Bathymetry and water column correction
LANDSAT 8 OLIP 
at SanLorenzoChannel, Baja california
Image courtesy of the U.S. Geological Survey
 
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 july 2018
GSD 15 m
 


 
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
 
The data

Mulroy Bay is located between
Sheephaven Bay (west) and Lough Swilly (east)
 

TCC over the full Landsat scene
 
co-registration

BLUE band LC82070222016153LGN01_B2
co-registration

NIR band LC82070222016153LGN01_B5

PAN band LC82070222016153LGN01_B8
PAN-MULTI co-registration :
can it be improved
in view of PAN sharpening?

Origin of the PAN band
is offset by
  • +15.0 m in XUTM
  • -    7.5 m in YUTM



Most of the "noise" is actually
well correlated sun glint,
NOT SO for the PAN band though
NOISE
As it collects all photons
between 500 and 675 nm,
one would expect the PAN band
to be much less noisy
than narrow bands.


NOT SO!

WHY?


Can this be improved?



I wish the response curve would include
the 675-850 nm range.
 
Deglinting

Shapefiles for estimating the glint regressions
Profile_blue is also shown
Backdrop is a TCC of the deglinted data

Glint regressions are quite good,
based on the SWIR1 band

Deglinted NIR band
  • excellent over open waters
  • cannot beat the backscatter by suspended solids over inland waters

Deglinted RED band
  • exhibits some striping over open waters
  • waters not quite homogeneous
  • muddy waters of Lough Swilly stand out
  • inter-tidal waters at Lough Foyle stand out

Deglinted PAN band
  • Lough Foyle waters pour out NorthWestwards

Deglinted GREEN band
  • Lough Foyle waters pour out NorthWestwards

Deglinted BLUE band
  • Lough Foyle waters pour out NorthWestwards

Deglinted COASTAL band
  • striping over open waters
  •  
  • Lough Foyle waters pour out NorthWestwards


 
Radiances along Profile Blue
  • Backscatter is much stronger
    • West of waypoint 2 and
    • East of watypoint 5
  • Between waypoint 2 and waypoint 5, waters exhibit  lowest backscatter
  • Between waypoints 3 and 4, a distinct shoal is not detected in RED band, quite bright in the GREEN band, and much less in both BLUE bands
  • At waypoint 6, one can notice the striping all bands except SWIR1 and NIR
  • Lsw: altogether, estimating deep water radiances for the simplified optical model can seem an un-reasonable challenge
    • therefore, for a conservative standpoint, thresholds should be set a high values




 
Calibration

Calibration diagram 6, 5, 3, 2
for Mulroy Bay
  • Coastal 1 water type
  • Kblue/Kgreen=0.90: this is clearly evidenced in X2,3 vs X5 diagram.
    • Kblue=0.244 m-1
    • Kgreen=0.271 m-1
  • There is hardly any color separation for the pair Blue vs Green: 4SM modeling is ~impossible

Calibration diagram 6, 5, 4, 2
for Mulroy Bay
  • Coastal 1 water type
  • Exhibits significant color separation  for the pair Blue vs PAN
    • therefore 4SM modeling is possible, down to in excess of ~10 m