Evaluating different satellite-derived bathymetry models: A case study of Tagwai Dam.

Abstract

Bathymetry entire process involves in a typical survey which is somewhat cumbersome as it involves the general assemble of echo sounder, GPS, Boat (vessel and crew etc. data manipulation and processing is also another rigour to overcome owing to the fact that water surface is dynamic as the water level continuously fluctuates in elevation during the survey. The aim of this research is to determine the underwater topography of Tagwai dam using two empirical methods of Satellite Derived Bathymetry (SDB). The data used where Insitu data (measurements) obtained from EchoMap 50s and accessories and Satellite based data (Landsat 8 OLI 2022). Atmospheric, radiometric and sunglint corrections were performed. OLI and TIRS at Sensor Spectral Radiance was also carried out. Land Use Land Cover classification was performed using the maximum likelihood classification algorithm. The Optimal Band Image Analysis (OBIA) and Stumpf models were employed in this research. The Root Mean Square Error analysis was also carried out to validate the results of the two methods. The results of this research show. both models had a close depth estimation, but the OBIA model had a better depth estimation with RMSE of 0.801124974 against 0.802449293. In estimating maximum and minimum depths from both models, where INSITU max is 8.8m, Stumpf had 1.290463 and OBIA had 1.318488m. where INSITU min was 0.9, both models estimated the min value of 1.314712m. On the general note, Stumpf model had maximum and minimum depth estimation of 1.35107 and 1.203045m respectively. On the other had OBIA had maximum and minimum depths of 1.3561 and 1.11067m. OBIA is recommended for this type of exercise since it give ample opportunity to vary the bands combinations. Although the depths obtained did not exactly correspond with the INSITU data obtained from Echo Sounder, but this research has clearly demonstrated the applicability of Satellite remote sensing in the Hydrographic Surveying.