Lake Nubia sediment capacity estimation based on satellite remotely sensed detected bathymetry

Abstract

This research is concerned with detecting the depths of the Aswan High Dam Lake (AHDL) (particularly Nubia Lake) using Landsat-8 multi-spectral bands with resolution 30m. The method of depths estimating from the satellite images is based on establishing Multiple Linear Regression (MLR) model using logarithms of reflectance bands. Quantifying the error between in-situ measurements and satellite estimated lake depths values. The results indicated that the correlation coefficient (R= 0.5) and the Root Mean Square Error (RMSE= 0.6m) for testing data set. The computed depths from the satellite remote sensing data were used to estimate the sediment and erosion capacities in Lake Nubia (LN) since the beginning of the storage process in the lake at the year 1964 till the year 2012. The results are comparable to those of the traditional method (cross section method) and the method of building 3D profile from the field measurements with additional merits of the present approach that it is more flexible and cost-effective tool for bathymetry mapping. Moreover, the results indicated that the computed amount of sediment using the estimated depths from the RS data (the present approach) overestimated the results by 4.16% compared to those computed using the cross-section method adopted by AHDA. On the other hand, the results indicate that the present approach overestimates sediment amount by about 3.93% compared to the results derived from the generated 3D profiles of the Lake based on the field measurements.