Confidence Measure of the Shallow-Water Bathymetry Map Obtained through the Fusion of Lidar and Multiband Image Data

Abstract

With the advancement of Lidar technology, bottom depth ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mi>H</mml:mi> </mml:math> ) of optically shallow waters (OSW) can be measured accurately with an airborne or space-borne Lidar system ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mrow> <mml:mi>H</mml:mi> </mml:mrow> <mml:mrow> <mml:mtext>Lidar</mml:mtext> </mml:mrow> </mml:msub> </mml:math> hereafter), but this data product consists of a line format, rather than the desired charts or maps, particularly when the Lidar system is on a satellite. Meanwhile, radiometric measurements from multiband imagers can also be used to infer <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mi>H</mml:mi> </mml:math> ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mrow> <mml:mi>H</mml:mi> </mml:mrow> <mml:mrow> <mml:mtext>imager</mml:mtext> </mml:mrow> </mml:msub> </mml:math> hereafter) of OSW with variable accuracy, though a map of bottom depth can be obtained. It is logical and advantageous to use the two data sources from collocated measurements to generate a more accurate bathymetry map of OSW, where usually image-specific empirical algorithms are developed and applied. Here, after an overview of both the empirical and semianalytical algorithms for the estimation of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mi>H</mml:mi> </mml:math> from multiband imagers, we emphasize that the uncertainty of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mrow> <mml:mi>H</mml:mi> </mml:mrow> <mml:mrow> <mml:mtext>imager</mml:mtext> </mml:mrow> </mml:msub> </mml:math> varies spatially, although it is straightforward to draw regressions between <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mrow> <mml:mi>H</mml:mi> </mml:mrow> <mml:mrow> <mml:mtext>Lidar</mml:mtext> </mml:mrow> </mml:msub> </mml:math> and radiometric data for the generation of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mrow> <mml:mi>H</mml:mi> </mml:mrow> <mml:mrow> <mml:mtext>imager</mml:mtext> </mml:mrow> </mml:msub> </mml:math> . Further, we present a prototype system to map the confidence of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mrow> <mml:mi>H</mml:mi> </mml:mrow> <mml:mrow> <mml:mtext>imager</mml:mtext> </mml:mrow> </mml:msub> </mml:math> pixel-wise, which has been lacking until today in the practices of passive remote sensing of bathymetry. We advocate the generation of a confidence measure in parallel with <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mrow> <mml:mi>H</mml:mi> </mml:mrow> <mml:mrow> <mml:mtext>imager</mml:mtext> </mml:mrow> </mml:msub> </mml:math> , which is important and urgent for broad user communities.

Topics & Concepts