Abstract: Two core power distribution reconstruction methods based on spatial statistical theory were proposed. Ordinary Kriging method is an optimal interpolation method based on spatial autocovariance and Kalman filter method is one of the data assimilation methods which could combine theoretical computation and measurements effectively. Measurement data from Unit 3 reactor of Qinshan No.2 Nuclear Power Plant and Unit 1 reactor of Daya Bay Nuclear Power Plant were used to verify the accuracy of these methods, and comparisons were made between these methods and the coupling coefficients method. The results show that the reconstruction errors of two methods satisfy the requirement of engineering and the reconstruction accuracies of two methods are better than the coupling coefficients method.