Assessing the potential of MODIS/Terra version 5 images to improve near shore lake bathymetric surveys

• Bathymetric surveys miss to accurately represent shore storage characteristics due to operational limitations.
• Storage characteristic equations are often extrapolated to model the shore area.
• MODIS images are used to reconstruct the shore bathymetry.
• The accuracy of the bathymetric model generated using MODIS versus existing bathymetry are compared through the estimation of lake levels for selected dates.
• RMSE of 0.14 m in lake level estimate was achieved as compared to 0.82 m using the existing bathymetric model.

Estimation of storage volume of inland water bodies is made using bathymetric surveys. Such surveys are often terminated far inward from shores due to operational limitations. The gap in depth data for the shore area is often overcome by extrapolating the established volume-area relationship beyond the survey boundary. This introduces large errors in volume estimation. In this study, the potential of the Moderate Resolution Imaging Spectroradiometer MODIS version-5 images to improve an existing bathymetric model is assessed over Lake Tana, Ethiopia. MODIS images are used to generate a new bathymetric model of the unsurveyed shore. The new model is validated through its ability to reproduce measured lake levels. The result is compared with lake level estimates made from an existing bathymetric survey. Results showed that MODIS improved bathymetric model reproduced lake water level with a RMSE of 0.14 m as compared to 0.82 m using the existing bathymetric model. The Mean Absolute Error (MAE) has also improved from 0.78 m to 0.13 m. These results suggest that despite their unavailability on cloudy days, MODIS images provide significant advantages for improving bathymetric models.