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=1, Blue=2, Green=3, PAN=4, Red=5, NIR=6 and SWIR1=7 |
![]() |
Dunes![]() | Lee Stocking Island![]() |
Discoloured waters in the Bahamas I hypothesize that, under soft modern sedimentation, there is a giant karst system and a fresh water lens which is subject to tidal pumping, through a great number of dolines, sinkholes and fractures. Black waters seep up here and there from below and might even entertain some kind of transient blooms. This is the cause of bad artifacts in 4SM results. |
![]() | ![]() | July 2013 SAM mapping Robustness: RMSE=1.47 m | ![]() | ![]() |
![]() | ![]() | September 2013 A bad image remains a bad image Robustness: RMSE=2.18 m | ![]() | ![]() |
![]() | ![]() | November 2013 Robustness: RMSE=0.85 m Winter is the time of choice | ![]() | ![]() |
reference image | reference image | January 2014 Reference image | ![]() | ![]() |
![]() | ![]() | October 2014 Bad image Robustness: RMSE=1.29 m | ![]() | ![]() |
![]() | ![]() | November 2014 Black hole discoloured water Robustness: RMSE=0.87 m | ![]() | ![]() |
![]() | ![]() | December 2014 Almost no discoloured waters Vegetation is dormant Robustness: RMSE=0.96 m Very low tide | ![]() | ![]() |
![]() | ![]() | February 2015 Robustness: RMSE=1.53 m | ![]() | ![]() |
![]() | ![]() | March 2015 Robustness: RMSE=0.93 m | ![]() | ![]() |
![]() | ![]() | August 2015 Extremely dense and variable atmosphere Robustness: RMSE=2.14 m BAD, DISCOLOURED, LUT | ![]() | ![]() |
![]() | ![]() | October 2015 Robustness: RMSE=1.07 m | ![]() | ![]() |
![]() | ![]() | February 2016 Winter time of choice Thin homogeneous atmosphere Homogeneous waters Very low tide RMSE=1.02 m Best calibration | ![]() | ![]() |
![]() | ![]() | April 2016 Poor deglinting by NIR Robustness RMSE=1.12 m | ![]() | ![]() |
![]() | ![]() | May 2016 Robustness RMSE=1.27 m | ![]() | ![]() |
![]() | Freeport December 2014 Freeport South December 2014 Whitings cause artifacts Ooid sands look pinkish | ![]() | ![]() | |
Achieved the optimization of the calibration | SAM results through the seasonal cycle appear to make sense | Now ready for computing a Combined Depth raster for LSI scene | Collected good reflectance signatures for coral sand enmember | Streamlined the processing of Landsat 8 (or WV2) data |