Lakes are integrator of environmental changes occurring at a regional to global scale and present a high variety of behaviors on a variety of time scale. Their crucial importance as water stocks and retaining given the significant environmental changes occurring worldwide at many anthropocentric levels has increased the necessity of monitoring all its morphodynamic characteristics i.e. water level, surface area and volume. In this paper a 57 years water level data obtained by combining Bahir Dar gauge station and satellite altimetry were analyzed for detection of trend using the non-parametric Mann-Kendall test in combination with the trend-free pre-whitening approach for correcting time series data sets for serial correlation. Sen’s slope method, a non-parametric alternative for estimating a slope for a univariate time series was used to determine the magnitude of trends.