A Comparative Study of Remote Sensing Techniques for Bathymetric Data Acquisition in Lokoja, Kogi State Nigeria
Abstract
This study successfully performed a comparative study of remote sensing techniques for bathymetric data acquisition in Lokoja, Kogi state Nigeria. The following objectives were achieved by acquiring the multi bands of Landsat 8, Sentinel -2 and MODIS imagery used for bathymetric processing, acquiring remotely sensed images and extracting bathymetric information using different remote sensing techniques and comparatively analyzing the accuracies obtained between the remote sensing techniques and ground-based bathymetry. The methodology involved the acquisition of multi-sensor images, acquisition of sounding data, image preprocessing, water depth extraction models and validation using sounding data and correlation coefficient. The results revealed that the bathymetric mapping using Stumpf’s ratio on Landsat 8 imagery gave depth values ranging from 0 – 0.18m classified as very shallow, 0.18m – 0.43m classified as shallow, 0.43m – 0.55m classified as deep and 0.55m – 1m classified as very deep. Results from using Stumpf’s ratio on Sentinel-2 imagery gave depth values ranging from 0 – 0.23m classified as very shallow, 0.23m – 0.45m classified as shallow, 0.45m – 0.57m classified as deep and 0.57m – 1m classified as very deep. Results from using Stumpf’s ratio on MODIS gave depth values ranged from 0 – 0.12m classified as very shallow, 0.12m – 0.29m classified as shallow, 0.29m – 0.46m classified as deep and 0.46m – 58m classified as very deep. Also, the results from bathymetric mapping using Dierssen’s ratio on Landsat 8 imagery gave depth values ranging from 0 – 0.21m classified as very shallow, 0.21m – 0.38m classified as shallow, 0.38m – 0.58m classified as deep and 0.58m – 1m classified as very deep. Results from using Dierssen’s ratio on Sentinel-2 imagery gave depth values ranging from 0 – 0.25m classified as very shallow, 0.25m – 0.48m classified as shallow, 0.48m – 0.57m classified as deep and 0.57m – 1m classified as very deep. Results from using Dierssen’s ratio on MODIS gave depth values ranging from 0 – 0.13m classified as very shallow, 0.13m – 0.30m classified as shallow, 0.30m – 0.49m classified as deep and 0.49m – 1m classified as very deep. The very shallow zone covered an area of 562 m2, shallow covered an area of 679m2, deep zone covered an area of 972m2, while very deep covered an area of 175m2. The validation results using correlation coefficient and sounding data, revealed that Stumpf’s Band ratio performed better that Dierssen’s Band ratio. Therefore, Stumpf’s band ratio method is recommended as a viable method for satellite based bathymetric mapping as it performed better than Diersson’s band ratio when compared with the validation data.
