| dc.description.abstract | Bathymetry is to water what topography is to land. Just as it is hard to set up anything significant on land without knowledge of the topography, the same applies to water without on the bathymetry. As such, it is very vital to have knowledge on the depth of water at different points on the lake to support activities like fishing, water resources management, navigation and routing and recreation to mention but a few. Water depth also affects different biochemical processes occurring on and in water bodies like growth of microorganisms. Despite the importance of upto-date and accurate reliable bathymetry data, Lake Victoria, which is the largest lake in East Africa and a source of livelihood to many people, falls short of this information. This is mostly due to the complex and costly nature of the conventional methods of acquiring bathymetry data like echo sounding and the LiDAR and RADAR based techniques.
This project explored the use of GIS and remote sensing of optical imagery as a cheap and fast alternative to the costly and time-consuming methods of bathymetry data collection. The project evaluated the use of the semi empirical models of Lyzenga, Stumpf and the IOPLM; and Planetscope satellite imagery for predicting water depth of shallow areas of Lake Victoria. Three
areas were selected as sample study areas; the River Kagera Delta, the Entebbe Bay and the Nile Headwaters. Of these, data from two were used to train models and datasets from all the three areas were employed for validation of the models.
According to this study, none of the models performed consistently in all the study areas; and none gave accuracies that were considered sufficient for prediction of water depth. In certain areas, the Lyzenga model gave the highest levels of accuracy compared to the other models of Stumpf and the IOPLM. This report goes to present the research done in coming up with this
work, the methodology followed in project execution and the findings of the study. | en_US |
