FIG Peer Review Journal

Back

Satellite-Derived Bathymetry Modelling in Shallow Water: A Case Study of Lighthouse Creek, Lagos (8059)

Dupe Olayinka and Chukwuma Okolie (Nigeria)
Dr Dupe Olayinka
Lecturer
Department of Surveying and Geoinformatics
Faculty of Engineering
University of Lagos, Nigeria
Dept of Surveying and Geoinformatics
University of Lagos, Nigeria
Yaba
23401
Nigeria
 
Corresponding author Dr Dupe Olayinka (email: dsaka[at]unilag.edu.ng, tel.: +2348111112569)
 

[ abstract ] [ paper ] [ handouts ]

Published on the web 2016-03-01
Received 2015-11-10 / Accepted 2016-02-01
This paper is one of selection of papers published for the FIG Working Week 2016 in Christchurch, New Zealand and has undergone the FIG Peer Review Process.

FIG Working Week 2016
ISBN 978-87-92853-52-3 ISSN 2307-4086
http://www.fig.net/resources/proceedings/fig_proceedings/fig2016/index.htm

Abstract

There is great demand for accurate and high-resolution seafloor topography data. However, present availability of such data remains spatially incomplete and limited. Bathymetry in shallow water areas are the most costly and challenging for ship-based bathymetric surveys and this zone is where remote sensing typically outperforms traditional methods. Specific uses of satellite-derived bathymetry include characterisation of the underwater environment and monitoring seafloor changes that have occurred since the last hydrographic survey was done. The obvious advantages over conventional echo sounding methods include the wide data availability, synoptic surface coverage, and high spatial resolution. Bathymetric models are very important in studying the underwater bottom morphology. The limitations of earlier models inspired Stumpf et al (2003) to develop an alternative model for transforming the reflectance in an attempt to determine the depth. A bathymetric survey by acoustic method was carried out at Lighthouse Creek, Lagos to extract reference data for evaluating the remotely sensed bathymetry from a time series of Landsat imagery (2002, 2006 and 2015). To derive depths from the imageries, they were first preprocessed and atmospherically corrected. Next, the optic depth limit for inferring bathymetry (the extinction depth) was determined (~5m). The final bathymetry was gotten after corrections gotten from tidal prediction values connected by a shape preserving interpolant were applied. The standard errors in the estimated depths were gotten as 0.29m for 2002, 0.31m for 2006 and 0.27m for 2015. The absolute differences between the actual depths and estimated depths at these points ranged from 0.1 – 0.49m. In a final step, we evaluated and modelled the image spectral contribution to the estimated bathymetry by the application of a least squares solution to derive a linear mathematical model.This was done to evaluate the contribution of the significant Blue, Green and NIR bands in the depth estimation. Accuracy tests of Stumpf’s model and validation of the subsequent linear model to relate bathymetry with the spectral bands of Landsat showed very good performance.
 
Keywords: Bathymetry; Satellite-Derived Bathymetry; Shallow Water; Echo Sounding; Least Squares

Back