Bathymetry and water column correction
LANDSAT 8 OLIP 
at Heron Island, GBR, Australia
7621*7731 , 30 m pixel size, UTM zone 56

Work done december 2013

see also CASI at Heron Island
see also ETM at Heron Island
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
 


 
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

Seven bands are used: Purple, Blue, Green, PAN, Red, NIR and SWIR1

Panchromatic band, sum-aggregated to 30 m,
exhibits good radiometric quality

TCC
Landsat OLI exhibits good radiometric quality

SWIR1 band
is badly banded
Using SWIR1 for deglinting
transfers this banding to other bands!

NIR band
has much less system noise than SWIR1
Better use NIR than SWIR1 for deglinting
Images of the Great Barrier Reef often exhibit strong plumes of turbid waters that come from rivers along the coast.
This the case here==>

Deglinted TCC zoom: turbid waters





 


Deglinting

Seven bands are used: Purple, Blue, Green, PAN, Red, NIR and SWIR1

Glint regression based on band 7
glint regressions are sampled
in the surrounding of Heron Island

Glint regression based on band 6
TCC
TCC deglinted

TCC raw: Location of Profile GREEN



Glint regression must be sampled in that area
in order to best represent the surrounding of these islands
 

TCC deglinted: Location of Profile GREEN



Glint regression must be sampled in that area
in order to best represent the surrounding of these islands

Profile GREEN
<==Waters are not really homogeneous,
as seen for the Purple and Blue bands.
 







Calibration
Seven bands are used: Purple, Blue, Green, PAN, Red, NIR and SWIR1
Calibration diagram
for bands 2, 3, 5 and 6
water type OII+0.0 of Jerlov
in the surrounding of the islands
Calibration diagram
for bands 2, PAN, 5 and 6
Mask mSE
This calibration only accounts
for the Red and Orange areas,
as green areas are eliminated from the calibration data.
  • Note in the bidimensional diagrams that the shallow substrates in the coastal strip RED-area exhibit a bottom brightness which is just about 5 times less bright than the brightest coral sand substrates in the island ORANGE-area.
  • Also note that the slope K2/K3=0.52 is also correct in the coastal strip (RED): waters there are approximately the same as in the island area: at least in the southern part of the image, this calibration applies from the coastal strip to the open waters.

 
 

 
 


 



Profiles
Seven bands are used: Purple, Blue, Green, PAN, Red, NIR and SWIR1
Modeling uses the "RED solution" and  the "PAN solution"

Profile YELLOW
  • up:      section B crosses Heron Island and the shoal
  • down: section A crosses the two islands

Z4SM retrieved depth
with location of Profile YELLOW




image WZ
  • RED: the "RED solution" is applied
  • GREY: the "PAN solution" is applied

Profile PURPLE
 

Z4SM retrieved depth
with location of Profile PURPLE


This area does not seem
to be affected by water turbidity


Profile RED using OLIP



Water column corrected  OLIP TCC
with location of Profile_RED

 

heron_20020702_profile_red_CASI_PlotZB
Profile RED using CASI
the red profile is the "seatruth" DTM from U. of Queensland


Water column corrected CASI TCC
with location of Profile_RED




Modeling

Seven bands are used: Purple, Blue, Green, PAN, Red, NIR and SWIR1
Modeling uses the "PAN solution"

Z4SM retrieved depth

Z4SM retrieved depth
with overlay of ZDTM: CASI model

Retrieved depth
The northern part of the image exhibits turbid waters

image B
 

 
 


The command line
nice -20  4SM   -Process  -Origin/USGS                   
-DB/heron_20130730/40_3S_7S_0s/7_44/7621_7731/207.915_7558.585_0_0/1_1           
-Mis/Australia/HeronIsland/Landsat/OLIP/TIF/UTM_-56_008/0.030_-0.030/30_SEP_2013   
-LS/030000.0/30000.0/30000.0/30000.0/30000.0/30000.0/30000.0               
-WLM/450.0/510.0/590.0/680.0/670.0/880.0/1650.           
-WLm/430.0/450.0/530.0/500.0/640.0/850.0/1570.           
@WL/0470.0/480.0/560.0/590.0/610.0/865.0/1610.           
-cWL/0.500/0.500/0.775/0.500/0.500/0.500/0.500           
-import/data*LC80910762013211LGN00               
-SCL/01496/01472/01565/01569/01682/02196/00993_new       
-LsM/169.0/198.1/226.4/240.9/233.9/252.4/217.5           
-Lsw/062.59/049.74/024.46/020.08/011.98/03.60/01.80       
-dLsw000.0/000.0/000.7/000.0/000.0/000.0/000.0_new       
-Lw/0020.0/020.0/005.0/002.0/000.0/000.0/000.0_new       
-M/@000001/@0002/00003/00004/00005/@0006/00007           
-Lm/0001.0/001.0/255.7/000.5/020.0/255.0/255.0_new       
-KK2_3_0.5200/Knir16.5/Bm440_BG500_GR600_RN700/mask_3       
-Z/0.00d/cSL1.50/N_3/NDR_43_1000_zDTM/mask_3           
-B/TclNe/cNorm5.00/cLM1 0.2                    
-deglint/vRbad/q/F1/L6/GlintM10.0               
-extract/v/rawBDH/fullBDH/NIRband7/NIRmax5.0/mBPL3 /mask_4   
-calibrate/v/BdSMpzg/plot_BPL/BC_6_5_4_2/BDh_12_13_14       
-Model/v/runWV2/makeClouds/mask_3               
-Smooth/1/D/Smart+     -TPsv/2/4/Text/Bmin150/panCode -LL10    
@-E1/7621/1/7731  @-E1/7621/4300/7731               
@-DN300/100/3/comment @-DDshp @-DNshp @-Eshp           
@ProfileAB/LVW/profile_red/chAB_5 _2_3_4 _5_6_7       
-ProfileAB/ZrvWB/profile_red_hyphoon/dTM/chAB_1_2_3 _4 _5          
@ProfileAB/RvWb/profile_red/dTM/5_6/0_201/0_50_50/*       
@Recode/2_InChannel_8_into_3_OutChannel_8 /mask_3