The problem of estimating the log return power in a lidar system in the presence of multiplicative noise ( speckle and turbulence noise) is discussed. The partitioning approach is applied to adaptive filter ,and the formulation of adaptive filters is given in the case where model parameter uncertainty exists.By simulations ,the adaptive filter algorithm is shown significantly superior to the filter algorithm only considering speckle noise as multiplicative noise in the presence of speckle and turbulence noise.