Landsat 8 OLIP at Shuwayhat Island, UAE
from Qatar towards Abu Dhabi
images courtesy of USGS
6747*6082, pixel size=30 m
work done in october 2014
Waters not so clear: watch these Youtube video  and Youtube video
see also the Marawaah island studycase

 
 
 
shuwayhat_20131207
shuwayhat_20140329
shuwayhat_20140820
shuwayhat_20140905
shuwayhat_20140921

Four results
Error on retrieved depth


home


shuwayhat_20140820

FCC
Grey: Shuwayhat Island AOI

TCC deglinted

TCC deglinted
 bad deglinting

 

ZPAN
bad deglinting

Retrieved depth by PAN solution
Thresholds LmPAN=0.5

 

Retrieved depth by GREEN solution
Thresholds LmGREEN=0.5



 
Profile Black
RED=PAN_solution
BLACK=GREEN_solution
along Profile Black
  • Smart_Smoothing is enabled
  • Discrepancies are regionalised
  • They appear to be caused by variations of water column corrected bottom reflectance signature
    • more than by system noise
  • We can assume that the PAN solution is more accurate than the GREEN solution
Profile Black

Z4=PAN solution
Z3=GREEN solution
along Profile Black

Local variations of water optical properties
probably cause some of the above scatter

see location of profile_black

Retrieved depth ZPAN
 

BGR color composite
of TOA water column corrected radiances
for Coastal, Blue and Green bands

 




   
shuwayhat_20140905

This is discussed in depth
at Fakava atoll



The only way that retrieved depths for this image
are compatible with the above bathymetric maps
is to
 
  • leave WLgreen =560 nm at mid-waveband
    • which yields a Oceanic III water type for Kblue/Kgreen=0.85 for the 0-7 m surface layer
    • with  Kblue =  0.212 m-1 and Kgreen=0.251 m-1 
  • and consequently set WLred at ~615 nm which is required
    • this yields Kred = 0.543 m-1
    • which satisfies (Kblue/Kred)/(Kgreen=/Kred)=0.85
  • therefore it achieves a nice fit in the calibration diagrams
This yields Kred=0.54 m-1 in the 0-7 m depth range: OIII
 
  • I have already seen that case at Caicos Bank (tmnov tutorial),
    • where the TM Red band clearly exhibits bottom detection down to 8 m of depth over bright bottoms, as evidenced by ground truth data
    • which would be impossible if the wavelength were left at 655 nm for the Red band
    • this of course warrants a through investigation using Hydrolight
  • This reminds me of Jerlov's statement
         
Waters get clearer at depth: OII+0.35
Watertype    Z   Ki/Kj   Ls1   Ls2   Ls3   Ls4   Ls5   Ls6   Ls7   
 O2+0.81   0.00  0.847   417.1 443.1 460.9 500.2 518.0 503.9 1233.5
 O2+0.81   4.46  0.847   209.7 231.1 185.1 159.7  73.7   0.0   0.0
 O2+0.16  13.64  0.687   155.7 145.5  74.7  54.8  30.5   0.0   0.0
 O2+0.16  39.86  0.687   132.3  97.7  51.0  44.0  30.3 503.9  14.0

Watertype    Z   Ki/Kj    K1    K2    K3    K4    K5    K6    K7  
 O2+0.81   0.00  0.847   0.292 0.212 0.251 0.308 0.543 0.000 17.500
 O2+0.81   4.46  0.847   0.292 0.212 0.251 0.308 0.543 0.000 17.500
 O2+0.16  13.64  0.687   0.183 0.143 0.209 0.275 0.563 0.000 17.500
 O2+0.16  39.86  0.687   0.183 0.143 0.209 0.275 0.563 0.000 17.500
K[ O2+0.16]_JERLOV       0.150 0.143 0.209 0.415 0.622   -     -  

Watertype    Z   Ki/Kj    ZM1   ZM2   ZM3   ZM4   ZM5   ZM6   ZM7  
 O2+0.81   0.00  0.847   27.25 38.40 33.19 27.36 15.65   inf  0.54
 O2+0.81   4.46  0.847   27.25 38.40 33.19 27.36 15.65   inf  0.54
 O2+0.16  13.64  0.687   43.55 56.82 39.86 30.68 15.08   inf  0.54
 O2+0.16  39.86  0.687   43.55 56.82 39.86 30.68 15.08   inf  0.54

 
Then, in order to get a nice fit
when comparing the Green and the PAN solutions
is to set CoefPath=1.15
for the PAN band

 

Z3 vs Z4 along profile_red
  • the PANsolution Z4 yields approximately  the same depth as the Green solution Z3
  • part of the scatter is caused by local variations in water optical properties

Retrieved depth by the PAN solution
  • Black line: approximate 18-20 m isobath contour
  • Red line: Profile_red has 22800 points which sample the whole scene along a 760 km path
This illustrates:
  • the PAN solution yields approximately the same depth as the Green solution,
  • but the PAN solution is expected to be more accurate
    • and therefore the water column corrected reflectances should be better
 

FCC
Grey: Shuwayhat Island AOI

TCC deglinted

Retrieved depth by PAN solution
Thresholds LmPAN=0.5

Retrieved depth by GREEN solution
Thresholds LmGREEN=0.5


 

Profile Black
RED=PAN_solution
BLACK=GREEN_solution
along Profile Black
  • Smart_Smoothing is enabled
  • Discrepancies are regionalised
  • They appear to be caused by variations of water column corrected bottom reflectance signature
    • more than by system noise
  • We can assume that the PAN solution is more accurate than the GREEN solution

Profile Black

Z4=PAN solution
Z3=GREEN solution
along Profile Black

Local variations of water optical properties
probably cause some of the above scatter
see location of profile_black

AOI
Retrieved depth



WZ over the AOI
Grey: PAN solution
Green_ Green solution

 

AOI
BGR color composite, linear enhancement
of water column corrected OLI reflectances
for Coastal, Blue and Green bands

OLI reflectances are calibrated (0-1)

 



Créer un site
Créer un site