then for an extension six months later, then for a further extension two years later.


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

1994 to 1999
  • In an earlier life, I was involved for years in deep-sea minerals prospecting: particularly in charge of 3D modeling of a typical manganese nodules mining site in all its details: multibeam echosounding, deep-tow sonar and cameras, data reduction, fusion and management.
  • To start with with RS in 1994, as the head of IFREMER's Remote Sensing Facility in Tahiti, French Polynesia
    • I flipped through the technical specifications of SPOT sensors over several long weekends
    • then I flipped through JERLOV's and KIRK's books on light in water,
    • then some of Andre MOREL's and Preisendorfer's books and papers
    • then I was granted a copy of Maritorena's thesis...
  • Then I was granted a field trip with the French SHOM, to gather precision depth soundings in the lagoons of Moorea
    • and worked it out with a student
  • Then I was denied by my employer any support for this very personal research which ventured in someone else's territory. To say it plainly: I was suppressed.
  • Then in 1996: Bill PATZERT noted my presentation in Canberra, hosted me for a visit at JPL, where my simplistic approach failed to impress.
  • Then in 1998: Walton CAMPBELL noted my presentation in 1998 in San Diego, hosted me for a visit to NRL at Stennis Space Center, where my simplistic approach failed to impress.
  • Then in 1999, I was granted the right to pursue my goals, on a private basis.


1999 to 2012
  • I feel bad that visitors to this website find it hard "to re-evaluate the current situation each time"
    • excited in 1999: the truth is: I've been stalled for so long:
      • starting in 1999, my canadian friend with his CASI was unaware of the particulars of  collecting/processing RS data for shallow water purpose,
      • and ended up being ousted from his own business in 2005 for overselling my 4SM application to the Canadian Navy... 
      • At least, I was so excited, and he kept me alive with some very valuable datasets
      • and even gave me some valuable $$$.
    • then in 2002,  Denis Toustou showed up, and I was so excited,
      • with a pan-sharpened SPOT 5 image.
      • That was an enlightment to me
      • such a nice S/N that I could only dream of:
        • Pleiades must be even better.
      • pan-sharpened works:                                
        • Pleiades must be even better.
    • then in 2002, Gavin Manson showed up, and I was so excited,
      • as it proved that using Jerlov's data and hyperspectral bands for 4SM shallow water work
      • yielded the true retrieved depth which only needed tide correction. 
      • Then he was relocated to working on the coasts of North Alaska and that was it!
    • then in 2002, Serge Andrefouet showed up with four Ikonos images, one of which is at LSI, and I was so excited,
      • but 9/11 happened and he disappeared from Florida to resurface at IRD, Noumea, and remained silent since then.
    • then in 2011 Richard Chiles showed up, and I was so excited,
    • then in 2011 Ron Abileah showed up, and I was so excited,
      • and he gave me access to his Jmonster computer on the cloud
      • and  a few WV2 images to play with.
      • But his interest is in off-nadir-viewing heavily glinted images, and radar data!
    • then in 2012 Gregory Miecznik showed up, and I was so excited,
    • then in 2012 Simon Allen showed up, and I was so excited,
      • but Geraldton/Marmion/Perth areas are such a difficult case because of extremely dark bottoms
      • that even CSIRO's Arnold Dekker appears to have given up on this case.
    • then in 2013 Landsat 8 OLIP showed up, and I got overwhelmed!!!
      • with its nice PAN channel,
      • with  its purple band
      • with its SWIR1 band
      • with its much improved S/N ratio,
      • and FREE.
  • That's why I was able to pursue the multiple dreams that haunted me since the start in 1994.
    • I keep digging!


December 2013
  • At present, I have reached to the point where
    • I have demonstrated a quantum leap forward by showcasing the "PAN solution" for shallow water work.
    • This should ring as thunderbolt
      • in the data provider's      community, as it adds a new band to the multispectral bandset of Landsat 8, WV2, SPOT, PLEIADES, IKONOS, etc, for shallow water work.
      • in the defense                 community, as it extends by a factor of ~4 in clear waters the reliability of the retrieved shallow depth range for the purpose of remote environment assessment.
      • in the hydrographer's      community, as it extends by a factor of ~4 in clear waters the reliability of the retrieved shallow depth range for the purpose of nautical chart updating.
      • in the coastal manager's community, as it extends by a factor of ~4 in clear waters the reliability of the water column corrected shallow depth range for the purpose of "habitat" mapping and monitoring.
    • All I can hope for in the future is
      • improvement of the imagerie's S/N specifications shall allways be welcome.
      • more spectral bands shall never hurt.
      • narrower bands would not hurt.
End december 2013
I joined HYPHOON to process a CASI image of Heron Island, Australia
then in april 2014 I posted a demonstration for this studycase

Gulf of Laganas, Greece extremely dark bottoms february 2016 WV2P
Sabah, Borneo, Malaysia 15 m GSD PAN sharpened time series november 2016 OLIP
Gulf of Laganas, Greece 15 m GSD PAN sharpened time series november 2016 OLIP
San Lorenzo Channel, B. California 15 m GSD PAN sharpened time series october     2016 OLIP
Lee Stocking Island, Bahamas time series of images, robustness september 2016 OLIP
Caicos Bank, Bahamas time series of 18 scenes PAN+MULTI july-sept   2016 OLIP
La Parguera, Puerto Rico PAN, LIDAR may          2016 OLIP

"I don't understand
if you do also use the other bands + Pan"

Next steps forward
until semi-analytical methods can actually take over
Over the last four years,
4SM has become far more capable
(I know: hard to keep up!!).
  • OLI metadata_MTL.txt allow me to convert TOA radiances into TOA reflectances http://landsat.usgs.gov/Landsat8_Using_Product.php
    • then, as I extract an estimate of the atmospheric path radiance from the image, I then convert from TOA reflectance into BOA reflectance: this for consistent water column corrected reflectances for bottom type classifications.
  • WV2: now I do the same
===========System noise===========
Time has come when the major step forward on the part of data providers should be:
Please give us access to images with decent Noise-equivalent change in radiance
for the purpose of shallow water work.
See below: this striping in OLI images is a shame!

This is a shame!!