A time series of landsat 8 images
at San Lorenzo Channel, Baja California

Using the Panchromatic band for water column correction
to derive water depth and spectral bottom signature:

Landsat 8 OLIP bandset used for this work

Purple=1Blue=2Green=3PAN=4Red=5NIR=6 and SWIR1=7

work done december 2016


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

scene LC80340432013292LGN00, October 19th 2013  
New processing done in december 10th 2016
15 m GSD pan-sharpened "on the fly" by 4SM; NO smoothing
Uses CombinedDepth as a seatruth DTM
RMSE=1.05 m 
over Isla Espiritu Santo and San Lorenzo Channel
Data and Deglinting
Bottom typing
A lot of novelties have been introduced in 4SM of late.
Time is needed until these are shown to be consolidated, and stand the test of time!
  • 15 m GSD: processing pan-sharpened Landsat 8 scenes should become routine, at very low computing cost
  • Bottom typing: SAM2 and NDRblues should help for thematic bottom typing
  • Combined Depth DTM at 15 m GSD from time series should become routine at low cost
GSD 15 m
  • Import:
    • Oversampling of MULTI bands from 30 to 15 m upon import in working database.pix
    • with weighing of 15 m pixels with their 3 immediate neighbours
    • Strict co-registration to the extent possible (quite difficult)
  • Processing:
    • PAN sharpening if required,  "on the fly" on-demand while processing the scene
    • note: oversampling and weighing can be implemented here instead of upon Import
    • NO smoothing needed
Normalized Difference Ratio for Blue bands, inspired from Collin et al
see legend for NDRblues
see http://www.watercolumncorrection.com/slcoli-sam.php
FCC  BOA  WCC normalized
The Red band is not used in the PAN solution; and it does not exhibit bottom detection beyond 4-6 m.
So I have changed for the following, which is a False Color Composite view:
  • B=CoastalWCC "normalized" (0-255)
  • G=GreenWCC   "normalized" (0-255)
  • R=BlueWCC      "normalized" (0-255)
I use the Combined Depth as a seatruth DTM
This gives a better estimate of the RMSE of the 4SM method: <=1 m
This scene shows signs of water flowing from East to West in the deeper channel.
  • A current of slightly turbid water is seen to flow westward through the channel
  • things do not come into place until the deep water reflectance is sampled under the blue rectangle

Seatruth: RMSE=1.05 m
Seatruth below is limited to Channel and Isla Espiritu Santo

October 19th 2013    vs   DTM


ZDTM-0.29m - Z4SM
DTM - New_October 19th 2013
see legend 

DTM : CombinedDepth 15m GSD


Data and Deglinting
Full scene

TOA TCC: raw image
logarithmic enhancement
Note the swirling plumes
coming westward out of the channel

BOA TCC deglinted image
linear enhancement


BOA TCC: water column corrected

Retrieved depth
see legend

Bottom typing

SAM classified image
see legend for image SAM2

Average bottom brightness
see legend for image B

see legend for NDRblues

  • A current of slightly turbid water is seen to flow westward through the channel
  • things do not come into place until the deep water reflectance is sampled under the blue rectangle

BOA CC normalized

location of profile_fabio1 and profile_black are shown


In the profiles below:
RED profile is ZDTM: the Combined Depth DTM
BLACK profile is Z4SM: retrieved depth




see profile_fabio1 for scene October 27th 2016